Res Chem Intermed (2017) 43:783–799

DOI 10.1007/s11164-016-2664-y

Extractive determination of bioactive flavonoids

from butterfly pea (Clitoria ternatea Linn.)

Jayanti Makasana1 • Bharatkumar Z. Dholakiya1 •

Narendra A. Gajbhiye2 • Saravanan Raju3

Received: 21 May 2016 / Accepted: 11 July 2016 / Published online: 18 July 2016
 Springer Science+Business Media Dordrecht 2016

Abstract Recently, in the field of natural product drug discovery, there has been
increasing interest in effective extraction and isolation of bioactive phytomolecules from
plants for use as important starting materials or chemical intermediates for new drug
development. In this investigation, various solvent extracts of butterfly pea (Clitoria
ternatea Linn.) were evaluated in terms of phenolic content and antioxidant activity.
Moreover, various parameters influencing optimal flavonoid extraction were studied
based on determination of the quercetin and kaempferol yields by an improved high-
performance liquid chromatography (HPLC) method. The results indicated that the total
phenolics and 2,20 -azino-bis-3-ethylbenzothiazoline-6-sulfonic acid (ABTS) scavenging
activity were maximum for the 80 % aqueous methanol extract, while the 1,1-diphenyl-
2-picrylhydrazyl (DPPH) reducing power and flavonoid content were maximum in the
methanol extract. Different solvent extracts showed significantly (p \ 0.05) different
phytochemical yield and antioxidant capacity. Furthermore, the total phenol and flavo-
noid contents revealed good (p \ 0.001) correlations with scavenging potency. The
maximum flavonoid content was found for methanol concentration, time, temperature,
and solid-to-liquid ratio of 80 %, 60 min, 80 C, and 1:20 g/mL, respectively. It is
concluded that butterfly pea leaf can be considered a potential source of flavonoids with
good antioxidant properties for use in dietary applications. The extraction protocols
developed in this study can be readily scaled up for industrial applications.

& Bharatkumar Z. Dholakiya

bzd.svnit@gmail.com
1
Applied Chemistry Department, Sardar Vallabhbhai National Institute of Technology,
Ichchanath, Surat, Gujarat 395007, India
2
ICAR-Directorate of Medicinal and Aromatic Plants Research, Boriavi, Anand, Gujarat, India
3
ICAR-Central Tuber Crops Research Institute, Sreekariyam, Thiruvananthapuram, Kerala,
India

123
784 J. Makasana et al.

Graphical Abstract

Keywords Antioxidant activity  Butterfly pea  Clitoria ternatea  Extractive effect

of solvent  Flavonoid  HPLC analysis

Introduction

Methods for extraction of phenolic compounds from plant sources vary widely due
to the complex nature of the sample matrix and because different plant materials
contain different forms of flavonoid compounds [1]. A natural source with high
phenolic content and high biological activity is a prerequisite for industrial-scale
production. Phytochemical yield and antioxidant capacity are significantly influ-
enced by the solvents used for extraction, due to their varying dielectric properties
and extraction efficiency, as well as the structural variability of the compounds
targeted in the process [2, 3]. Traditional heating reflux is widely used for extraction
of phenolics from plant sources because of its easy operation. However, proper
selection of solvents, optimum solid-to-liquid ratio, and duration and temperature of
extraction is imperative to achieve high efficiency. Hence, such procedures should
be standardized for maximum extraction to provide the real composition and
optimal level of flavonoids from plants [1, 4]. Flavonoids are the most widely
distributed, ubiquitous plant secondary metabolites, mainly existing in the form of
glycoside conjugates. Their chromatographic analysis generates a large number of
peaks, making individual flavonoid estimation difficult due to the lack of standard
reference compounds. To overcome this problem, acid hydrolysis is routinely
applied to reduce flavonoid glycosides to respective aglycones by removal of sugar
moieties before quantitative HPLC analysis. The efficiency of conversion and
degradation of aglycones are influenced by critical factors such as the acid
concentration and the temperature and duration of hydrolysis [5].

123
Extractive determination of bioactive flavonoids from… 785

Reactive oxygen species (ROS), viz. hydroxyl radical (OH) and superoxide free
radical (O2-), are generated as byproducts of human metabolism. However,
excessive amounts of these species may cause various diseases including cancer,
cardiovascular disease, ageing, atherosclerosis, inflammatory injury, diabetes, brain
dysfunction, weakened immune system, etc. [6, 7]. Antioxidant compounds can
quench ROS and ameliorate their harmful effects. However, synthetic antioxidants
such as butylated hydroxytoluene (BHT), butylated hydroxyanisole (BHA), and
propyl gallate (PG) could lead to cancer risk [8]. Considerable attention is therefore
being paid to natural antioxidants obtained from dietary sources to ward off various
diseases and improve human health without negative side-effects [9]. Phenolic
compounds are currently being increasingly investigated as new sources of natural
antioxidants available from herbs, and as functional foods and nutraceuticals.
Phenolic constituents (phenolic acids, flavonoids, quinones, coumarins, lignans,
stilbenes, tannins, etc.) are widely distributed in the plant kingdom. Among these,
flavonoids are an interesting class of phytochemicals, strongly related to antioxidant
activity and promising health benefits [10]. Kaempferol (KF) and quercetin (QR) are
major flavonoids with positive reducing power and ability to scavenge free radicals
and donate H-atom to peroxy radical. These bioactive constituents (Fig. 1) and
some of their glycosides exhibit a wide range of biological activities, such as
antioxidant, analgesic, antiinflammatory, antimicrobial, antihypertension, gastro-
protective, immunity-promoting, antiarthritis, anti-interstitial cystitis, anticancer,
cardioprotective, neuroprotective, antidiabetic, antiobesity, antiestrogenic, anxi-
olytic, and antiallergic activities [11, 12]. Interestingly, metaanalysis of several
epidemiological studies revealed that KF- and QR-rich diets prevent growth of
cancer cells and reduce cancer risk [13].
Nowadays, plant extracts and isolated pure bioactive phytocompounds are
leading to a new era in pharmacological research, becoming standard treatments for
diseases, and forming a platform for new drug synthesis or as models for
therapeutically active agents [14]. Medicinal plants will play a key role in the future,
especially given the impressive application of synthetic chemistry to develop new
active drugs [15]. Thus, in the field of natural product drug discovery, work aims at

