identifying bioactive phytochemicals in spent coffee
TRANSCRIPT
IDENTIFYING BIOACTIVE PHYTOCHEMICALS IN SPENT COFFEE GROUNDS
FOR COSMETIC APPLICATION THROUGH GLOBAL METABOLITE ANALYSIS
_______________________________________
A Thesis
Presented to
the Faculty of the Graduate School
University of Missouri-Columbia
_______________________________________________________
In Partial Fulfillment
of the Requirements for the Degree
Master of Science
_____________________________________________________
by
JIHYUN PARK
Dr. Chung-Ho Lin,
Thesis Supervisor
DECEMBER 2017
The undersigned, appointed by the dean of the Graduate School,
have examined the thesis entitled:
IDENTIFYING BIOACTIVE PHYTOCHEMICALS IN SPENT COFFEE GROUNDS
FOR COSMETIC APPLICATION THROUGH GLOBAL METABOLITE
ANALYSIS
presented by
JIHYUN PARK
a candidate for the degree of Master of Science and hereby
certify that in their opinion it is worthy of acceptance
Professor Chung-Ho Lin
Professor Shibu Jose
Professor Gary Stacey
Professor Minviluz Stacey
ii
ACKNOWLEDGEMENTS
I would like to thank my advisor, Professor Chung-Ho Lin about his guidance with
huge effort, and the Center for Agroforestry, without their considerable assistant, I could
not have done this research. Conducting this project provided me a great opportunity to
meet and work with many people who have a comprehensive mind and passion. It is very
pleasure for me to offer thanks to them.
In addition, my thesis committee members, Dr. Shibu Jose, Gary Stacey, and Bing
Stacey, were the most responsible professors in their field and they leading me to focus
on this research. They realized the novelty of my thesis topics paying attention to this
project and encouraging me to accomplish the experiment on time.
Also I would like to say thanks to Nahom Taddese Ghile who gave me the first
help beginning this project and the laboratory members Van Ho, Danh Vu, Phuc Vo, who
provided assistance in several steps of my experiment with familiarity.
Special thanks to many capable professionals Zhentian Lei, Lloyd Summer,
Michael Greenlief, Brian P. Mooney in metabolomics center and department of
chemistry.
Lastly I would like to thank to my family and friends who continuously supported
me with their love.
iii
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ............................................................................................................. ii
LIST OF TABLES ........................................................................................................................... v
LIST OF FIGURES ........................................................................................................................ vi
Chapter 1. Identifying the bioactive compounds in SCG through global metabolomics
analysis (XCMS) ............................................................................................................................ 1
1. Introduction .............................................................................................................................. 2
2. Objectives .............................................................................................................................. 11
3. Materials and Methods ........................................................................................................... 11
3.1. Sample preparation ..............................................................................................12
3.2. LCMS Analysis ...................................................................................................12
3.3 Global metabolite profiling using XCMS and METLIN Library ........................15
4. Results .................................................................................................................................... 18
4.1. Multi-group Comparison between three SCG cultivars using low resolution
data .............................................................................................................................19
4.2. Identified bioactive compounds through global metabolomics analysis using
high resolution data ....................................................................................................28
5. Discussion .............................................................................................................................. 42
5.1. Global metabolomics approach (XCMS Online) vs traditional analytical
techniques ...................................................................................................................42
5.2. Phenolic acids and caffeine in SCG ....................................................................45
5.3. Glycosylated form of flavonoids and their activity .............................................47
5.4. Flavonoids (flavonols, flavones, isoflavones, flavanones, anthocyanidins, and
flavanols) in SCG for cosmetic application ................................................................48
6. Conclusion ............................................................................................................................. 54
Chapter 2. Isolation of the antioxidant and antimicrobial compounds with bioassay-guided
fractionation ................................................................................................................................. 56
1. Introduction ............................................................................................................................ 57
2. Objectives .............................................................................................................................. 67
3. Materials and Methods ........................................................................................................... 67
3.1. Explore new application (antioxidant and antimicrobial activities) through
bioassay-guided purification ......................................................................................70
3.2. Determination of anti-oxidant activities ..............................................................73
iv
3.3. Anti-bacterial analysis .........................................................................................73
3.4. High resolution (HR) metabolomics analysis......................................................74
3.5. Statistical analysis ...............................................................................................76
4. Results .................................................................................................................................... 76
4.1 Antioxidant study .................................................................................................76
4.2 Antibacterial activity against Methicillin-resistant Staphylococcus aureus
(MRSA) ......................................................................................................................89
5. Discussion .............................................................................................................................. 93
5.1. Antioxidant activity of SCG for cosmetic application ........................................93
5.2. Antimicrobial activity of SCG for cosmetic application .....................................96
5.3. Importance of anti-inflammatory and anti-carcinogenesis activity in cosmetics 99
5.4. High resolution metabolomics analysis followed by bioassay guided
purification ...............................................................................................................100
6. Conclusion ........................................................................................................................... 101
References ................................................................................................................................... 103
Appendix A. ................................................................................................................................ 123
Appendix B. ................................................................................................................................ 147
1. Introduction .......................................................................................................................... 147
2. Objectives ............................................................................................................................ 150
3. Materials and Methods ......................................................................................................... 150
4. Results and discussion ......................................................................................................... 151
v
LIST OF TABLES
Table Page
1.1. Major compounds exist in natural resources that have biological function related to
skin care. The list of bioactive compounds have been reported, showing antioxidant,
Anti-fungal, skin-whitening, anti-inflammatory, anti-aging, and antimicrobial properties
or other skin-care applications…………………………………………………….............8
2.1. Antioxidant activity of SCG methanol extracts of three cultivars including Ethiopian
Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona ....................................................79
2.2. Identified compounds in purified fraction 14 through XCMS Online HR
metabolomics analysis. Each row indicates feature, compound name, m/z, retention time,
maximum intensity and its known bioactivity. Using XCMS Online platform, chemical
profiling was compiled. .....................................................................................................85
2.3. FeCl2 concentration (µM) of bioactive compounds which were reported in previous
study (unpublished data by Nahom Taddese Ghile and Lin) .............................................88
Appendix A. Phenolic acids, flavonoids, and other bioactive compounds including coffee
alkaloids, phenylpropanoids, stilbenoids, and anthraquinones extracted from three
different coffee wastes (Ethiopian Yirgacheffe, Costa Rican Tarrazu, and Hawaiian
Kona). Maximum intensity of peaks detected by XCMS online comparing the relative
intensity between the three cultivars of SCG are shown .................................................123
vi
LIST OF FIGURES
Figure Page
1.1. Bioactive compounds related to UV protection efficacy..............................................7
1.2. Flow chart of the SCG metabolomics approach project using Waters Xevo UPLC-
MS for non-VOC analysis and XCMS global metabolomics platform. ............................14
1.3. Flow chart of metabolomics data processing………………………………………..15
1.4a. Box-plot generated from interactive cloud plot ......................................................21
1.4b. Composite data base after mouse click on interactive cloud plot. Seven compounds
were matched with METLIN database based on ion chromatogram of each circle.
Compounds in composite DB linked to METLIN library it will show compound formula,
structure, and spectrum ......................................................................................................22
1.4c. Multi-group cloud plot. Metabolite feature which varies significantly (p < 0.01)
from each groups are presented on cloud plot depending on their retention time (x-axis)
and m/z (y-axis). Each circle indicates each compound feature. The color intensity of
circle represents statistical significance (p-value). The size of the circle indicates feature
intensity. If you move the mouse on each circle, p-value, q-value, m/z, RT, Max intensity
will pop-up. With mouse click each circle, Box-plot, EIC, Post-HOC, and Composite
database (METLIN) are displayed under the cloud plot. Compounds in composite DB are
linked to METLIN library showing compound formula, structure, and spectrum. ...........23
1.5a, Interactive principal component analysis. A Scores plot shows the correlation
between submitted samples. Three clusters represent each SCG cultivars which includes
Hawaiian Kona (ESI(-)_K1), Costa Rican Tarrazu (ESI(-)_T1, and Ethiopian Yirgacheffe
(ESI(-)_Y1) ........................................................................................................................25
1.5b. Principal Component Analysis (PCA) results: SCG samples from three different
coffee cultivars (Ethiopian Yirgacheffe, Costa Rican Tarrazu, Hawaiian Kona) clustered
by each groups. Individuals are projected in a reduced dimension space defined by
component 1 (x-axis) and component 2 (y-axis). Each color corresponds to a specific
cluster. ................................................................................................................................26
1.5c. Non-metric multi-dimensional Scaling of different coffee cultivars (Ethiopian
Yirgacheffe, Costa Rica Tarrazu, and Hawaiian Kona) shows that the chemical profiling
is significantly different between three cultivars………………………………………...27
1.6. Chemical structures of phenolic acid. .........................................................................29
1.7a. Extracted ion chromatogram of Ethiopian Yirgacheffe (E) and Hawaiian Kona (H).
The peak represents relative intensity of quercetin pentaacetate in SCG from Yirgacheffe
(E), and Kona (H) ..............................................................................................................39
vii
1.7b. Extracted ion chromatogram of Yrigacheffe (E) and Costa Rican Tarrazu (C). The
peak represents relative intensity of candidate compounds such as cyanidin-3-(6'-
malonylglucoside) or apigenin 7-O-(6-O-malonylglucoside) at RT 34.86 .......................40
1.8. Cloud Plot results: A Cloud Plot displays features whose intensities are altered
between sample groups. Up-regulated features are represented as circles on the top of the
plot and down-regulated features are represented as circles on the bottom of the plot,
where the size corresponds to the (log) fold change of the feature. Circles with black
outlines indicate hits in the METLIN database. The lighter or darker the color of the
circle relates directly to the significance of the p-value. 46 features were generated by
pairwise comparison between Ethiopian Yirgacheffe and Costa Rican Tarrazu showing
the relationship between retention time and m/z. Green color represents upregulated
compounds which means abundant in Tarrazu and red color circle represents down
regulated compounds which means the compounds exist more in Yirgacheffe. With a
mouse click, EIC, Box and Whisker, and Spectrum appears under the cloud plot ...........41
2.1. Common bioactive compounds in SCG ......................................................................66
2.2. Flow chart of the SCG project using bioassay guided purification………………....69
2.3. Flow chart of the bioassay-guided purification for identifying antioxidant activity
and antimicrobial assay. Same approach was applied for antimicrobial test. ....................72
2.4. Standard curve of FRAP assay. Each dot in X axis represents around 0, 31.25, 62.5,
125, 250, 500, 1000 FeCl2 concentration of standards (STD; n = 7)…………………….78
2.5. Reducing power of methanol extract (10mg/mL) from three SCG cultivars including
Ethiopian Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona were compared.
Concentration of FeCl2 was calculated from FRAP value (absorbance in 560nm) using y
= 0.002x - 0.0054. Hawaiian Kona had highest antioxidant compared to other cultivars.
Means in the same column followed by different letters (A-B) represent statistical
significance (P <0.05). Sextuplicates were analyzed using one-way analysis of variance
(ANOVA) followed by pairwise comparisons of means using Tukey’s range test was
performed to assess the difference in three cultivars (significance level P < 0.05). R: The
R Project for Statistical Computing 3.3.1 was used for statistical computing. ..................80
2.6. Antioxidant activities of the fractions from flash chromatography (1:10 dilution). The
fraction number 4 showed highest antioxidant activity as 475.16 µM FeCl2 condition
among 43 fractions so F4 was injected into HPLC for further purification. .....................81
2.7. 26 fractions was obtained from HPLC and tested through FRAP assay. Fraction 14
(vial 12) was revealed as bioactive fractions with high antioxidant activity .....................82
2.8. Total ion chromatogram (TIC) of fraction 14 purified by HPLC. The TIC was
generated by XCMS Online positive ion mode. The highest peak intensity 618,624 was
shown at RT 11.83 with m/z 679.9 and cafestol was the best possibility among other
compound candidates .........................................................................................................86
viii
2.9. SCG samples (Number from 47 to 55) had strong anti-microbial activity which is
Ethiopian Yirgacheffe, Costa Rican Tarrazu, Hawaiian Kona. The size of inhibition zone
was measured three times for all samples. Among three cultivars, Kona showed the
highest anti-microbial activity having 8.85 mm inhibition zone. ......................................90
2.10. Antimicrobial activities of the E (Ethiopian Yirgacheffe), C (Costa Rican Tarrazu),
H (Hawaiian Kona). Using zone of inhibition assay for assessing anti-microbial activity
against S. aureus, we found the following order of efficiency among coffee Arabica
cultivars: Kona (8.85) > Yirgacheffe (8.19) >> Tarrazu (5.76). DMSO did not show any
inhibitory activity. Inhibition zone: 8 mm or greater — good antibacterial activity; 6–7
mm — moderate antibacterial activity; 4–5 mm — weak antibacterial activity; 2–3 —
very weak antibacterial activity. Mean values followed by different letters (A and B)
represent statistical different (P <0.05). Zone of inhibition is expressed as the diameter
(mm). Values represent means ± standard deviations of inhibition zones from
experiments, each using triplicate plates and was measure three times. Means with the
same superscripts (A, and B) are not significantly different (P < 0.05) from each other.
One-way analysis of variance (ANOVA) followed by pairwise comparisons of means
was performed to assess the difference in three cultivars (significance level P < 0.05). R:
The R Project for Statistical Computing 3.3.1 was used………………………………..91
2.11. Antimicrobial activity of 43 fractions using flash chromatography from Ethiopian
Yirgacheffe. Fraction number 7 and 12 showed highest zone of inhibition value. The rest
of fraction number over 20 that have no activity doesn’t represent in the chart ...............92
Appendix B1. Mode of action of anti-pigmentation ........................................................148
Appendix B2. Melanin synthesis pathway.......................................................................149
Appendix B3. Calibration curve of the positive control (arbutin). Each dot in X axis
represents around 0, 1.1, 11, 110, 1100 µg/ml concentration of arbutin (STD; n = 5) ....152
1
Chapter 1. Identifying the bioactive compounds in SCG through global
metabolomics analysis (XCMS)
Abstract: Annually, more than 6 million tons of spent coffee grounds (SCG) are
generated worldwide. The present study explores the possible use of spent coffee grounds
as the raw materials for cosmetics industry. The main objective of this project are to
investigate the chemical profiles and identify the bioactive compounds for cosmetics
application through global metabolite analysis. The compounds extracted from SCG of
Ethiopia coffee (Yirgacheffe), Costa Rican coffee (Tarrazu) and Hawaiian coffee (Kona
Blend) were analyzed by ultra-high pressure liquid chromatography coupled with mass
spectrometry (UPLC-MS). The ion chromatograms were submitted to XCMS platform
operated by Center for Metabolomics at the Scripps Research Institute. The peak
detection, peak grouping, spectra extraction, and retention alignment were processed by
XCMS. The spectra were annotated and the compounds were identified and categorized
by integration with METLIN, the world's largest metabolite database. Multivariate and
univariate statistical analysis including PCA and cloud-plot were performed by XCMS to
compare the chemical profiles between the three coffee cultivars. These analyses
indicated that each cultivar showed a specific cluster. Over 300 compounds related to
anti-oxidant, anti-inflammatory, anti-tyrosinase and anti-tumor for skin care application
were identified by XCMS. Therefore, the presence of bioactive compounds in SCG
makes it a potential source of raw material for cosmetic application (e.g., anti-oxidant,
anti-inflammatory, skin-whiting, and anti-aging).
2
1. Introduction
In USA alone, more than 4000 tons of coffee are consumed every day. Annually, more
than 6 million tons of spent coffee grounds (SCG) are generated worldwide (Mussatto et
al., 2011b). Conventionally, the uses of the SCG has been only limited to home gardening
(Cruz et al., 2012), compost (Adi and Noor, 2009) and landfills. Recent studies have
suggested that SCG could be an excellent source of high value phenolic compounds
which have wide applications in personal care products (Mussatto et al., 2011a). For
examples, commercialized extracts which contain chlorogenic acids, major compounds in
SCG, with proanthocyanidins, quinic acids, and ferulic acid have shown positive results
for facial skin care purpose due to their antioxidant, anti-aging, anti-inflammatory
activities (Esquivel and Jiménez, 2012). Several compounds such as phenolic acids that
have shown strong anti-microbial effects were also found in the extracts (Table 1.1).
Table 1.1 shows variety of major compounds in extracts reported in the previous studies
that have biological function related to skin care. Among these, SCG contain large
amounts of phenolic acids such as caffeoylquinic acids, dicaffeoylquinic acids,
feruloylquinic acids, and p-coumaroylquinic acids (Zuorro and Lavecchia, 2012) showing
strong antioxidant, anti-fungal, skin-whitening, anti-aging, and antimicrobial properties
or other skin-care applications. Especially, antioxidant reaction, as a result of free radical
scavenging activity, was attributed to the presence of these phenolic compounds in SCG.
This is because the brown pigments such as phenolic polymers generated from roasting
process protect cells from oxidative damage (Ramalakshmi et al., 2009). Additionally,
several flavonoids in SCG are known to be responsible for the antioxidant activity
(Mussatto et al., 2011a). Figure 1.1 shows flavonoids such as cyanidin, resveratrol,
3
quercetin, apigenin, hesperetin with strong antioxidant effects against UV for reduced
skin damage. These flavonoids also possess anti-aging and anti-melanogenesis efficacy
as well (Arung et al., 2011; Kitagawa et al., 2011; Wang et al., 1999). In addition to
phenolic acids and flavonoids, caffeine, one of the major bioactive alkaloid in SCG
contributes significantly in reducing interfacial tension between water and oil phase,
important in defining the emollient characteristics of cosmetics.(Campos-Vega et al.,
2015). Likewise, galactomannans acts as excellent stiffeners and emulsion stabilizers in
cosmetics as a crude flour form (Silveira and Bresolin, 2011). This polysaccharide mostly
remains as insoluble material bound to the SCG matrix during coffee brewing (Campos-
Vega et al., 2015). Due to considerable amount of these bioactive compounds, it provides
an excellent opportunity to utilize the organic-based SCG as a potential valuable raw
material in cosmetics industry. Although several cosmetics applications of SCG have
been reported and identified, the bioactive compounds in SCG have never been
systematically studied or characterized. To date, most of the bioactive compounds were
isolated and identified through the conventional labor-intensive and time-consuming
bioassay-guided fractionation followed by targeted analysis and NMR structural
elucidation.
Recent advances of metabolomics science, mass spectrometry, computation capacity,
and compound libraries have allowed systematic identification and characterization of
bioactive phytochemicals. Over the last decades, untargeted global metabolomics
analysis has been utilized as a powerful tool for natural product research through
processing raw spectra information acquired from mass spectrometry such as liquid
chromatography-mass spectrometry (LC-MS) and gas chromatography-mass
4
spectrometry (GC-MS). Currently, several freely available or commercial metabolomics
software platforms such as Metab, Mzmatch, MZmine and XCMS have emerged.
However, each software platform has some technical limitations due to installation
requirement, or requirement of command line interface which demand advanced
computer skills, or lack of the data analytical tools in the package. For example, both
Metab and Mzmatch do not provide any statistics analytical tools, such as principal
component analysis (PCA) in the software package, and therefore users may need
additional statistical analysis. For the accurate mass LC/MS data, XCMS shows higher
detection percentages than MZmine (Coble and Fraga, 2014). The technical challenges of
several platforms and user unfriendly interfaces often require scientists or clinicians
collaborate closely with bioinformatics laboratory for data processing (Tautenhahn et al.,
2012). One way to solve these problems is the development of user-friendly interfaces
with strong analytic algorithms and comprehensive libraries and database.
As a solution, XCMS Online, an open sourced and web-based metabolomics platform
was launched with powerful analyzing algorithms and comprehensive MS and MS/MS
metabolite database through improvement of XCMS R packaging version. Newly
developed online platform allows easy file uploading, parameter setting, and result
interpretation (Tautenhahn et al., 2012). As a result, XCMS Online has been known as
the “World’s most cited metabolomics software” cited in literature over 1000 times (Ubhi
et al, 2014). It has detection algorithm, quality assessment, metabolite identification
(Wen et al., 2017) coupled with METLIN database.
Compared to other platforms, XCMS Online offers a more comprehensive package
that integrates the signal deconvolution, library matching, and statically analysis
5
including interactive cloud plot (univariate) and interactive PCA (multivariate). XCMS
Online includes peak detection, profile alignment, feature annotation, and exploratory
statistical analyses (Libiseller et al., 2015) with customized parameter setting. According
to its peak detection algorithm, it covers both low-resolution, and high-resolution using
matchedFilter method and centWave method respectively (Smith, 2013). Interactive
platform can monitor the statistical output of both univariate (cloud plot) and multivariate
(PCA) in real time by adjusting various parameters (Gowda et al., 2014). For identifying
distinct metabolites between two classes, it considers which intensities are significantly
different from sample class to sample class and rank the metabolites by p-value (Smith et
al., 2006). These statistical results are valuable when researchers are looking for distinct
metabolites between two treatments (basically control versus sample) or visualized
cluster between treatments. The seemly integration of METLIN library into XCMS
allows rapid identifying list of candidates of compounds for each extracted ion
chromatogram based on accurate mass (Smith et al., 2006). The METLIN and METLIN
mobile now include more than 961,820 molecules providing information for each
compound such as name, systematic name, structure, elemental formula, mass, CAS
number, KEGG ID and link, HMDB ID and link, PubChem ID and link, commercial
availability and direct search options on the molecule itself (https://metlin.scripps.edu/).
With the development of non-targeted metabolomics algorithm and computational
capacity, analytical instrument such as high resolution mass spectrometry enable high-
throughput chemical profiling in crude samples for natural product research. Compare to
targeted analysis, non-targeted analysis requires high resolution mass spectrometer
(HRMS) to screen (Un)known chemicals in the samples (Ruff et al., 2015). Several
6
concerns about non-targeted analysis due to massive amount of data processing, are being
solved by recent advancement in computation power, deconvolutions software algorithms
and the design of mass spectrometers. Since HRMS can measure molecular weights more
accurately with several decimal, compounds with similar mass can be distinguished by
HRMS that provides precision of molecular formula to be narrowed down to only a few
possibilities. There are several mass spectrometers that have been widely used for
untargeted analysis; time-of-flight (TOF) or orbitrap (Ruff et al., 2015). In the interest of
screening unknown or less common compounds, liquid chromatography coupled to mass
spectrometry (LC-MS) with time-of-flight (TOF) analyzers (TOF-MS or QTOF-MS) is
most suitable, due to its accurate mass analysis, resolving power, enhanced selectivity
and high sensitivity (García-Reyes et al., 2007). A large number of targets could be
screened simultaneously without sensitivity loss and unknown peaks can be identified
through accurate mass evaluation. Besides TOF-MS having 5000 FWHM (term of mass-
resolving power) (Petrovic et al., 2006), Orbitrap-MS is the most recent instrument for
HRMS-based non-target screening providing much higher mass-resolving with 280,000
FWHM (Kaufmann, 2014).
These innovational technologies especially XCMS Online metabolomics platform
could be extremely helpful to reduce time spent in processing the huge data sets for
health-promoting compound identification in each SCG cultivars and the SCG cultivar
comparison analysis in our study. These qualitative and quantitative analysis could
quickly help translate the benchtop experiments into real-world practical application in
the cosmetic industry.
7
1. Cyanidin 2. Resveratrol 3. Quercetin
4. Apigenin 5. Hesperetin 6. 3,5-Di-O-caffeoylquinic acid
Figure 1.1. Bioactive compounds related to UV protection efficacy
1 Cyanidin Anti-oxidant, Anti-inflammatory, Anti-cancer, Skin protection, anti-aging
2 Resveratrol Anti-inflammatory, antioxidant, anti-aging, photo protection
3 Quercetin Anti-bacterial, anti-oxidant, skin protection, Anti-melanogenesis
4 Apigenin Anti-inflammatory, antioxidant and anticancer, UVA/UVB-induced skin carcinogenesis
5 Hesperetin Photoprotective, Anti-inflammatory, antioxidant
6 3,5-Di-O-caffeoylquinic acid Antioxidant, anti-tyrosinase
8
Table 1.1. Major compounds exist in natural resources that have biological function related to skin care.
The list of bioactive compounds have been reported, showing antioxidant, Anti-fungal, skin-whitening,
anti-inflammatory, anti-aging, and antimicrobial properties or other skin-care applications.
Compound name Biological function
1 Quercetin1,2 Anti-bacterial, anti-oxidant, skin protection, anti-
melanogenesis
2 Resveratrol2,3,16 Anti-inflammatory, antioxidant, anti-aging, photo
protection
3 Cyanidin4,5 Antioxidant, anti-inflammatory, anti-cancer, skin
protection, anti-aging
4 Apigenin6,7 anti-inflammatory, antioxidant and anticancer,
UVA/UVB-induced skin carcinogenesis
5 Curcumin8,9 Anti-inflammatory, photo-protective, anti-cancer
6 Delphinidin10,11 Anti-proliferative, anti-inflammatory, antioxidant, anti-
mutagenic, anti-angiogenic
7 Malvidin12 Antioxidant, anti-inflammatory
8 Pelargonidin13, 14 Anti-inflammatory, antioxidant
9 Isorhamnetin14 Anti-inflammatory, anticancer
10 Kaempferol14 Anti-inflammatory ,hypoglycemic and antioxidant
11 Daidzein14 Anti-inflammatory, anti-carcinogenic, anti-inflammatory
12 Genistein14 Anti-inflammatory, antioxidant and anticancer
13 Caffeic acid15 Antioxidant
14 Chlorogenic
acid15,16,41
Anti-inflammatory, antioxidant, anti-aging, anti-skin
cancer, antimicrobial
15 Luteolin17, 30 UV protective, antioxidant, anti-inflammatory, anti-
melanogenic
16 Tangeritin18 Anti-inflammatory and anti-skin cancer
17 Myricetin19 UVB-induced skin cancer, anti-wrinkle
18 Proanthocyanidins20 Skin lightening, antioxidant
19 Hesperetin21 Photo protective, anti-inflammatory, antioxidant
20 Naringenin14,21, 47 Anti-inflammatory, anti-oxidant and anti-tumor, photo-
protective
21 Gallic acid22,23 Anti-skin tumor, antioxidant, anti-tyrosinase
22 Vanillic acid24 Antioxidant, anti-tyrosinase
23 Protocataechuic
acid25
Antioxidant
24 Dicaffeoylquinic
acid26
Antioxidant, anti-tyrosinase
25 Tannic acid22, 27 Anti-skin tumor, anti-bacterial
26 p-Coumaric acid28 Anti-pigmentation
9
27 Taxifolin29, 30 Anti-skin carcinogenesis, anti-tyrosinase
28 Luteolin30 Anti-melanogenic
29 Naringin31 Skin photo protection, antioxidant, anti-carcinogenic
30 Chrysin32 Skin protection, anti-inflammation and anti-oxidation
31 Rutin33 Anti-inflammation, anti-oxidation
32 Ethylguaiacol34 Antimicrobial
33 Vinylguaiacol34 Antimicrobial
34 Cinnamic acid35 Anti-tyrosinase, antioxidant and antimicrobial
35 Vanillin36 Antioxidants, antibacterial, antimicrobial anti-
inflammatory
36 Glycitin38,39,40 Anti-photo aging, anti-inflammatory, anti-aging
37 Genistin42 Prevent UV-induced skin damage
38 Caffeine44,45,46 Antioxidant, anti-skin cancer, antimicrobial 1: (Ramos et al., 2006)
2: (Choquenet et al., 2008)
3. (Alarcon De La Lastra and Villegas, 2005)
4: (Choi et al., 2010)
5: (Wang et al., 1999)
6: (Svobodová et al., 2003)
7: (Kris-Etherton et al., 2004)
8: (Saraf and Kaur, 2010)
9:(García-Bores and Avila, 2008)
10: (Chamcheu et al., 2015)
11: (Bin Hafeez et al., 2008)
12: (Bognar et al., 2013)
13: (Noda et al., 2002)
14: (Hämäläinen et al., 2007)
15: (Sato et al., 2011)
16: (Kitagawa et al., 2011)
17: (Wölfle et al., 2011)
18: (Yoon et al., 2010)
19: (Jung et al., 2010)
10
20: (Yamakoshi et al., 2003)
21: (Bonina et al., 1996)
22: (Perchellet et al., 1992)
23: (Kim, 2007)
24: (Chou et al., 2010)
25: (Tseng et al., 1996)
26: (Iwai et al., 2004)
27: (Akiyama et al., 2001)
28: (Seo et al., 2011)
29: (Oi et al., 2012)
30: (An et al., 2008)
31: (Ren et al., 2016)
32: (Wang et al., 2011)
33: (Choi et al., 2014)
34: (Banister and Cheetham, 2001):
35: (Kong et al., 2008)
36: (Makni et al., 2011)
37: (Jin et al., 2012)
38: (Seo et al., 2014)
39: (García-Lafuente et al., 2009)
40: (Kim et al., 2015)
41: (Zhao et al., 2010)
42: (Russo et al., 2006)
43: (Arung et al., 2011)
44: (Devasagayam et al., 1996)
45: (Lu et al., 2002)
46: (Almeida et al., 2006)
47: (An et al., 2010)
11
2. Objectives
Over the past several decades, most of the bioactive compounds identified in coffee and
its by-products were through the labor-intensive conventional bio-assay guided
purification schemes followed by structural analysis (e.g., NMR and MS/MS). Although
high-resolution and tandem mass spectrometry have been utilized for structural
elucidation process, the process often requires manual database searching for compound
identification. Due to these technical limitations and time-consuming processes, it has
been challenging to systematically characterize the bioactive compounds in coffee
extracts. Recent advances in metabolomics science, mass spectrometry, computation
capacity, and compound libraries have made this task possible.
