The Journal of Supercritical Fluids 147 (2019) 1–8

Contents lists available at ScienceDirect

The Journal of Supercritical Fluids

journal homepage: www.elsevier.com/locate/supflu

Assessment of the bioactive capacity of extracts from Leptocarpha rivularis T

stalks using ethanol-modified supercritical CO2
⁎
Edgar Uquiche , Claudia Campos, Claudia Marillán
Department of Chemical Engineering, Center of Food Biotechnology and Bioseparations, BIOREN, Universidad de La Frontera (UFRO), P.O. Box 54-D, Temuco, Chile

G R A P H I C A L A B S T R A C T

A R T I C LE I N FO A B S T R A C T

Keywords: Supercritical extraction from L. rivularis stalks using ethanol-modified supercritical CO2 was studied. The effects
Supercritical extraction of ethanol concentration (0.5―1.5 wt.%) and pressure (20―40 MPa) on the total extract yield (Y1, g/kg d.s.),
Leptocarpha rivularis phenolics extraction yield (Y2, mg GAE/kg d.s.), antioxidant (Y3, mmol TE/kg d.s.) and anti-inflammatory ac-
Bioactive properties tivities (Y4=IC50, mg/mL) were studied. The greatest total extract yield, phenolics extraction yield and anti-
Modifier ethanol
oxidant activity were reached at 1 wt.% and 40 MPa. Anti-inflammatory activity improved with the addition of
ethanol, being higher at 1.5 wt.%. The following results were obtained at the selected condition (1 wt.%,
40 MPa): Y1 = 34.15 g/kg d.s, Y2 = 658.44 mg GAE/kg d.s., Y3 = 0.103 mmol TE/kg d.s. and Y4 = 3.38 mg/mL.
Important bioactive compounds like caryophyllene oxide, quercetin, kaempferol and resveratrol were quantified
in the selected extract. The extract showed inhibitory capacity against α-amylase and α-glucosidase, demon-
strating important bioactive properties.

1. Introduction
Leptocarpha rivularis is a medicinal plant endemic to southern Chile
belonging to the Asteraceae family. L. rivularis is a Chilean medicinal
plant and its application forms part of the ancestral medicine of the
Mapuche people (pre-Hispanic inhabitants of Chile) [1]. Nowadays it is
marketed in pharmacies and alternative medicine shops as a product to
make infusions. Bioactive compounds present in the plant have shown
an inhibitory and cytotoxic effect on cancer cells [2]. Supercritical fluid
extraction is a technology for extraction of high quality natural com-
pounds, and provides a more selective and efficient extraction by
controlling temperature and pressure, which determine its solvent
power by regulating the density of the supercritical fluid. Carbon di-
oxide (CO₂) is the solvent most commonly used for supercritical

⁎
Corresponding author.
E-mail address: edgar.uquiche@ufrontera.cl (E. Uquiche).