Fig. 1 HPLC chromatograms of commercial standard compounds (A), 80 % aqueous methanol extract
before acid hydrolysis (B), and 80 % aqueous methanol extract after acid hydrolysis (C), with ultraviolet–
visible (UV–Vis) detector set at 370 nm

123
786 J. Makasana et al.

improving the speed of preliminary screening for selection of high-quality raw

materials as well as effective extraction and isolation of bioactive phytocompounds
from plants used in traditional medicine systems [16]. This procedure should be
standardized to deal with issues involved in large-scale production of active
compounds for industrial applications and high-quality products.
Clitoria ternatea Linn. (CT) is an extensively used Ayurvedic medicinal herb.
This plant is also known as butterfly pea, conch flower or Shankaphushpi. It is a
perennial climber with white or blue, butterfly-like flowers, belongs to the legume
family, and often grows in the wild or in gardens. The plant has high antioxidant
activity [17]. It is a potent central nervous system (CNS) stimulant and memory-
enhancing herb and considered as a brain tonic [17, 18]. The plant is used against
enlargement of abdominal viscera and swollen joints. The roots are used as a
laxative and diuretic. It is also used against burning sensation, leprosy, inflamma-
tion, bronchitis, and asthma, and the leaves are useful in otalgia and hepatopathy.
The root, stem, and flower are recommended for treatment of snakebite and scorpion
sting in India. Various bioactive molecules such as flavonol glycosides, triter-
penoids, anthocyanins, and sterols have been identified from butterfly pea [18]. The
kaempferol content has been quantified in five different accessions of the plant [19].
Despite the above-mentioned facts and pharmaceutical importance of flavonoids
to date, no systematic studies have been conducted on maximum extraction of
phenolics from butterfly pea. Hence, the present study, as a first attempt,
investigated: (1) suitable solvents for extraction of total phenol and flavonoid
contents, (2) optimization of extraction parameters for maximum yield of
flavonoids, (3) examination of hydrolysis conditions for effective conversion of
aglycones, and (4) simultaneous analysis of QR and KF by HPLC. All experiments
were conducted using butterfly pea leaves, because the leaves can be sustainably
harvested without endangering plant survival. Data were analyzed statistically by
calculating the Pearson correlation between the phenolic content and radical
scavenging capacity. The presented extraction method could be readily employed
for scaling up production of phenolics from butterfly pea.

Materials and methods

Chemical reagents and plant material

Leaves were collected from fully grown butterfly pea plants (about 18 months old)
from experimental plots at ICAR-DMAPR, Boriavi, Anand, Gujarat, India (22.5N,
73.0E), having 800 mm average rainfall, and minimum and maximum temperature
of 12.7 and 42 C, respectively. Collected fresh leaves were washed in tap water to
remove sand and soil, dried in an oven below 45 C to constant weight, then made
into fine powder using a cyclone mill (UDY Corporation). Gallic acid, Folin–
Ciocalteu phenol reagent, anhydrous sodium carbonate, and hydrochloric acid were
purchased from Merck (India). Standard reference chemicals [kaempferol, quercetin,
2,20 -azino-bis-3-ethylbenzothiazoline-6-sulfonic acid (ABTS), and 1,1-diphenyl-2-
picrylhydrazyl (DPPH)] were purchased from Sigma-Aldrich (Germany). During all

123
Extractive determination of bioactive flavonoids from… 787

experiments, Millipore ultrapure water with resistivity greater than 18 MX was used,
with analytical-grade solvents used for extraction and HPLC-grade chemicals for
chromatographic analysis.

Extraction procedure

Ten grams dried butterfly leaf powder was extracted in 250 mL of solvents with
different polarity, viz. hexane, chloroform, ethyl acetate, acetone, methanol,
acetonitrile, water, and 80 % methanol, for 60 min on a water bath at 80 ± 1 C.
The extract was cooled, then vacuumed-filtered through qualitative grade 1 filter
paper (Whatman, Germany) and centrifuged at 10,000 rpm for 10 min to remove
any floating matter. Solvent was completely removed by rotary evaporation under
reduced pressure at temperature below 40 C, and the extract residue was weighed
(% extraction yield = residue weight 9 100/dry leaf powder weight). The exper-
iment was carried out in triplicate. The extract of phenolics was dissolved in
appropriate solvent for determination of antioxidant activity and total phenol and
flavonoid content.