The goal of this project is to investigate the chemical profiles of SCG derived from three
coffee cultivars through global metabolite analysis and gauge the potential use of SCG
from these cultivars cosmetics application.
The specific objectives of this study are as following:
1. Identify and characterize the health-promoting polyphenols, and other health-
promoting secondary metabolites in SCG using global metabolomics approach (XCMS).
2. Compare the chemical profiles between the SCG extracts from Ethiopia coffee
(Yirgacheffe), Costa Rican coffee (Tarrazu) and Hawaiian coffee (Kona)
3. Materials and Methods
Figure 1.3 illustrated the flowchart of the SCG research using a non-targeted
metabolomics approach. In this scheme, the compounds were extracted, and analyzed by
12
UPLC-MS and UPLC-HRMS (Bruker maXis Q-TOF), followed by XCMS Online
analysis coupled with METLIN compound database. The list of identified bioactive
compounds through this global metabolomics approach were compiled.
3.1. Sample preparation
SCG from three Arabica coffee cultivars including Ethiopian Yirgacheffe, Costa Rican
Tarrazu, and Hawaiian Kona were selected for this study. Roasted and ground coffee
were purchased from the local Lakota Coffee (Columbia, Missouri). After brewing for 5
minutes, SCG were collected. Twenty-five grams of wet SCG (78% water, about 5.5 g
DW equivalent) were homogenized with blender and the compounds were extracted with
200 ml methanol (HPLC-grade, purchased from Fisher Scientific, Pittsburg, PA, USA).
The extract was then sonicated for 60 mins and filtered through Whatman phase
separators (silicone treated filter paper, diameter 125mm). Each SCG extracts from three
cultivars was prepared as independent triplicates.
3.2. LCMS Analysis
3.2.1. LC-Low Resolution conditions: The LCMS analysis was performed by a
Waters Ultra-High Performance Liquid Chromatography system coupled with Waters
Xevo TQ triple quadrupole mass spectrometer (HPLC-MS/MS). The compounds were
separated by a Phenomenex (Torrance, CA) Kinetex C18 (100mm x 4.6 mm; 2.6 µm
particle size) reverse-phase column. The mobile phase consisted of 10mM ammonium
acetate and 0.1% formic acid in water (A) and 100% acetonitrile (B). The gradient
conditions were 0–0.5 min, 2% B; 0.5–7 min, 2–80% B; 7.0–9.0 min, 80–98% B; 9.0–
13
10.0 min, 2% B; 10.0–15.0 min, 2% B at a flow rate of 0.5 ml/min. The MS system was
operated using electrospray ionization in the either negative ion mode (ES-) or positive
(ES+) with capillary voltage of 1.5 kV (ES-). The ionization source was programmed at
150C and the desolvation temperature was programmed at 750C.
14
Figure 2.2. Flow chart of the SCG metabolomics approach project using Waters Xevo UPLC-MS for non-
VOC analysis and XCMS global metabolomics platform.
15
3.2.2 The LC-HRMS conditions: LC-MS analysis was performed on a Bruker
maXis impact quadrupole-time-of-flight mass spectrometer coupled to a Waters
ACQUITY UPLC system. Separation was achieved on a Waters C18 column (2.1x 100
mm, BEH C18 column with 1.7-um particles) using a linear gradient of 95%:5% to
30%:70% eluent A:B (A: 0.1% formic acid, B: acetonitrile) over 30 min. Between 30-
33min, the a linear gradient was increased from 70% to 95% B, and maintained at 95%B
for 3 min. The percentage of B was maintained at 5% from 36-40 min. The flow rate was
0.56 mL/min and the column temperature was 60 C. Mass spectrometry was performed in
the negative (or positive) electrospray ionization mode with the nebulization gas pressure
at 43.5 psi, dry gas of 12 l/min, dry temperature of 250C and a capillary voltage of
4000V. Mass spectral data were collected from 100 to 1500 m/z and were auto-calibrated
using sodium format after data acquisition.
3.3 Global metabolite profiling using XCMS and METLIN Library
Figure 1.3. Flow chart of Metabolomics Data Processing.
16
XCMS Online data processing includes peak detection, retention time correction,
alignment, annotation, and statistical analyses (Figure 1.3). First, determining the
boundaries, centres and intensities of the two-dimensional peak signals in the LC/MS raw
data is called feature detection (Tautenhahn et al., 2008). According to its peak detection
algorithm, it covers both low-resolution, and high-resolution using matchedFilter method
and centWave method respectively (Smith, 2013). After peak detection, XCMS Online
uses nonlinear methods to compensate for retention time drifts between samples
(Tautenhahn et al., 2012). All total ion chromatogram (TIC)s should be aligned to
compare compound quantities between fractions after retention time correction,. XCMS
Online uses the OBI-Warp (Ordered Bijective Interpolated Warping) retention time
alignment method based on dynamic time warping with a one-to-one (bijective) smooth
warp-function, especially for ESI-LC-MS data (Smith et al., 2006). Next step is ion
annotation. By annotating at least two ions, the molecular mass can be calculated which
is necessary to search compound libraries (Kuhl et al., 2011). After preprocessing,
statistical analyses were conducted. For identifying distinct metabolites between two
classes, it considers which intensities are significantly different from sample class to
sample class and rank the metabolites by p-value (Smith et al., 2006). XCMS Online
provides statistical result of data processing such as interactive cloud plot, principle
component analysis (PCA), and non-metric multidimensional scaling (MDS).
3.3.1. Low resolution data processing through XCMS
The ion chromatograms generated from MS (.cdf) files including triplicate samples for
each SCG cultivars were submitted to XCMS platform operated by Center for
17
Metabolomics at the Scripps Research Institute. The peak detection, peak grouping,
spectra extraction, retention alignment, were processed by XCMS (Figure 1.3). The job
was defined a multi-group job for comparative analysis to distinguish chemical profiling
between three cultivars and the nearest instrument option, HPLC / Single Quad – HPLC
with ~60 min gradient, Single Quadrupole MS was selected. For feature detection, the
parameters were set as the following, mass tolerance of 30 ppm in consecutive scans,
peak width of 10-60 s, S/N threshold of 6, and matched Filter algorithm was selected for
low resolution data analysis. Both positive and negative polarity were analyzed. For the
retention time alignment, the peakgroups method with m/z width 0.25 s, 5 (50%) of
minimum fraction of samples was selected to make at least one of the sample groups
valid. In every XCMS data processing, ‘plant’ was selected as a biosource. Post-hoc
analysis in multi-group job was performed. Multivariate analysis and principle
component analysis (PCA) were performed by XCMS to compare the chemical profiles
between the three coffee cultivars. The spectra were annotated and the compounds were
identified and categorized by the integration of the METLIN, the world's largest
metabolite database. Each identified compound was assigned and related to its biological
pathway through XCMS biological pathway/network analysis.
3.3.2. High resolution data processing through XCMS
Pairwise job was selected for two-group comparison study. First, we compared among
cultivars without control. For example, Ethiopian Yirgacheffe versus Costa Rican
Tarrazu, Costa Rican Tarrazu versus Hawaiian Kona, Hawaiian Kona versus Ethiopian
Yirgacheffe. In second two-group comparison analysis, we used control (MeOH) with
18
each cultivar such as control versus Ethiopian Yirgacheffe, control versus Costa Rican
Tarrazu, control versus Hawaiian Kona. Each file group was submitted as triplicates. In
the instrument selection, UPLC/Bruker QTOF POS was marked for high resolution data
analysis, which automatically activates centWave algorithm, a newly published algorithm
designed for centroid and high-resolution data analysis for both two-group comparison
study. Parameters in high resolution analysis were set as follows: maximal tolerance of
10 ppm between successive measurements, peak width of 5-20 s, S/N threshold of 6 and
obiwarp algorithm for retention time correction and alignment (width of overlapping m/z
slices 0.015, minimum fraction of samples 0.5 for group validation). To visualize
differences between two samples in pairwise job, 0.001 of p-value threshold and 1.5 of
fold change was selected to be considered highly significant.
4. Results
XCMS Online allows for visualization of statistical results such as interactive cloud plot
and principal component analysis (PCA). Both functions could be applied to multi-group
and two-group comparison. Through multi-job using low resolution data, three clusters in
PCA and non-metric multidimensional scaling were used to closely examine and
differentiate the chemical profiles and composition between the SCG extracts from
Ethiopian Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona. Pairwise job using high
resolution data enables comparisons between cultivars and provides statistical analysis as
well. Lastly, pairwise analyses were performed to identify the bioactive compounds and
evaluate their relative concentrations between the three cultivars. The results of pairwise
analyses were compiled in table 3.1 (see Appendix A.) sorted by their compound groups
19
including phenolic acids, flavonoids and other compounds with their activities such as
antioxidant, anti-inflammatory, and other biological effects relevant to skin protection.
4.1. Multi-group Comparison between three SCG cultivars using low
resolution data
XCMS provides information about a comparison of features extracted from the
chromatograms using principal component analysis (PCA), non-metric multi-dimensional
scaling and metabolomic cloud plot (Figure 1.4c) to visualize statistical results.
For multi-job comparison analysis using XCMS, if more than three groups should be
analyzed, multiple group analysis is more suitable. It enables the comparison between
three SCG cultivars based on their metabolite features when they show significant
difference. To distinguish each group as a specific cluster, Post-HOC multiple
comparison was selected in parameter setting (Gowda et al., 2014).
Multi-group cloud plot allows visualization of metabolite features if they vary
significantly among each group. Cloud Plot is showed based on their retention time and
m/z ratio. In figure 1.4c, statistical significance threshold (Kruskal–Wallis p-value) was
entered as 0.01 in main panel. In addition to p-value, feature intensity, m/z, retention
time, and circle radius ratio range were selected as 29706525, 993, 15, 0.1 default values,
respectively. With each selected circle, the box-plot, EIC, Post-HOC, and METLIN were
illustrated under the cloud plot. Among features, ID: 280 was selected which includes
quercetin 3,7-di-O-sulfate which is our candidate of interest in composite database.
Figure 1.4a represents the sulfate conjugate of quercetin which is significantly up-
regulated in the Hawaiian Kona (ESI(_)_K1) among other cultivals such as CostaRican
Tarazzu (ESI(_)_T1), and Ethiopian Yirgacheffe (ESI(_)_Y1). Therefore, multi-group
20
analysis helps to compare relative feature intensity among different groups (in this case,
among different cultivars)
21
Figure 1.4a. Box-plot generated from interactive cloud plot
The relative concentration between three cultivars for ID 280 (in interactive cloud plot) were represented by Box-plot. ID 280 includes quercetin 3,7-di-
O-sulfate which is our candidate of interest in composite database. Each SCG extracts was shown in x-axis and its relative concentration was indicated
in y-axis as signal intensity.
Cultivar
Inte
nsi
ty
22
Figure 1.4b. Composite data base after mouse click on interactive cloud plot. Seven compounds were matched with METLIN database based on ion
chromatogram of each circle. Compounds in composite DB linked to METLIN library it will show compound formula, structure, and spectrum.
23
Figure 1.4c. Multi-group cloud plot. Metabolite feature which varies significantly (p < 0.01) from each groups are presented on cloud plot depending on
their retention time (x-axis) and m/z (y-axis). Each circle indicates each compound feature. The color intensity of circle represents statistical significance
(p-value). The size of the circle indicates feature intensity. If you move the mouse on each circle, p-value, q-value, m/z, RT, Max intensity will pop-up.
With mouse click each circle, Box-plot, EIC, Post-HOC, and Composite database (METLIN) are displayed under the cloud plot. Compounds in
composite DB are linked to METLIN library showing compound formula, structure, and spectrum.
24
XCMS Online provides interactive multivariate principal component analysis (PCA). The
results of the PCA analysis suggests that SCG samples from three different coffee
cultivars (Ethiopian Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona) clustered by
each group indicating chemical composition of cultivars is significantly different each
other (Figure 1.5). The multivariate besides univariate analysis (cloud plot) are often
required, when the differences between groups cannot be distinguished by simple
separating method like univariate tests (Gowda et al., 2014). PCA is a major statistical
tools in untargeted metabolomics for chemical profiling (Bartel et al., 2013). PCA
reduces the dimensionality of the data while remaining as much of variation so it can
detect differences between the groups. In figure 1.5, the three SCG cultivars were divided
into each cluster by PCA. XCMS Online also provides scree plot, scores plot and loading
plot in Interactive PCA tab (Figure 1.5a). The scree plot represents how many principal
components were used in PCA. The scores plot visualizes cluster of submitted samples.
Three clusters are examined through PCA which means three SCG cultivars have distinct
compounds contents compared to each other. Non-metric Multidimensional Scaling
(NMDS) plot also shows variation in the composition of SCG metabolite profiles.
25
Figure 1.5a, Interactive principal component analysis. A Scores plot shows the correlation between
submitted samples. Three clusters represent each SCG cultivars which includes Hawaiian Kona
(ESI(-)_K1), Costa Rican Tarrazu (ESI(-)_T1, and Ethiopian Yirgacheffe (ESI(-)_Y1).
26
Figure 1.5b. Principal Component Analysis (PCA) results: SCG samples from three different coffee
cultivars (Ethiopian Yirgacheffe, Costa Rican Tarrazu, Hawaiian Kona) clustered by each groups.
Individuals are projected in a reduced dimension space defined by component 1 (x-axis) and component 2
(y-axis). Each color corresponds to a specific cluster.
27
Figure 1.5c. Non-metric multi-dimensional Scaling of different coffee cultivars (Ethiopian Yirgacheffe,
Costa Rica Tarrazu, and Hawaiian Kona) shows that the chemical profiling is significantly different
between three cultivars.
28
4.2. Identified bioactive compounds through global metabolomics analysis
using high resolution data
4.2.1 Two -group comparison study between control and each cultivar
Through XCMS Online, around 200 bioactive compounds were identified including 22
phenolic acids, 161 flavonoids 2 coffee alkaloids, 1 stilbenoid, and 3 anthraquinones (see
table 3.1). Free forms of these identified compounds have bioactivity related to skin care
such as antioxidant, anti-microbial, anti-inflammatory, anti-skin cancer and skin whitening
effect. In table 3.1, each row represents compound name, m/z, retention time, relative
concentration between cultivars (Yirgacheffe, Tarrazu, and Kona) and bioactivity of the
free form. Comparison of maximum peak intensity among the three SCG cultivars was
accomplished by combination of 3 two-group comparison analysis. Several compounds
which shares same molecular weight with the same m/z and max intensity indicates
candidate compounds of certain ion chromatogram. Most of the compounds are derivatives
and glycosides of well-known anti-oxidant, anti-inflammatory, anti-bacterial or anti-photo
aging compounds. These bioactive compounds include the derivatives or free form of
cinnamic acid, hydroxycinnamic acid, glycosidic form of isorhamnetin, hesperetin,
theaflavin derivatives, glycosidic form of chrysin, epigallocatechin coumaroylquinic acid,
coumaric acid glycosides, caffeoylquinic aicd, dicaffeoylquinic acid, glycosidic form of
quercetin, kaempferol, cyanidin, pelargonidin, malvidin, apigenin, fujikinetin, genistin,
isovitexin, viexin, puerarin, daidzein, dihydroxyflavone, free form of glycitin, obtusifolin
glycosides, formononetin glycosides and procyanidin glycosides were identified in SCG
by XCMS Online.
29
Figure 1.6. Chemical structures of phenolic acid.
30
Many types of phenolic acids ,caffeine and flavonoids (flavonols, flavones, isoflavones,
flavanones, anthocyanidins and flavanols) were identified in SCG through XCMS
Online.For example, cinnamic acid, hydrocinnamic acid, p-coumaroylquinic acid, p-
coumaric acid glucosides, vanillic acid derivatives, 1,4-Dicaffeoylquinic acid
(Isochlorogenic acid), chlorogenic acid were detected as a phenolic acid group.
Phenolic acids: For the cinnamic acids, a wide range of derivatives were identified as
described in Table 3.1. Among them, hydrocinnamic acid was detected at RT 23.79 and
RT 14.21. Generally, Costa Rican Tarrazu has highest concentration of hydrocinnamic acid
among the three cultivars. Likewise, this cultivar shows the highest amount of 4-Methoxy-
(2E)-cinnamic acid at RT 9.73. The detection of O-Methoxyhydrocinnamic acid in SCG is
reported for the first time from this study. However, its bioactivity has not been carefully
examined. The ion chromatogram showed that O-methoxyhydrocinnamic has m/z of
181.0848 with the RT 6.03 min. The relative signal intensities of O-
methoxyhydrocinnamic were 29,768, 34,924, and 27,494 for Ethiopian Yirgacheffe, Costa
Rican Tarrazu and Hawaiian Kona respectively. P-coumaroylquinic acid, only appeared in
Hawaiian Kona variety, has been known for its strong antioxidant, and antibacterial activity
(Pereira et al., 2007). As candidate compounds, trans-p-coumaric acid 4-glucoside and
coumaric acid at RT 4.92, m/z 349.0898 were only identified in cultivar Tarrazu. Vanillic
acid 5-(4-carboxy-2-methoxyphenoxy)-, was first identified by this study having m/z
357.0585, RT 3.47. The free form of vanillic acid has been known for its strong bioactivity
including antioxidant (Han et al., 1981), antibacterial and anti-fungal activity (Aziz et al.,
1998). Relative concentration of vanillic acid was the highest in Ethiopian Yirgacheffe as
68,432, 14,468, 20,036 in each cultivars (4.73 times higher than Costa Rican Tarrazu
31
varieties). Epigallocatechin 3-O-vanillate is first discovered from our research. Through
XCMS Online data processing, the compound was identified at RT 5.92 with m/z 457.1109
in Costa Rican Tarrazu and Hawaiian Kona. In case of caffeoylquinic acid, only Costa
Rican possesses 1,4-Dicaffeoylquinic acid and Cis-5-Caffeoylquinic acid with m/z
517.1335 and 355.1025 respectively. On the other hand, 1-feruloyl-5-caffeoylquinic acid
concentration appear higher in Ethiopian Yirgacheffe than Costa Rican Tarrazu.
Isochlorogenic acid A (3,5-Di-O-caffeoylquinic acid), a compound with anti-oxidant and
anti-tyrosinase activity so that it has value in and cosmetics (Iwai et al., 2004) showed
highest concentration in Costa Rican Tarrazu varieties. In XCMS, the compound was
identified with m/z 539.1149 at RT 6.67. XCMS also detected an anti-pigmentation
compound from the methanol extract of SCG, which was identified as 4,5-𝑂-
dicaffeoylquinic acid (4,5-diCQA).
Flavonoids: Quercetin, kaempferol, and isorhamnetin exist in glycosylated forms in SCG
samples with various types of attached sugars, such as glucose, rhamnose, galactose,
arabinose, and rutinose. As flavones, apigenin, chrysin, and hydroxyluteolin derivatives
and dihydroxyflavone appeared in XCMS Online results table. Glycosylated flavanones
such as naringin, and hesperetin glycosides were also detected. Isoflavones are commonly
found in legume plants (e.g., soybean) (Amaral et al., 2006). However, XCMS Online
detected the three major soybean isoflavones including genistein, daidzein, and naringenin
attached with variety of glycosides in SCG samples. Also, isoflavones such as retusin,
formononetin and fujikinetin glycosides were identified as candidate compounds. In
addition, antocyanins including malvidin, cyanidin, pelargonidin glycosides, and
glucopyranosylprocyanidin were detected through XCMS Online chemical profiling.
32
Relative concentration of flavonoids among SCG cultivars:
A remarkable difference in the concentrations of 8-C-ascorbylepigallocatechin 3-gallate
(m/z 633.1122, RT 35.86) were found between the SCG varieties. Previously, 8-C-
ascorbylepigallocatechin 3-gallate was identified in oolong tea with no known
bioactivities. Through XCMS Online chemical profiling, this compound was found in SCG
extracts. The highest concentration was found in Ethiopian Yirgacheffe, approximately
3.24 times higher than that in Costa Rican Tarrazu.
We found that caffeine, was the most abundant compounds in Ethiopian Yirgacheffe
cultivar followed by isorhamnetin glycosides, isorhamnetin 3-(3'''-ferulylrobinobioside),
E-cinnamic acid and isorhamnetin 3-glucosyl-(1->2)-galactoside-7-glucoside. Especially,
isorhamnetin 3-glucosyl-(1->2)-galactoside-7-glucoside was found only in Ethiopian
Yirgacheffe SCG. In case of E-cinnamic acid detected at RT 16.76, Hawaiian Kona showed
the highest concentration of the compound among the three cultivars. Costa Rican Tarrazu
had the highest concentration of dicaffeoyquinic acid.
Flavonols such as quercetin and kaempferol are the most dominant flavonoids in foods and
often exist in glycosylated forms (Manach et al., 2004). A wide range of quercetin
glycosides including 3-sophoroside-7-glucuronide, 3-gentiobioside-7-glucuronide, and 7-
glucuronoside 3-sophoroside were detected as candidates for certain ion peak having m/z
820.2114, and RT 4.57 in SCG samples. The maximum intensity generated by XCMS
Online and high resolution SCG data files was 3,178 2,780 or 2,312 for Ethiopian
Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona respectively. Likewise kaempferol
exists in many types of glycosylated forms including 7-rhamonoside, diacetylrhamnoside,
33
p-coumarylrhamnoside, coumaroylrutinoside, galloylrutinoside, coumarylrobinobioside, -
galloyl-alpha-L-arabinopyranoside, arabinofuranosyl-(1->6)-glucoside. In the case of
kaempferol with coumarylrobinobioside and diacetylrhamnoside moieties, the highest
amount was found in Costa Rican Tarrazu. However, the levels of the glycosylated
kaempferol with 7-rhamnoside (at RT 32.85) were around 2.7 times higher in Ethiopian
Yirgacheffe than Costa Rican Tarrazu. The levels for the rest of kaempfrerol derivatives at
RT 19.09 and RT 35.68 are the highest in Hawaiian Kona especially for latter case, the gap
is 71% between Costa Rican Tarrazu. Isorhamnetin 5-glucoside that can be found in
Artemisia capillaris (Wu et al., 2001) also isolated in SCG Ethiopian Yirgacheffe.
Information about bioactivity related to cosmetics of Isorhamnetin 3-(4''-sulfatorutinoside)
was rare likewise isorhamnetin 5-glucoside. Basically, contents of isorhamnetin glycosides
are higher in Ethiopian Yirgacheffe. Among flavones, types of apigenin glycosides were
noticeable in table 3.1. In case of apigenin 7-O-(6-O-malonylglucoside), it exists only in
Ehiopian Yirgacheffe at RT 34.86. Apigenin 8-C-(6''-acetylgalactoside) and apigenin 7-
rhamnoside were not detected in Tarrazu and the concentration was rarely different
between Yirgacheffe and Kona. Specially, the concentration of apigenin contained with
acetylalloside in Ethiopian Yirgacheffe is 2.71 times higher than Tarrazu. The 7-O-beta-
D-glucoside of genistein was found at RT 5.57 only in Kona. Isoflavone glycosides, retusin
7-O-glucoside which has been reported exists in plant Pterocarpus marsupium;
leguminosae (Schmid et al., 2003) was identified in SCG Ethiopian Yirgacheffe varieties
with m/z 464.1547, RT 23.9. Formononetin 7-O-glucoside-6''-malonate (formononetin 7-
O-(6''-malonylglucoside) was detected as candidates at RT 5.83 and RT 6.67 accompanied
by isochlorogenic acid (dicaffeoyquinic acid) showing relatively high concentration in
34
Tarrazu. Isoflavone glycosides, fujikinetin 7-O-laminaribioside and fujikinetin 7-O-
glucoside were detected with quercetin 3-(4''-acetylrhamnoside)-7-rhamnoside at RT 32.85
and with 8-C-beta-D-glucofuranosylapigenin 2''-O-acetate at RT 5.83 respectively. In
former case, the concentration of this compound in Yirgacheffe is 2.71 times higher than
concentration in Tarrazu.
Among compounds described in table 3.1, anthraquinone glycosides such as physcion 8-
glucoside were only detected in Yirgacheffe. The concentration of another anthraquinone
derivatives chrysophanol 1-tetraglucoside was 57% higher in Kona compared to Tarrazu.
Lastly, various types of dihydroxyflavone (DHF) were selected in XCMS chromatogram
peak at RT 20.28 with m/z 255.0655. The concentration of the compound in variety
Ethiopian Yirgacheffe is 23 % higher than variety Costa Rican Tarrazu.
Biological function
Our analytical results have identified several major compounds having bioactivity related
to skin care application such as antioxidant, anti-inflammatory, anti-microbial and other
biological functions (see table 3.1). For example, among the 200 identified compounds,
133 antioxidant, 132 anti-inflammatory, 13 anti-microbial, 15 anti-pigmentation, and 69
anti-cancer agents were found. Additionally, 30 compounds which have been reported as
components for skin protection against UV or anti-aging were also detected through XCMS
Online.
Antioxidant effect: Several phenolic acids with antioxidant effect were found in SCG
through XCMS Online; caffeoyl quinic acids (Iwai et al., 2004), chlorogenic acids (CGA)
35
(Sato et al., 2011), cinnamic acid derivatives (Kong et al., 2008), and vanillic acids (Chou
et al., 2010) are well-known as their antioxidant. SCG also contains apigenin, chrysin,
quercetin, hesperetin, naringenin, kaempferol, cyanidin, pelargonidin, isovitexin and
malvidin glycosides and the free from of these flavonoids have strong antioxidant activity
(Saija et al., 1995) (Wu et al., 2011) (Torel et al., 1986) (Rice-Evans et al., 1996). Flavone
tectochrysin in Ethiopian Yirgacheffe at RT 8.42 and the flavanol theaflavin which were
detected in SCG by XCMS Online platform have been studied that it has antioxidant
activity (Lee et al., 2003b) (CHEN and HO, 1995). Also, there is one interesting fact that
7,8-dihydroxyflavone selected in XCMS chromatogram peak at RT 20.28 with m/z
255.0655 has an antioxidant effect inhibiting cell-death from hydrogen peroxide (Chen et
al., 2011). Isoflavones, formononetin derivatives containing malonate are detected at RT
6.67 and the free form is known as antioxidants (Mu et al., 2009). Interestingly, puerarin
4'-O-glucoside were reported as antioxidant as glycosides itself (Xiao et al., 2011). Among
several kinds of procyanidin included in SCG sample, procyanidin B2 (epicatechin-(4β-8)-
epicatechin) are known as antioxidant (Sakano et al., 2005). Coffee alkaloid caffeine which
is most abundant in SCG are also have antioxidant activity (Devasagayam et al., 1996).
Lastly, eugenol and resveratrol are also studied widely as antioxidant (Bendre et al., 2016)
(Mahal and Mukherjee, 2006).
Anti-inflammatory effect: In table 3.1, it is revealed that a lot of compounds that have anti-
inflammatory effect with anti-oxidant activity. For example, the isoflavones daidzein and
genistein, the flavonols isorhamnetin, kaempferol and quercetin, the flavanone naringenin,
and the anthocyanin pelargonidin existed in SCG samples as glycosylated forms and their
free forms are known as anti-inflammatory compounds inhibiting NF- κB activation (Rice-
36
Evans et al., 1996). Additionally, other flavonoids such as malvidin (Huang et al., 2014),
chrysin (Wu et al., 2011), isovitexin (Lv et al., 2016), and hydroxykempferol (Williams et
al., 1999) 7,8-dihydoxyflavone, (Park et al., 2012), procyanidin B1 (Xing et al., 2015) are
important anti-inflammatory plant flavonoids and were detected by XCMS Online with
attached with glycosides. Other types of compounds including phenylpropanoid,
stilbenoid, and anthraquinones were found in SCG with anti-inflammaotry activity (e.g.
eugenol, resveratrol, chrysophanol, physcion) (Bendre et al., 2016) (García-Sosa et al.,
2006). Puerarin 4'-O-glucoside antioxidant introduced in above, has anti-inflammatory
effect as well (Xiao et al., 2011).
Other activities related to skin care: Among phenolic acids, E-cinnamic acid (RT 16.76),
vanillic acid (RT 3.47), and dicaffeoylquinic acid (RT 6.67) has anti-tyrosinase effect
which means skin whitening efficacy (Lin et al., 2011) (Chou et al., 2010) (Iwai et al.,
2004). Likewise, coumaric acid which was detected at RT 4.92 also has anti-pigmentation
effect (An et al., 2010) while chlorogenic acid (5-Caffeoylquinic acid) affects to photo
aging (Kitagawa et al., 2011). In case of caffeic acid, there has been some reports that it
increases the chances of cancer. However, the International Agency for Research on
Cancer (IARC), convened by the World Health Organization has just concluded that there’s
no strong evidence that coffee increases the chances of cancer which have mooted in the
past (Loomis et al., 2016). In addition, the compound protects phospholipidic membranes
from UV light-induced peroxidation (Bonina et al., 2002) and showed a significant
protection to the human skin against UVB-induced erythema (Saija et al., 2000).
Flavonoids which have been widely studied for skin protection properties are found in
SCG. Among them, apigenin, genistin, genistein, daidzein, chrysin, eriodictyol, hesperetin,
37
and naringenin have been reported as photo protective agents against UV (Wang et al.,
2011) (Bonina et al., 1996) (An et al., 2010). Especially quercetin are notable for having
anticancer, anti-melanogenesis, skin protection, and antibacterial effects (Svobodová et al.,
2003). Additionally, genistin detected in Kona at RT 5.57 was also reported to have
protective effect on UV-induced DNA damage in cell-free condition. This compound
reduces the M14 cell viability due to anti-carcinogenic effects (Russo et al., 2006).