https://doi.org/10.1016/j.supflu.2019.02.005
Received 24 November 2018; Received in revised form 6 February 2019; Accepted 7 February 2019
Available online 11 February 2019
0896-8446/ © 2019 Elsevier B.V. All rights reserved.
E. Uquiche, et al.
extraction applications in the food industry. CO₂ is a solvent suitable for
extraction of nonpolar organic compounds. Solubility of polar organic
compounds in supercritical CO₂ is low, however could improve by in-
creasing pressure or adding a polar modifier, such as ethanol, methanol
or water at a low concentration. Modifiers also can improve extraction
yield by reducing solute/solid matrix interactions, facilitating the re-
moval of solutes [3].
Studies on supercritical extraction using L. rivularis leaves were
carried out by Uquiche and Martínez [4] and Uquiche and Garcés [5].
Nowadays, the preparation of infusions of the plant is done with the
mixture of stalks and leaves (90:10). There are no previous studies on
supercritical extraction using stalks. A review on supercritical extrac-
tion of polyphenols from plant substrates indicates a range of pressures
used between 10―50 MPa, and temperature range between 318―348 K
(45―75 °C) [6]. In current work were selected pressures between 20
and 40 MPa because this range covers a large part of the pressure
conditions used in different studies. Likewise 60 °C was chosen as an
intermediate value in the temperature range reported in studies in the
review. In the case of the modifier, to obtain a balanced three-level
design, was used 0.5–1.0 - 1.5% by weight.
Studies have been reported on the use of ethanol as a modifier in
supercritical CO2 extraction from medicinal plants. Daukšas et al. [7]
informed that extract recovery increased with 1 wt.% of ethanol,
compared to extraction without modifier. However, extract recovery
decreased when ethanol was increased to 2 wt.%. Michielin et al. [3]
observed a decrease in extraction yield when an excess of ethanol as
modifier was used in supercritical CO2 extraction from Cordia verbe-
nacea. They pointed out that the decrease in extraction yield was be-
cause the addition of ethanol in 8 wt.% decreased the interactions be-
tween solute and CO2. Thus, increase of modifier concentration does
not necessarily mean a more efficient extract recovery.
Lipoxygenase is an enzyme involved in the regulation of in-
flammatory responses by biosynthesis of leukotrienes and lipoxins,
important mediators in inflammation. Leukotrienes have been related
to inflammatory diseases including cancer [8]. Phenolic compounds
exert an anti-inflammatory effect by lipoxygenase inhibition, through
their actions at the enzyme active site, and by chelating iron or redu-
cing it to its ferrous form [9]. Phenolic compounds are known for their
antioxidant properties. Free radicals can cause damage to molecules
such as DNA and proteins, resulting in pathologies associated with in-
flammatory disorders, causing cancer, diabetes and cardiovascular
diseases. Thus, between the production of free radicals and inflamma-
tion exist a close relationship. Diabetes mellitus type 2 is a disease
caused by deficiency in the action and/or secretion of insulin, causing
the accumulation of glucose in the blood (hyperglycaemia). α-amylase
hydrolyzes the α-1,4 glycosidic linkages of complex carbohydrates (e.g.
starch); while α-glucosidase catalyzes the final step of carbohydrate
hydrolysis, releasing glucose. Thus inhibition of α-amylase and α-glu-
cosidase activity may potentially offer a means of regulating blood
glucose levels. The aim of our work was to evaluate the effect of
modifier (ethanol) concentration and extraction pressure on the total
extract yield, total phenolics extraction yield, antioxidant activity and
anti-inflammatory activity of extract from L. rivularis stalks.
2. Materials and methods
2.1. Materials
Ethanol and methanol both analytical grade were procured from
J.T. Baker (Phillipsburg, NJ, USA). DPPH (2,2-diphenyl-1-picrylhy-
drazyl) was purchased from Calbiochem Co. (San Diego, CA, USA).
Acetonitrile, methanol and water, all grade for chromatography
(LiChrosolv Reag. Ph Eur) and DMSO (dimethyl sulfoxide) and formic
acid were obtained from Merck KGaA (Darmstadt, Germany). Iron (II)
sulfate heptahydrate and iron (III) chloride anhydrous were purchased
from ACROS Organics (Morris, NJ, USA). Potassium sodium tartrate
2
The Journal of Supercritical Fluids 147 (2019) 1–8
tetrahydrate, sodium hydroxide, lipoxygenase from Glycine max, α-
amylase from Bacillus licheniformis, α-glucosidase from Saccharomyces
cerevisiae, TROLOX (6-hydroxy-2,5,7,8-tetramethylchromane-2-car-
boxylic acid), DNSA (3,5-Dinitrosalicylic acid), pNPG (p-Nitrophenyl β-
D-glucopyranoside), p-Nitrophenol, 1,2,4,5-tetramethylbenzene, car-
yophyllene oxide, gallic acid, catechin, naringenin, quercetin, resvera-
trol, maltose monohydrate from potato were procured from Sigma-