Determination of total phenol yield (TPY)

The TPY was determined by the Folin–Ciocalteu reagent method as described

previously by Mhiri et al., with some modifications [20]. Five hundred microliters
of plant extract (1000 lg/mL) and 2.5 mL Folin–Ciocalteu reagent (10 %, v/v)
were mixed, and the mixture was incubated at room temperature for 6 min. Then,
2.0 mL saturated sodium carbonate solution (7.5 %, w/v) was added, followed by
further incubation for 15 min at temperature of 45 C, then the absorbance was read
at 760 nm using a digital spectrophotometer (Systronic, India) against a blank
sample. TPY is expressed as milligrams of gallic acid equivalents (GAE) per gram
of dried extract.

Determination of flavonoid content (FC)

The FC is expressed as the total sum of QR and KF contents determined in the plant
extract by acid hydrolysis. The extract of phenolics from leaves was analyzed by
high-performance liquid chromatography (HPLC) [21]. Sample extract (200 mg)
was dissolved in 25 mL 80 % methanol. The resulting solution was centrifuged, and
the supernatant was hydrolyzed (12 mL) with 5 M HCl (8 mL) at 80 C on a water
bath for 30 min. The hydrolyzed sample was cooled at room temperature and
diluted up to 50 mL by methanol, then filtered through a 0.45-lm membrane filter
(Sartorius) prior to HPLC quantification. The analysis was carried out three times.
Flavonoid compositions are presented as milligrams per gram of dried extract.

DPPH radical scavenging capacity

The 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical scavenging capacity of

butterfly leaf extracts was measured according to a slight modification of the method

123
788 J. Makasana et al.

of Rao et al. [22]. Briefly, various concentration of leaf extract (3 mL) were mixed
with 1 mL 0.1 mM DPPH solution in methanol. The reaction mixture was vortexed
for 30 s, then incubated at room temperature in the dark for 30 min. Blank sample
was prepared as per the above procedure, replacing leaf extract with methanol. After
that, the absorbance was read at 517 nm using a UV–visible spectrophotometer. The
free radical scavenging assay was carried out for each extract in triplicate. Radical
scavenging capacity is expressed as percent inhibition activity by the sample,
calculated using the formula
% DPPH radical scavenging activity
½Absorbance of blank ðAcÞ  Absorbance of extract ðAsÞ
¼  100:
Absorbance of blank ðAcÞ

ABTS radical scavenging activity

The scavenging activity for the 2,20 -azino-bis-3-ethylbenzothiazoline-6-sulfonic

acid (ABTS) free radical was determined according to the method reported by Re
et al., with slight modification [23]. ABTS reagent was generated by mixing 7 mM
ABTS solution (5 mL) and 2.45 mM potassium persulfate solution (5 mL). The
mixture was kept at room temperature in the dark for 12 h. The generated ABTS
radical was diluted using methanol up to absorbance of 0.700 at 734 nm. Freshly
prepared ABTS radical was used for each experiment. Plant extract (150 lL) at
different concentrations was mixed with ABTS radical solution (2850 lL) and
instantly vortexed and incubated for 6 min in the dark at room temperature. Control
sample was prepared similarly, containing methanol instead of leaf extract. The
absorbance of the reaction mixture was read at 734 nm by UV–visible spectropho-
tometer. The assay was done in triplicate. The percent inhibition was calculated as
per the formula mentioned for the DPPH assay.

HPLC instrument and chromatographic conditions

A modular HPLC (Shimadzu Corporation, Kyoto, Japan), LC system consisting of

two LC-20AD pumps, SPD-20A UV–Vis detector, DGU-20A3 degasser, SIL-20AC
HT autosampler, CTO-10ASvp column oven, and CBM-20 communication bus
module was used for the analysis. In previous study, a HPLC method was applied
for quantification of the KF content from butterfly pea plant [19]. As the baseline
separation with this previously reported chromatographic method was inadequate to
resolve QR and KF, modified chromatographic conditions were developed as
follows: reverse-phase C18 chromatographic column (Altima, 4.6 9 250 mm,
5 lm, Grace, UK) with column temperature of 40 C, mobile phase of isocratic
acetonitrile (45 %) and 0.1 % formic acid in water (55 %), flow rate of
1.0 mL min-1, and detection wavelength of 370 nm.

123
 Table 1 Effect of extraction solvent on yield of phytochemicals from CT leaf
Solvent Dielectric constant (20 C) Extraction yield (%) TPC (mg GAE/g extract) QR (mg/g extract) KF (mg/g extract) FC (mg/g extract)

Hexane 1.9 2.82 ± 0.22g 12.34 ± 0.59h n.d. 0.46 ± 0.10f 0.46 ± 0.10g
f g g f
Chloroform 4.8 4.53 ± 0.13 20.76 ± 0.45 0.19 ± 0.02 1.85 ± 0.15 2.04 ± 0.13f
e f f e
Ethylacetate 6.02 5.26 ± 0.27 47.79 ± 0.17 1.77 ± 0.20 15.27 ± 1.78 17.04 ± 1.86e
e c c a
Acetone 20.6 5.87 ± 0.25 71.68 ± 0.35 6.90 ± 0.21 53.60 ± 0.95 60.50 ± 0.89b
c b a a
Methanol 32.6 20.29 ± 0.23 73.03 ± 0.12 8.85 ± 0.13 55.00 ± 0.66 63.84 ± 0.67a
d e d b
Acetonitrile 37.5 6.46 ± 0.20 62.56 ± 0.30 5.61 ± 0.23 47.71 ± 1.11 53.32 ± 1.03c
a d e d
Water 79.7 29.83 ± 0.54 64.38 ± 0.27 3.80 ± 0.09 23.37 ± 0.40 27.17 ± 0.48d
b a b c
Extractive determination of bioactive flavonoids from…