Flavanones naringin was studied as photo protection, and anti-carcinogenic compound
(Ren et al., 2016) mostly found in Yirgacheffe in this study. Cyanidin also has multiple
health-promoting effects such as anticancer, skin protection, and antiaging (An et al.,
2010). Glycitin (García-Bores and Avila, 2008) which is a natural isoflavone, was detected
as a free form in SCG at RT 23.9 and it is known as antioxidant, anti-allergic agent which
has been reported to increase dermal fibroblast cell proliferation and migration linked to
anti aging effect (Kim et al., 2015). The protective effect on UV-induced skin photo aging
was also reported by Seo and Park. (Seo et al., 2014). Both 6-hydroxykaempferol and 6-
methoxyeriodictyol identified in SCG using metabolomics analysis has anti-tyrosinase by
itself (Gao et al., 2007) (Lee et al., 2006). Compared to other flavonoids whether free form
or glycosylated form, a great amount of hesperetin laurate which has anti-fungal effect
strongly inhibiting the mycelial growth of P. expansum (Razzaghi-Abyaneh and Rai, 2013)
was shown in variety Kona at RT 36.05. The compound also acts as skin conditioning agent
(Slavtcheff et al., 2005). In addition, it is interesting that caffeine has anti-skin cancer (Lu
et al., 2002) and antimicrobial (Fardiaz, 1995) with emollient properties. Besides, several
compounds including guaiacol, chrysophanol, and physcion shares anti-microbial activities
and three anthroquinones (chrysophanol, soranjidiol, and physcion) listed at RT 35.68, RT
38
20.28, and RT 23.9 respectively have anti-skin cancer effects (García-Sosa et al., 2006;
Vittar et al., 2014). Eugenol detected through XCMS Online in SCG are well known
aromatic compound responsible for “woodsy” flavor in coffee. It has multiple bioactivities
such as antimicrobial, and anticancer with antioxidant and anti-inflammatory effects
(Bendre et al., 2016). Lastly, resveratrol is well known for anti-aging and photo protection
effects (Alarcon De La Lastra and Villegas, 2005). Such reports about bioactivity can
make SCG as a promising raw material in the dermatology field.
4.2.2 Two-Group Comparisons between SCG cultivars
Comparing two groups is the basic strategy in every research to determine certain
bioactive metabolite features that one group has significant difference compared to the
other. Here, pairwise analyses were performed to compare the chemical profiles and the
concentrations of the bioactive compounds between Ethiopian Yirgacheffe vs. Costa
Rican Tarrazu, Costa Rican Tarrazu vs. Hawaiian Kona, and Hawaiian Kona vs.
Ethiopian Yirgacheffe. For each group, data files (.cdf) from three independent replicates
were uploaded and processed by XCMS. More than 11,932 features generated from
comparison between Ethiopian Yirgacheffe and Hawaiian Kona, and several of them
were identified as the potential bioactive compounds for skin-care application. For
example, strong anti-inflammatory compounds quercetin pentaacetate was identified with
m/z 530.1253, with retention time (RT) 11.70 min. The concentration of quercetin
pentaacetate was significantly higher (p = 0.00147) in Ethiopian Yirgacheffe with 2.3
folds higher as compared to that in Hawaiian Kona. In another experiment, Ethiopian
versus Costa Rican Tarrazu, significant feature (p = 0.02081) detected at RT 34.86. The
candidate compounds were cyanidin-3-(6'-malonylglucoside), apigenin 7-O-(6-O-
39
malonylglucoside) with m/z 558.0949, four times higher concentration in Ethiopian
rather than Costa Rican Tarrazu (Figure 1.7b). The maximum intensity was 242,958.
Comparison between two groups were also visualized by cloud plots. 21 features were
acquired from Ethiopian versus Hawaiian Kona. Nine and 46 features were detected from
Costa Rican Tarrazu versus Hawaiian Kona and Ethiopian Yirgacheffe versus Costa
Rican Tarrazu respectively. Similar to multi-job interactive cloud plot, users can view
m/z, RT, p-value, fold change, max intensity, and METLIN library matches by moving
mouse. Interactive cloud plot enables filtering p-value and fold change by adjusting the
bar in main panel. By zooming, mouse can detect every circle without any inconvenience
even though the circles are overlapped.
Figure 1.7a. Extracted ion chromatogram of Ethiopian Yirgacheffe (E) and Hawaiian Kona (H). The peak
represents relative intensity of quercetin pentaacetate in SCG from Yirgacheffe (E), and Kona (H).
40
Figure 1.7b. Extracted ion chromatogram of Yrigacheffe (E) and Costa Rican Tarrazu (C). The peak
represents relative intensity of candidate compounds such as cyanidin-3-(6'-malonylglucoside) or apigenin
7-O-(6-O-malonylglucoside) at RT 34.86.
41
Figure 1.8. Cloud Plot results: A Cloud Plot displays features whose intensities are altered between sample
groups. Up-regulated features are represented as circles on the top of the plot and down-regulated features
are represented as circles on the bottom of the plot, where the size corresponds to the (log) fold change of
the feature. Circles with black outlines indicate hits in the METLIN database. The lighter or darker the
color of the circle relates directly to the significance of the p-value. 46 features were generated by pairwise
comparison between Ethiopian Yirgacheffe and Costa Rican Tarrazu showing the relationship between
retention time and m/z. Green color represents upregulated compounds which means abundant in Tarrazu
and red color circle represents down regulated compounds which means the compounds exist more in
Yirgacheffe. With a mouse click, EIC, Box and Whisker, and Spectrum appears under the cloud plot.
42
5. Discussion
5.1. Global metabolomics approach (XCMS Online) vs traditional analytical
techniques
Over the past decades, scientists have been relying on traditional analytical techniques
such as targeted analysis to identify compounds in samples. Although bioassay-guided
fractionation followed by targeted metabolite analysis and NMR structural elucidation
has been broadly used in identifying novel bioactive natural compounds, these techniques
are time-consuming research. For example, bioassay-guided fractionation involves
multiple steps such as repetitive preparative-scale fractionation and assessment of
biological activity until pure compound with the bioactivity of interest have been isolated
(Marston and Hostettmann, 2009). However, natural compound research has been
improved by modern separation and structural elucidation techniques such as HPLC-MS
and NMR (Seger et al., 2005; Steinmann and Ganzera, 2011). Probing the compound
composition, content, and structure in time and space often need accurate, sensitive,
selective, stable, fast, automated, high-throughput processes (Ju, 2013). These
characteristics are the eternal goals of analytical chemistry and are challenges even in
modern industry.
Since metabolomics is a powerful tool for profiling of largely unknown chemicals,
global metabolomics analysis has been used as a modern approach for identification of
bioactive compounds in natural products such as cooked rice, small berries and whole
grain cereals (Heuberger et al., 2010) (Badjakov et al., 2008) (Gani et al., 2012). In these
papers, chemical comparison and profiling were achieved globally and rapidly by
untargeted metabolomics-driven strategy (UMDS) trying to meet these requirements
43
described above. By introducing the combination of modern data processing online
platform (XCMS) and global compound data bank (METLIN), we overcome challenges
of traditional analytical chemistry which often are time consuming and labor intensive to
distinguish the bioactive compounds between sample treatments or groups. XCMS
provides PCA, non-metric multi-dimensional scaling, and extracted ion chromatogram
results (Figure 1.5 and 1.7) showing significant differences between sample groups based
on their chemical profiling. These results showing distinct characteristics between
cultivars matched with our expectation the three varieties of coffee are likely comprised
of different metabolites. Since they have distinct and typical characteristics in flavor
profile coming from different growing conditions such as geographic regions, elevation,
soil composition and temperature, we anticipated different chemical profiling results for
the three varieties.
Table 1.2. Growing condition of coffee cultivars (Ethiopian Yirgacheffe, Costa Rican Tarrazu, and
Hawaiian Kona)
Species Growing
Altitude
Harvest
period
Roasting
Ethiopian
Yirgacheffe
Coffea Arabica 1,700 – 2,200 M October –
December
Medium
Costa Rican
Tarrazu
Coffea Arabica 1,200 – 1,800 M December –
February
Medium
Hawaiian Kona Coffea Arabica 250 – 1000 M August –
January
Medium
44
For example, Ethiopian Yirgacheffe coffee is grown at elevations from 1,700 to 2,200
meters above sea level in southern Ethiopia where coffees grow slowly due to the altitude
(the town of Yirgacheffee in Ethiopia’s Sidamo area), allowing additional time for the
tree to deliver nutrients to the coffee and develop the best flavors. They were harvested
from October to December. Compared to this, Costa Rican Tarrazu coffees were grown
in lower regions which is called Tarrazu Coffee harvested between December and
February. Lastly, the Hawaiian Kona coffee is grown in the Kona district in Hawaii. The
farms are on the volcanic land with mild temperatures and located in lower altitude than
some of other Arabica coffees. Their harvesting time is from late August to late January.
Although shade and altitude are major factors known to have benefit on coffee quality,
there are contradictory results obtained from literatures on the chemical composition
related to these influence. In case of total lipid content, it was reported to decrease or not
to be influenced by altitude increases. Chlorogenic acid contents were shown inconsistent
results in several studies (Guyot et al., 1996). Additionally, the bigger factor than
elevation is reported as maturation rate of coffee fruit on the plant. In usual, low
temperature slow down the ripening process and leads to higher accumulation of aromatic
compounds (Joët et al., 2010). During the roasting process, not only aromas but also
chemical composition of coffee will change. At medium roast level, the coffee will create
a balance between acidity and body showing a darker brown color. According to Vignoli
(2011), increasing roasting will degrade 5-CQA and form high molecular weight
polymers, which is called melanoidins. However, they concluded that the antioxidant
activities under different conditions were equal and the activity depends more on
chemical composition than roasting degree (Vignoli et al., 2011). In terms of antibacterial
45
activities, Daglia (1994) found that 1) the lack of this activity in green coffee and 2)
degree of roasting affected the activities in roasted coffee extracts (Daglia et al., 1994a).
To quantify the contents of specific compounds, high resolution data from UPLC Bruker
Q-TOF was submitted to XCMS Online. Compare to low- resolution data which provides
visualized statistical results (PCA, and nonmetric-dimensional scaling), high resolution
data generates the relative intensity providing more detailed and numerical results of
compound content (table 3.1). From the XCMS Online result table of pairwise job
submission, maximum intensity was summarized in table 1.2 which shows significantly
different concentration between three cultivars. Additionally in table 3.1, 200 bioactive
compounds are compiled to compare the relative intensity of compounds which have
bioactivity related to skin care (phenolic compounds). While traditional target analysis
would take much more time and labor, this study has clearly demonstrated the advantages
of global non-targeted analysis over the traditional method for systematic analysis of the
bioactive compounds in SCG. Various compounds were detected in coffee samples in
XCMS result table, from simple linear and branched molecule structures to complex
glycosylated forms.
5.2. Phenolic acids and caffeine in SCG
From XCMS Online result, 22 of the phenolic acids were detected including cinnamic
acids, vanillic acids, and caffeoylquinic acids. Their concentration were relatively higher
than other groups of compounds. Specially, the concentration of cinnamic aicd was most
abundant followed by caffeine in the SCG. The majority of the phenolic acids identified
in this study were in free form. Phenolic acids including gallic acid, p-coumaric, caffeic,
46
and ferulic acids are rarely found in the free form, except in processed food that has
undergone freezing, sterilization, or fermentation (Raman et al., 1995). The bound forms
are glycosylated derivatives or esters of quinic acid. Especially caffeoylquinic acids
(CQAs) which is found in high concentrations in coffee can be formed by combining
caffeic acid and quinic acid (El Gharras, 2009). Their biological function have been
widely studied for the anti-oxidant and anti-pigamentation effects for potential skin
whitening agent. According to the Bravo, even though 5-CQA is the most abundant
caffeoylquinic acids in SCG, the total amount of CQA including 3-CQA, 4-CAQ, and 5-
CQA is less than the total Di-CQA including 3,4-diCQA, 3,5-diCQA, and 4,5-diCQA .
Similar aspect was detected in XCMS Online result table (table 3.1). Interestingly, in
Yrgacheffe, CQA was not detected although several types of Di-CQA such as 1,3-
diCQA, 1,4-diCQA, 1,5- diCQA, 3,4- diCQA, 3,5- diCQA, 4,5- diCQA, were identified.
The reason why diCQA contents are higher than monoCQAs in SCG is that monoCQAs
are extracted during coffee brewing, whereas the diCQAs need more time to be extracted
(Bravo et al., 2012). Other research on this phenomenon explained that diCQAs are
extracted slowly compared to monoCQAs from coffee (Blumberg et al., 2010; Ludwig et
al., 2012). In addition, a previous study revealed that caffeine and chlorogenic acids
contents, including diCQA and monoCQA in arabica SCG were slightly higher or similar
with reported concentrations in arabica coffee brew (Monente et al., 2015). Therefore,
spent coffee extracts which pass through the brewing process can be considered as a
better source of diCQAs and caffeine (compare to coffee brew) which are used in
cosmetic fields due to their antioxidant and antimicrobial effects (Almeida et al., 2006;
Devasagayam et al., 1996).
47
Our results are not consistent with the studies previously conducted by Belguidoum
(Belguidoum et al., 2014). In their studies, the concentration of cinnamic acid was much
less than chlorogenic acids in roasted coffee and green coffee samples whereas a higher
amount of E-cinnamic acid was detected by XCMS Online compared to other phenolic
acids such as coumaric acid, vanilic acid derivatives and (di)-caffeoylquinnic acids.
When the levels of the bioactive phenolic acids were compared between the cultivars,
the levels of E-cinnamic acid (the third highest concentration in SCG samples) in variety
Kona has almost 1.4 times higher than Tarrazu, suggesting the SCG from the variety
Hawaiian Kona is an possible source for antioxidant and anti-pigmentation products.
However, other anti-pigmentation compounds such as coumaric acid derivatives, vanillic
acid derivatives, and (di)-caffeoylquinic acids were produced as highest amount in
Tarazzu even though their concentrations were much lesser compared to E-cinnamic acid
(minimum 4.48 times) (see table 3.1).
5.3. Glycosylated form of flavonoids and their activity:
In this study, we identified a 139 glycosylated flavonoids with potential applications for
personal care product industry. In cosmetics, flavonoids are becoming popular due to
their antioxidant, anti-aging activities related to fibroblast proliferation (Kim et al., 1997)
or reduction of collagen breakdown (Schmid and ZÜLLI, 2002). Most of the flavonoids
in plants are found in glycosylated form so they can be dissolved in water phase and kept
in plant vacuoles (Manach et al., 2004). The sugar is often glucose or rhamnose, but
galactose, arabinose, xylose, glucuronic acid also can be attached (Manach et al., 2004).
As flavonoids do not exist in plants as aglycones except for catechins, consistent with our
48
findings that most of the compounds were in the glycosylated form (Table 3.1). Even
though glycosides are generally inactive form in humans because of low cellular uptake
due to inadequate size for enzyme binding sites and receptors (Schmid et al., 2003) most
cosmetic ingredients containing flavonoids are glycoside forms (Schmid et al., 2003).
Therefore, after topical application, they should pass through enzymatic hydrolysis at
their target location to become biologically active aglycones (the molecular from without
sugar residues) (Schmid et al., 2003). For example, one study revealed that, the
antioxidant capacity of soy bean flavonoids increased 33 % after enzymatic treatment
(Sansanelli et al., 2014). Other studies using mulberry leaves demonstrated that
improved bioactivity is due to enzyme reaction, hydrolysis of the glycoside polyphenols
rutin, soquercitrin, and astragalin into the aglycone polyphenols quercetin and
kaempferol producing a remarkable increse in anti-oxidant and anti-tyrosinase inhibition
activities by 219.5% and 230.9% respectively. Additionally, a significantly improved
skin permeability (513%) was observed by the reaction due to β-glucosidase enhancing
hydrophobicity (239.41%) and reduced molecular weight (from 387.3 to 291.4), which
are critical factors for absorption of bioactive ingredients (Do et al., 2009). Enzymatic
treatment is more important in cosmetic application rather than nutraceutical intake
because the latter undergoes hydrolysis of glycoside to aglycones by fermentation during
the manufacturing of certain foods or by intestinal bacterial metabolism (Manach et al.,
2004).
5.4. Flavonoids (flavonols, flavones, isoflavones, flavanones,
anthocyanidins, and flavanols) in SCG for cosmetic application
49
Through the global metabolomics analysis, more than 180 bioactive flavonoids, including
flavonols, flavones, isoflavones, flavanones, anthocyanidins, and flavanols were
identified. These compounds are strong anti-bacterial, anti-oxidant, skin protection, anti-
melanogenesis, and anti-inflammatory agents for skin care applications (Ramos et al.,
2006) (Choquenet et al., 2008) (Hämäläinen et al., 2007). According to Manach,
flavonols are the most abundant flavonoids in foods, and the main representatives are
quercetin and kaempferol (Manach et al., 2004). These compounds are also found in SCG
as well with relatively higher concentrations than other flavonoids.Compared to flavonols
such as quercetin and kaempferol, flavones which are mostly glycosylated apigenin,
luteolin, and chrysin are less abundant in fruit or vegetables (Manach et al., 2004). As
expected, these three flavones existed in SCG attached to several sugars such as glucose,
galactose, glucopyranose, coumaryl glucose, rhamnose and malonylglucose. Another
type of flavonoids, flavanones (eriodictyol, hesperetin, naringenin) are generally
glycosylated at position 7 (Tomás‐Barberán and Clifford, 2000). These flavanones can be
found in tomatoes, mint and but are only present at high concentration in citrus fruits.
XCMS Online identified naringenin and naringenin-7-O-glucodside in SCG. Although
isoflavones are known as major compounds which can be found in leguminous plants
(Tomás‐Barberán and Clifford, 2000), daidzein 4'-O-glucoside, daidzein 6''-O-acetate,
daidzein 7,4'-di-O-glucoside, genistin (genistein-7-glucoside) and genistein 7-O-
rutinoside were identified by XCMS Online in SCG as well. Isoflavones glycosides are
sensitive to heat. However, XCMS Online detected them in SCG that passed through the
roasting process. In case of flavanols such as catechin and epicatechin the most common
compounds in tea, exist in both monomer and polymer forms (El Gharras, 2009). The
50
ratio of monomers and polymers would change based on the fermentation (heating)
process producing more complex polyphenols known as theaflavins (dimers) and
thearubigins (polymers) (Cloughley, 1980). Therefore, black tea that is a more highly
processed product has more complex polyphenols compared to green tea due to oxidation
process leading to orange-brown color of black tea. Similarly, the theaflavins and its
derivatives were detected in SCG which passed through the roasting process (see table
3.1). However, the monomer form of flavonoids such as 8-C-Ascorbylepigallocatechin 3-
gallate still remained in SCG. In summary, some polyphenols such as quercetin are found
in all plant products including fruit, vegetables, cereals, leguminous plants, tea, wine, etc
(Pandey and Rizvi, 2009). However, other flavonoids can be found only in certain foods.
For example, flavanones are in citrus fruit at high concentration, while isoflavones are in
high concentration in soy protein. In SCG, flavonoids including flavanone, flavanonol,
anthocyanidins, and isoflavones were found by METLIN database matches based on their
accurate mass.
Our results reveal a existence of many kinds of flavonoids in SCG, which is consistent
with the studies previously conducted by Mussatto (Mussatto et al., 2011a). In their
studies, they also have reported high levels of total flavonoids and antioxidant properties
in SCG extracted using methanol. They agreed that SCG could be of great interest for
application in food and pharmaceutical fields due to antioxidant activities from phenolic
compounds in SCG. Another research claimed that rutin, and quercetin were not detected
in studied coffee samples (Belguidoum et al., 2014). The likely reason why quercetin was
51
not detected in their study is that the compound exist as glycosylated form not as free
form as shown in our study.
When the levels of the bioactive flavonoids were compared among the varieties, the
levels of 8-C-ascorbylepigallocatechin 3-gallate in variety Ethiopian Yirgacheffe was
almost 3.24 times higher than Costa Rican Tarrazu. Although this compound was
reported to have anti-HIV and anti-diabetic effects (Lee et al., 2003a) (Fukui and
Iwashita, 2013), dermatological activity has not been analyzed. Interestingly, the
concentration of dihydroxyflavone which has strong anti-cancer effects was highest in
Yirgacheffee suggesting the SCG from the variety Yirgacheffe is the most possible
candidate among three cultivars for skin protection products against UV-induced cancer.
Additionally, the antiaging agent, naringenin chrysin, which can provide UV protection,
was detected only in Yirgacheffe. Likewise, the levels of the antiaging compound,
glycosylated cyanidin, was 2.7 times higher in cultivar Yirgacheffe compared to Costa
Rican Tarrazu. The concentration of isorhamnetin, which has antioxidant and anti-
inflammatory activity, was highest in Kona about 20 times than found in the Tarrazu
cultivar. However, the concentration of other bioactive compounds such as quercetin,
apigenin, and keampferol varied between the three cultivars.
The levels of 6-methoxyeriodictyol, which has skin-whitening activity, was significant
only in Tarazzu, whereas the other anti-pigmentation agent such as 6-hydroxykaempferol
3,6-diglucoside 7-glucuronide concentration were not present. However, the
concentration gap between cultivars is quite small.
52
Anti-oxidants, including hesperetin derivatives, were primarily found in Hawaiian Kona
mostly. Of note is hesperetin 5,7-laurate which is used for emollient in skin conditioning
products.
Only three flavonols (quercetin, myricetin, and kaempferol) and two flavones (apigenin
and luteolin) are well studied with regard to the positive effects of dietary flavonoid
intake. Bioactive compounds which exist in SCG have never been systematically studied
using metabolomics approach for skin care application. Through this study, variety of
flavonoids were detected in SCG, even though further studies is need to measure absolute
quantity of these bioactive compounds for SCG to become a potential raw material in
cosmetic formulation. Basically, the idea was that SCG is a by-product of roasted coffee
which has variety of bioactive compounds including polyphenols (Nichols and Katiyar,
2010) and SCG may still retain those compounds. Several types of antioxidant
compounds has more uses than aroma including caffeic acid melanoidins and vitamins,
are generated through the roasting process (Yanagimoto et al., 2004). In many cases,
plant extracts contained natural antioxidants such as polyphenols, flavonoids, flavanols,
stilbens, and terpenes are used in commercial cosmetics. Some commercial products
contain pure natural compounds such as quercetin, and resveratrol in their formulation
(Kusumawati and Indrayanto, 2013). These compounds were detected in SCG as well.
The bioflavonoids, including quercetin and hesperidin, are known as anti-inflammatory
compounds inhibiting histamine release and mast cell degranulation. Quercetin (found in
high levels in SCG through XCMS) inhibits phospholipase A2 (Lanni and Becker, 1985)
and lipoxygenase enzymes (Kato et al., 1983). Other compounds such as tannins are also
53
major component in soothing compresses from their astringent, and cooling efficacy
(Graf, 2000). Several cosmetic products and soothing compresses includes bioflavonoids
such as quercetin, hesperidin, and kaempferol which are the best known therapeutically
useful bioflavonoids (Kenner and Requena, 2001). In addition, flavonoids such as
apigenin and luteolin are major components in chamomile and documented to have anti-
inflammatory and soothing effects in creams. This study observed that SCG contains
several main active compounds with antioxidant or anti-inflammatory activities used in
real cosmetic formulations. Since the early 2000s, a lot of studies have revealed that
flavonoids have photo protective effect on skin due to their strong anti-oxidant activity.
From XCMS analysis, we found that SCG contains various bioactive compounds such as
anti-oxidant, anti-inflammatory anti-tumor, and anti-tyrosianse. These results support that
the idea that SCG can be used as a source of major active materials in cosmetic industry.
In addition to identifying health-promoting compounds, our study has importance in
terms of using novel untargeted method to identify these compounds in SCG.
As described in table 3.1, several listed bioactive compounds have being known as agents
with strong antioxidants, antimicrobial, antitumor and anti-tyrosinase activities. In
addition, relative intensity of each compound was summarized through XCMS pairwise
analysis.
Even though the compounds were qualified and the relative concentrations of these
compounds were analyzed between three cultivars through computational technology,
synergistic or antagonist effects between compounds that may affect their human health-
promoting efficacy was not determined in this study. Therefore, the results can’t show
which cultivar is better among the three varieties in antioxidant, anti-inflammatory, anti-
54
microbial or anti-tyroisnase activity. Several kinds of bioassays that are listed above will
address this limitation and allow the selection of one variety with the highest efficacy in
vitro and in vivo. For example, synergy effect between p-coumaric acid and ferulic acid
has been reported by Maillard and Berset (1995) in antioxidant activity assays. Also,
caffeic acids, which are abundant in SCG, can be affected by the presence of certain
compound such as ascorbic acid. Compared to synergy effects, antagonist effects in
antioxidant activity have been reported between ellagic acid and catechin, which were
also identified in SCG through XCMS. Many studies are focusing on natural bioactive
compounds from residual sources for cosmetics and these effects have been reported.
Synergistic effects of phenols from grape seeds and pomace polyphenols have been
reported. Even though flavonoids and phenolic compounds interact with each other
(Meyer, Heinonen et al., 1998), the choice of the right combination of compounds for
their stability and synergistic effect in cosmetics should be achieved to meet customer’s
demands. That is why crude extracts of the three SCG cultivars should be tested through
bioassay besides purified compounds even though the relative intensity was qualified for
each compounds through metabolomics approach.
6. Conclusion
This project utilizes a modern metabolomics approach to identify the bioactive
compounds in SCG and explore the new applications for skin-care application. The
findings from this project will help turn the low value renewable materials into high-
value products. It will not only benefit consumers (organic based, safe, effective),
producers (low cost and abundant materials), and the environment (green, sustainable),
55
but also foster the development of rural economies. According to Industry Trends Report
on Organic Personal Care Maket size, organic based personal care market size will reach
USD 25.11 billion by 2025. Therefore, the vision of this project completely aligns to
global trends and provide the organic-based products without concerns of toxicity. In this
study, methanol was used as a solvent to extract SCG. However, for practical large-scale
process, supercritical fluid extraction (SFE) method should be used to reduce the
extraction cost, toxic solvent use, and produce large quantity of the bioactive extracts. In
this SCG project, there are more than 200 bioactive compounds were identified including
phenolic acids, flavonoids, coffee alkaloids, stilbenoids, and anthraquinones. The toxicity
of the bioactive compounds identified in SCG should be further evaluated.
56
Chapter 2. Isolation of the antioxidant and antimicrobial compounds
with bioassay-guided fractionation
Abstract: Today, the importance of using organic-based anti-oxidant and anti-microbial
compounds for cosmetic formulation is emerging. Oxidative stress is closely linked to
skin aging processes. Antioxidants are valuable antiaging component by helping the skin
fight against free radicals produced during oxidative stress. In addition, multiple
antimicrobial ingredients are included to protect the products against the growth of
micro-organisms such as bacteria and fungi. By formulating anti-microbial compounds in
cosmetics, the products can be used to treat skin conditions associated with microbial
infection, such as acne. It will also help prolong the shelf-life of the products and keep
the products from causing skin infections, corruption or discoloration of the product itself
during the usage. The present research examines SCG as the bioactive ingredient having
anti-oxidant and anti-microbial activity for cosmetics formulation. To create value-added
product from SCG, the objectives of this project are to 1) isolate and identify the
bioactive compounds in the SCG of three cultivars (Ethiopian Yirgacheffe, Costa Rican
Tarazzu, and Hawaiian Kona) using bioassay guided-purification, and 2) explore their
application in cosmetics. The compounds in SCG were extracted and tested by FRAP
antioxidant assay, and zone of inhibition anti-microbial assay. The bioactive compounds
were then isolated by bioassay-guided flash chromatography followed by HPLC
purification for the structural analysis by high resolution mass spectrometer (HRMS). In
anti-oxidant assay, the SCG extracts from Hawaiian Kona showed the highest activity
among the three cultivars. Antimicrobial activity from the SCG extracts from Hawaiian
Kona and Ethiopian Yirgacheffe were much higher than Costa Rican Tarrazu. The
57
findings from bioassay and high-resolution analysis showed that SCG is an excellent
potential raw material for cosmetic application (e.g., anti-oxidant, anti-microbial, skin-
whitening, and anti-aging). High resolution analysis using systematic approach was
conducted to identify the compound in three fractions which were purified through
HPLC. The SCG from Ethiopia coffee (Yirgacheffe) was fractionated by flash
chromatography and HPLC for further purification and structural analysis. Fraction
number 4 from flash chromatography showed the highest anti-oxidant activity and was
analyzed further by HPLC. In addition, fractions 7 and 12 from flash chromatography
showed the highest anti-microbial activity. After injecting fraction number 4 from Flash
chromatography, 27 purified fractions were collected through HPLC and purified fraction
14 was selected for high resolution analysis. The results suggested that cafesterol, coffee
alkaloid is the bioactive compound responsible for the observed strong anti-oxidant
activities.
1. Introduction
Recent studies have suggested that SCG could be an excellent source of high value
phytochemicals for cosmetics and skin-care application. Some of examples include
chlorogenic acid (antioxidant), caffeic acid (antioxidant), isochlorogenic acid
(antioxidant, anti-inflammatory), protocatechuic acid (antioxidant, anti-fungal), 4,5-
dicaffeoylquinic acid (pigmentation inhibitor) and quercetin (anti-fungal, anti-oxidant,
skin protection, anti-melanogenesis) (Figure 2.1)
According to Mussatto, the content of the total phenolic compound in methanol extract
of SCG is similar with the contents in raspberry, blackberry (Wang and Lin, 2000), and
58
almond shells. Also, they revealed that SCG includes chlorogenic acid, protocatechuic acid
and flavonoids leading to antioxidant activity higher than the other fruits such as apple,
peach, pear, and kiwi (Moure et al., 2007). Therefore, the coffee is laden with bioactive
components including polyphenols and flavonoids with their antioxidant activities making
them an attractive candidate for studies on naturally occurring “personal care products”
(Kilmartin and Hsu, 2003). In addition to antioxidant activity, the use of SCG is emerging
as the trends in the dermatological field. For example, chlorogenic derivatives 4,5-
dicaffeoylquinic acid have been known to be a pigmentation inhibitor in both melanocytes
and the zebrafish vertebrate model (Russo et al., 2006). Specially, the inhibitory activities
against melanogenesis of this compound was superior to arbutin, and kojic acid, which
were well-known inhibitors against melanogenesis (Takagi et al., 2014). According to
Tabassum, the compound has no toxicity so that it can be considered as an attractive
candidate for further development in pharmaceutical or cosmetic depigmentation
(Tabassum et al., 2016). Besides, SCG showed effectiveness in UV-induced photo-aging
even though specific compounds have not been studied for this reaction (Choi et al., 2015).