Aldrich (St. Louis, MO, USA). Starch soluble for analysis (potato starch,
solubility in water at 90 °C: 50 g/L) was acquired from Scharlau Chemie
S.A. (Barcelona, Spain). Phosphate buffer pH 7 was obtained from
Winkler Ltda. (Santiago, Chile).
The L. rivularis stalks used as substrate were provided by Los Esteros
Company (La Union, Chile) (39°52′S, 73°14′W). The material was dried
in an oven (Memmert UF110, Schwabach, Germany) at 50 °C for 10 h.
Substrate was placed in a freezer (–60 °C) for 2 days and stalks were
ground in a Moulinex chopper (model AD5661AR, Moulinex, Ecully,
France). Substrate was screened through a series of sieves using a Ro-
Tap testing sieve shaker (model RX-29-10, W.S. Tyler, Mentor, OH,
USA). Average particle diameter (dp) was calculated using standard
method S319.3 [10].
2.2. Supercritical extraction
Supercritical extraction was carried out in a Spe-ed SFE unit
(Applied Separations, Allentown, PA) loading 12 g of substrate (feed) in
a 50 cm3 extraction vessel (14 mm inner diameter). The substrate had a
moisture content of 7.53 ± 0.55 g/100 g dry substrate, determined
gravimetrically by drying in an air-convection oven (Memmert model
UM-400, WTB Binder, Tuttlingen, Germany) set at 102 °C to a constant
final weight (10 h) [11]. Dry substrate (d.s.) is the substrate loaded into
the extractor vessel discounting its moisture. Depending on the ex-
traction conditions (20―40 MPa, at 60 °C), between 3.7 and 4.6 L NPT
per min of CO2 (Linde Chile S.A.) was used as solvent (superficial ve-
locity, 1 mm/s). The static 15-min extraction period was followed by a
dynamic extraction period, which varied between 44 and 54 min in
order to obtain a total solvent consumption of 30 kg CO2/kg d.s. In
supercritical extractions involving the use of modifier, ethanol was
pumped into the CO2 feed line at a flow rate between 0.04 and 0.11 mL/
min so as to achieve three levels of modifier concentration (0.5, 1.0 and
1.5 wt.%) by using an HPLC pump (Knauer, model K-501, Berlin, Ger-
many) (flow accuracy < 1%, at 1 mL/min, 12 MPa). The extracts were
collected during extraction in pre-weighed glass vials (60 cm3 capa-
city). The extracts were subjected to a gentle stream of nitrogen (Linde
Chile S.A.) to evaporate the ethanol. Extraction yield (Y1) was ex-
pressed as grams of extract per kilogram of dry substrate (g/kg d.s.).
2.3. Analysis of extracts
2.3.1. Total phenolics extraction yield
Total phenolics content was determined according to Folin-
Ciocalteu’s method [12] with minor modifications as described pre-
viously [4]. Extract solution (5 mg/mL) was prepared by dissolving
10 mg of extract in 2 mL of ethanol. Quantification was done using a
calibration curve constructed with gallic acid solution and expressed as
milligrams of gallic acid equivalents (GAE) per gram of extract (Cph, mg
GAE/g). Total phenolics extraction yield (Y2, mg GAE/kg d.s.) was
calculated as Y1×Cph.
2.3.2. Antioxidant activity
Antioxidant activity was measured by DPPH radical scavenging
assay based on the standard calibration curve, according to Brand-
Williams [13] with minor modifications [4]. Extract solution (10 mg/
mL) was prepared by dissolving 50 mg of extract in 5 mL of ethanol.
Antioxidant activity was expressed as mmol TROLOX equivalent (TE)
per kilogram of dried substrate (Y3, mmol TE/kg d.s.). Additionally,
antioxidant activity was measured by Ferric-Reducing Antioxidant
E. Uquiche, et al.
Power (FRAP) assay according to Benzie and Strain [14]. Extract so-
lution (10 mg/mL) was prepared by dissolving 50 mg of extract in 5 mL
of ethanol. A standard calibration curve was constructed using aqueous
solutions of FeSO4·7H2O at different concentrations. The FRAP value
was calculated and expressed as mmol Fe+2 equivalents per kilogram of
dried substrate (mmol Fe+2/kg d.s.) based on the standard calibration
curve.
2.3.3. Anti-inflammatory activity
Anti-inflammatory activity was measured by lipoxygenase inhibi-
tion assay according to Ahmed et al. [15], with some modifications [4].
Anti-inflammatory activity (Y4) was expressed as IC50 value (mg/mL),
which represents the extract concentration sufficient to obtain 50%
inhibition of maximum capacity for lipoxygenase activity.
2.3.4. Individual quantification of flavonoids
Quantification of catechin, quercetin and resveratrol was carried
out according to Kim [16], using a HPLC–DAD 1260 infinity system
(Agilent Technologies, Santa Clara, CA, USA), equipped with a qua-
ternary pump. Stock solution of samples were prepared at the con-
centration of 10 mg/mL in methanol. Chromatographic separations
were performed on an column Agilent Zorbax rapid resolution high-
definition (RRHD) SB-C18 (2.1 mm i.d. × 100 mm, 1.8 μm, catalog
number: 858758-902). The column temperature was 30 °C, the flow
rate was 0.3 mL/min and injection volume was 20 μL. Mobile phase A
(water with 0.1% formic acid) and B (acetonitrile with 0.1% formic
acid) were used for gradient elution.