80 % methanol 41.46 25.92 ± 0.40 75.21 ± 0.23 7.96 ± 0.08 46.07 ± 0.57 54.03 ± 0.65c

All values presented as mean ± standard deviation (n = 3)

n.d. not detected
Different superscript letters indicate significant (p \ 0.05) difference in the same column, with results presented in descending order: a [ b [ c [ d [ e [ f [ g [ h
GAE gallic acid equivalents, TPC total phenol content, QR quercetin content, KF kaempferol content, FC flavonoid content represented as total sum of quercetin and
kaempferol content
Dielectric constant value (20 C) from Handbook of Organic Solvent Properties, Ian M. Smallwood, Butterworth–Heinemann (2012) and Gosta A., Journal of the
American Chemical Society 54:4130 (1932)
789

123
790 J. Makasana et al.

Optimization of flavonoid extraction and acid hydrolysis conditions

The solvent concentration, solid-to-liquid ratio, time, and temperature were

standardized for flavonoid extraction from butterfly pea plant. Leaf powder (5 g)
was extracted by reflux with continuous heating on a water bath using different
methanol concentrations (0, 20, 40, 60, 80, and 100 %) with varying solid-to-liquid
ratio (1:5, 1:10, 1:15, 1:20, 1:25, and 1:30 g/mL), extraction time (30, 60, 90, 120,
150, and 180 min), and water bath temperature (40, 50, 60, 70, 80, and
90 ± 1.0 C). The resulting extract was filtered, centrifuged, and used for further
study. The effects of hydrochloric acid (HCl) concentration, temperature, and
duration were investigated for optimum conversion of glycosides to aglycone
content from butterfly pea leaf extract. The leaf extract solution from the previous
stage was subjected to hydrolysis using varying HCl concentrations (2, 3, 4, 5, and
6 M), times (15, 30, 60, 90, and 120 min), and temperatures (50, 60, 70, 80, and
90 C). Ascorbic acid (2, 5, and 10 mg) was added prior to hydrolysis to determine
the effect of antioxidant for preventing flavonoid degradation.

Statistical analyses

Data were analyzed using SPSS Statistics (USA). Difference was considered
statistically significant for p value below 0.05 (p \ 0.05). Pearson correlation was
performed, and the correlation coefficients (r) were analyzed to determine the
relationship between phytochemical (TPY, FC) content and antioxidant (DPPH,
ABTS) activity.

Results and discussion

Effect of solvent on phytochemical yield from butterfly pea plant

and antioxidant activity

Extraction yield

Plant extraction involves separation of medicinally active portions of a plant from

inactive or inert components using selected solvents and standard extraction
procedures, prior to phytocompound purification [24]. The extraction yield is
expressed as the percentage to crude extract derived from dried plant material. In the
present study, butterfly pea leaves were extracted using eight different solvents
ranging from nonpolar to aqueous and with varying dielectric constant values. The
total extractable yield is given in Table 1. The highest extraction yield
(29.83 ± 0.54 %) was obtained in water, followed by 80 % methanol in aqueous
(25.92 ± 0.40 %). The extraction yield was strongly influenced by the solvent used,
increasing significantly with increasing solvent polarity. This is in agreement with
Zang et al. [7] and also supports the results of the investigation carried out by
Banerjee and Bonde [3].

123
Extractive determination of bioactive flavonoids from… 791

Total phenol yield (TPY)

Several studies have reported strong correlation between phenolic content from
herbs and reducing capacity [25, 26, 30, 32]. Such results motivate determination of
the total phenolic content and antioxidant activity of traditionally listed medicinal
plants to provide scientific evidence of their medicinal importance. In the TPY
assay, phenolic compounds undergo oxidation, and the Folin–Ciocalteu reagent is
reduced to blue color whose spectrophotometric intensity indicates the amount of
phenolics in the plant extract. The TPY of the various plant extracts (Table 1)
showed significant (p \ 0.05) difference according to the extraction solvent used.
The poor phenolic yield in nonpolar organic solvent could be due to low solubility
of polyphenols. Aqueous methanol (80 %) showed the significantly highest TPY,
followed by methanol extract with 75.21 ± 0.23 and 73.03 ± 0.12 mg GAE/g
extract, respectively. This phenomena may be attributed to presence of water-
soluble compounds such as carbohydrates in the plant sample. Meanwhile, hexane
extract showed the lowest TPY value of 12.34 ± 0.59 mg GAE/g extract. These
results are in agreement with previously reported phenolic extractions from
Matricaria pubescens, in which the maximum phenolic content was found in
aqueous methanol (50 %) extract, followed by pure organic solvents [25]. For
Artemisia selengnesis, the TPY was as follows (in descending order): aqueous
ethanol (50 %) [ methanol [ acetone [ water [ ethyl acetate [7]. The higher TPY
in the hydro-alcoholic extract indicates that phenolic compounds present in butterfly
pea plant are mainly hydrophilic in nature.