However, the study considered caffeine and chlorogenic acids as health-promoting factors
because they have protection efficacy against UV-induced skin damage (Huang et al.,
1997) (Kitagawa et al., 2011) and as most abundant compounds in SCG. Even though
specific flavonoid contents except quercetin in SCG have not been reported yet, the total
flavonoid concentration was estimated by colorimetric assay. Also, the effect of roasted
coffee has been extensively studied and widely marketed with its antibacterial activity. The
activity which is not present in green coffee was only shown in roasted coffee beans
depends on the roasting degree (Daglia et al., 1994a). During the roasting process, various
59
types and concentrations of compounds were obtained having antibacterial activity against
a wide range of bacteria (Daglia et al., 1994b) and they have been considered as
nontraditional food antimicrobial agents (SHIBASAKI, 1982). Furthermore, a prior study
reported that the lipid fraction of SCG including fatty acids such as Palmitic and Linoleic
extracted with supercritical carbon dioxide showed improved skin lipids and hydration
qualities (Ribeiro et al., 2013).
These antioxidant and antibacterial properties are extremely important in the cosmetic
industry especially in natural cosmetics. The most important factor in the initiation of
several skin disorders from dryness, wrinkling, localized pigment abnormalities
including, hypopigmentation and hyperpigmentation to skin cancer is the chronic
exposure of the skin to solar UV radiation (Lee et al., 2008) (Nichols and Katiyar,
2010). The exposure to UVA and UVB induce the generation of singlet oxygen and
hydroxyl-free radicals. This oxidative stress can cause damage to cellular molecules,
such as proteins, lipids and DNA (Lee et al., 2008) leading to premature aging of the
skin (Nichols and Katiyar, 2010), called photo-aging, and multiple effects on the
immune system (Ullrich, 1995). The increasing interest in flavonoids is due to their broad
efficacy in pharmacological and dermatological field. The mechanism of action of
flavonoids as antioxidants includes their ability to scavenge free radicals directly, activate
antioxidant enzymes, chelate metals, reduce alpha-tocopheryl radicals, inhibit oxidases
(Procházková et al., 2011). Free radical scavenging ability is highly related to chemical
structure of flavonoids such as high number of hydroxyl groups (Burda and Oleszek,
2001). In terms of activating antioxidant enzymes, flavonoids can induce detoxifying
enzymes which regulate protective gene expression (Lee-Hilz et al., 2006). For examples,
60
flavonoids quercetin, myricetin containing 3-position OH induce detoxifying enzymes
against toxic chemical reaction (Ferrali et al., 1997). Quercetin also interferes with the
development of free radicals by chelating active form of metal (Ferrali et al., 1997). In
addition, flavonoids including kaempferol, myricetin, quercetin, and catechin can reduce
prooxidant such as alpha-tocopherol radicals by interacting with them even though the
activity was varied each other (Zhu et al., 2000). Lastly, flavonoids (quercetin, and
luteolin) inhibit the enzymes related to superoxide so that reducing oxidative injury in
cells (Hanasaki et al., 1994).
Related to antimicrobial effect, antimicrobial ingredients should be formulated to
protect the personal care products against the growth of micro-organisms such as
bacteria, and fungi. By adding anti-microbial compounds in cosmetic formulation, the
products can be safe from causing skin infections, or discoloration of the product itself
during the usage. Furthermore, antimicrobial agents in the products make consumers can
be protected from bacterial disease. Anti-microbial agents against gram positive bacteria
such as Propionibacterium acnes, Staphylococcus aureus, and Staphylococcus
epidermidis helps to prevent skin illnesses, from minor skin infections including pimples,
impetigo, boils, cellulitis, folliculitis, carbuncles, and abscesses, to life-threatening
diseases such as toxic shock syndrome, bacteremia, and sepsis (Hanasaki et al., 1994) .
These bacterial strains are the major causes of acne in human skin (Nishijima et al., 2000)
and acne is the most common skin issue in cosmetics, affecting up to 50 million
Americans annually (Bickers et al., 2006). Numerous research groups have sought to
elucidate the antibacterial mechanisms of selected polyphenols. Inhibition of nucleic acid
synthesis is one of the modes of actions because B ring of the flavonoids can form
61
interaction with nucleotide bases in microorganism. Flavonoids such as myricetin and
epigallocatechin, can affect RNA synthesis of S. aureus (Mori et al., 1987). Enzyme
binding of flavonoids also contributes to antimicrobial activity by inhibiting DNA gyrase
in bacteria (e.g. quercetin and apigenin) (Ohemeng et al., 1993). Glycosylated rutin are
also inhibitor of nucleic acid synthesis in E.coli (Bernard et al., 1997). Another way to
prevent bacterial infection is by reducing membrane fluidity of bacterial cells.
Epigallocatechin gallate shows strong antibacterial activity inducing aggregation of cell
membrane (Ikigai et al., 1993). In addition, flavonoids are reported to inhibit bacterial
respiratory electron transport chain, called energy metabolism (Haraguchi et al., 1998).
Even though non-flavonoids also shows antimicrobial activity, it is weaker than
flavonoids. However, some phenolic acids including caffeic acid, ferulic acid and gallic
acid are notified that they have antibacterial activity against both Gram-positive such as
S. aureus and L. monocytogenes and Gram-negative such as E. coli and Pseudomonas
aeruginosa showing stronger activity compared to general antibiotics including
gentamicin and streptomycin (Saavedra et al., 2010). It is critical to evaluate the
antimicrobial activities of SCG against gram positive stains to explore its potential for
cosmetics product formulation for acne treatment or product protection.
These days, a lot of studies are being conducted related to antioxidant derived from
natural resources. Therefore, the bioactive compounds such as polyphenolic acids,
flavonoids, vitamins having antioxidant activity are identified (Núñez-Sellés, 2005). To
evaluate their ability to inhibit oxidation, several assays are developed. Based on research
types, adequate tests should be selected between lipid peroxidation and electron/radical
scavenging associated with oxidation inhibition. Since reactive oxygen species (ROS) are
62
hard to measure due to their extremely short half-lives, other oxidative stress products
can be measured. As a result of oxidative stress, thiobarbituric acid reactive substances
(TBARS) are formed. Using thiobarbituric acid, TBARS can be detected and this test is
called TBA assay (Hodges et al., 1999). Another antioxidant assay associated lipid
peroxidation is beta-carotene bleaching assay; beta-carotene can react with peroxyl
radical which is formed from lipids such as linoleic acid in the presence of ROS.
Antioxidant activity is measured by reduction amount of beta-carotene (Tsuchihashi et
al., 1995). Conjugated diene assay is monitoring conjugated diene which occurs by the
action related to ROS and oxygen (Moore and Roberts, 1998). This method has weakness
because natural compounds have similar absorbance wavelength with conjugated diene
(Moon and Shibamoto, 2009). Antioxidant assays associated with electron and radical
scavenging include DPPH (2,2-diphenyl-1-picrylhydrazyl), ABTS (2,2'-azino-bis(3-
ethylbenzothiazoline-6-sulphonic acid)), and FRAP (ferric reducing antioxidant power)
assay (MacDonald‐Wicks et al., 2006). DPPH•, one of few stable and commercially
available radicals accepts hydrogen from an antioxidant based on the theory that
antioxidant act as a hydrogen donor. Antioxidant activity of many natural compounds are
evaluated by ABTS assay as well due to accurate and easy method especially for fruit and
vegetable extracts (Sanchez-Moreno, 2002). By measuring ABTS radical cation which
absorbs blue-green wavelength, antioxidant ability can be determined (Re et al., 1999). In
FRAP assay which gives fast, reproducible results, antioxidant activity can be observed
by looking transformation of Fe3+-TPTZ complex to Fe2+ form by an antioxidant (Moon
and Shibamoto, 2009). The FRAP has been recognized as one of the most accurate and
reliable methods to evaluate the antioxidant activities of the natural product. Multiple
63
natural antioxidant studies using FRAP assay were showed consistent result with DPPH
and ABTS assays (Katalinic et al., 2006).
In terms of antimicrobial assays, there are several methods to view the activities.
Very common assays include disk diffusion and agar dilution (Balouiri et al., 2016).
Since microbial resistance is getting more important in drug discovery, dermatology, and
cosmetics, in vitro antimicrobial test is prevalently used. Both crude extracts and pure
compounds could be evaluated for their application in personal care industry (Runyoro et
al., 2006). Among several methods, agar disk diffusion testing is officially used in many
clinics and laboratories (Balouiri et al., 2016). Once antimicrobial agent diffuses into
plates and inhibits growth of microorganism, the zone of inhibition are measured (Ahmad
and Beg, 2001). Even though the agar disk diffusion method cannot quantify the
antimicrobial agent in agar plates, this is the most common method due to its low cost,
simplicity, availability of vast number of testing samples in one plate and the ease of
result interpretation (Korgenski and Daly, 1998). Other techniques including flow
cytofluorometric method and time-kill test have pros and cons. For example, the former
technique is useful to give reproducible results rapidly (Ramani and Chaturvedi, 2000)
with information on cell damage of the tested microorganism but unlikely used due to
required equipment.
Bioassay-guided purification has been often used as an approach to isolate the
bioactive compounds in the natural extracts such as Chinese green tea (Si et al., 2006),
Brazilian propolis (Castro et al., 2009), and lemon balm (Awad et al., 2009). With this
approach, the compounds were first separated by their hydrophobicity and polarity based
on their interaction with the functional group on the surface of the solid-phase on the
64
liquid chromatography column. The eluted fractions were collected, concentrated and
tested for their bioactivity until the lead compounds are isolated (Evans, 2009). Since
plant extracts are being considered for their potential use as natural preservatives in
dietary supplements or personal care products, it is needed to study biological properties
of the compounds in extracts (Seeram et al., 2005). Unfortunately, purifying pure
compound from crude extract for compound structure studies such as Nuclear Magnetic
Resonance (NMR) requires a lot of time and intensive labor using chromatographic
technique. As a solution, high-resolution MS metabolomics and natural product libraries
was introduced for fast identification of the compounds in the bioactive fractions without
labor intensive purification required in NMR structural analysis. The “World’s most cited
metabolomics software called XCMS Online” cited in literature over 1000 times (Ubhi et
al, 2014), which includes adequate algorithm in peak detection, retention time correction,
alignment, annotation, and statistical analyses was introduced for compound
identification. Only basic steps are needed for metabolomics analysis after the
fractionation of the extract using liquid chromatography coupled with mass spectrometer
(LC-MS). Once researchers obtain ion chromatogram of fractionated sample from MS,
users can submit it to web-based metabolomics platform to accomplish chemical
profiling and acquire structural information. According to its peak detection algorithm, it
covers both low-resolution, and high-resolution using matchedFilter method and
centWave method respectively (Smith, 2013). After peak detection, XCMS Online uses
nonlinear methods to compensate for retention time drifts between samples (Tautenhahn
et al., 2012). All total ion chromatogram (TIC)s should be in alignment to compare
compound quantities between fractions after retention time correction (Christin et al.,
65
2008). If there is significant difference in the peak between submitted groups, one group
can be distinguished from another group. XCMS Online uses minimum fraction method
to process this step. For example, minfrac value 0.5 (50%) means, two groups can be
considered as two distinct groups when they show at least 50% difference. In annotation
step, information about at least two ions is need to calculate the molecular mass which is
necessary to be matched with compound library (Kuhl et al., 2011) called METLIN. The
fingerprint of compounds are matched with the database which has more than 961,820
molecules providing information for each compound such as name, systematic name,
structure, elemental formula, and mass (https://metlin.scripps.edu/).
Through bioassay (e.g. antioxidant and antimicrobial assay) -guided purification
method followed by global metabolomics approach, SCG can be assessed as potential
raw material in cosmetic field. Especially, high resolution MS metabolomics was
introduced to identify pure compound in fraction while reducing time and intensive labor.
66
1. Chlorogenic acid 2. Caffeic acid 3. Isochlorogenic acid
4. Protocatechuic acid 5. 4,5-dicaffeoylquinic acid 6. Quercetin
Figure 2.1. Common bioactive compounds in SCG
1 3-Caffeoylquinic (Chlorogenic) acid Antioxidant1
2 3,4-Dihydroxycinnamic (Caffeic acid) Antioxidant1
3 3,4-Di-O-caffeoylquinic acid (Isochlorogenic acid) Antioxidant1, Anti-inflammatory8
4 Dihydroxybenzoic acid (protocatechuic acid) Antioxidant2, Anti-fungal4
5 4,5-dicaffeoylquinic acid Pigmentation inhibitor3
6 Quercetin Anti-fungal4, Anti-oxidant5, Skin protection6, Anti-melanogenesis7
1: (Sroka and Cisowski, 2003)
2: (Li et al., 2011)
3: (Russo et al., 2006)
4: (Aziz et al., 1998)
5: (Boots et al., 2008)
6: (Choquenet et al., 2008)
7: (Arung et al., 2011)
8: (Park et al., 2009)
67
2. Objectives
In chapter 2, the bioassay- guided fractionation approach was used including anti-oxidant
and anti-microbial assay to purify the bioactive compounds from SCG extraction.
The specific objectives of this study are described as following
1. Perform bioassay guided-purification to explore new skin-care application.
2. Isolate the bioactive compounds in SCG by flash liquid chromatography fractionation
followed by HPLC.
3. Perform FRAP anti-oxidant assay and zone of inhibition anti-microbial assay to
confirm and evaluate the activity.
4. Compare the difference in antioxidant and antimicrobial activities between selected
three cultivars.
5. Identify potential compounds in active fractions using high-resolution MS
metabolomics.
3. Materials and Methods
The health-promoting properties of coffee are often attributed to its rich phytochemistry,
including caffeine, chlorogenic acid, caffeic acid, hydroxyhydroquinone (HHQ). Since
coffee contains multiple bioactive compounds such as anti-oxidant, anti-inflammatory,
anti-microbial, and anti-tumor, we need to isolate and identify each compound from
whole extract by checking each fractions from chromatography. Liquid chromatography
including Flash chromatography and HPLC enable to generate fractions from extract
having different purpose. Flash chromatography helps to fractionate the extract into vials
which contains around 100 compounds. Further steps are need to purify the one hundred
68
compounds using HPLC. In this stage, we can narrow down into only one or two
compounds in one vial.
For this study, the SCG extract from three cultivars, including Ethiopian Yirgacheffe,
Costa Rican Tarrazu, Hawaiian Kona) were evaluated for their antioxidant and
antimicrobial activities by FRAP antioxidant assay and zone of inhibition assay
respectively.
As described in the Figure 2.3, the bioactive compounds in SCG were purified by HPLC
after fractionated by flash liquid chromatography. The purified compounds were
characterized by ultra-pressure liquid chromatogram (UPLC) coupled with high-
resolution mass spectrometry (HRMS). The ion chromatograms were submitted to
XCMS, and the identified signals were annotated by comparing with the METLIN
metabolomics library.
69
Figure 2.2. Flow chart of the SCG project using bioassay guided purification
UPLC-HRMS analysis
Identify non-volatile chemical composition in
active fractions
Metabolomics approach (XCMS Online)
Perform metabolite profiling in active fractions
Antioxidant, antimicrobial assay
Perform assays of each fraction to identify the
active one
Extraction
Extract non-volatile compounds from SCG
using methanol
Flash chromatography, HPLC
Collect fractions from liquid chromatography
SCG extracts
70
3.1. Explore new application (antioxidant and antimicrobial activities)
through bioassay-guided purification
Anti-oxidant analysis
3.1.1. Sample preparation
SCG from three Arabica coffee cultivars including Ethiopian Yirgacheffe, Costa Rican
Tarrazu, and Hawaiian Kona were selected for this study. Roasted and grinded coffee were
purchased from the local Lakota Coffee (Columbia, Missouri). After brewing for 5
minutes, SCG were collected. Twenty-five grams of wet SCG (78% water, about 5.5 g DW
equivalent) were homogenized with blender and the compounds were extracted with 200
ml methanol (HPLC-grade, purchased from Fisher Scientific, Pittsburg, PA, USA). The
extract was then sonicated for 60 mins and filtered through Whatman phase separators
(silicone treated filter paper, diameter 125mm). The methanol was evaporated by rotary
evaporator. The phenolic compounds were extracted by a sequential water: chloroform
liquid/liquid extraction with 200ml of chloroform (HPLC-grade, purchased from Fisher
Scientific, Pittsburg, PA, USA). Following the liquid/liquid extraction, the chloroform
fraction was collected and the chloroform was removed by a rotary evaporator (BUCHI,
Rotavapor R 110). The powder form of final extract was loaded on the flash
chromatography column packed with 34g of C18 resin (Specific surface area: 460-520 m2
/g, average particle size: 47-60 µm, average pore diameter: 60-87 Å)
3.1.2. Fractionation and bioassay-guided purification
The compounds in the extract was fractionated by a Jones Flash Master II Flash
Chromatography connected with a fraction collector. The mobile phases consist of
71
methanol and deionized water. Compounds were separated by a gradient described as
following: 25% in the first 40 min, followed by a linear gradient from 25% to 50% solvent
B (Methanol) for 50 min, then followed by a linear gradient from 50% to 75% B for 30
min, and then remain 75% for 60 min before achieve the 100% conditions. The flow rate
was maintained at 2.5 mL/min.
For further purification, the bioactive fractions collected from flash chromatography were
injected into a Shimadzu SCL-10Avp HPLC system with a SPD-M10Avp photodiode
array detector coupled with a Shimadzu FRC-10A fraction collector. The compounds were
separated by a Phenomenex (Torrance, CA) Columbus C8 (250mm x 4.6 mm; 5 µm
particle size) reverse-phase column. The mobile phases consist of ACN and deionized
water by following gradient: 20% to 80% for 16 min with a linear gradient, then followed
by a linear gradient from 80% to 98% solvent (ACN) for 2 min, and then remain 98% for
1 min before go back to 20% at 20 min. The gradient remain 20% for 9 min. The total
running time was 30 min with 0.5 mL/min of the flow rate. The signals were monitored at
wavelength of 220 and 254 nm. The injection volume was 50 µL. Each fraction was
collected, and the fraction with the same retention time window were combined after
injection of the extracts 30 times.
72
Figure 2.3. Flow chart of the bioassay-guided purification for identifying antioxidant activity and
antimicrobial assay. Same approach was applied for antimicrobial test.
73
3.2. Determination of anti-oxidant activities
Each fraction was tested for the anti-oxidant activities using FRAP™ Detection Kit for
anti-oxidant activity. All fractions were prepared with three dilution factors including
1:10, 1:100, and 1:1000 and raw extracts were prepared with two dilution factors
including 1:25 and 1:50 to be within the range of the calibration curve (Figure 4). 20ul of
samples or STDs into wells were pipetted in 96 well. 20ul of buffer was used as Zero
STD. After all samples was loaded, 75ul of Color Solution to each well was added using
repeater pipet. The absorbance value at optical density 560 nm was read after incubating
at room temperature for 30 minutes. FRAP assay uses antioxidants as reductants in
redox-linked colorimetric method. Ferric (Ⅲ) [colorless] to Ferrous (Ⅱ) [blue] can be
monitored by measuring absorbance at 560 nm. The absorption readings are related to the
reducing power of the electron-donating antioxidants present in the test compound.
Hence the FRAP assay can rank the reducing power and the antioxidant potential of a
wide range of test compounds. Calibration curve represents FeCl2 concentration (µM) in
X-axis and optical density (nm) in Y-axis. Seven standards solutions including 0, 31.25,
62.5, 125, 250, 500, and 1000 (µM) were used for the development of the calibration
equation.
3.3. Anti-bacterial analysis
3.3.1. Sample preparation
Twenty-five grams of wet SCG (78% water, about 5.5 g DW equivalent) were
homogenized with blender and extracted with 200 ml methanol. The extract was then
sonicated for 60 mins and filtered using phase separators (silicone treated filter paper,
74
diameter 125mm). Methanol was evaporated by nitrogen evaporator and samples were
redissolved in 0.5ml of DMSO (550 mg/ml). Raw extracts including Ethiopian
Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona using same method as anti-
oxidant assay for bioassay-guided purification were used for zone of inhibition assay.
Fractionated 43 samples through Flash chromatography dissolved in 100 µl of DMSO
after evaporating methanol were also tested. With agar diffusion method, approximately
one million cells from a single strain of Staphylococcus aureus were spread over an agar
plate, then incubated in the presence of 10 µl of the extracts or fractions. If the bacterial
strain is susceptible to the antimicrobial agent, then a zone of inhibition appears on the
agar plate. If it is resistant to the antimicrobial agent, then no zone is evident. DMSO was
used as a control. The size of the zones were measured to quantify and evaluate the
antibacterial activities.
3.4. High resolution (HR) metabolomics analysis
The compounds in the purified bioactive fractions were analyzed by UPLC-HRMS and
data was processed by XCMS Online. The signal was annotated by comparing the
METLIN metabolomics library. The purified fraction was analyzed by a Bruker maXis
impact quadrupole-time-of-flight mass spectrometer coupled to a Waters ACQUITY
UPLC system. Separation was achieved on a Waters C18 column (2.1x 100 mm, BEH
C18 column with 1.7-um particles) using a linear gradient of 95%:5% to 30%:70% eluent
A:B (A: 0.1% formic acid, B: acetonitrile) over 30 min. Between 30-33min, the a linear
gradient was increased from 70% to 95% B, and maintain at 95%B for 3 min. The
percentage of B was maintained at 5% from 36-40 min. The flow rate was 0.56 mL/min
75
and the column temperature was 60 C. Mass spectrometry was performed in the negative
(or positive) electrospray ionization mode with the nebulization gas pressure at 43.5 psi,
dry gas of 12 l/min, dry temperature of 250C and a capillary voltage of 4000V. Mass
spectral data were collected from 100 to 1500 m/z and were auto-calibrated using sodium
formate after data acquisition. The generated spectra were submitted to XCMS analysis.
The pairwise analysis between the control (ACN) and purified fraction was performed for
the background subtraction. Parameters in high resolution analysis were: maximal
tolerance of 10 ppm between successive measurements, peak width of 5-20 s, S/N
threshold of 6 and obiwarp algorithm for retention time correction and alignment (width
of overlapping m/z slices 0.015, minimum fraction of samples 0.5 for group validation).
To visualize differences between two samples in pairwise job, 0.001 of p-value threshold
and 1.5 of fold change was selected to be considered highly significant. The bioactive
compounds were identified by comparing the accurate mass of the molecular ions with
the candidates in the METLIN database.
In metabolomics, two group comparison is very common way to assess difference between
treated versus control. In the study, the control indicates ACN (purchased from Fisher
Scientific, Pittsburg, PA, USA), solvent which is used in UPLC Bruker Q-TOF (High
resolution MS) system. In XCMS Online pairwise test, comparison between ACN and
purified F13 was analyzed. This process was conducted in both positive and negative
modes. Once UPLC/Bruker QTOF (HRMS) was marked, it automatically activates
centWave algorithm, which is new published algorithm designed for centroid and high
resolution data for both two-group comparison study. Parameters in high resolution
analysis were: maximal tolerance of 10 ppm between successive measurements, peak width
76
of 5-20 s, S/N threshold of 6 and obiwarp algorithm for retention time correction and
alignment (width of overlapping m/z slices 0.015, minimum fraction of samples 0.5 for
group validation). To visualize differences between two samples in pairwise job, 0.001 of
p-value threshold and 1.5 of fold change was selected to be considered highly significant.
3.5. Statistical analysis
All data including antioxidant and antimicrobial assay are presented as mean values
(±S.E). Values within columns followed by different superscript letters (A and B) are
statistically different (P < 0.05). One-way analysis of variance (ANOVA) followed by
pairwise comparisons of means was performed to assess the difference in three cultivars.
Pairwise comparison of means was analyzed by Tukey’s range test. R: The R Project for
Statistical Computing 3.3.1 was used for both statistical analysis. Samples were analyzed
with 2 dilution factors and sextuplicate (n = 6) for antioxidant assay. In zone of inhibition
assay, the diameter was measure three times for each samples with triplicate. However,
the maximum and minimum value was eliminated for statistical analysis. Therefore,
Septuplicate (n = 7) were used for each cultivars in ANOVA.
4. Results
4.1 Antioxidant study
The FRAP analysis was successfully utilized to evaluate the antioxidant activities of raw
SCG extracts and the isolated fractions, suggesting the correlation coefficient of 0.9997
with the calibration range from 0-1000 µM of FeCl2 (Figure 2.4). The results of the study
suggested that the SCG from Hawaiian Kona had highest antioxidant, about 23% higher
77
as compared to Yirgacheffe (Figure 2.5 and Table 2.1). The FRAP (Ferric Reducing
Antioxidant Power) assay measures the ability of an antioxidant to reduce ferric (III) to
ferrous (II) in a redox-linked colorimetric reaction (Grover and Patni, 2013). Antioxidant
induces the reaction producing ferrous (II) which has blue color from ferric (III). The
reducing power of a compound serves as a significant indicator of its potential
antioxidant activity. In investigated antioxidant results, the reducing power was higher in
higher concentration of the extracts (Table 2.1). Interestingly, SCG extracts from Costa
Rican Tarrazu and Hawaiian Kona have significantly higher antioxidant activity (p <
0.05) compared to extracts from Ethiopian Yirgacheffe even though there was no
significant difference between Tarrazu and Kona in 1:25 concentration. Same aspect was
assessed in lower concentration (1:50). The antioxidant result value in absorbance was
converted into value of FeCl2 concentration using equation y = 0.002x - 0.0054 generated
by standard curve (Figure 2.4). Figure 2.5 shows FeCl2 concentration among three
cultivars and Hawaiian Kona has highest reducing capacity.
Among the 43 fractions of the Ethiopian Yirgacheffe SCG extracts collected from the
flash chromatography fractionation process, fraction number 4 showed the highest
antioxidant activity (Figure 2.6). The bioactive compounds in the fraction 4 was further
purified by the HPLC and 26 sub-fractions were obtained. After all fractions were tested
through FRAP assay, sub-fraction 14 collected from HPLC separation showed the highest
antioxidant activity among 26 fractions (Figure 2.7).
78
Figure 2.4. Standard curve of FRAP assay. Each dot in X axis represents around 0, 31.25, 62.5, 125, 250,
500, 1000 FeCl2 concentration of standards (STD; n = 7).
y = 0.002x - 0.0054R² = 0.9997
-0.5
0
0.5
1
1.5
2
2.5
0 200 400 600 800 1000 1200
OD
(nm
)
FeCl2 concentration (µM)
FRAP assay
Series1
Linear (Series1)
79
Table 2.1. Antioxidant activity of SCG methanol extracts of three cultivars including Ethiopian
Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona
FRAP (abs. at 560 nm)
Dilution factors
Cultivar 1:25 1:50
Ethiopian Yirgacheffe 0.469±0.023a 0.310±0.008a
Costarican Tarrazu 0.553±0.008b 0.350±0.008b
Hawaiian Kona 0.551±0.016b 0.353±0.008b
Means in the same column followed by different letters (a-b) represent statistical significance (P <0.05).
Samples were analyzed with 2 dilution factors (1:25 and 1:50) and sextuplicate (n = 6). One-way analysis
of variance (ANOVA) followed by pairwise comparisons of means using Tukey’s range test was performed
to assess the difference in three cultivars (significance level P < 0.05). R: The R Project for Statistical
Computing 3.3.1 was used for statistical computing.
80
Figure 2.5. Reducing power of methanol extract (10mg/mL) from three SCG cultivars including Ethiopian
Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona were compared. Concentration of FeCl2 was
calculated from FRAP value (absorbance in 560nm) using y = 0.002x - 0.0054. Hawaiian Kona had highest
antioxidant compared to other cultivars.
Means in the same column followed by different letters (A-B) represent statistical significance (P <0.05).
Sextuplicates were analyzed using one-way analysis of variance (ANOVA) followed by pairwise
comparisons of means using Tukey’s range test was performed to assess the difference in three cultivars (n
= 6, significance level P < 0.05). R: The R Project for Statistical Computing 3.3.1 was used for statistical
computing.
4479.37 ± 235.55A
5530.94 ± 235.55AB 5596.56 ± 266.49B
0
1000
2000
3000
4000
5000
6000
7000
Ethiopian Costa Rican Hawaiian
FeC
l2co
nce
ntr
atio
n (
µM
)
Cultivar
FRAP assay
81
Figure 2.6. Antioxidant activities of the fractions from flash chromatography (1:10 dilution). The fraction
number 4 showed highest antioxidant activity as 475.16 µM FeCl2 condition among 43 fractions so F4 was
injected into HPLC for further purification.
475.16
0
50
100
150
200
250
300
350
400
450
500
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43
FeC
l 2(u
M)
Fraction Number
Frap Assay from Flash Chromatography
82
Figure 2.7. 26 fractions was obtained from HPLC and tested through FRAP assay. Fraction 14 (vial 13)
was revealed as bioactive fractions with high antioxidant activity.