Determination of flavonoid content (FC)

Total flavonoid content determined by the UV–visible spectrometric method is not

accurate, as the color development is nonspecific, resulting in incorrect quantifi-
cation. A specific and sensitive HPLC method is required for detection and
quantification of individual flavonoids [27]. Hence, in this study, the HPLC method
was preferred for determination of flavonoid content from butterfly pea, using QR
and KF as reference. Flavonoid analysis generally requires acid hydrolysis to
convert flavonoid glycosides to respective aglycones for easy separation and
detection [5, 19, 21], therefore hydrolysis was applied to leaf extract prior to HPLC
analysis. Chromatographic separation of QR (4.25 min) and KF (5.16 min) was
achieved within short runtime of 5.5 min (Fig. 1). Identification of these compounds
from plant extract was confirmed by comparing retention times and UV spectra, and
by spiking with reference standards. Using this improved method, calibration curves
showed good linearity (not less than 0.999) in the ranges of 0.20–200 lg/mL for QR
and 0.39–400 lg/mL for KF. The limit of detection (LOD) was 0.03 and 0.021 lg/
mL for QR and KF, while the limit of quantification (LOQ) was found to be 0.082
and 0.054 lg/mL, respectively, indicating good sensitivity of this method. The
intra- and interday precision were found to lie in the ranges of 0.10–0.21 and
0.48–0.82 % relative standard deviation (RSD) for QR, and 0.13–0.20 and
0.30–0.54 % RSD for KF, respectively, revealing good precision of the proposed

123
792 J. Makasana et al.

method. The recoveries of QR and KF were found to lie in the range of

99.3–103.3 %, revealing good accuracy of the developed method.
HPLC analysis of leaf extract in 80 % methanol generates a chromatographic
fingerprint with different peaks. Furthermore, the same extract subjected to acid
hydrolysis indicated abundant KF and QR peaks, revealing that the flavonoids
present in the leaf are mainly in glycosylated form of KF and QR. Meanwhile, the
increase in KF peak compared with QR in these results confirms that the major
compounds present in the leaf were kaempferol derivatives, as also found in earlier
isolation from butterfly pea [17]. The flavonoid content (FC) in different solvent
extracts (Table 1) differed significantly (p \ 0.05) in the range from 0.46 ± 0.10 to
63.84 ± 0.67 mg/g extract. The FC extraction efficiency was affected by the
solvent polarity, lying in the following order: methanol [ acetone [ aqueous
methanol (80 %) [ water [ acetonitrile [ ethyl acetate [ chloroform [ hexane.
Solvent polarity played a key role in the extraction efficiency of flavonoids from
CT. The flavonoid content on dry weight basis was maximum when aqueous
methanol was used, because of its higher extractive capacity compared with pure
methanol. This is in agreement with previous findings by Ghasemzadeh et al. [28].
Therefore, in further optimization study, methanol was used for flavonoid extraction
from CT in combination with water as binary solvent system.

DPPH scavenging activity (DRSA)

The free radical scavenging capacity of the different extracts of CT leaf was studied
using the DPPH assay introduced by Blois [29]. The DPPH radical is a stable free
radical with absorption at 517 nm. It exhibits decreased absorption when accepting
an electron or free radical, and the reaction mixture undergoes a color change from
purple to yellowish, suggesting that the sample contained free radical scavengers.
The DRSA of the plant extracts was measured using the 50 % inhibitory
concentration (IC50) value and the percentage (%) inhibition; lower IC50 and
higher % inhibition correspond to greater antioxidant capacity. The IC50 value of an
extract is inversely associated with its radical scavenging potency; the value
represents the concentration of antioxidant compounds required for 50 % inhibition
of DPPH radical. The DRSA of the different extracts is presented in Table 2. The
strongest scavenging activity was observed for the pure methanol extract, followed
by the 80 % methanol extract, with 83.64 ± 0.63 and 72.48 ± 0.50 % inhibition,
respectively. IC50 values ranging between 89.70 ± 1.87 and 344.13 ± 1.01 lg/mL
of CT leaf extract were found for the different solvents used. Potent DRSA was
observed in the methanol followed by the 80 % methanol extract, being
significantly (p \ 0.05) higher than for the other solvent extracts studied. The
hexane and chloroform extracts required higher concentration for 50 % inhibition of
DPPH compared with other extracts. These results are in accordance with reported
data, according to which the methanol extract of Paramignya trimera revealed
higher DPPH antioxidant activity than when using other solvents [2]. The
experimental results showed that the DRSA was positively influenced by different
extracts, and it is clear that the different types of solvent resulted in different
amounts of soluble antiradical compounds.

123
Extractive determination of bioactive flavonoids from… 793

Table 2 DPPH and ABTS radical scavenging potency of CT leaf extracts

Solvent DPPH antioxidant activity ABTS antioxidant activity

% Inhibition IC50 (lg/mL) % Inhibition IC50 (lg/mL)

Hexane 21.79 ± 0.06h 344.13 ± 1.01h 46.99 ± 0.06h 2128.06 ± 2.60h

Chloroform 25.09 ± 0.13g 298.88 ± 1.55g 58.95 ± 0.03g 1696.42 ± 0.90g
f f f
Ethylacetate 39.03 ± 0.30 192.17 ± 1.49 62.79 ± 0.08 1592.59 ± 1.94f
c c c
Acetone 65.16 ± 0.42 115.11 ± 0.74 77.19 ± 0.13 1295.51 ± 2.18c
a a b
Methanol 83.64 ± 0.63 89.70 ± 1.87 82.93 ± 0.22 1205.91 ± 3.18b
e e e
Acetonitrile 46.45 ± 0.42 161.49 ± 1.46 72.25 ± 0.22 1384.10 ± 4.28e
d d d
Water 52.53 ± 0.36 142.79 ± 0.99 75.04 ± 0.04 1332.56 ± 0.69d
b b a
80 % methanol 72.48 ± 0.50 103.46 ± 1.33 86.14 ± 0.14 1160.95 ± 1.87a

All values presented as mean ± standard deviation (n = 3)

Different superscript letters indicate significant (p \ 0.05) difference in the same column, with results
presented in descending order: a [ b [ c [ d [ e [ f [ g [ h. Higher % inhibition and lower IC50
value indicate greater antioxidant activity

ABTS radical scavenging assay (ARSA)

A free radical scavenging assay of the different solvent extracts of CT leaf was
also carried out using the ABTS radical [30] to confirm the results obtained by the
DPPH method. The reduction ability of ABTS radical was determined by the
decrease in absorbance at 734 nm with discoloration of the reaction mixture,
which is induced by hydrogen donor as antioxidant. According to the results in
Table 2, the difference among the ABTS scavenging capacities of the different
solvent extracts was significant at p \ 0.05. The IC50 values varied from
1160.95 ± 1.87 to 2128.06 lg/mL among the solvents used for extraction. The
ARSA of the different solvent extracts from CT showed solvent dependency and
decreased as the polarity of the solvent decreased. The 80 % methanol extract
exhibited the strongest radical scavenging activity (86.14 ± 0.14 % inhibition),
followed by the 100 % methanol extract (82.93 ± 0.22 % inhibition). The other
solvent extracts showed ARSA in the following, decreasing order: acetone [ wa-
ter [ acetonitrile [ ethyl acetate [ chloroform [ hexane. This result can be
explained by the fact that the selected solvent determined the solubility of
compounds contributing antioxidant activity. A similar study reported that 50 %
methanol had higher ABTS radical scavenging capacity compared with pure
methanol extract of Matricaria pubescens [25]. These results are also in
agreement with the finding by Addai et al. [31] that the highest ABTS inhibition
was shown by aqueous methanol (80 %) extract of Carica papaya when compared
with pure methanol, acetone or water. These results demonstrate that the polarity
of the solvent used significantly influenced the ARSA, with compounds recovered
in polar solvents contributing most antioxidant activity.

123
794 J. Makasana et al.

Correlation study

In this work, the Pearson correlation matrix (Table 3) for different solvent extracts
with respect to phenolic content and antioxidant properties was obtained. Among the
eight extracts, the phenolic and flavonoid contents showed a statistically significant
influence on the observed antioxidant capacities (Table 3). The positive correlation
(r = 0.939) between the DPPH and ABTS assays of the different extracts was highly
significant (p \ 0.001). This is in agreement with results for Matricaria pubescens
extracts [25]. These results demonstrate that the phytoconstituents of CT have high
free radical scavenging properties, making it an excellent source of natural
antioxidants. The correlation coefficient (r) between the DPPH scavenging activity
and TPY and FC was 0.913 and 0.907, respectively, being significant at p \ 0.001.
The ABTS radical scavenging activity showed significant (p \ 0.001) correlation of
0.959 and 0.888 with TPY and FC, respectively. A similar, significant and positive
relationship was observed between the antioxidant activities and the contents of
phytochemicals (TPY and FC) present in different solvent extracts of Artemisia
selengnesis, Matricaria pubescens, and Citrus mitis [7, 25, 26]. In the present study,
the TPY showed the strongest correlation with the DPPH and ABTS radical
scavenging activities, and the positive correlation (r = 0.916) between TPY and FC
reflects a potent contribution of flavonoids to the antioxidant activities of the CT leaf
extracts. In addition, the different solvent extracts showed good correlation
(r = 0.878, p \ 0.01) between extraction yield and the solvent dielectric properties.

Effect of different extraction parameters on flavonoid yield from butterfly

pea leaves

Effect of methanol concentration

To study the effects of different concentrations of methanol, a range of

combinations in water, namely 0, 20, 40, 60, 80, and 100 %, were used for
extraction, with solid-to-liquid ratio of 1:20 g/mL and extraction time of 60 min in a
water bath at 80 C (±1 C). It was found that the methanol concentration had a
significant effect on flavonoid extraction from CT. The yield increased linearly with
increasing methanol concentration in water, then decreased after reaching a peak
value at 80 % methanol (Fig. 2). The highest QR (2.06 ± 0.01 mg/g) and KF

Table 3 Pearson correlation matrix (r) between antioxidants and antioxidant activities of various CT
extracts
DPPH radical ABTS radical Total phenol content Flavonoid content

DPPH radical 1 – – –
ABTS radical 0.939 1 – –
Total phenol content 0.913 0.959 1 –
Flavonoid content 0.907 0.888 0.916 1

Correlation significant at 0.001 level

123
Extractive determination of bioactive flavonoids from… 795

Fig. 2 Effect of different parameters: methanol concentration (a, b), extraction time (c, d), extraction
temperature (e, d), and solid-to-liquid ratio (g, h) on extraction yield of quercetin and kaempferol,
respectively

123
796 J. Makasana et al.

(11.94 ± 0.11 mg/g) content was found when using 80 % methanol for extraction.
Therefore, the preferred methanol concentration is 80 %, for which the maximum
flavonoid yield was obtained, in line with earlier work [31]. This maximum
extraction efficiency in aqueous methanol is attributed to increased swelling of the
plant matrix by water with increased contact surface area between the solute
components and solvent, which might improve the extraction yield [33].