0
10
20
30
40
50
60
70
80
90
Via
l0
Via
l1
Via
l2
Via
l3
Via
l4
Via
l5
Via
l6
Via
l7
Via
l8
Via
l9
Via
l10
Via
l11
Via
l12
Via
l13
Via
l14
Via
l15
Via
l16
Via
l17
Via
l18
Via
l19
Via
l20
Via
l21
Via
l22
Via
l23
Via
l24
Via
l25
Via
l26
FRAP assay from HPLC
83
4.1.2. High resolution metabolomics analysis of flash chromatography
fractions
Purified bioactive fractions 13 was used to be analyzed for identification of compound in
using global metabolomics platform, XCMS Online. The reason why fraction 14 was
selected for high resolution metabolomics analysis is it showed highest anti-oxidant
activity among 26 HPLC fractions.
Table 2.2 represents identified bioactive compounds in fraction 14 purified by HPLC.
The result was generated through XCMS Online high-resolution metabolomics analysis.
In fraction 14, several compounds which were used as fragrances in cosmetics. For
example, linalyl cinnamate was detected at RT 12.75 and piperonal acetone was found at
RT 16.43. Additionally, coumarin derivative including hydroxycoumarin, and
umbelliferone were detected. These bioactive compounds have anti-viral (Cingolani et
al., 1969), anti-oxidant (Bailly et al., 2004), and photo protective (de Araujo Leite et al.,
2015) effects. Antioxidant (Ramesh and Pugalendi, 2006), anti-inflammatory (Lino et al.,
1997) and anti-mutagenic (Ohta et al., 1983) effects were reported from studies on
umbelliferone. In fraction 14, another bioactive compound were found at RT 11.05 with
m/z 151.0754. Especially, hydrocinnamic acid is widely used in food and perfume
industry as fixative agent, or a preservative to maintain the original scents. It also acts as
an emulsifier. In cosmetics, it is used frequently as softeners, and soaps providing floral
scent (Grover and Patni, 2013). Another compound cafestol was identified as highly
possibility among other candidate compounds such as fragrant compounds, coumarin
derivatives, and hydrocinnamic acid. The peak intensity was the highest for cafestol
compared to other compounds in same fraction. Cafestol is one of the coffee-specific
diterpenes which has been reported as antioxidant agents. According to the Lee and
84
Jeong, cafestol is markedly effective in protection against H2O2- induced oxidative stress
and DNA damage due to its free radical scavenging activity as strong antioxidants.
85
Table 2.2. Identified compounds in purified fraction 14 through XCMS Online HR metabolomics analysis.
Each row indicates feature, compound name, m/z, retention time, maximum intensity and its known
bioactivity. Using XCMS Online platform, chemical profiling was compiled.
F14POS_Hconc_HR m/z RT
Max
intensity
bioactivity
1329 Cafestol 679.9 11.83 618,624 Antioxidant
5 Linalyl cinnamate 285.1853 12.75 16,010 Fragrance
6 Piperonal acetone 191.0704 16.43 85,550 Fragrances
28 3-Hydroxycoumarin 163.0391 13.19 43,960
Anti-viral,
antioxidant,
photo protective
28 Umbelliferone 163.0391 13.19 43,960
Antioxidant,
anti-
inflammatory,
anti-mutagenic
29 Hydrocinnamic acid 151.0754 11.05 11,982 Anti-tyrosinase
86
Figure 2.8. Total ion chromatogram (TIC) of fraction 14 purified by HPLC. The TIC was generated by XCMS Online positive ion mode. The highest
peak intensity 618,624 was shown at RT 11.83 with m/z 679.9 and cafestol was the best possibility among other compound candidates.
87
Besides SCG extracts samples, we also evaluated the pure compounds that have been
reported in the SCG or coffee using FRAP assay. Table 2.3 shows ferric reducing power
of bioactive compounds which were reported as antioxidant in previous studies.
Interestingly, some compounds including quinic acid, resveratrol, chlorogenic acid, 4-
hydroxybenzoic acid, caffein, chryin, apigenin, rutin hydrate, nargine P-coumaric acid,
and hesperetin had no reducing ability through FRAP assay. Guaiacol showed highest
reducing power followed by walnut sample, gallic acid, quercetin, and myricetin.
88
Table 2.3. FeCl2 concentration (µM) of bioactive compounds which were reported in previous study
(unpublished data by Nahom Taddese Ghile and Lin).
List of samples analized Ferrous Chloride Con.(µM)
Quinic acid 0
Resveratrol 0
Chlorogenic Acid 0
Gallic Acid 71.7
Quercetin 61.1
4-Hydroxybenzoic acid 0
Caffein 0
Chryin 0
Apigenin 0
Guaiacol 109.29
Rutin Hydrate 0
Naringin 0
Myricetin 18.62
P-coumaric acid 0
Hesperetin 0
walnut 74.81
Guaiacol showed highest Ferrous Chloride concentration which means highest antioxidant activity among
samples. Walnut samples, Gallic acid, and Quercetin showed an intermediate antioxidant activity, followed
by myricetin. quinic acid, resvertarol, chlorogenic acid, 4-hydroxbenzoic acid, caffein, chryin, apigenin,
rutin hydrate, nargine, P-cormaric acid, and hesperetin did not have detectable value from the FRAP assay.
89
4.2 Antibacterial activity against Methicillin-resistant Staphylococcus aureus
(MRSA)
All the SCG extracts from the 3 varieties, have shown a strong or moderate antibacterial
activity against Methicillin-resistant Staphylococcus aureus (MRSA) in our anti-
microbial activity in zone of inhibition test (Figure 2.9). For this assay, the inhibition
zone diameters around each of the disc (diameter of inhibition zone minus diameter of the
disc) were measured. The inhibition zones were 8 mm or greater, suggesting strong
antibacterial activity.
The SCG extracts from Hawaiian Kona and Yirgacheffe showed the highest
antimicrobial activity. The mean values of the antibacterial activities of the SCG extracts
of all the investigated cultivars is shown in Figure 2.10. The mean value of zone of
inhibition was 8.85 mm in Hawaiian Kona sample followed by 8.19 and 5.76 in
Yirgacheffe and Tarrazu respectively. The difference between these Hawaiian Kona and
Yirgacheffe was not significant (P < 0.05).
Among the 43 collected fractionated Ethiopian Yirgacheffe SCG extracts generated by
flash chromatography, fraction number 7 and 12 showed the highest antimicrobial
activity (Figure 2.11). The rest of fractions beyond fraction number 20 that have no
activity were not presented in the Figure 2.11. Unlike antioxidant result, Tarrazu showed
the lowest antimicrobial activity. Although it is ideal to use multiple bacterial strains to
examine anti-microbial activity of SCG, S. aureus causing dermatological issue such as
acne was presented in this study due to its importance in cosmetic field.
90
Figure 2.9 SCG samples (Number from 47 to 55) had strong anti-microbial activity which is Ethiopian
Yirgacheffe, Costa Rican Tarrazu, Hawaiian Kona. The size of inhibition zone was measured three times
for all samples. Among three cultivars, Kona showed the highest anti-microbial activity having 8.85 mm
inhibition zone.
91
Figure 2.10. Antimicrobial activities of the E (Ethiopian Yirgacheffe), C (Costa Rican Tarrazu), H
(Hawaiian Kona). Using zone of inhibition assay for assessing anti-microbial activity against S. aureus, we
found the following order of efficiency among coffee Arabica cultivars: Kona (8.85) > Yirgacheffe (8.19)
>> Tarrazu (5.76). DMSO did not show any inhibitory activity. Inhibition zone: 8 mm or greater — good
antibacterial activity; 6–7 mm — moderate antibacterial activity; 4–5 mm — weak antibacterial activity; 2–
3 — very weak antibacterial activity.
Mean values followed by different letters (A and B) represent statistical different (P <0.05). Zone of
inhibition is expressed as the diameter (mm). Values represent means ± standard deviations of inhibition
zones from experiments, each using triplicate plates and was measure three times. Means with the same
superscripts (A, and B) are not significantly different (P < 0.05) from each other. One-way analysis of
variance (ANOVA) followed by pairwise comparisons of means was performed to assess the difference in
three cultivars (n = 7, significance level P < 0.05). R: The R Project for Statistical Computing 3.3.1 was
used.
8.19± 0.97B
5.76 ± 0.66A
8.85 ± 1.57B
00
2
4
6
8
10
12
Yirgacheffe Tarrazu Kona DMSO
Zon
e o
f In
hib
itio
n (
mm
)
Cultivars
Zone of Inhibition
92
Figure 2.11. Antimicrobial activity of 43 fractions using flash chromatography from Ethiopian Yirgacheffe.
Fraction number 7 and 12 showed highest zone of inhibition value. The rest of fraction number over 20 that
have no activity doesn’t represent in the chart.
5.55.94
5.4
6.055.83
5.66.05
5
5.66
0
1
2
3
4
5
6
7
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
mm
Fraction number
zone of inhibition from flash chromatography
93
5. Discussion
5.1. Antioxidant activity of SCG for cosmetic application
5.1.1. Comparison between three cultivars (Yirgacheffe, Tarrazu, and Kona)
in antioxidant properties
Among three cultivars of SCG samples, Hawaiian Kona showed the highest antioxidant
activities (Figure 2.5). Even though the activity doesn’t show much difference between
Tarazzu and Kona, their reducing power were significantly (significance level P < 0.05)
higher than Yirgacheffe (Table 2.1). All three SCG extracts showed dose-dependent
aspect in FRAP anti-oxidant assay. The reducing power was increased with the
increasing concentrations. It means that at higher concentration (1:25), the reducing
power was stronger than low concentration (1:50). The concentration of raw extracts was
solvent ratio 100 (ml/g) and their FRAP value (mM Fe(II)/g SCG) was from 0.814 mM to
1.0 mM of FeCl2. Our results showed linear correlation to the other research. According
to Mussatto, in methanol extracted SCG, sample solvent/solid ratio 10 (ml/g) showed
0.021 (mM Fe(II)/g SCG) FRAP value and sample solvent/solid ratio 30 (ml/g) showed
0.116 (mM Fe(II)/g SCG) FRAP value (Mussatto et al., 2011a). When the antioxidant
activities of SCG were compared to other research on tea antioxidants which is widely
known as strong antioxidant source containing large amounts of polyphenolic
compounds, SCG have similar or much better reducing power. Their study showed that
different types of tea had widely different antioxidant activities, ranging from 132 µmol/g
for one type of black tea (“Premium Yun-Nan Puerh Tea”) to 1144 µmol/g for one of the
green teas (“China Green Tea”) tested. Also they estimated that the a cup of tea with the
highest antioxidant which has similar reducing power with SCG can be act as 100-200mg
94
of pure ascorbic acid (vitamin C). This highlights the enormous potential of SCG as a
strong antioxidant material (Benzie and Szeto, 1999).
5.1.2. High-resolution analysis result of purified fractions 14; possible anti-
oxidant effect
Cafestol was identified as the major antioxidant agent in the purified fraction 14, which
had the strongest antioxidant activities among all the purified fractions. Previous research
have demonstrated the strong antioxidant activities of cafestol (Lee and Jeong, 2007).
Oxidative stress plays an important role in skin aging. This process is deeply related to
reactive oxygen species (ROS), induced by UV light. Antioxidants are major protector
against the free radicals coming from oxidative stress. Especially, plant-derived
polyphenols are considered as strong antioxidants and the use of them are increasing in
cosmetic formulation (Potapovich et al., 2013). Our results are consistent with other
studies conducted previously. According to Lee, and Jeong (Lee and Jeong, 2007),
cafestol has shown to rapidly react with hydroxyl radicals to protect DNA damage and
scavenge superoxide radicals.
5.1.3. FRAP antioxidant assay with the analytical standards
In table 2.3, guaiacol showed highest Ferrous Chloride concentration suggesting the
highest antioxidant activity among samples. There are several assays that have been
developed to measure antioxidant capacities of the natural products (Thaipong et al.,
2006). For example, Thaipong and Kriengsak (2006), compare and evaluate ABTS,
95
DPPH, FRAP, and ORAC assays in measuring the antioxidant ability in plant extract.
Among them FRAP assay was revealed as appropriate method due to high
reproducibility, simplicity, rapidity in performance and highest correlation with both
ascorbic acid and total phenolics (Thaipong et al., 2006). In other research using DPPH
antioxidant assay which measures radical scavenging activity, guaiacol known as
commercial antioxidants showed strong activity. Similar to their findings, our study also
showed that the compound has the highest antioxidant activity when it was evaluated by
FRAP assay (Table 2.3). This strong antioxidant activity associated with the ortho
substitution with the electron donor methoxy groups, which is an important factor to
stabilize the phenoxy radical leading to antioxidant capacity. Walnut extracts, gallic acid,
and quercetin showed an intermediate antioxidant activity, followed by myricetin. On the
other hand, quinic acid, resvertarol, chlorogenic Acid, 4-hydroxbenzoic acid, caffein,
chryin, apigenin, rutin hydrate, nargine, P-cormaric acid, and hesperetin did not show
detectable value from the FRAP assay even though they are studied as antioxidant. This
is because quinic acid is pro-metabolite for synthesis of tryptophan and nicotinamide in
the gastrointestinal tract possessing DNA repair properties instead of showing direct
antioxidant activity in FRAP assay (Pero et al., 2009). One study assessing antioxidant
activity from FRAP assay validated quercetin and tannic acid showed the has high
antioxidant activity, followed by gallic and caffeic acids. In addition, resveratrol showed
the lowest reducing ability of all the studied antioxidants among quercetin, tannic acid,
gallic acid, caffeic acid, BHA, rutin, trolox, catechin, ferulic acid, and ascorbic acid. It
agrees with the results reported by this paper. Multiple studies demonstrated that
resveratrol has less activity than quercetin or epicatechin toward reduction of hexanal
96
generation from oxidized LDL (Pulido et al., 2000). Furthermore in evaluation of the
antioxidant activity of flavonoids using FRAP assay, -OH flavone, 7-OH flavone,
chrysin, apigenin, naringenin, and hesperetin showed no or very low activity in the FRAP
assay (Firuzi et al., 2005) which is similar aspect as this study. In other antioxidant
property test, activity of even naringenin which was higher than its glycoside naringin
(Cavia‐Saiz et al., 2010). In most of comparison study for antioxidant activity, quercetin
is located in high-level among flavonoids and phenolic acids. It acts as iron chelator,
which maintains antioxidant activity in mechanism and usually included in sunscreen
such as EltaMD UV Physical Broad-Spectrum, SPF41, and EltaMD by Swiss-American
(TM/®). Unlike quercetin, the effect of resveratrol in skin wrinkling and skin photo aging
has been demonstrated in cellular level (Lee et al., 2016). Several cosmetic products
contains resveratrol as plant extracts, or pure compound form in Moisture Reservoir
Hand Cream, Love Life Skin, 46 Graphic Place Moonachie, NJ Estee Lauder Re-Nutriv
Ultimate Youth Crème Reviews (Kusumawati and Indrayanto, 2013). The result of this
study supports that SCG can be used in cosmetic formula as active ingredient due to its
high concentration of antioxidant phenolic acid and flavonoids.
5.2. Antimicrobial activity of SCG for cosmetic application
Zone of inhibition assay indicates that Hawaiian Kona and Ethiopian Yirgacheffe SCG
extract can be considered as great anti-bacterial source compared to Costa Rican Tarrazu
cultivars (Figure 2.10). These days, anti-microbial activities become more important for
safe use of the products prohibiting germ spread in using cosmetics. By formulating anti-
97
microbial compounds in cosmetics, the products can be safe from causing skin infections,
or discoloration of the product itself during the usage, and after they are opened. Since
coffee has been reported as antimicrobial material due to several compounds such as
caffeine, chlorogenic acid, protocatechuic acid (Almeida et al., 2006), and melanoidins
(Rufian-Henares and de la Cueva, 2009), we expect that SCG still contains bioactive
compounds for preventing spoilage or skin disease from microbial contamination.
Thus, zone of inhibition assay was conducted to evaluate the activity of raw SCG
samples and flash chromatography fractions against S. aureus, one of the most common
pathogen causing acne. Since disease from S. aureus occurs in hospital starting from skin
infection, S. aureus are used prevalently these days for antimicrobial activity evaluation.
In this study, S. aureus strain which is called MRSA that resistant to antibiotics was used
for testing anti-microbial activity of SCG. S. aureus major strain of skin infection causing
a range of illnesses, from minor skin infections, such as pimples, impetigo, boils,
cellulitis, folliculitis, carbuncles, and abscesses, to life-threatening diseases such as toxic
shock syndrome, bacteremia, and sepsis. In cosmetics, acne is the most common skin
issue in the United States, affecting up to 50 million Americans annually (Bickers et al.,
2006). Specially, S. aureus is known as one of the major bacteria causing acne in human
skin. Since the acne-causing bacteria are usually gram positive (Daud et al., 2013), it is
reasonable to believe that the SCG could be also used to treat acne pathogenesis and
other skin conditions associated with similar infection by gram positive bacterial, such as
P. acnes, S. epidermidis, S. aureus, Klebsiella pneumoniae, Streptococcus, and
Enterobacter.
98
Similar to the findings reported by Monente (2015), SCG samples were shown
antimicrobial activity against Gram-positive bacteria using the inhibition zones of the
agar-well diffusion method (Figure 2.10). Another studies evaluate that spent coffee has
more strong antibacterial activity against S. aureus than coffee brew in Arabica and
Robusta species (Monente et al., 2015) indicating SCG as a potential by-product with
high value. Recent studies evaluated that caffeine, trigonelline and protocatechiuic acids
can act as natural antimicrobial compounds (Almeida et al., 2006). Anti-fungal effect of
caffeine is related to inhibition of aflatoxin production (Buchanan and Fletcher, 1978)
and its antimicrobial effect is from DNA repair mechanism inhibition . Aromatic and
phenolic compounds exert antimicrobial activity by altering the structure and function of
the cytoplasmic membrane, inhibiting electron transport (Buchanan and Fletcher, 1978).
In addition, bioactivity can be influenced by the heat treatment during roasting process
forming coffee melanoidins which have antibacterial activity. It is known that
antimicrobial activity of melanoidins could be mediated by metal chelating mechanisms
(Rufian-Henares and de la Cueva, 2009). This process called Maillard reaction are
responsible for antibacterial activity of roasted coffee which shows better antimicrobial
activity than raw coffee extracts. One study on “antimicrobial activity of Psidium
guajava and Juglans regia leaf extracts to acne-developing organisms” presented the in
vitro inhibitory effect of the leaf extracts on acne lesions. They figured out that the acne
legion constitutes Propionibacterium acnes, Staphylococcus aureus, and Staphylococcus
epidermidis account for 47%, 13%, and 24% respectively. The zones of inhibition due to
15% of Juglans regia leaf extracts showed 10mm against S. aures, (Qa'dan et al., 2005)
similar result as Hawaiian Kona SCG (maximum 12mm). Further studies such as high
99
resolution metabolomics are need to identify the compounds in fraction 7 and 12 obtained
from flash chromatography. Besides caffeine, trigonelline and protocatechiuic, other
compounds such as quercetin glycosides, guaiacol, and cinnamic acid can be candidates
of antimicrobial compounds in fraction 7 and 12. Selected compounds which can be
found in coffee for antimicrobial activity were based on prior study by (Aga et al., 1994)
and (Arima and Danno, 2002). This translational research projects could help turn the
low-value bioactive SCG into bioactive, and organic active ingredients in personal care
product, as an excellent alternative, to the synthetic preservatives. As a result, it provides
safe and long-lasting “green “cosmetics product for customers for skin conditioning or
treatment, such as inhibition of the acne-causing bacteria or acne prevention.
Due to the strong inhibitory effects of the SCG extracts against gram positive bacteria,
the active ingredients cane be formulated into other personal products, such as toothpaste,
lotions, spray, sanitizers, shampoo, wound healing patches.
Other skin-care application
5.3. Importance of anti-inflammatory and anti-carcinogenesis activity in
cosmetics
In table 2.2, positive modes, umbelliferone was found in SCG which have also anti-
inflammatory effect (Rauf et al., 2014). It is a potential candidate compound that can be
found in purified fraction 14 (Table 2.2). Skin is always subject to inflammation, and the
immediate effects on skin is redness, cracked skin, dry patches, spots (post-inflammatory
hyperpigmentation) and aging. The anti-inflammatory properties can be related to
inhibiting of the cyclooxygenase and lipoxygenase activity and pro-inflammatory
cytokines inhibition. The skin irritation is related to releasing of cyclooxygenase and
100
prostaglandin in the skin. This phenomenon increases in trans-epidermal water loss and
skin pH value. The active compounds in SCG extracts specially flavonoids and phenolic
acid showing anti-inflammatory activity, can attribute to the reduction of the erythema
and also help to improve the skin barrier function (decrease in trans-epidermal water loss
and pH value), which is associated with the fast skin return to homeostasis after exposure
on irritants. Usually, exposure to UVB radiation induces oxidative and inflammatory skin
damage, thereby increasing the risk of skin carcinogenesis (Choi et al., 2014). The study
by Cooper Bowden, activation of two transcription factor such as NF-κB, and AP-1 by
prolonged UV exposure have been identified to be involved in the processes of cell-
proliferation, cell differentiation and cell survival and therefore play important roles in
tumorigenesis (Cooper and Bowden, 2007).
5.4. High resolution metabolomics analysis followed by bioassay guided
purification
Even though glycosylated form of flavonoids show weaker bioactivities compared to free
form, this study demonstrates potential antioxidants which are from SCG flavonoids such
as coumarin derivatives. By chemical profiling through XCMS Online high resolution
analysis, we revealed that strong ferric reducing power might from flavonoids including
umbelliferone which is contained in fraction 14. One study revealed that umbelliferone
has a remarkable antioxidants from hydrogen donating or radical-scavenging ability,
deoxyribose, chelating power, reducing power and lipid peroxidation assay (Singh et al.,
2010). It is therefore, imperative to study the plant material for its potential as a
scavenger of free radicals. Flavonoids are the active ingredients of many herbal
medicines which can exhibit anti-inflammatory, anti-oxidant, immuno-regulatory, anti-
101
tumor and anti-radiation effects (Grover and Patni, 2013; Oh et al., 1997). These days,
studies on natural compounds with bioactivity and effort to apply to cosmetic approach
are being conducted widely. Apparently, the aglycon forms (the molecular form without
sugar residues) of flavonoids are the biologically most active compounds in mammal
metabolism. Commonly the bioactive ingredients of plants are always exist in mixture of
aglycones and hydrophilic glycosides (Ryu et al., 2013). Aglycones are hydrophobic and
can be permeable through skin easier due to low molecular weight whereas glycosides
have disadvantages to be a cosmetic material (Ryu et al., 2013). This phenomenon also
can be explained that penetration enhancers have low molecular weight (Bos and
Meinardi, 2000). In addition to bioavailability difference between flavonoid aglycones
and glycosides, difference in bioactivity is important as well. Antioxidant activity of
quercetin and quercetin monoglucosides shows that it is apparent that the activity of
quercetin is much higher than that of quercetin glucosides due to structural effect to
radical scavenging activity (Ioku et al., 1995). Since bioactive fractions evaluated by
zone of inhibition assay have not been purified through HPLC, further studies are need to
identify compounds which has antimicrobial activity.
6. Conclusion
Bioassay-guided purification method coupled with global metabolomics was utilized to
identify possible antioxidant candidates in SCG samples. We identified several
antioxidant agents such as cafestol, and coumarin derivaties in purified fractions 13.
Also, comparison analysis of FRAP antioxidant assay was conducted to evaluate
reducing power between three SCG cultivars including Ethiopian Yirgacheffe, Costa
102
Rican Tarrazu and Hawaiian Kona. According to antioxidant assay, Hawaiian Kona
showed the highest antioxidant activities among SCG varieties. In antibacterial test,
Ethiopian Yirgacheffe and Hawaiian Kona showed higher activities compared to Costa
Rican Tarrazu. Besides antioxidant and antibacterial properties, multiple bioactive
compounds were found in SCG related to skin care application such as anti-
inflammatory, anti-carcinogenesis, and anti-tyrosinase. These findings from SCG
suggesting the promising opportunity of SCG as a raw materials that can be formulated to
cosmetic products. Future plan for this project is conducting target analysis to quantify
the bioactive compounds for structural study (NMR) to confirm our results.
Since SCG has been considered as an abundant, and sustainable resources, and
distributed freely throughout the world, identifying a new application of SCG for
cosmetic industry not only benefit consumer and environment, it also help foster a new
industry to stabilize the rural economy. It can be achieved by maximizing supply chain of
coffee production from sustainability, and by generating economic values in both coffee
production and consumption countries in global scale. Through utilizing the modern
global metabolomics approach to identify bioactive compounds, this project demonstrate
a novel approach to quickly identify the bioactive ingredients in the SCG.
103
References
Adi, AJ and ZM Noor 2009. Waste recycling: Utilization of coffee grounds and kitchen waste in
vermicomposting. Bioresource Technology 100: 1027-1030.
Aga, Hajime, Takashi Shibuya, Toshiyuki Sugimoto, Masashi Kurimoto and Shuhei Nakajima
1994. Isolation and identification of antimicrobial compounds in Brazilian propolis. Bioscience,
Biotechnology, and Biochemistry 58: 945-946.
Ahmad, Iqbal and Arina Z Beg 2001. Antimicrobial and phytochemical studies on 45 Indian
medicinal plants against multi-drug resistant human pathogens. Journal of ethnopharmacology
74: 113-123.
Akiyama, Hisanori, Kazuyasu Fujii, Osamu Yamasaki, Takashi Oono and Keiji Iwatsuki 2001.
Antibacterial action of several tannins against Staphylococcus aureus. Journal of antimicrobial
chemotherapy 48: 487-491.
Alarcon De La Lastra, Catalina and Isabel Villegas 2005. Resveratrol as an anti‐inflammatory
and anti‐aging agent: Mechanisms and clinical implications. Molecular nutrition & food research
49: 405-430.
Almeida, Ana Amélia P, Adriana Farah, Daniela AM Silva, Elzíria A Nunan and M Beatriz A
Glória 2006. Antibacterial activity of coffee extracts and selected coffee chemical compounds
against enterobacteria. Journal of agricultural and food chemistry 54: 8738-8743.
Amaral, Leonard, Miguel Viveiros and Jette E Kristiansen 2006. " Non-Antibiotics": alternative
therapy for the management of MDRTB and MRSA in economically disadvantaged countries.
Current drug targets 7: 887-891.
An, Sang Mi, Hyo Jung Kim, Jung‐Eun Kim and Yong Chool Boo 2008. Flavonoids, taxifolin
and luteolin attenuate cellular melanogenesis despite increasing tyrosinase protein levels.
Phytotherapy Research 22: 1200-1207.
An, Sang Mi, Jae‐Sook Koh and Yong Chool Boo 2010. p‐coumaric acid not only inhibits
human tyrosinase activity in vitro but also melanogenesis in cells exposed to UVB. Phytotherapy
Research 24: 1175-1180.
Arima, Hidetoshi and Gen-ichi Danno 2002. Isolation of antimicrobial compounds from guava
(Psidium guajava L.) and their structural elucidation. Bioscience, Biotechnology, and
Biochemistry 66: 1727-1730.
104
Arung, Enos Tangke, Shoko Furuta, Hiroya Ishikawa, Irawan Wijaya Kusuma, Kuniyoshi
Shimizu and Ryuichiro Kondo 2011. Anti-melanogenesis properties of quercetin-and its
derivative-rich extract from Allium cepa. Food Chemistry 124: 1024-1028.
Awad, Rosalie, Asim Muhammad, Tony Durst, Vance L Trudeau and John T Arnason 2009.
Bioassay‐guided fractionation of lemon balm (Melissa officinalis L.) using an in vitro measure of
GABA transaminase activity. Phytotherapy Research 23: 1075-1081.
Aziz, NH, SE Farag, LA Mousa and MA Abo-Zaid 1998. Comparative antibacterial and
antifungal effects of some phenolic compounds. Microbios 93: 43-54.
Badjakov, I, M Nikolova, R Gevrenova, V Kondakova, E Todorovska and A Atanassov 2008.
Bioactive compounds in small fruits and their influence on human health. Biotechnology &
Biotechnological Equipment 22: 581-587.
Bailly, Fabrice, Cédric Maurin, Elisabeth Teissier, Hervé Vezin and Philippe Cotelle 2004.
Antioxidant properties of 3-hydroxycoumarin derivatives. Bioorganic & medicinal chemistry 12:
5611-5618.
Balouiri, Mounyr, Moulay Sadiki and Saad Koraichi Ibnsouda 2016. Methods for in vitro
evaluating antimicrobial activity: A review. Journal of Pharmaceutical Analysis 6: 71-79.
Banister, Nigel and Peter Cheetham 2001. Skin lightening agents. In Skin lightening agents:
Google Patents.
Bartel, Jörg, Jan Krumsiek and Fabian J Theis 2013. Statistical methods for the analysis of high-
throughput metabolomics data. Computational and structural biotechnology journal 4: 1-9.
Belguidoum, K, H Amira-Guebailia, Y Boulmokh and O Houache 2014. HPLC coupled to UV–
vis detection for quantitative determination of phenolic compounds and caffeine in different
brands of coffee in the Algerian market. Journal of the Taiwan Institute of Chemical Engineers
45: 1314-1320.
Bendre, Ratnamala S, Jamatsing D Rajput, Suresh D Bagul and PS Karandikar 2016. Outlooks
on medicinal properties of eugenol and its synthetic derivatives. Nat Prod Chem Res 4: 2.
Benzie, Iris FF and YT Szeto 1999. Total antioxidant capacity of teas by the ferric
reducing/antioxidant power assay. Journal of agricultural and food chemistry 47: 633-636.
Bernard, Francois-Xavier, Serge Sable, Beatrice Cameron, Jean Provost, Jean-Francois
Desnottes, Joel Crouzet and Francis Blanche 1997. Glycosylated flavones as selective inhibitors
of topoisomerase IV. Antimicrobial agents and chemotherapy 41: 992-998.
105
Bickers, David R, Henry W Lim, David Margolis, Martin A Weinstock, Clifford Goodman, Eric
Faulkner, Ciara Gould, Eric Gemmen and Tim Dall 2006. The burden of skin diseases: 2004: A
joint project of the American Academy of Dermatology Association and the Society for
Investigative Dermatology. Journal of the American Academy of Dermatology 55: 490-500.
Bin Hafeez, Bilal, Mohammad Asim, Imtiaz Ahmad Siddiqui, Vaqar Mustafa Adhami, Imtiyaz
Murtaza and Hasan Mukhtar 2008. Delphinidin, a dietary anthocyanidin in pigmented fruits and
vegetables: a new weapon to blunt prostate cancer growth. Cell cycle 7: 3320-3326.
Blumberg, Simone, Oliver Frank and Thomas Hofmann 2010. Quantitative studies on the
influence of the bean roasting parameters and hot water percolation on the concentrations of
bitter compounds in coffee brew. Journal of agricultural and food chemistry 58: 3720-3728.
Bognar, Eszter, Zsolt Sarszegi, Aliz Szabo, Balazs Debreceni, Nikoletta Kalman, Zsuzsanna
Tucsek, Balazs Sumegi and Ferenc Gallyas Jr 2013. Antioxidant and anti-inflammatory effects in
RAW264. 7 macrophages of malvidin, a major red wine polyphenol. PLoS One 8: e65355.
Bonina, Francesco, Maria Lanza, Lucia Montenegro, Claudio Puglisi, Antonio Tomaino,
Domenico Trombetta, Francesco Castelli and Antonella Saija 1996. Flavonoids as potential
protective agents against photo-oxidative skin damage. International journal of pharmaceutics
145: 87-94.
Boots, Agnes W, Guido RMM Haenen and Aalt Bast 2008. Health effects of quercetin: from
antioxidant to nutraceutical. European journal of pharmacology 585: 325-337.
Bos, Jan D and Marcus MHM Meinardi 2000. The 500 Dalton rule for the skin penetration of
chemical compounds and drugs. Experimental dermatology 9: 165-169.
Bravo, Jimena, Isabel Juaniz, Carmen Monente, Bettina Caemmerer, Lothar W Kroh, M Paz De
Pena and Concepcion Cid 2012. Evaluation of spent coffee obtained from the most common
coffeemakers as a source of hydrophilic bioactive compounds. Journal of agricultural and food
chemistry 60: 12565-12573.
Buchanan, RL and AM Fletcher 1978. Methylxanthine inhibition of aflatoxin production. Journal
of Food Science 43: 654-655.
Burda, Stanislaw and Wieslaw Oleszek 2001. Antioxidant and antiradical activities of
flavonoids. Journal of agricultural and food chemistry 49: 2774-2779.
Campos-Vega, Rocio, Guadalupe Loarca-Pina, Haydé A Vergara-Castañeda and B Dave Oomah
2015. Spent coffee grounds: A review on current research and future prospects. Trends in Food
Science & Technology 45: 24-36.
106
Castro, Myrella Lessio, Walquíria Ribeiro Vilela, Rogéria Cristina Zauli, Masaharu Ikegaki,
Vera Lúcia Garcia Rehder, Mary Ann Foglio, Severino Matias de Alencar and Pedro Luiz
Rosalen 2009. Bioassay guided purification of the antimicrobial fraction of a Brazilian propolis
from Bahia state. BMC complementary and alternative medicine 9: 25.
Cavia‐Saiz, Monica, Maria D Busto, Maria Concepción Pilar‐Izquierdo, Natividad Ortega,
Manuel Perez‐Mateos and Pilar Muñiz 2010. Antioxidant properties, radical scavenging activity
and biomolecule protection capacity of flavonoid naringenin and its glycoside naringin: a
comparative study. Journal of the Science of Food and Agriculture 90: 1238-1244.
Chamcheu, Jean Christopher, Harish C Pal, Imtiaz A Siddiqui, Vaqar M Adhami, Seyoum
Ayehunie, Brendan T Boylan, Felicite K Noubissi, Naghma Khan, Deeba N Syed and Craig A
Elmets 2015. Prodifferentiation, anti-inflammatory and antiproliferative effects of delphinidin, a
dietary anthocyanidin, in a full-thickness three-dimensional reconstituted human skin model of
psoriasis. Skin pharmacology and physiology 28: 177-188.
CHEN, CHUNG‐WEN and CHI‐TANG HO 1995. Antioxidant properties of polyphenols
extracted from green and black teas. Journal of food lipids 2: 35-46.
Chen, Jing, Kao-Wei Chua, Chu C Chua, Hailong Yu, Aijie Pei, Balvin HL Chua, Ronald C
Hamdy, Xingshun Xu and Chun-Feng Liu 2011. Antioxidant activity of 7, 8-dihydroxyflavone
provides neuroprotection against glutamate-induced toxicity. Neuroscience letters 499: 181-185.
Choi, Hyeon-Son, Eu Ddeum Park, Yooheon Park and Hyung Joo Suh 2015. Spent coffee
ground extract suppresses ultraviolet B-induced photoaging in hairless mice. Journal of
Photochemistry and Photobiology B: Biology 153: 164-172.
Choi, Ki-Seok, Joydeb Kumar Kundu, Kyung-Soo Chun, Hye-Kyung Na and Young-Joon Surh
2014. Rutin inhibits UVB radiation-induced expression of COX-2 and iNOS in hairless mouse
skin: p38 MAP kinase and JNK as potential targets. Archives of biochemistry and biophysics
559: 38-45.
Choi, Mi Jin, Boh Kyung Kim, Kun Young Park, Takako Yokozawa, Young Ok Song and Eun
Ju Cho 2010. Anti-aging effects of cyanidin under a stress-induced premature senescence
cellular system. Biological and pharmaceutical bulletin 33: 421-426.
Choquenet, Benjamin, Céline Couteau, Eva Paparis and Laurence JM Coiffard 2008. Quercetin
and rutin as potential sunscreen agents: determination of efficacy by an in vitro method. Journal
of natural products 71: 1117-1118.
Chou, Tzung‐Han, Hsiou‐Yu Ding, Wei Jing Hung and Chia‐Hua Liang 2010. Antioxidative
characteristics and inhibition of α‐melanocyte‐stimulating hormone‐stimulated melanogenesis of
vanillin and vanillic acid from Origanum vulgare. Experimental dermatology 19: 742-750.
107
Christin, Christin, Age K Smilde, Huub CJ Hoefsloot, Frank Suits, Rainer Bischoff and Peter L
Horvatovich 2008. Optimized time alignment algorithm for LC− MS data: correlation optimized
warping using component detection algorithm-selected mass chromatograms. Analytical
chemistry 80: 7012-7021.
Cingolani, Gian, Fulvio Gualtieri and Maria Pigini 1969. Researches in the field of antiviral
compounds. Mannich bases of 3-hydroxycoumarin. Journal of medicinal chemistry 12: 531-532.
Cloughley, John B 1980. The effect of temperature on enzyme activity during the fermentation
phase of black tea manufacture. Journal of the Science of Food and Agriculture 31: 920-923.
Coble, Jamie B and Carlos G Fraga 2014. Comparative evaluation of preprocessing freeware on
chromatography/mass spectrometry data for signature discovery. Journal of chromatography A
1358: 155-164.
Cooper, SJ and GT Bowden 2007. Ultraviolet B regulation of transcription factor families: Roles
of nuclear factor-kappa B (NF-κB) and activator protein-1 (AP-1) in UVB-induced skin
carcinogenesis. Current cancer drug targets 7: 325-334.
Cruz, Rebeca, Paula Baptista, Sara Cunha, José Alberto Pereira and Susana Casal 2012.
Carotenoids of lettuce (Lactuca sativa L.) grown on soil enriched with spent coffee grounds.
Molecules 17: 1535-1547.
Daglia, Maria, Maria Teresa Cuzzoni and Cesare Dacarro 1994a. Antibacterial activity of coffee.
Journal of Agricultural and Food Chemistry 42: 2270-2272.
Daglia, Maria, Maria Teresa Cuzzoni and Cesare Dacarro 1994b. Antibacterial activity of coffee:
relationship between biological activity and chemical markers. Journal of Agricultural and Food
Chemistry 42: 2273-2277.
Daud, Farhat S, Gauri Pande, Mamta Joshi, Ruchita Pathak and Shubhangi Wankhede 2013. A
study of antibacterial effect of some selected essential oils and medicinal herbs against acne
causing bacteria. J. Pharm. Sci. Invent 2: 27-34.
de Araujo Leite, Jocelmo Cássio, Tainá Myra Xavier de Castro, José Maria Barbosa-Filho, José
Pinto de Siqueira-Junior and Luis Fernando Marques-Santos 2015. Photoprotective effect of
coumarin and 3-hydroxycoumarin in sea urchin gametes and embryonic cells. Journal of
Photochemistry and Photobiology B: Biology 146: 44-51.
Devasagayam, TPA, JP Kamat, Hari Mohan and PC Kesavan 1996. Caffeine as an antioxidant:
inhibition of lipid peroxidation induced by reactive oxygen species. Biochimica et Biophysica
Acta (BBA)-Biomembranes 1282: 63-70.
108
Do, Youn-Kyung, Jong-Min Kim, Sang-Mok Chang, Ji-Hwan Hwang and Woo-Sik Kim 2009.
Enhancement of polyphenol bio-activities by enzyme reaction. Journal of Molecular Catalysis B:
Enzymatic 56: 173-178.
El Gharras, Hasna 2009. Polyphenols: food sources, properties and applications–a review.
International journal of food science & technology 44: 2512-2518.
Esquivel, Patricia and Víctor M Jiménez 2012. Functional properties of coffee and coffee by-
products. Food Research International 46: 488-495.
Fardiaz, Srikandi 1995. Antimicrobial activity of coffee (Coffea robusta) extract. ASEAN Food
Journal 10: 103-106.
Ferrali, M, C Signorini, B Caciotti, L Sugherini, L Ciccoli, D Giachetti and M Comporti 1997.
Protection against oxidative damage of erythrocyte membrane by the flavonoid quercetin and its
relation to iron chelating activity. FEBS letters 416: 123-129.
Firuzi, Omidreza, Antonio Lacanna, Rita Petrucci, Giancarlo Marrosu and Luciano Saso 2005.
Evaluation of the antioxidant activity of flavonoids by “ferric reducing antioxidant power” assay
and cyclic voltammetry. Biochimica et Biophysica Acta (BBA)-General Subjects 1721: 174-184.
Fukui, Yuko and Takashi Iwashita 2013. Epigallocatechin gallate trimer and α-glucosidase
inhibitor containing epigallocatechin gallate polymer. In Epigallocatechin gallate trimer and α-
glucosidase inhibitor containing epigallocatechin gallate polymer: Google Patents.
Gani, Adil, SM Wani, FA Masoodi and G Hameed 2012. Whole-grain cereal bioactive
compounds and their health benefits: a review. J Food Process Technol 3: 146-156.
Gao, Hong, Jun Nishida, Shizuka Saito and Jun Kawabata 2007. Inhibitory effects of 5, 6, 7-
trihydroxyflavones on tyrosinase. Molecules 12: 86-97.
García-Bores, AM and J GUILLERMO Avila 2008. Natural products: Molecular mechanisms in
the photochemoprevention of skin cancer. Rev Latinoamer Quím 36: 83-102.
García-Lafuente, Ana, Eva Guillamón, Ana Villares, Mauricio A Rostagno and José Alfredo
Martínez 2009. Flavonoids as anti-inflammatory agents: implications in cancer and
cardiovascular disease. Inflammation Research 58: 537-552.
García-Reyes, Juan F, M Dolores Hernando, Antonio Molina-Díaz and Amadeo R Fernández-
Alba 2007. Comprehensive screening of target, non-target and unknown pesticides in food by
LC-TOF-MS. TrAC Trends in Analytical Chemistry 26: 828-841.
García-Sosa, Karlina, Ninibe Villarreal-Alvarez, Petra Lübben and Luis M Peña-Rodríguez
2006. Chrysophanol, an antimicrobial anthraquinone from the root extract of Colubrina greggii.
Journal of the Mexican Chemical Society 50: 76-78.
109
Gowda, Harsha, Julijana Ivanisevic, Caroline H Johnson, Michael E Kurczy, H Paul Benton,
Duane Rinehart, Thomas Nguyen, Jayashree Ray, Jennifer Kuehl and Bernardo Arevalo 2014.
Interactive XCMS Online: simplifying advanced metabolomic data processing and subsequent
statistical analyses. Analytical chemistry 86: 6931.
Graf, J 2000. Herbal anti-inflammatory agents for skin disease. Skin Therapy Lett 5: 3-5.
Grover, Neha and Vidya Patni 2013. Phytochemical characterization using various solvent
extracts and GC-MS analysis of methanolic extract of Woodfordia fruticosa (L) Kurz. Leaves.
International Journal of Pharmacy and Pharmaceutical Sciences 5: 291-295.
Guyot, Bernard, Dominique Gueule, Jean-Claude Manez, Jean-Jacques Perriot, J Giron and Luc
Villain 1996. Influence de l'altitude et de l'ombrage sur la qualité des cafés Arabica. Plantations,
recherche, développement 3: 272-283.
Hämäläinen, Mari, Riina Nieminen, Pia Vuorela, Marina Heinonen and Eeva Moilanen 2007.
Anti-inflammatory effects of flavonoids: genistein, kaempferol, quercetin, and daidzein inhibit
STAT-1 and NF-κB activations, whereas flavone, isorhamnetin, naringenin, and pelargonidin
inhibit only NF-κB activation along with their inhibitory effect on iNOS expression and NO
production in activated macrophages. Mediators of inflammation 2007.
Han, Byung Hoon, Myung Hwan Park and Yong Nam Han 1981. Studies on the antioxidant
components of Korean ginseng (III). Archives of pharmacal research 4: 53-58.
Hanasaki, Yukiko, Shunjiro Ogawa and Shozo Fukui 1994. The correlation between active
oxygens scavenging and antioxidative effects of flavonoids. Free Radical Biology and Medicine
16: 845-850.
Haraguchi, Hiroyuki, Katsuhito Tanimoto, Yukiyoshi Tamura, Kenji Mizutani and Takeshi
Kinoshita 1998. Mode of antibacterial action of retrochalcones from Glycyrrhiza inflata.
Phytochemistry 48: 125-129.
Heuberger, Adam L, Matthew R Lewis, Ming-Hsuan Chen, Mark A Brick, Jan E Leach and
Elizabeth P Ryan 2010. Metabolomic and functional genomic analyses reveal varietal differences
in bioactive compounds of cooked rice. PLoS One 5: e12915.
Hodges, D Mark, John M DeLong, Charles F Forney and Robert K Prange 1999. Improving the
thiobarbituric acid-reactive-substances assay for estimating lipid peroxidation in plant tissues
containing anthocyanin and other interfering compounds. Planta 207: 604-611.
Huang, Mou-Tuan, Jian-Guo Xie, Zhi Yuan Wang, Chi-Tang Ho, You-Rong Lou, Chung-Xiou
Wang, Gordon C Hard and Allan H Conney 1997. Effects of tea, decaffeinated tea, and caffeine
on uvb light-induced complete carcinogenesis in SKH-1 mice: demonstration of caffeine as a
biologically important constitutent of tea. Cancer Research 57: 2623-2629.
110
Huang, Wu-Yang, Jian Wang, Ya-Mei Liu, Qi-Sheng Zheng and Chun-Yang Li 2014. Inhibitory
effect of Malvidin on TNF-α-induced inflammatory response in endothelial cells. European
journal of pharmacology 723: 67-72.
Ikigai, Hajime, Taiji Nakae, Yukihiko Hara and Tadakatsu Shimamura 1993. Bactericidal
catechins damage the lipid bilayer. Biochimica et Biophysica Acta (BBA)-Biomembranes 1147:
132-136.
Ioku, Kana, Tojiro Tsushida, Yoko Takei, Nobuji Nakatani and Junji Terao 1995. Antioxidative
activity of quercetin and quercetin monoglucosides in solution and phospholipid bilayers.
Biochimica et Biophysica Acta (BBA)-Biomembranes 1234: 99-104.
Iwai, Kazuya, Noriaki Kishimoto, Yukari Kakino, Kyo Mochida and Tokio Fujita 2004. In vitro
antioxidative effects and tyrosinase inhibitory activities of seven hydroxycinnamoyl derivatives
in green coffee beans. Journal of agricultural and food chemistry 52: 4893-4898.
Jin, Seong Eun, You Kyung Son, Byung-Sun Min, Hyun Ah Jung and Jae Sue Choi 2012. Anti-
inflammatory and antioxidant activities of constituents isolated from Pueraria lobata roots.
Archives of pharmacal research 35: 823-837.
Joët, Thierry, Andréina Laffargue, Frédéric Descroix, Sylvie Doulbeau, Benoît Bertrand and
Stéphane Dussert 2010. Influence of environmental factors, wet processing and their interactions
on the biochemical composition of green Arabica coffee beans. Food Chemistry 118: 693-701.
Ju, Huangxian 2013. Grand challenges in analytical chemistry: towards more bright eyes for
scientific research, social events and human health. Frontiers in chemistry 1.
Jung, Sung Keun, Ki Won Lee, Ho Young Kim, Mi Hyun Oh, Sanguine Byun, Sung Hwan Lim,
Yong-Seok Heo, Nam Joo Kang, Ann M Bode and Zigang Dong 2010. Myricetin suppresses
UVB-induced wrinkle formation and MMP-9 expression by inhibiting Raf. Biochemical
pharmacology 79: 1455-1461.
Kanlayavattanakul, Mayuree and Nattaya Lourith 2013. Spent Arabica Coffee as a Rich Source
of Antioxidant Appraisal for Cosmetic Applications. Advanced Science, Engineering and
Medicine 5: 173-176.
Katalinic, V, M Milos, T Kulisic and M Jukic 2006. Screening of 70 medicinal plant extracts for
antioxidant capacity and total phenols. Food Chemistry 94: 550-557.
Kato, Ryuichi, Teruo Nakadate, Satoshi Yamamoto and Takashi Sugimura 1983. Inhibition of
12-O-tetradecanoylphorbol-13-acetate-induced tumor promotion and ornithine decarboxylase
activity by quercetin: possible involvement of lipoxygenase inhibition. Carcinogenesis 4: 1301-
1305.
111
Kaufmann, A 2014. Combining UHPLC and high-resolution MS: a viable approach for the
analysis of complex samples? TrAC Trends in Analytical Chemistry 63: 113-128.
Kenner, Dan and Yves Requena 2001. Botanical medicine: a European professional perspective:
Paradigm publications.
Khaiat, Alain and Claude Saliou 2015. Botanical Extracts. Cosmeceuticals and Active
Cosmetics: 385.
Kiattisin, Kanokwan, Thananya Nantarat and Pimporn Leelapornpisid 2016. Evaluation of
antioxidant and anti-tyrosinase activities as well as stability of green and roasted coffee bean
extracts from Coffea arabica and Coffea canephora grown in Thailand. Journal of
Pharmacognosy and Phytotherapy 8: 182-192.
Kilmartin, Paul A and Chyong F Hsu 2003. Characterisation of polyphenols in green, oolong,
and black teas, and in coffee, using cyclic voltammetry. Food Chemistry 82: 501-512.
Kim, Seong-Jin, Mi-Hye Lim, Inn-Ki Chun and Young-Ho Won 1997. Effects of flavonoids of
Ginkgo biloba on proliferation of human skin fibroblast. Skin pharmacology and physiology 10:
200-205.
Kim, You-Jung 2007. Antimelanogenic and antioxidant properties of gallic acid. Biological and
pharmaceutical bulletin 30: 1052-1055.
Kim, Young Mee, Jung Sik Huh, Yoongho Lim and Moonjae Cho 2015. Soy Isoflavone Glycitin
(4'‐Hydroxy‐6‐Methoxyisoflavone‐7‐D‐Glucoside) Promotes Human Dermal Fibroblast Cell
Proliferation and Migration via TGF‐β Signaling. Phytotherapy Research 29: 757-769.
Kitagawa, Shuji, Kenta Yoshii, Shin-ya Morita and Reiko Teraoka 2011. Efficient topical
delivery of chlorogenic acid by an oil-in-water microemulsion to protect skin against UV-
induced damage. Chemical and Pharmaceutical Bulletin 59: 793-796.
Kong, Yeon Hee, Youn Ock Jo, Chang-Won Cho, Dongwook Son, Soojin Park, Jeonghae Rho
and Sang Yoon Choi 2008. Inhibitory effects of cinnamic acid on melanin biosynthesis in skin.
Biological and pharmaceutical bulletin 31: 946-948.
Korgenski, E Kent and Judy A Daly 1998. Evaluation of the BIOMIC video reader system for
determining interpretive categories of isolates on the basis of disk diffusion susceptibility results.
Journal of clinical microbiology 36: 302-304.
Kris-Etherton, PM, M Lefevre, GR Beecher, MD Gross, CL Keen and TD Etherton 2004.
Bioactive compounds in nutrition and health-research methodologies for establishing biological
function: the antioxidant and anti-inflammatory effects of flavonoids on atherosclerosis. Annu.
Rev. Nutr. 24: 511-538.
112
Kuhl, Carsten, Ralf Tautenhahn, Christoph Bottcher, Tony R Larson and Steffen Neumann 2011.
CAMERA: an integrated strategy for compound spectra extraction and annotation of liquid
chromatography/mass spectrometry data sets. Analytical chemistry 84: 283-289.
Kusumawati, Idha and Gunawan Indrayanto 2013. Natural antioxidants in cosmetics. Stud. Nat.
Prod. Chem 40: 485-505.
Lanni, C and EL Becker 1985. Inhibition of neutrophil phospholipase A2 by p-bromophenylacyl
bromide, nordihydroguaiaretic acid, 5, 8, 11, 14-eicosatetrayenoic acid and quercetin.
International Archives of Allergy and Immunology 76: 214-217.
Lee-Hilz, Yee Y, Anne-Marie JF Boerboom, Adrie H Westphal, Willem JH van Berkel, Jac
MMJG Aarts and Ivonne MCM Rietjens 2006. Pro-oxidant activity of flavonoids induces EpRE-
mediated gene expression. Chemical research in toxicology 19: 1499-1505.
Lee, Jin Young, Yeon Kyung Kim, Jin Young Seo, Chong Won Choi, Jae Sung Hwang, Byeong
Gon Lee, Ih Seop Chang and Jin Ho Chung 2008. Loss of elastic fibers causes skin wrinkles in
sun-damaged human skin. Journal of dermatological science 50: 99-107.
Lee, Kuo-Hsiung, Hideji Itokawa and Mutsuo Kozuka 2003a. Oriental herbal products: the basis
for development of dietary supplements and new medicines in the 21st century. ACS
Publications.
Lee, Kyung Jin and Hye Gwang Jeong 2007. Protective effects of kahweol and cafestol against
hydrogen peroxide-induced oxidative stress and DNA damage. Toxicology letters 173: 80-87.
Lee, Mei-Hsien, Yi-Pei Lin, Feng-Lin Hsu, Gui-Rong Zhan and Kun-Ying Yen 2006. Bioactive
constituents of Spatholobus suberectus in regulating tyrosinase-related proteins and mRNA in
HEMn cells. Phytochemistry 67: 1262-1270.
Lee, Sanghyun, Kyoung Soon Kim, Youmie Park, Kuk Hyun Shin and Bak-Kwang Kim 2003b.
In vivo anti-oxidant activities of tectochrysin. Archives of pharmacal research 26: 43-46.
Lee, Taek Hwan, Lalita Subedi, Hussain Mustatab Wahedi, Yong Un Park and Sun Yeou Kim
2016. Resveratrol-enriched rice protects human skin against UVB-induced photoaging. The
FASEB Journal 30: lb557-lb557.
Li, Xican, Xiaozhen Wang, Dongfeng Chen and Shuzhi Chen 2011. Antioxidant activity and
mechanism of protocatechuic acid in vitro. Functional Foods in Health and Disease 1: 232-244.
Libiseller, Gunnar, Michaela Dvorzak, Ulrike Kleb, Edgar Gander, Tobias Eisenberg, Frank
Madeo, Steffen Neumann, Gert Trausinger, Frank Sinner and Thomas Pieber 2015. IPO: a tool
for automated optimization of XCMS parameters. BMC bioinformatics 16: 118.
113
Lin, Chih-Chien, Chao-Hsun Yang, Pey-Shiuan Wu, Chang-Chin Kwan and Yi-Shyan Chen
2011. Antimicrobial, anti-tyrosinase and antioxidant activities of aqueous aromatic extracts from
forty-eight selected herbs. Journal of Medicinal Plants Research 5: 6203-6209.
Lino, CS, ML Taveira, GSB Viana and FJA Matos 1997. Analgesic and antiinflammatory
activities of Justicia pectoralis Jacq and its main constituents: coumarin and umbelliferone.
Phytotherapy Research 11: 211-215.
Lu, Yao-Ping, You-Rong Lou, Jian-Guo Xie, Qing-Yun Peng, Jie Liao, Chung S Yang, Mou-
Tuan Huang and Allan H Conney 2002. Topical applications of caffeine or (−)-epigallocatechin
gallate (EGCG) inhibit carcinogenesis and selectively increase apoptosis in UVB-induced skin
tumors in mice. Proceedings of the National Academy of Sciences 99: 12455-12460.
Ludwig, Iziar A, Lidia Sanchez, Bettina Caemmerer, Lothar W Kroh, M Paz De Peña and
Concepción Cid 2012. Extraction of coffee antioxidants: impact of brewing time and method.
Food Research International 48: 57-64.
Lv, Hongming, Zhenxiang Yu, Yuwei Zheng, Lidong Wang, Xiaofeng Qin, Genhong Cheng and
Xinxin Ci 2016. Isovitexin exerts anti-inflammatory and anti-oxidant activities on
lipopolysaccharide-induced acute lung injury by inhibiting MAPK and NF-κB and activating
HO-1/Nrf2 pathways. International journal of biological sciences 12: 72.
MacDonald‐Wicks, Lesley K, Lisa G Wood and Manohar L Garg 2006. Methodology for the
determination of biological antioxidant capacity in vitro: a review. Journal of the Science of
Food and Agriculture 86: 2046-2056.
Mahal, HS and T Mukherjee 2006. Scavenging of reactive oxygen radicals by resveratrol:
antioxidant effect. Research on chemical intermediates 32: 59-71.
Makni, Mohamed, Yassine Chtourou, Hamadi Fetoui, El Mouldi Garoui, Tahia Boudawara and
Najiba Zeghal 2011. Evaluation of the antioxidant, anti-inflammatory and hepatoprotective
properties of vanillin in carbon tetrachloride-treated rats. European journal of pharmacology 668:
133-139.
Manach, Claudine, Augustin Scalbert, Christine Morand, Christian Rémésy and Liliana Jiménez
2004. Polyphenols: food sources and bioavailability. The American journal of clinical nutrition
79: 727-747.
Momtaz, S, BM Mapunya, PJ Houghton, C Edgerly, A Hussein, S Naidoo and N Lall 2008.
Tyrosinase inhibition by extracts and constituents of Sideroxylon inerme L. stem bark, used in
South Africa for skin lightening. Journal of ethnopharmacology 119: 507-512.
114
Monente, Carmen, Jimena Bravo, Ana Isabel Vitas, Leire Arbillaga, M Paz De Peña and
Concepción Cid 2015. Coffee and spent coffee extracts protect against cell mutagens and inhibit
growth of food-borne pathogen microorganisms. Journal of Functional Foods 12: 365-374.
Moon, Joon-Kwan and Takayuki Shibamoto 2009. Antioxidant assays for plant and food
components. Journal of agricultural and food chemistry 57: 1655-1666.
Moore, Kevin and L Jackson Roberts 1998. Measurement of lipid peroxidation. Free radical
research 28: 659-671.
Mori, Akihisa, Chikao Nishino, Nobuyasu Enoki and Shinkichi Tawata 1987. Antibacterial
activity and mode of action of plant flavonoids against Proteus vulgaris and Staphylococcus
aureus. Phytochemistry 26: 2231-2234.
Moure, Andrés, Manuel Pazos, Isabel Medina, Herminia Domínguez and Juan Carlos Parajó
2007. Antioxidant activity of extracts produced by solvent extraction of almond shells acid
hydrolysates. Food Chemistry 101: 193-201.
Mu, H, Y-H Bai, S-T Wang, Z-M Zhu and Y-W Zhang 2009. Research on antioxidant effects
and estrogenic effect of formononetin from Trifolium pratense (red clover). Phytomedicine 16:
314-319.
Mussatto, Solange I, Lina F Ballesteros, Silvia Martins and José A Teixeira 2011a. Extraction of
antioxidant phenolic compounds from spent coffee grounds. Separation and Purification
Technology 83: 173-179.
Mussatto, Solange I, Ercília MS Machado, Silvia Martins and José A Teixeira 2011b.
Production, composition, and application of coffee and its industrial residues. Food and
Bioprocess Technology 4: 661.
Nichols, Joi A and Santosh K Katiyar 2010. Skin photoprotection by natural polyphenols: anti-
inflammatory, antioxidant and DNA repair mechanisms. Archives of dermatological research
302: 71-83.
Nishijima, Setsuko, Ichiro Kurokawa, Naoki Katoh and Kunitomo Watanabe 2000. The
bacteriology of acne vulgaris and antimicrobial susceptibility of Propionibacterium acnes and
Staphylococcus epidermidis isolated from acne lesions. The Journal of dermatology 27: 318-323.