Effect of duration of extraction

The duration of extraction is critical to minimize the energy and cost of the
extraction process. The extraction process was carried out for durations of 15, 30,
60, 90, 120 and 150 min with solid-to-liquid ratio of 1:20 g/mL using 80 %
methanol at 80 C (±1 C). The QR and KF contents increased gradually with
increasing duration of extraction from 15 to 60 min (Fig. 2), decreasing thereafter.
The optimal duration for flavonoid extraction was determined to be 60 min. The
highest flavonoid yield was obtained when extracting for 60 min, while longer
durations resulted in reduced extraction efficiency and poor yield, possibly due to
loss or degradation of flavonoids [4].

Effect of extraction temperature

The extraction temperature is one of the most important parameters affecting

flavonoid recovery. To study the effect of different temperatures on the flavonoid
yield from CT, extraction was conducted at different temperatures, viz. 40, 50, 60,
70, 80, and 90 C, with solid-to-liquid ratio of 1:20 g/mL, using 80 % aqueous
methanol for 60 min. The content of QR and KF gradually increased as the
temperature was increased from 40 to 80 C, reached a maximum at 80 C, and
declined thereafter (Fig. 2). The achievement of the highest flavonoid yields at
elevated temperatures may be due to increased molecular agitation [34], while the
decrease in the extraction yield at temperatures above 80 C may be due to thermal
degradation of flavonoids [4]. Oxidation of flavonoids at elevated temperatures is a
cause for concern, hence 80 C is considered to be optimum for extraction from CT.

Effect of solid-to-liquid ratio

The solid-to-liquid ratio is another important factor for reducing costs and
environmental hazards, as it determines the solvent consumption during extraction.
Use of the optimum solid-to-liquid ratio is also essential for maximum extraction
efficiency. To study the effect of the solid-to-liquid ratio on the flavonoid yield from
CT, the extraction process was carried out using different solid-to-liquid ratios (g/
mL) of 1:5, 1:10, 1:15, 1:20, 1:25, and 1:30, with 80 % aqueous methanol for
60 min at 80 C (Fig. 2). The content of QR and KF gradually increased with the
solid-to-liquid ratio from 1:5 to 1:20. Reduction of the solid-to-liquid ratio greatly
improved the ability of the solvent to dissolve the constituents, in turn influencing
the dissolution of solutes from the sample matrix into the solvent, resulting in
increased flavonoid yield [35]. The present optimization process leads to a reduction

123
Extractive determination of bioactive flavonoids from… 797

of the solvent volume and overall cost of the process. An optimum solid-to-liquid
ratio of 1:20 g/mL is recommended for efficient extraction.

Optimization of acid hydrolysis

The effects of various acid hydrolysis parameters were studied for formation of
aglycones of flavonoids while avoiding degradation of aglycones. Aqueous
methanolic (80 %) leaf extract was hydrolyzed using various concentrations (2, 3,
4, 5, and 6 M) of HCl. The QR and KF contents were maximum at 5 M HCl
(Fig. 3). Various temperatures, namely 50, 60, 70, 80, and 90 C, were compared
for effective conversion of glycosides to aglycones. The yield of aglycones
gradually increased with temperature, reaching a maximum at 80 C (Fig. 3), while
further rise in temperature decreased the yield, presumably due to thermal

Fig. 3 Effect of hydrolysis parameters: acid concentration (a, b), time (c, d), and temperature (e, f) on
yield of quercetin and kaempferol, respectively

123
798 J. Makasana et al.

degradation of flavonoids [36]. The QR and KF yields were maximum up to 30 min

extraction, while the yield decreased for longer durations (Fig. 3). This is in
agreement with earlier reports on flavonoids extracted from vegetables and fruits
[37]. In this study, we have increased the acid concentration and reduced the
hydrolysis time. With respect to reported results. We recommend optimum acid
hydrolysis conditions of 5 M HCl at 80 C for 30 min, since this represents an easy,
cost-effective, and efficient process for complete recovery of aglycones from CT
leaves.

Conclusions

Preliminary studies are required to screen for active principles and their in vitro
biological activities to provide scientific evidence of the pharmaceutical importance
of medicinal plants used in traditional medicinal systems. Nowadays, use of
medicinal plants and their bioactive phytocompounds is increasing worldwide, so
high-quality raw materials are always in demand by consumers. This study aimed to
determine methods for maximum flavonoid recovery from Clitoria ternatea using a
standardized extraction protocol. Such flavonoid-enriched extracts containing
valuable bioactive compounds may be used in dietary supplements or for
development of natural drugs with effective human health benefits. Moreover, this
study also widely assessed the phytochemical (TPY and FC) yields and antioxidant
(ABTS and DPPH radical scavenging) capacities of different solvent extracts. The
results reveal that the choice of solvent plays a key role, significantly (p \ 0.05)
affecting the extraction yield of bioactive constituents from butterfly pea leaf. Based
on correlation study, it is clear that flavonoids were more closely related to the
radical scavenging abilities of C. ternatea leaf extract. An improved HPLC method
for simultaneous determination of quercetin and kaempferol in short runtime has
been developed for quality assessment of the raw drug of the species, which may
also be adopted for formulations from this and other, similar herbs.

Acknowledgments We sincerely thank the Director of SVNIT, Surat and Director of ICAR-DMAPR,
Boriavi, Anand for extending the facilities necessary for this work. The authors are grateful to Dr. (Ms) K.
A. Geetha for providing C. ternatea leaf material, Dr. V. Ravi for critically going through the manuscript,
and Dr. (Ms) Aarti Kavane for technical support.