Noda, Yasuko, Takao Kaneyuki, Akitane Mori and Lester Packer 2002. Antioxidant activities of
pomegranate fruit extract and its anthocyanidins: delphinidin, cyanidin, and pelargonidin.
Journal of agricultural and food chemistry 50: 166-171.
Núñez-Sellés, Alberto J 2005. Antioxidant therapy: myth or reality? Journal of the brazilian
Chemical society 16: 699-710.
115
Oh, Kyung-Tack, Jae-Won Ka, Hyon-Pyo Kim and Chang-Hee Lee 1997. Syntheses of
Porphyrin-Isoflavonoid Conjugates and Octa-Substituted Porphyrins as a Building Blocks for 3-
Dimensional Array of Porphyrins. BULLETIN-KOREAN CHEMICAL SOCIETY 18: 907-910.
Ohemeng, KA, CF Schwender, KP Fu and JF Barrett 1993. DNA gyrase inhibitory and
antibacterial activity of some flavones (1). Bioorganic & Medicinal Chemistry Letters 3: 225-
230.
Ohta, Toshihiro, Kazuko Watanabe, Masaaki Moriya, Yasuhiko Shirasu and Tsuneo Kada 1983.
Anti-mutagenic effects of coumarin and umbelliferone on mutagenesis induced by 4-
nitroquinoline 1-oxide or UV irradiation in E. coli. Mutation Research/Genetic Toxicology 117:
135-138.
Oi, Naomi, Hanyong Chen, Myoung Ok Kim, Ronald A Lubet, Ann M Bode and Zigang Dong
2012. Taxifolin suppresses UV-induced skin carcinogenesis by targeting EGFR and PI3K.
Cancer Prevention Research 5: 1103-1114.
Pandey, Kanti B and Syed Ibrahim Rizvi 2009. Current understanding of dietary polyphenols
and their role in health and disease. Current Nutrition & Food Science 5: 249-263.
Park, Hye Young, Gi-Young Kim, Jin Won Hyun, Hye Jin Hwang, Nam Deuk Kim, Byung-Woo
Kim and Yung Hyun Choi 2012. 7, 8-Dihydroxyflavone exhibits anti-inflammatory properties by
downregulating the NF-κB and MAPK signaling pathways in lipopolysaccharide-treated
RAW264. 7 cells. International journal of molecular medicine 29: 1146-1152.
Park, Kwan Hee, Min Park, Sun Eun Choi, Mi Sook Jeong, Joo Hee Kwon, Myung Hwan Oh,
Hyung Kyun Choi, Seong Jun Seo and Min Won Lee 2009. The anti-oxidative and anti-
inflammatory effects of caffeoyl derivatives from the roots of Aconitum koreanum R.
RAYMOND. Biological and pharmaceutical bulletin 32: 2029-2033.
Perchellet, Jean-Pierre, Hala U Gali, Elisabeth M Perchellet, Darren S Klish and Andrew D
Armbrust 1992. Antitumor-promoting activities of tannic acid, ellagic acid, and several gallic
acid derivatives in mouse skin. In Plant Polyphenols, 783-801: Springer.
Pereira, José Alberto, Ivo Oliveira, Anabela Sousa, Patrícia Valentão, Paula B Andrade, Isabel
CFR Ferreira, Federico Ferreres, Albino Bento, Rosa Seabra and Letícia Estevinho 2007. Walnut
(Juglans regia L.) leaves: phenolic compounds, antibacterial activity and antioxidant potential of
different cultivars. Food and Chemical Toxicology 45: 2287-2295.
Pero, Ronald W, Harald Lund and Tomas Leanderson 2009. Antioxidant metabolism induced by
quinic acid. Increased urinary excretion of tryptophan and nicotinamide. Phytotherapy Research
23: 335-346.
116
Petrovic, Mira, Meritxell Gros and Damia Barcelo 2006. Multi-residue analysis of
pharmaceuticals in wastewater by ultra-performance liquid chromatography–quadrupole–time-
of-flight mass spectrometry. Journal of chromatography A 1124: 68-81.
Potapovich, Alla I, Vladimir A Kostyuk, Tatyana V Kostyuk, Chiara de Luca and Liudmila G
Korkina 2013. Effects of pre-and post-treatment with plant polyphenols on human keratinocyte
responses to solar UV. Inflammation Research 62: 773-780.
Procházková, Dagmar, I Boušová and N Wilhelmová 2011. Antioxidant and prooxidant
properties of flavonoids. Fitoterapia 82: 513-523.
Pulido, Raquel, Laura Bravo and Fulgencio Saura-Calixto 2000. Antioxidant activity of dietary
polyphenols as determined by a modified ferric reducing/antioxidant power assay. Journal of
Agricultural and Food Chemistry 48: 3396-3402.
Qa'dan, Fadi, Abdul-Jalil Thewaini, Dalia A Ali, Rana Afifi, Abdalla Elkhawad and Khalid Z
Matalka 2005. The antimicrobial activities of Psidium guajava and Juglans regia leaf extracts to
acne-developing organisms. The American Journal of Chinese Medicine 33: 197-204.
Ramalakshmi, K, L Jagan Mohan Rao, Yuko Takano-Ishikawa and Masao Goto 2009.
Bioactivities of low-grade green coffee and spent coffee in different in vitro model systems.
Food Chemistry 115: 79-85.
Raman, A, U Weir and SF Bloomfield 1995. Antimicrobial effects of tea‐tree oil and its major
components on Staphylococcus aureus, Staph. epidermidis and Propionibacterium acnes. Letters
in Applied Microbiology 21: 242-245.
Ramani, Rama and Vishnu Chaturvedi 2000. Flow cytometry antifungal susceptibility testing of
pathogenic yeasts other than Candida albicans and comparison with the NCCLS broth
microdilution test. Antimicrobial agents and chemotherapy 44: 2752-2758.
Ramesh, B and KV Pugalendi 2006. Antioxidant role of umbelliferone in STZ-diabetic rats. Life
sciences 79: 306-310.
Ramos, Freddy A, Yoshihisa Takaishi, Miki Shirotori, Yousuke Kawaguchi, Koichiro Tsuchiya,
Hirofumi Shibata, Tomihiko Higuti, Tetsuo Tadokoro and Minoru Takeuchi 2006. Antibacterial
and antioxidant activities of quercetin oxidation products from yellow onion (Allium cepa) skin.
Journal of agricultural and food chemistry 54: 3551-3557.
Rauf, Abdur, Rehan Khan, Haroon Khan, Samreen Pervez and Abdul Saboor Pirzada 2014. In
vivo antinociceptive and anti-inflammatory activities of umbelliferone isolated from Potentilla
evestita. Natural product research 28: 1371-1374.
117
Razzaghi-Abyaneh, Mehdi and Mahendra Rai 2013. Antifungal metabolites from plants:
Springer.
Re, Roberta, Nicoletta Pellegrini, Anna Proteggente, Ananth Pannala, Min Yang and Catherine
Rice-Evans 1999. Antioxidant activity applying an improved ABTS radical cation decolorization
assay. Free Radical Biology and Medicine 26: 1231-1237.
Ren, Xiaolin, Yuling Shi, Di Zhao, Mengyu Xu, Xiaolong Li, Yongyan Dang and Xiyun Ye
2016. Naringin protects ultraviolet B-induced skin damage by regulating p38 MAPK signal
pathway. Journal of dermatological science 82: 106-114.
Ribeiro, Helena, Joana Marto, Sara Raposo, Maria Agapito, Vera Isaac, Bruna G Chiari, Pedro F
Lisboa, Alexandre Paiva, Susana Barreiros and Pedro Simões 2013. From coffee industry waste
materials to skin‐friendly products with improved skin fat levels. European journal of lipid
science and technology 115: 330-336.
Rice-Evans, Catherine A, Nicholas J Miller and George Paganga 1996. Structure-antioxidant
activity relationships of flavonoids and phenolic acids. Free Radical Biology and Medicine 20:
933-956.
Ruff, Matthias, Miriam S Mueller, Martin Loos and Heinz P Singer 2015. Quantitative target and
systematic non-target analysis of polar organic micro-pollutants along the river Rhine using
high-resolution mass-spectrometry–identification of unknown sources and compounds. Water
research 87: 145-154.
Rufian-Henares, Jose A and Silvia P de la Cueva 2009. Antimicrobial Activity of Coffee
Melanoidins A Study of Their Metal-Chelating Properties. Journal of agricultural and food
chemistry 57: 432-438.
Runyoro, Deborah KB, Mecky IN Matee, Olipa D Ngassapa, Cosam C Joseph and Zakaria H
Mbwambo 2006. Screening of Tanzanian medicinal plants for anti-Candida activity. BMC
complementary and alternative medicine 6: 11.
Russo, Alessandra, Venera Cardile, Laura Lombardo, Luca Vanella and Rosaria Acquaviva
2006. Genistin inhibits UV light-induced plasmid DNA damage and cell growth in human
melanoma cells. The Journal of nutritional biochemistry 17: 103-108.
Ryu, Jae Sik, Hyun Jung Lee, Song Hwan Bae, Sun Young Kim, Yooheon Park, Hyung Joo Suh
and Yoon Hwa Jeong 2013. The bioavailability of red ginseng extract fermented by Phellinus
linteus. Journal of ginseng research 37: 108.
118
Saavedra, Maria J, Anabela Borges, Carla Dias, Alfredo Aires, Richard N Bennett, Eduardo S
Rosa and Manuel Simões 2010. Antimicrobial activity of phenolics and glucosinolate hydrolysis
products and their synergy with streptomycin against pathogenic bacteria. Medicinal Chemistry
6: 174-183.
Saghaie, Lotfollah, Morteza Pourfarzam, A Fassihi and B Sartippour 2013. Synthesis and
tyrosinase inhibitory properties of some novel derivatives of kojic acid. Research in
pharmaceutical sciences 8: 233.
Saija, Antonella, Mario Scalese, Maria Lanza, Daniela Marzullo, Francesco Bonina and
Francesco Castelli 1995. Flavonoids as antioxidant agents: importance of their interaction with
biomembranes. Free Radical Biology and Medicine 19: 481-486.
Sakano, Katsuhisa, Mika Mizutani, Mariko Murata, Shinji Oikawa, Yusuke Hiraku and Shosuke
Kawanishi 2005. Procyanidin B2 has anti-and pro-oxidant effects on metal-mediated DNA
damage. Free Radical Biology and Medicine 39: 1041-1049.
Sanchez-Moreno, C 2002. Methods used to evaluate the free radical scavenging activity in foods
and biological systems. Revista de Agaroquimica y Tecnologia de Alimentos 8: 121-137.
Sansanelli, Sabrina, Dario Zanichelli, Alessandro Filippini, Maura Ferri and Annalisa Tassoni
2014. Production of free and glycosylated isoflavones in in vitro soybean (Glycine max L.)
hypocotyl cell suspensions and comparison with industrial seed extracts. Plant Cell, Tissue and
Organ Culture (PCTOC) 119: 301-311.
Saraf, S and CD Kaur 2010. Phytoconstituents as photoprotective novel cosmetic formulations.
Pharmacognosy reviews 4: 1.
Sato, Yuki, Shirou Itagaki, Toshimitsu Kurokawa, Jiro Ogura, Masaki Kobayashi, Takeshi
Hirano, Mitsuru Sugawara and Ken Iseki 2011. In vitro and in vivo antioxidant properties of
chlorogenic acid and caffeic acid. International journal of pharmaceutics 403: 136-138.
Schmid, Daniel, F Zulli, Hans-Peter Nissen and Hanspeter Prieur 2003. Penetration and
metabolism of isoflavones in human skin. Cosmetics and toiletries 118: 71-76.
Schmid, Daniel and Fred ZÜLLI 2002. Topically applied soy isoflavones increase skin
thickness. Cosmetics and toiletries 117: 45-50.
Seeram, N, Rupo Lee, M Hardy and D Heber 2005. Rapid large scale purification of
ellagitannins from pomegranate husk, a by-product of the commercial juice industry. Separation
and Purification Technology 41: 49-55.
Seo, Ga Young, Sanggyu Park, Jung-Sik Huh and Moonjae Cho 2014. The protective effect of
glycitin on UV-induced skin photoaging in human primary dermal fibroblast. Journal of the
Korean Society for Applied Biological Chemistry 57: 463-468.
119
Seo, YK, SJ Kim, YC Boo, JH Baek, SH Lee and JS Koh 2011. Effects of p‐coumaric acid on
erythema and pigmentation of human skin exposed to ultraviolet radiation. Clinical and
experimental dermatology 36: 260-266.
SHIBASAKI, ISAO 1982. Food preservation with nontraditional antimicrobial agents. Journal of
Food Safety 4: 35-58.
Si, Weiduo, Joshua Gong, Rong Tsao, Milosh Kalab, Raymond Yang and Yulong Yin 2006.
Bioassay-guided purification and identification of antimicrobial components in Chinese green tea
extract. Journal of chromatography A 1125: 204-210.
Silveira, Joana Léa Meira and Tania Mari Bellé Bresolin 2011. Pharmaceutical use of
galactomannans. Química Nova 34: 292-299.
Singh, Rajbir, Bikram Singh, Sukhpreet Singh, Neeraj Kumar, Subodh Kumar and Saroj Arora
2010. Umbelliferone–An antioxidant isolated from Acacia nilotica (L.) Willd. ex. Del. Food
Chemistry 120: 825-830.
Slavtcheff, Craig, Michael Brennan, Sandra Meraviglia, Jessica Goldberg and Michael Massaro
2005. Self-supporting aerosol cleansing composition. In Self-supporting aerosol cleansing
composition: Google Patents.
Smith, Colin A 2013. LC/MS Preprocessing and Analysis with xcms. In LC/MS Preprocessing
and Analysis with xcms.
Smith, Colin A, Elizabeth J Want, Grace O'Maille, Ruben Abagyan and Gary Siuzdak 2006.
XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak
alignment, matching, and identification. Analytical chemistry 78: 779-787.
Sroka, Z and W Cisowski 2003. Hydrogen peroxide scavenging, antioxidant and anti-radical
activity of some phenolic acids. Food and Chemical Toxicology 41: 753-758.
Svobodová, Alena, Jitka Psotová and Daniela Walterová 2003. Natural phenolics in the
prevention of UV-induced skin damage. A review. Biomed Pap Med Fac Univ Palacky Olomouc
Czech Repub 147: 137-145.
Tabassum, Nadia, Ji-Hyung Lee, Soon-Ho Yim, Galzad Javzan Batkhuu, Da-Woon Jung and
Darren R Williams 2016. Isolation of 4, 5-O-dicaffeoylquinic acid as a pigmentation inhibitor
occurring in artemisia capillaris thunberg and its validation in vivo. Evidence-Based
Complementary and Alternative Medicine 2016.
120
Takagi, Mio, Yosuke Tachi, Jie Zhang, Takuro Shinozaki, Kenta Ishii, Takashi Kikuchi,
Motohiko Ukiya, Norihiro Banno, Harukuni Tokuda and Toshihiro Akihisa 2014. Cytotoxic and
Melanogenesis‐Inhibitory Activities of Limonoids from the Leaves of Azadirachta indica
(Neem). Chemistry & biodiversity 11: 451-468.
Tautenhahn, Ralf, Christoph Boettcher and Steffen Neumann 2008. Highly sensitive feature
detection for high resolution LC/MS. BMC bioinformatics 9: 504.
Tautenhahn, Ralf, Gary J Patti, Duane Rinehart and Gary Siuzdak 2012. XCMS Online: a web-
based platform to process untargeted metabolomic data. Analytical chemistry 84: 5035.
Thaipong, Kriengsak, Unaroj Boonprakob, Kevin Crosby, Luis Cisneros-Zevallos and David
Hawkins Byrne 2006. Comparison of ABTS, DPPH, FRAP, and ORAC assays for estimating
antioxidant activity from guava fruit extracts. Journal of food composition and analysis 19: 669-
675.
Tomás‐Barberán, Francisco A and Michael N Clifford 2000. Flavanones, chalcones and
dihydrochalcones–nature, occurrence and dietary burden. Journal of the Science of Food and
Agriculture 80: 1073-1080.
Torel, Joseph, Josiane Cillard and Pierre Cillard 1986. Antioxidant activity of flavonoids and
reactivity with peroxy radical. Phytochemistry 25: 383-385.
Tseng, Tsui-Hwa, Chau-Jong Wang and Erl-Shyh Kao 1996. Hibiscus protocatechuic acid
protects against oxidative damage induced by tert-butylhydroperoxide in rat primary
hepatocytes. Chemico-Biological Interactions 101: 137-148.
Tsuchihashi, Hideyasu, Makoto Kigoshi, Misato Iwatsuki and Etsuo Niki 1995. Action of β-
carotene as an antioxidant against lipid peroxidation. Archives of biochemistry and biophysics
323: 137-147.
Ullrich, SE 1995. Potential for immunotoxicity due to environmental exposure to ultraviolet
radiation. Human & experimental toxicology 14: 89-91.
Vignoli, JA, DG Bassoli and MT Benassi 2011. Antioxidant activity, polyphenols, caffeine and
melanoidins in soluble coffee: The influence of processing conditions and raw material. Food
Chemistry 124: 863-868.
Vittar, Natalia Belen Rumie, Laura Comini, Ivana Maria Fernadez, Elizabeth Agostini, Susana
Nuñez-Montoya, Jose Luis Cabrera and Viviana Alicia Rivarola 2014. Photochemotherapy using
natural anthraquinones: Rubiadin and Soranjidiol sensitize human cancer cell to die by apoptosis.
Photodiagnosis and photodynamic therapy 11: 182-192.
121
Wang, Haibo, Muraleedharan G Nair, Gale M Strasburg, Yu-Chen Chang, Alden M Booren, J
Ian Gray and David L DeWitt 1999. Antioxidant and antiinflammatory activities of anthocyanins
and their aglycon, cyanidin, from tart cherries. Journal of natural products 62: 294-296.
Wang, Jun, Lu-Lu Zhao, Guo-Xia Sun, Yao Liang, Fu-An Wu, ZI Chen and SM Cui 2011. A
comparison of acidic and enzymatic hydrolysis of rutin. African Journal of Biotechnology 10:
1460-1466.
Wang, Shiow Y and Hsin-Shan Lin 2000. Antioxidant activity in fruits and leaves of blackberry,
raspberry, and strawberry varies with cultivar and developmental stage. Journal of agricultural
and food chemistry 48: 140-146.
Wen, Bo, Zhanlong Mei, Chunwei Zeng and Siqi Liu 2017. metaX: a flexible and
comprehensive software for processing metabolomics data. BMC bioinformatics 18: 183.
Williams, Christine A, Jeffrey B Harborne, Hans Geiger and J Robin S Hoult 1999. The
flavonoids of Tanacetum parthenium and T. vulgare and their anti-inflammatory properties.
Phytochemistry 51: 417-423.
Wölfle, Ute, Philipp R Esser, Birgit Simon-Haarhaus, Stefan F Martin, Jürgen Lademann and
Christoph M Schempp 2011. UVB-induced DNA damage, generation of reactive oxygen species,
and inflammation are effectively attenuated by the flavonoid luteolin in vitro and in vivo. Free
Radical Biology and Medicine 50: 1081-1093.
Wu, Nan-Lin, Jia-You Fang, Marcelo Chen, Chia-Jung Wu, Chieh-Chen Huang and Chi-Feng
Hung 2011. Chrysin protects epidermal keratinocytes from UVA-and UVB-induced damage.
Journal of agricultural and food chemistry 59: 8391-8400.
Wu, Tian-Shung, Zhih-Jing Tsang, Pei-Lin Wu, Fu-Wen Lin, Chia-Yin Li, Che-Ming Teng and
Kuo-Hsiung Lee 2001. New constituents and antiplatelet aggregation and anti-HIV principles of
Artemisia capillaris. Bioorganic & medicinal chemistry 9: 77-83.
Xiao, Cheng, Jian Li, Xinxin Dong, Xiaojuan He, Xuyan Niu, Cha Liu, Guoyue Zhong, Rudolf
Bauer, Dajian Yang and Aiping Lu 2011. Anti-oxidative and TNF-α suppressive activities of
puerarin derivative (4AC) in RAW264. 7 cells and collagen-induced arthritic rats. European
journal of pharmacology 666: 242-250.
Xing, Jing, Rui Li, Nan Li, Jian Zhang, Yueqing Li, Ping Gong, Dongna Gao, Hui Liu and Yu
Zhang 2015. Anti-inflammatory effect of procyanidin B1 on LPS-treated THP1 cells via
interaction with the TLR4–MD-2 heterodimer and p38 MAPK and NF-κB signaling. Molecular
and cellular biochemistry 407: 89-95.
122
Yamakoshi, Jun, Fujio Otsuka, Atsushi Sano, Shoichi Tokutake, Makoto Saito, Mamoru Kikuchi
and Yoshiro Kubota 2003. Lightening effect on ultraviolet‐induced pigmentation of guinea pig
skin by oral administration of a proanthocyanidin‐rich extract from grape seeds. Pigment Cell &
Melanoma Research 16: 629-638.
Yanagimoto, Kenichi, Hirotomo Ochi, Kwang-Geun Lee and Takayuki Shibamoto 2004.
Antioxidative activities of fractions obtained from brewed coffee. Journal of agricultural and
food chemistry 52: 592-596.
Yoon, Ji Hye, Tae-Gyu Lim, Kyung Mi Lee, Ae Ji Jeon, Su Yeon Kim and Ki Won Lee 2010.
Tangeretin reduces ultraviolet B (UVB)-induced cyclooxygenase-2 expression in mouse
epidermal cells by blocking mitogen-activated protein kinase (MAPK) activation and reactive
oxygen species (ROS) generation. Journal of agricultural and food chemistry 59: 222-228.
Zhao, Mouming, Haiyan Wang, Bao Yang and Hong Tao 2010. Identification of cyclodextrin
inclusion complex of chlorogenic acid and its antimicrobial activity. Food Chemistry 120: 1138-
1142.
Zhu, Qin Yan, Yu Huang and Zhen-Yu Chen 2000. Interaction between flavonoids and α-
tocopherol in human low density lipoprotein. The Journal of nutritional biochemistry 11: 14-21.
Zuorro, Antonio and Roberto Lavecchia 2012. Spent coffee grounds as a valuable source of
phenolic compounds and bioenergy. Journal of Cleaner Production 34: 49-56.
123
Appendix A.
Table 3.1. Phenolic acids, flavonoids, and other bioactive compounds including coffee alkaloids, phenylpropanoids, stilbenoids, and anthraquinones
extracted from three different coffee wastes (Ethiopian Yirgacheffe, Costa Rican Tarrazu, and Hawaiian Kona). Maximum intensity of peaks detected by
XCMS online comparing the relative intensity between the three cultivars of SCG are shown.