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflicts of interest.

References
1. A. Mustafa, C. Turner, Anal. Chim. Acta 703, 8 (2011)
2. V.T. Nguyen, M.C. Bowyer, Q.V. Vuong, I.A.V. Altena, C.J. Scarlett, Ind. Crop. Prod. 67, 192
(2016)
3. S.K. Banerjee, C.G. Bonde, J. Med. Plants Res. 5, 817 (2011)
4. Z.L. Sheng, P.F. Wan, C.L. Dong, Y.H. Li, Ind. Crop. Prod. 43, 778 (2013)

123
Extractive determination of bioactive flavonoids from… 799

5. A.M. Nuutila, K. Kammiovirta, Food Chem. 76, 519 (2002)

6. B. Halliwell, Nutr. Rev. 55, 44 (1997)
7. L. Zhang, Z.C. Tu, T. Yuan, H. Wang, Z.F. Fu, Q.H. Wen, X.Q. Wanga, Ind. Crop. Prod. 56, 223
(2014)
8. A.A.M. Botterweck, H. Verhagen, R.A. Goldbohm, J. Kleinjans, P.A. Van den Brandt, Food Chem.
Toxicol. 38, 599 (2000)
9. A. Scalbert, C. Manach, C. Morand, C. Remesy, L. Jimenez, Crit. Rev. Food Sci. Nutr. 45, 287
(2005)
10. J.A. Ross, C.M. Kasum, Annu. Rev. Nutr. 22, 19 (2002)
11. M.C.J. Montano, B.E. Morón, P.C. Guerrero, L.M. Lázaro, Mini Rev. Med. Chem. 11, 298 (2011)
12. L. Parul, K.R. Deepak, IJMU 2, 22 (2007)
13. J.D. Tena, E. Burgos-Morón, J. Calderón-Montaño, I. Sanz, J. Sainz, M. Lopez-Lazaro, Web Med
Cent. CANCER 4, WMC004264 (2013). doi:10.9754/journal.wmc.2013.004264
14. R.R. Mendonça Filho, Modern phytomedicine: turning medicinal plants into drugs, (Weinheim,
Weley, 2006), pp. 1–25
15. G.M. Cragg, D.J. Newman, Biochim. Biophys. Acta 1830, 3670 (2013)
16. L. Mouhssen, Pharmacol. Pharm. 4, 17 (2013)
17. P.K. Mukherjee, V. Kumar, N.S. Kumar, M. Heinrich, J. Ethnopharmacol. 120, 291 (2008)
18. N.K. Sethiya, A. Nahata, S.H. Mishra, V.K. Dixit, J. Chin. Integr. Med. 7, 1001 (2009)
19. Z. Ali, S.H. Ganie, A. Narula, M.P. Sharma, P.S. Srivastava, Ind. Crop. Prod. 43, 768 (2013)
20. M’hiri, I. Ioannou, N. Mihoubi Boudhrioua, M. Ghoul et al., Food Bioprod. Proc. 96, 161 (2015)
21. Q. Xu, Y. Shen, H. Wang, N. Zhang, S. Xu, L. Zhang, Food Chem. 138, 2122 (2012)
22. K.S. Rao, P.K. Chaudhury, A. Pradhan, Food Chem. Toxicol. 48, 729 (2010)
23. R. Re, N. Pellegrini, A.S. Protegenete, A. Pannala, M. Yang, C. Rice-Evans, Free Radic. Biol. Med.
26, 1231 (1999)
24. O. Sticher, Nat. Prod. Rep. 25, 517 (2008)
25. H.M. Amir, C.M.M. Duarte, F. Maiza, Ind. Crop. Prod. 67, 249 (2015)
26. S.N. Lou, Y.S. Hsu, C.T. Ho, J. Food Drug Anal. 22, 290 (2014)
27. C. Yin, D. Wan, Pharmacogn. Mag. 8, 273 (2012)
28. A. Ghasemzadeh, H.Z. Jaafar, E. Karimi, A. Rahmat, B.M.C. Complemen, Altern. Med. 14, 318
(2014)
29. B.W. Williams, M.E. Cuvelier, C. Berset, LWT Food Sci. Tech. 28, 25 (1995)
30. M. Ozgen, R.N. Reese, A.Z. Tulio Jr., J.C. Scheerens, A.R. Miller, J. Agric. Food Chem. 54, 1151
(2006)
31. Z.R. Addai, A. Abdullah, S.A. Mutalib, J. Med. Plants Res. 7, 3354 (2013)
32. Q.V. Nguyen, V.B. Nguyen, J.B. Eun, S.L. Wang, D.H. Nguyen, T.N. Tran, A.D. Nguyen, Res.
Chem. Intermed. 42, 5859 (2016)
33. F. Kormin, I. Ahmed, R.M. Yunus, Z.A.M. Yusof, Int. J. Eng. Technol. 10, 7 (2010)
34. J. Dai, R. Mumper, Molecules 15, 7313 (2010)
35. T. Sathishkumar, R. Baskar, M. Aravind, S. Tilak, S. Deepthi, V.M. Bharathikumar, ISRN Biotech. 1,
1 (2013)
36. S. Deng, B.J. West, C.J. Jensen, Adv. J. Food Sci. Technol. 3, 155 (2011)
37. M.G.L. Hertog, P.C.H. Hollman, D.P. Venema, J. Agric. Food Chem. 40, 1591 (1992)

123