Compound name m/z RT Ethiopian Costa
Rican Hawaiian Bioactivity
E-Cinnamic acid 149.0604
16.76 1,351,108 1,099,826 1,535,870 Anti-tyrosinase,
antioxidant,
antimicrobial
19.35 41,746 x 40,944
19.61 34,294 x 34,294
4.25 37,560 38,420 x
3,5-Di-tert-butyl-4-
hydroxycinnamic acid, (E)- 299.1617 21.59 346,552 329,470 334,008
3,5-Di-tert-butyl-4-
hydroxycinnamic acid, (E)- 299.1617 23.77 418,452 419,996
3,4-Difluorohydrocinnamic acid 187.0569 0.74 927,880 927,880 x
2-ISOPROPYL-3-
METHOXYCINNAMIC ACID 221.1168 21.59 120,608 120,608 116,090
Hydrocinnamic acid 151.0753 14.21 49,678 x 52,386 Flavoring agent
3,5-Di-tert-butyl-4-
hydroxycinnamic acid, (E)- 299.1617 33.5 20,938 20,790 23,652
Hydrocinnamic acid 151.0755 23.79 40,730 42,226 42,226 Flavoring agent
Hydrocinnamic acid, p-(3-(N-
ethylacetamido)-2,4,6-
triiodobenzamido)-
749.8794 18.84 2,894 2,894 3,226
4-(Trifluoromethyl)hydrocinnamic
acid 236.0895 2.63 36,746
124
2-ISOPROPYL-3-
METHOXYCINNAMIC ACID 221.1168 20.38 36,354 36,386 36,354
2-ISOPROPYL-3-
METHOXYCINNAMIC ACID 221.1173 17.57 19,396 19,396 20,822
3,5-Di-tert-butyl-4-
hydroxycinnamic acid, (E)- 277.1786 22.45 34,334 34,334 33,802
o-Methoxyhydrocinnamic acid 181.0848 6.03 29,768 34,924 27,494
3,5-Di-tert-butyl-4-
hydroxycinnamic acid, (E)- 277.1786 21.6 19,322 19,322 19,322
3,5-Di-tert-butyl-4-
hydroxyhydocinnamic acid 279.1944 16.73 28,956
4-Methoxy-(2E)-cinnamic acid 179.0705 9.73 17,846 14,142
Vanillic acid, 5-(4-carboxy-2-
methoxyphenoxy)- 357.0585 3.47 68,432 14,468 20,036
Antioxidant,
anti-tyrosinase
P-coumaroylquinic acid 339.106 1.94 8,388
Coumaric acid 349.0898 4.92 42,442 Anti-
pigmentation,
antioxidant
trans-p-Coumaric acid 4-glucoside 349.0898 4.92 42,442
Anti-
pigmentation,
antioxidant,
antimicrobial
1-Feruloyl-5-caffeoylquinic acid 531.1492 7.72 25,668 22,098
1,4-Dicaffeoylquinic acid 539.1146 5.83 71,610 78,570
1,4-Dicaffeoylquinic acid 539.1149 6.67 78,602 85,892 74,576
1-Feruloyl-5-caffeoylquinic acid 553.1315 7.71 142,308 119,366
1,4-Dicaffeoylquinic acid 517.1335 5.81 16,768
1,3-Dicaffeoylquinic acid 539.1149 6.67 78,602 85,892 74,576
1,5-Dicaffeoylquinic acid 539.1149 6.67 78,602 85,892 74,576
3,5-Di-O-caffeoylquinic acid 539.1149 6.67 78,602 85,892 74,576 Antioxidant,
anti-tyrosinase
125
4,5-Di-O-caffeoylquinic acid 539.1149 6.67 78,602 85,892 74,576 Antioxidant,
antipigmentation
1,4-Di-O-caffeoylquinic acid 539.1149 6.67 78,602 85,892 74,576
1,4-Dicaffeoylquinic acid 539.1146 5.83 71,610 78,570 61,832
Isochlorogenic acid A (3,5-
Dicaffeoylquinic acid) 539.1146 5.83 71,610 78,570 61,832
Antioxidant,
anti-tyrosinase
1,5-Dicaffeoylquinic acid 539.1146 5.83 71,610 78,570 61,832
1,3-Dicaffeoylquinic acid 539.1146 5.83 71,610 78,570 61,832
Isochlorogenic acid A 539.1149 6.67 78,602 85,892 74,576 Antioxidant,
anti-tyrosinase
Isochlorogenic acid b (3,4-
Dicaffeoylquinic acid) 539.1149 6.67 78,602 85,892 74,576
Chlorogenic acid (5-Caffeoylquinic
acid) 355.1025
1.51 x 85,878 x Anti-
inflammatory,
antioxidant,
anti-aging, anti-
skin cancer,
antimicrobial
1.82 x 245,560 200,058
2.29 x x x
3.07 x 134,046 94,340
1-O-Caffeoylquinic acid 355.1025
2.29 x 245,560 x
1.51 x x 200,058
3.07 x 245,560 x
1.82 x 134,046 94,340
Apigenin 8-C-(6''-
acetylgalactoside) 497.1064 6.69 20,560 x 20,746
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 8-C-(6''-
acetylgalactoside) 497.108 5.83 27,832 24,098 22,526
anti-
inflammatory,
antioxidant,
126
anticancer,
UVinduced skin
carcinogenesis
Apigenin 4'-[feruloyl-(->2)-
glucuronyl-(1->2)-glucuronide] 7-
glucuronide
992.2384 22.72 2,538 2,208 3,108
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-[feruloyl-(->2)-
glucuronyl-(1->2)-glucuronide] 4'-
glucuronide
992.2384 22.72 2,538 2,208 3,108
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-[feruloyl-(->2)-
[glucuronyl-(1->3)]-glucuronyl-(1-
>2)-glucuronide]
992.2384 22.72 2,538 2,208 3,108
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-rutinoside-4'-trans-
caffeate 779.1633 27.69 2,454 3,200 3,194
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-rhamnoside 439.0999 10.98 21,104 x 19,424
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
127
Apigenin 7-(2'',6''-di-p-
coumarylglucoside) 763.1482 20.24 2,718 4,206 3,402
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-(3'',6''-di-p-
coumarylglucoside) 763.1482 20.24 2,718 4,206 3,402
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-(4'',6''-di-p-
coumarylglucoside) 763.1482 20.24 2,718 4,206 3,402
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-(3'',6''-Di-E-p-
coumaroylgalactoside) 763.1482 20.24 2,718 4,206 3,402
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-(2'',3''-
diacetylglucoside) 539.1149 6.67 78,602 85,892 74,576
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-(3'',4''-
diacetylglucoside) 539.1149 6.67 78,602 85,892 74,576
anti-
inflammatory,
antioxidant,
128
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-O-(6-O-
malonylglucoside) 558.0948 34.86 242,958 x x
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-(2'',3''-
diacetylglucoside) 539.1146 5.83 71,610 78,570 61,832
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-(3'',4''-
diacetylglucoside) 539.1146 5.83 71,610 78,570 61,832
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-neohesperidoside-4'-
sophoroside 925.254 35.68 40,232 36,984 63,418
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Apigenin 7-(6'''-acetylallosyl-(1-
>2)glucoside) 659.1566 32.85 75,804 27,948 55,172
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
129
Apigenin 7-(6''-acetylalloside)-4'-
alloside 659.1566 32.85 75,804 27,948 55,172
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
8-C-beta-D-
Glucofuranosylapigenin 2''-O-
acetate
497.108 5.83 27,832 24,098 22,526
anti-
inflammatory,
antioxidant,
anticancer,
UVinduced skin
carcinogenesis
Fujikinetin 7-O-glucoside 497.108 5.83 27,832 24,098 22,526
Genistin 6''-O-acetate 497.108 5.83 27,832 24,098 22,526 anti-
carcinogeninc
Vitexin 2''-acetate 497.108 5.83 27,832 24,098 22,526
Antioxidative,
anti-
inflammatory
Isovitexin 6''-O-acetyl 497.108 5.83 27,832 24,098 22,526
Vitexin 3''-O-acetate 497.108 5.83 27,832 24,098 22,526
Antioxidative,
anti-
inflammatory
Vitexin 6''-O-acetate 497.108 5.83 27,832 24,098 22,526
Antioxidative,
anti-
inflammatory
Isovitexin 2''-O-(6'''-(E)-p-
coumaroyl)glucoside 779.1633 27.69 2,454 3,200 3,194
Antioxidant,
anti-
inflammatory
Vitexin 7-O-glucoside 2''-p-
coumarate 779.1633 27.69 2,454 3,200 3,194
Antioxidative,
anti-
inflammatory
130
Isovitexin 7-O-(6'''-O-E-p-
coumaroyl)glucoside 779.1633 27.69 2,454 3,200 3,194
Antioxidant,
anti-
inflammatory
Daidzein 4'-O-glucoside 439.0999 10.98 21,104 x 19,424
Anti-
inflammatory,
anti-carcinogenic
3',4'-Dihydroxyflavone 4'-glucoside 439.0999 21,104 x 19,424
Anti-
inflammatory,
anti-
carcinogenic,
anti-
inflammatory
Chrysin 8-C-beta-D-
glucopyranoside 439.0999 10.98 21,104 x 19,424
anti-
inflammation,
anti-cancer, and
anti-oxidation
Chrysin 6-C-beta-D-
glucopyranoside 439.0999 10.98 21,104 x 19,424
anti-
inflammation,
anti-cancer, and
anti-oxidation
Chrysin 7-galactoside 439.0999 10.98 21,104 x 19,424
anti-
inflammation,
anti-cancer, and
anti-oxidation
Chrysin-7beta-monoglucoside 439.0992 11 21,104 18,130 19,424
anti-
inflammation,
anti-cancer, and
anti-oxidation
Daidzein-4,7-diglucoside 601.1539 15.2 5,336 4,514 4,628
Anti-
inflammatory,
anti-carcinogenic
131
Vitexin-2''-rhamnoside 601.1539 15.2 5,336 4,514 4,628
Antioxidative,
anti-
inflammatory
Vitexin rhamnoside 601.1539 15.2 5,336 4,514 4,628
Antioxidative,
anti-
inflammatory
Puerarin 4'-O-glucoside 601.1539 15.2 5,336 4,514 4,628
Antioxidant,
anti-
inflammatory
Daidzein 7,4'-di-O-glucoside 601.1539 15.2 5,336 4,514 4,628
Anti-
inflammatory,
anti-carcinogenic
Genistein 7-O-rutinoside 601.1539 15.2 5,336 4,514 4,628
Anti-
inflammatory,
antioxidant and
anticancer
Genistein 4'-O-neohesperidoside 601.1539 15.2 5,336 4,514 4,628
Anti-
inflammatory,
antioxidant and
anticancer
Chrysin 7-gentiobioside 601.1539 15.2 5,336 4,514 4,628
Skin protection,
anti-
inflammation
and anti-
oxidation
Isovitexin 2''-O-rhamnoside 601.1539 15.2 5,336 4,514 4,628
Antioxidant,
anti-
inflammatory
Coumestrol 3-O-glucoside 469.0529 29.75 2,420 2,684 2,380 Anti-
inflammatory
Chrysin 7-glucuronide 469.0529 29.75 2,420 2,684 2,380
Skin protection,
anti-
inflammation
132
and anti-
oxidation
Daidzein 4'-O-glucuronide 469.0529 29.75 2,420 2,684 2,380
Anti-
inflammatory,
anti-carcinogenic
Daidzein 7-O-glucuronide 469.0529 29.75 2,420 2,684 2,380
Anti-
inflammatory,
anti-carcinogenic
7,8-DIHYDROXYFLAVONE 255.0655 20.28 8,512 6,930 7,272
Anti-
inflammatory,
antioxidant
3,7-DIHYDROXYFLAVONE 255.0655 20.28 8,512 6,930 7,272 anticancer
Eriodictyol (5,7,3′,4′-
tetrahydroxylflavanone)
Anti-
inflammatory,
Anticancer,
antioxidant
6-Methoxyeriodictyol 319.0811 3.78 30,916 28,356 Anti-tyrosinase
Homoeriodictyolchalcone 2'-
glucoside 465.1413 24.3 15,284 19,588
Daidzein 255.0655 20.28 8,512 6,930 7,272
Anti-
inflammatory,
anti-carcinogenic
Chrysin 255.0655 20.28 8,512 6,930 7,272
Skin protection,
anti-
inflammation
and anti-
oxidation
3,2'-Dihydroxyflavone 255.0655 20.28 8,512 6,930 7,272
2',3'-Dihydroxyflavone 255.0655 20.28 8,512 6,930 7,272
3,3'-DIHYDROXYFLAVONE 255.0655 20.28 8,512 6,930 7,272
6,2'-Dihydroxyflavone 255.0655 20.28 8,512 6,930 7,272
133
Tectochrysin 291.0627 8.42 22,418 x x Antioxidant,
anticancer
Glycitin 464.1547 23.9 3,750 x x
Anti-photoaging,
Anti-
inflammatory,
Anti-aging
Retusin 7-O-glucoside 464.1547 23.9 3,750 x x
8-Hydroxyluteolin 4'-methyl ether
7-(6'''-acetylallosyl)(1->2)(6''-
acetylglucoside)
763.1482 20.24 2,718 4,206 3,402
Isorhamnetin 3-(4''',6'''-
diacetylglucosyl) (1->3)-
galactoside
763.1482 20.24 2,718 4,206 3,402
Anti-
inflammatory,
Anticancer
2',2'-BISEPIGALLOCATECHIN
MONOGALLATE 763.1482 20.24 2,718 4,206 3,402
8-Hydroxyluteolin 4'-methyl ether
7-(6'''-acetylallosyl)(1->2)(6''-
acetylglucoside)
747.1806 6.48 5,678 5,356 4,544
Isorhamnetin 3-(4''',6'''-
diacetylglucosyl) (1->3)-
galactoside
747.1806 6.48 5,678 5,356 4,544
Anti-
inflammatory,
Anticancer
Rhamnocitrin 3-(5'''-p-
coumarylapiosyl)-(1->2)-glucoside 779.1633 27.69 2,454 3,200 3,194
anti-
inflammatory
Formononetin 7-O-glucoside-6''-
malonate 539.1149 6.67 78,602 85,892 74,576 antioxidants
Formononetin 7-(6''-
malonylglucoside) 539.1149 6.67 78,602 85,892 74,576 antioxidants
Formononetin 7-O-glucoside-6''-O-
malonate 539.1149 6.67 78,602 85,892 74,576 antioxidants
Formononetin 7-O-glucoside-6''-
malonate 539.1149 6.67 78,602 85,892 74,576 antioxidants
2'',6''-Di-O-Acetyl isovitexin 539.1149 6.67 78,602 85,892 74,576
134
Formononetin 7-O-glucoside-6''-
malonate 539.1146 5.83 71,610 78,570 61,832 antioxidants
2'',6''-Di-O-Acetyl isovitexin 539.1146 5.83 71,610 78,570 61,832
Formononetin 7-O-glucoside-6''-O-
malonate 539.1146 5.83 71,610 78,570 61,832 antioxidants
Formononetin 7-(6''-
malonylglucoside) 539.1146 5.83 71,610 78,570 61,832 antioxidants
Fujikinetin 7-O-laminaribioside 659.1566 32.85 75,804 27,948 55,172
Quercetin 3-sophoroside-7-
glucuronide 820.2114 4.57 3,178 2,780 2,312
Anti-bacterial,
anti-oxidant,
skin protection,
anti-
melanogenesis
Quercetin 3-gentiobioside-7-
glucuronide 820.2114 4.57 3,178 2,780 2,312
Anti-bacterial,
anti-oxidant,
skin protection,
anti-
melanogenesis
Quercetin 7-glucuronoside 3-
sophoroside 820.2114 4.57 3,178 2,780 2,312
Anti-bacterial,
anti-oxidant,
skin protection,
anti-
melanogenesis
Quercetin 3-(3'',6''-di-p-
coumarylglucoside) 779.1633 27.69 2,454 3,200 3,194
Anti-bacterial,
anti-oxidant,
skin protection,
anti-
melanogenesis
Quercetin 3-(4''-acetylrhamnoside)-
7-rhamnoside 659.1566 32.85 75,804 27,948 55,172
Anti-bacterial,
anti-oxidant,
skin protection,
anti-
melanogenesis
135
Kaempferol 3-(4''-(Z)-p-
coumarylrobinobioside) 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 3-(4''-(E)-p-
coumarylrobinobioside) 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 3-rhamnosyl-(1-
>6)(2''-p-coumarylglucoside) 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 3- [ 6'''-p-
coumarylglucosyl- (1->2) -
rhamnoside ]
779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 3-(2G-(E)-p-
coumaroylrutinoside) 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 2G-
coumaroylrutinoside 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 3-[2''-(p-
coumaroylglucosyl)rhamnoside] 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 3-[4''-(p-
coumaroylglucosyl)rhamnoside] 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 3-[6'''-p-
coumarylglucosyl-(1->2)-
rhamnoside]
779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory
,antioxidant
Kaempferol 3-(2''-galloyl-alpha-L-
arabinopyranoside) 588.139 12.4 4,606 3,658 3,146
Anti-
inflammatory
,antioxidant
6-Hydroxykaempferol 3,6-
diglucoside 7-glucuronide 820.2114 4.57 3,178 2,780 2,312
Anti-tyrosinase,
anti-
inflammatory
136
Kaempferol 3-[2''',3''',5'''-triacetyl-
alpha-L-arabinofuranosyl-(1->6)-
glucoside
707.1755 19.09 4,230 3,716 4,370
Anti-
inflammatory
,antioxidant
Kaempferol 3-(2''-(E)-p-coumaroyl-
alpha-L-arabinofuranoside)-7-
rhamnoside
711.1949 7.54 4,010 4,650 4,184
Anti-
inflammatory
,antioxidant
Kaempferol 3-(2'',4''-di-(E)-p-
coumarylrhamnoside) 763.1482 20.24 2,718 4,206 3,402
Anti-
inflammatory
,antioxidant
Kaempferol 3-(2'',4''-
diacetylrhamnoside) 539.1149 6.67 78,602 85,892 74,576
Anti-
inflammatory
,antioxidant
Kaempferol 3-(2Gal-
glucosylrobinobioside)-7-
rhamnoside
925.254 35.68 40,232 36,984 63,418
Anti-
inflammatory
,antioxidant
Kaempferol 3-rhamnosyl-(1-
>2)[rhamnosyl-(1->6)-galactoside]-
7-glucoside
925.254 35.68 40,232 36,984 63,418
Anti-
inflammatory
,antioxidant
Kaempferol 3-galactosyl-(1->6)-
glucoside-7-rhamnosyl-(1->3)-
rhamnoside
925.254 35.68 40,232 36,984 63,418
Anti-
inflammatory
,antioxidant
Kaempferol 3-O-alpha-L-
rhamnopyranosyl(1->6)-beta-D-
glucopyranosyl(1->2)-beta-D-
glucopyranoside-7-O-alpha-L-
rhamnopyranoside
925.254 35.68 40,232 36,984 63,418
Anti-
inflammatory
,antioxidant
Kaempferol 3-glucosyl-(1->3)-
rhamnosyl-(1->2)-[rhamnosyl-(1-
>6)-galactoside]
925.254 35.68 40,232 36,984 63,418
Anti-
inflammatory
,antioxidant
Kaempferol 3-(6''-
acetylgalactoside)-7-rhamnoside 659.1566 32.85 75,804 27,948 55,172
Anti-
inflammatory
,antioxidant
137
Kaempferol 3-(6''-acetylglucoside)-
7-rhamnoside 659.1566 32.85 75,804 27,948 55,172
Anti-
inflammatory
,antioxidant
Rhamnazin 3-[6''-(3-hydroxy-3-
methylglutaryl)glucoside] 659.1566 32.85 75,804 27,948 55,172
Anti-
inflammatory
,antioxidant
Kaempferol 3-(6''-
acetylgalactoside) 7-rhamnoside 659.1566 32.85 75,804 27,948 55,172
Anti-
inflammatory
,antioxidant
Rhamnazin 3-[6''-(3-hydroxy-3-
methylglutaryl)glucoside] 659.1566 32.85 75,804 27,948 55,172
Kaempferol 3-(2'',4''-
diacetylrhamnoside) 539.1146 5.83 71,610 78,570 61,832
Anti-
inflammatory
,antioxidant
Kaempferol 3-(3'',4''-
diacetylrhamnoside) 539.1146 5.83 71,610 78,570 61,832
Anti-
inflammatory
,antioxidant
Kaempferol 3-[glucosyl-(1->3)-
rhamnosyl-(1->2)-[rhamnosyl-(1-
>6)-galactoside]]
925.254 35.68 40,232 36,984 63,428
Anti-
inflammatory
,antioxidant
Cyanidin 3-O-(2''-O-galloyl-6''-O-
alpha-rhamnopyranosyl-beta-
galactopyranoside
747.1806 6.48 5,678 5,356 4,544
Antioxidant,
Anti-
inflammatory,
Anti-cancer,
Skin protection,
anti-aging
Cyanidin 3-(2G-galloylrutinoside) 747.1806
6.48 5,678 5,356 4,544
Antioxidant,
Anti-
inflammatory,
Anti-cancer,
Skin protection,
anti-aging
138
Cyanidin 3-O-(2''-O-galloyl-6''-O-
alpha-rhamnopyranosyl-beta-
galactopyranoside
764.203 15.53 4,478 5,480 4,478
Antioxidant,
Anti-
inflammatory,
Anti-cancer,
Skin protection,
anti-aging
Cyanidin 3-(2G-galloylrutinoside) 764.203 15.53 4,478 5,480 4,478
Antioxidant,
Anti-
inflammatory,
Anti-cancer,
Skin protection,
anti-aging
6-Glucopyranosylprocyanidin B2 779.1633 27.69 2,454 3,200 3,194 Antioxidant
6-Glucopyranosylprocyanidin B1 779.1633 27.69 2,454 3,200 3,194 Anti-
inflammatory
8-Glucopyranosylprocyanidin B2 779.1633 27.69 2,454 3,200 3,194 Antioxidant
8-Glucopyranosylprocyanidin B1 779.1633 27.69 2,454 3,200 3,194 Anti-
inflammatory
Procyanidin B3 7-glucoside 779.1633 27.69 2,454 3,200 3,194 antioxidant
Cyanidin-3-(6'-malonylglucoside) 558.0948 34.86 242,958 x x
Antioxidant,
Anti-
inflammatory,
Anti-cancer,
Skin protection,
anti-aging
Cyanidin 3-(3''-malonyl-glucoside) 558.0948 34.86 242,958 x x
Antioxidant,
Anti-
inflammatory,
Anti-cancer,
Skin protection,
anti-aging
139
Cyanidin 3-(4'''-acetylrutinoside) 659.1566 32.85 75,804 27,948 55,172
Antioxidant,
Anti-
inflammatory,
Anti-cancer,
Skin protection,
anti-aging
Cyanidin-3-(2'-acetylrutinoside) 659.1566 32.85 75,804 27,948 55,172
Antioxidant,
Anti-
inflammatory,
Anti-cancer,
Skin protection,
anti-aging
Pelargonidin 3-rhamnoside 439.0999 10.98 21,104 x 19,424
Anti-
inflammatory,
antioxidant
Pelargonidin 3-robinobioside 601.1539 15.2 5,336 4,514 4,628
Anti-
inflammatory,
antioxidant
Pelargonidin 3-neohesperidoside 601.1539 15.2 5,336 4,514 4,628
Anti-
inflammatory,
antioxidant
Pelargonidin 3-rutinoside 601.1539 15.2 5,336 4,514 4,628
Anti-
inflammatory,
antioxidant
Pelargonidin 3-rhamnoside-5-
glucoside 601.1539 15.2 5,336 4,514 4,628
Anti-
inflammatory,
antioxidant
Pelargonidin 3-[6-((Z)-p-
coumaroyl)glucoside]-5-glucoside 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory,
antioxidant
Pelargonidin 3-(6''-acetylglucoside) 497.1064 6.69 20,560 x 20,746
Anti-
inflammatory,
antioxidant
140
Pelargonidin 3-(6''-acetylglucoside) 497.108 5.83 27,832 24,098 22,526
Anti-
inflammatory,
antioxidant
Pelargonidin 3-O-β-D-glucoside 5-
O-(6-coumaroyl-β-D-glucoside) 779.1633 27.69 2,454 3,200 3,194
Anti-
inflammatory,
antioxidant
Pelargonidin 3-(6''-
acetylglucoside)-5-glucoside 659.1566 32.85 75,804 27,948 55,172
Anti-
inflammatory,
antioxidant
Pelargonidin 3-glucoside-5-(6''-
acetylglucoside) 659.1566 32.85 75,804 27,948 55,172
Anti-
inflammatory,
antioxidant
Malvidin 3-O-(6-O-(4-O-malonyl-
alpha-rhamnopyranosyl)-beta-
glucopyranoside
763.1482 20.24 2,718 4,206 3,402
Anti-
inflammatory,
antioxidant
Malvidin 3-O-(6-O-(4-O-malonyl-
alpha-rhamnopyranosyl)-beta-
glucopyranoside
747.1806 6.48 5,678 5,356 4,544
Anti-
inflammatory,
antioxidant
malvidin-3-O-glucoside-
acetaldehyde 539.1149 6.67 78,602 85,892 74,576
Anti-
inflammatory,
antioxidant
Malvidin 3-glucoside-4-
vinylguaiacol 677.1273 14.37 3,758 3,148 3,960
Anti-
inflammatory,
antioxidant
Malvidin 3-glucoside-4-
vinylguaiacol 639.1679 31.59 2,922 3,126 3,246
Anti-
inflammatory,
antioxidant
Malvidin 3-glucoside-4-
vinylguaiacol 661.1548 29.01 3,202 3,348 3,202
Anti-
inflammatory,
antioxidant
Dihydroisorhamnetin 319.081 9.7 62,062
141
Isorhamnetin 5-glucoside 499.1222 5.85 58,816
Anti-
inflammatory,
Anticancer
Isorhamnetin 3-(4''-
sulfatorutinoside) 705.1318 33.64 356,714 34,344 685,630
Anti-
inflammatory,
Anticancer
Isorhamnetin 3-(3'''-
ferulylrobinobioside) 839.1756 36.5 1,629,892 199,560
Anti-
inflammatory,
Anticancer
Isorhamnetin 3-glucosyl-(1->2)-
galactoside-7-glucoside 841.1747 36.49 1,093,388
Anti-
inflammatory,
Anticancer
Isorhamnetin 3-(6''-(E)-
sinapoylsophoroside) 864.2498 36.48 254,224
Anti-
inflammatory,
Anticancer
Isorhamnetin 3-rhamnosyl-(1->2)-
gentiobiosyl-(1->6)-glucoside 949.2854 36.65 372,300 397,468
Anti-
inflammatory,
Anticancer
Hesperetin
8-Hydroxyhesperetin 319.081 9.7 62,062 x x antioxidant
Hesperetin 5-O-glucoside 465.1395 24.23 15,284 19,588
Photo protective,
Anti-
inflammatory,
antioxidant
Hesperetin 5,7-laurate 667.4204 36.05 192,366 237,416 257,946
Antioxidant;
Emollient; Skin-
Conditioning
Agent -
Miscellaneous
Naringenin 273.0741 3.7 19,376 15,444
Photo protective,
Anti-
inflammatory,
142
anti-oxidant and
anti-tumor
3'-Geranylchalconaringenin 409.2032 28.44 35,230
Chalconaringenin 2',4'-di-O-
glucoside 619.1579 2.63 30,088
Naringenin 295.0583 5.12 11,234
Photo protective,
Anti-
inflammatory,
anti-oxidant and
anti-tumor
Naringenin-7-O-Glucoside 457.1098 7.95 45,070
Photo protective,
Anti-
inflammatory,
anti-oxidant and
anti-tumor
Naringenin 7-(2-p-
Coumaroylglucoside) 619.1167 33.84 354,694
Photo protective,
Anti-
inflammatory,
anti-oxidant and
anti-tumor
Chalconaringenin 2'-O-glucoside
4'-O-gentobioside 797.1941 36.2 59,314
Daidzein
Dihydrodaidzein diacetate 379.058 4.19 23,852
Daidzein 4'-O-glucoside 439.0994 8.27 21,104
Anti-
inflammatory,
anti-
carcinogenic,
anti-
inflammatory
Daidzein 6''-O-acetate 459.1266 24.31 27,846 Anti-
inflammatory,
143
anti-
carcinogenic,
anti-
inflammatory
Daidzein 7,4'-di-O-glucoside 579.1673 5.76 12,738
Anti-
inflammatory,
anti-
carcinogenic,
anti-
inflammatory
Genistein 7,4'-bis(O-
glucosylapioside) 859.2501 36.63 64,594
Anti-
inflammatory,
antioxidant and
anticancer
6,8-Diprenylgenistein 407.1835 24.31 141,212
Genistin 433.1105 5.57 59,032
Prevent UV-
induced skin
damage
Genistein 7-O-rutinoside 579.1673 5.76 12,738
Prevent UV-
induced skin
damage
Taxifolin
6-Methoxytaxifolin 357.0585 3.47 68,432 20,036
6-Methoxytaxifolin 357.0583 4.7 14,468
Naringin 4'-glucoside 743.2366 4.19 3,344
Skin photo
protection,
antioxidant, anti-
carcinogenic
Chrysin
Chrysin-7beta-monoglucoside 439.0994 8.27 21,104 18,130
Skin protection,
anti-
inflammation
144
and anti-
oxidation
2-ETHOXYCARBONYL-2-
ETHOXYOXALOYLOXYDIHYD
ROCHRYSIN DIMETHYL
ETHER
495.1057 9.02 4,408
Strychochrysine 601.3394 12.22 28,694
Helichrysin 449.1474 2.63 30,618
Chrysin 7-(4''-acetylglucoside) 459.1266 24.31 27,846
Skin protection,
anti-
inflammation
and anti-
oxidation
Chrysin 7-gentiobioside 579.1673 5.76 12,738
Skin protection,
anti-
inflammation
and anti-
oxidation
Epigallocatechin 3-O-vanillate 457.1109 5.92 39,556 31,318
8-C-Ascorbylepigallocatechin 3-
gallate 633.1122 35.86 799,104 246,502 691,586
Epigallocatechin-(4beta->8)-
epicatechin-3-O-gallate ester 747.1512 34.53 34.53 27,012 257,604
Epigallocatechin-(4beta->8)-
epicatechin-3-O-gallate ester 764.1797 33.57 74,912 20,244
Epigallocatechin 3-O-gallate-
(4beta->6)-epicatechin 3-O-gallate 921.1464 33.77 74,102
2',2'-BISEPIGALLOCATECHIN
DIGALLATE 932.185 36.66 74,360
Epigallocatechin-(4beta->8)-
epicatechin-3-O-gallate ester 764.18 34.03 400,350
145
Caffeine (alkaloid) 195.088
2 2.63 36,703,416 38,792,408 1,861,824
Antioxidant,
Anti-skin cancer,
antimicrobial,
emollient
Isocaffeine 195.088
2 2.63 36,703,416 38,792,408 1,861,824
Eugenol (phenylpropanoid) 179.105
7 14.22 329,800 238,938 343,726
Anti-microbial,
anti-
inflammatory,
analgesic,
antioxidant and
anticancer
Resveratrol (stilbenoid)
229.086
2 7.71 x x 106,592
Anti-
inflammatory,
antioxidant, anti-
aging, photo
protection 251.068
1 36.7 x 6,864 10,756
Guaiacol benzoate 229.086
2 7.71 x x 106,592 Anti-microbial
Epigallocatechin-(4beta->8)-
epicatechin-3-O-gallate ester 764.1809 36.2 323,184
Epigallocatechin 3-O-gallate-
(4beta->6)-epicatechin 3-O-gallate 921.1564 35.45 65,912
Epigallocatechin 3-O-gallate-
(4beta->6)-epicatechin 3-O-gallate 921.1551 36.23 49,962
Theaflavin
Theaflavin monogallate A 717.1419 3.7 13,048 Antioxidant
THEAFLAVIN DIGALLATE 891.1442 34.98 50,776 antioxidant
146
Chrysophanol (anthraquinone)
Anti-cancer,
anti-
inflammatory
antimicrobial
Chrysophanol 1-tetraglucoside 925.254 35.68 40,232 36,984 63,428
Anticancer, anti-
inflammatory,
anti-microbial
Soranjidiol 255.065
5 20.28 8,512 6,930 7,272 Anti-cancer
Physcion 8-glucoside 464.154
7 23.90 3,750 x x
Antibacterial,
anti-
inflammatory,
anti-skin cancer
147
Appendix B.
Whitening activities of the spent coffee grounds
1. Introduction
In addition to antioxidant and anti-microbial activity, skin whitening effect is emerging as
a major interest in cosmetic market. The pigmentation (melanogenesis) on the skin causes
multiple issues including lentigo, and freckles. Melanin is produced by oxidation reaction
of tyrosine by tyrosinase, oxidase controlling the melanin production (Saghaie et al.,
2013). Multiple natural whitening chemicals extracted from plants such as arbutin,
bisaborol, ricorice has been used in real product by inhibiting tyrosinase (Khaiat and
Saliou, 2015). Besides, tyrosinase inhibition, there are other methods in blocking skin-
pigmentation process. For example, vitamin C derivatives such as ascorbyl glucoside, and
ascorbyl tetraisopalmitate act as antioxidant inhibiting oxidation of tyrosine.
148
AppendixB1. Mode of action of anti-pigmentation
149
AppendixB.2. Melanin synthesis pathway
150
2. Objectives
In Appendix B, the tyrosinase inhibition assay was conducted to evaluate skin whitening
effect of SCG extracts from three cultivars.
3. Materials and Methods
Tyrosinase inhibition activities (skin-whitening property)
Tyrosinase inhibition activity will be determined as described by (Momtaz et al., 2008)
with tyrosine as substrates. Samples were dissolved in dimethyl sulfoxide (DMSO) to a
concentration of 20 mg/ml, and further diluted in potassium phosphate buffer (50 mM, pH
6.5) to 600 μg/ml. Assays were carried out in a Falcon 96 flat bottom clear plate and
SYNERGY HT microplate reader (BIO-TEK) was used. All the steps in the assay were
conducted at room temperature. In triplicate, each prepared sample (70 μl) was mixed with
30 μl of tyrosinase (333 Units/ml in phosphate buffer, pH 6.5). After 5 min incubation, 110
μl of substrate (2 mM L -tyrosine) was added to the reaction mixtures and incubated further
for 30 min. Arbutin (0-1100 µg/ml) was used as a positive control while a blank test was
used as each sample that had all the components except L-tyrosine. Results were compared
with a control consisting of DMSO instead of the test sample. Absorbance values of the
wells were then determined at 492 nm. The percentage tyrosinase inhibition was calculated
as follows:
%inhibition=[(Acontrol–Asample)/Acontrol]x100 where A control is the absorbance of
DMSO and A sample is the absorbance of the test reaction mixture containing extract or
arbutin.
151
4. Results and discussion
All SCG extracts including three cultivars had neglectable effect on tyrosinase inhibition
test. However, SCG had neglectable. From previous studies, several compounds that can
be found in SCG such as E-cinnamic acid, vanillic acid, 3,5-Di-O-caffeoylquinic acid are
reported as anti-tyrosinase agents (Kong et al., 2008) (Chou et al., 2010) (Iwai et al.,
2004). However, from out study, SCG extracts from Ethiopian Yirgacheffe, Costa Rian
Tarrazu and Hawaiian Kona did not show any anti-tyrosinase activities. Our findings are
not consistent with other studies which have reported that SCG and coffee bean can act as
a skin-whitening agents related to anti-tyrosinase properties (Kanlayavattanakul and
Lourith, 2013) (Kiattisin et al., 2016). In latter study they conducted anti-tyrosinase and
anti-oxidant assay using coffee bean including green and roasted. Even though they
found ethanolic extracts of green coffee bean which has the highest antioxidant activities
showed that highest anti-tyrosinase activity (44.27 inhibition %), all the ethanolic extracts
including green bean and roasted bean presented a lower anti-tyrosinase activity than
kojic acid and arbutin. From this result, they concluded that antioxidant compounds
might promote the tyrosinase inhibition activity due to their antioxidative synergistic
(Chang, 2009). Our result indicates that anti-tyrosinase effect is not a major mode of
action in SCG. Therefore, the skin-whitening effect of coffee extracts and SCG may more
related to mode of action in anti-pigmentation as antioxidant agents.
152
Appendix B3. Calibration curve of the positive control (arbutin). Each dot in X axis represents around 0,
1.1, 11, 110, 1100 µg/ml concentration of arbutin (STD; n = 5).
-20
0
20
40
60
80
100
120
-200 0 200 400 600 800 1000 1200
% in
hib
itio
n
µg/ml
Series1