The gradient elution program was used as follows: 0% B (0 min), 5%
(0―3.5 min), 15% (3.5―7.1 min), 40% (7.1―25 min), 40%
(25―26 min), 100% (26―27 min), 100% (27―29 min), and 0%
(29―35 min).
Flavonoids were identified by comparing their retention times with
those of standards. Solution of standards were prepared at the con-
centration of 0.1 mg/mL in methanol. Quantification was carried out by
the internal standard method [17] using naringenin. The HPLC-DAD
was controlled using the Agilent ChemStation Software.
2.3.5. Individual quantification of terpenes
Quantification of caryophyllene oxide was carried out according to
Uquiche and Martínez [4] using a GC-FID 6850 (Agilent Technologies,
Santa Clara, CA, USA) with a flame ionization detector equipped with a
HP-5 capillary column (0.25 mm i.d. × 30 m, 0.25 μm, catalog number:
19091S-433). Internal standard method [17] was used for quantifica-
tion with 1,2,4,5-tetramethylbenzene as the internal standard. The GC-
FID was controlled using the Agilent ChemStation Software.
2.3.6. Inhibition of α-amylase
The inhibition of α-amylase was carried out according to Rahali
et al. [18] with some modifications. Extract solution was prepared in
DMSO at different concentrations (2―20 mg/mL). For the color re-
agent, 20 mL of 96 mM solution of DNSA was mixed with a solution of
sodium potassium tartrate (12 g of potassium sodium tartrate dissolved
in 8 mL of 2 M sodium hydroxide), and was brought to a total volume of
40 mL with deionized water. Extract solution (200 μL) was mixed with
400 μL of starch solution (1% w/v in phosphate buffer) and incubated
for 3 min at 37 °C. Then was added 200 μL of enzyme solution (2 U/mL
in phosphate buffer) and incubated for another 3 min at 37 °C. DNSA
solution (400 μL) was added and the solution was heated at 95 °C for
15 min, and followed by cooling to ˜20 °C and addition of 3600 μL of
deionized water. The absorbance at 540 nm was measured in the
spectrophotometer. Maltose was quantified using a calibration curve. A
blank was prepared by replacing 200 μL of enzyme solution with 200 μL
of phosphate buffer. A control that represents 100% of the enzymatic
reaction was prepared by replacing 200 μL of extract solution with
200 mL of DMSO solution. The inhibition (%) was calculated with the
following Eq. (1):
3
The Journal of Supercritical Fluids 147 (2019) 1–8
mg maltosesample ⎤
% Inhibition = ⎡1 − × 100
⎢ mg maltosecontrol ⎥ (1)
⎣ ⎦
2.3.7. Inhibition of α-glucosidase
The inhibition of α-glucosidase was determined according to
Muccilli et al. [19] with some modifications. Extract solutions in DMSO
at different concentrations were prepared (1.0―3 mg/mL). A 3 mM
solution of pNPG was prepared as substrate. The enzyme solution was
prepared at a concentration of 1 U/mL in phosphate buffer. For the
inhibition test 50 μL of extract solution was mixed in a test tube with
125 μL of pNPG and 1250 μL of phosphate buffer, and incubated for
10 min at 37 °C. To start the reaction, 50 μL of the enzyme solution was
added and incubated for 20 min at 37 °C. Then the reaction was stopped
by adding 4000 μL of 0.1 M solution of Na2CO3. A blank solution was
prepared by replacing 50 μL of the enzyme solution with 50 μL phos-
phate buffer. A control assay representing 100% of the enzymatic re-
action was performed by replacing the 50 μL of the extract solution with
50 μL DMSO. The activity of α-glucosidase was determined by mea-
suring the p-nitrophenol released from the hydrolysis of pNPG at
400 nm in the spectrophotometer. The inhibition (%) was calculated
with the following Eq. (2):
Asample − Ablank ⎞ ⎤
Inhibition (%) = ⎡1 − ⎛ ⎜ ⎟ × 100
⎢ ⎝ Acontrol ⎠⎥ (2)
⎣ ⎦
2.4. Experimental design
A two-level full factorial design [22] with four replicates in the
central point was used to evaluate the effects of the independent vari-
ables coded modifier concentration (X1) and coded pressure (X2) on the
total extract yield (Y1, g/kg d.s.), total phenolics extraction yield (Y2,
mg GAE/kg d.s.), antioxidant (Y3, mmol TE/kg d.s.) and anti-in-
flammatory (Y4=IC50, mg/mL) activities. A second-order model (Eq.
(3)) was used to describe the response variable as a function of coded
modifier concentration (X1) and coded pressure (X2), where A0 is a
constant; A1 and A2 are linear coefficients; A12 is a cross-product
coefficient; and A12 and A22 are quadratic coefficients. Coefficients of the
second-order model were estimated using Design-Expert Software,
version 6.0.1 (Stat-Ease, Inc., Minneapolis, MN, USA). Goodness of fit of
the second-order model was evaluated by analysis of variance. Statis-
tical significance was based on the total error criteria with a confidence
level of 95%.
Y = A0 + A1 X1 + A2 X2 + A12 X1 X2 + A12 X12 + A22 X22 (3)
3. Results and discussion
3.1. Substrate
Dried and milled stalks had a moisture content of 7.53 ± 0.55/
100 g d.s. Average particle diameter (dp) was 0.95 mm and was con-
sidered suitable for supercritical fluid extraction in packed beds. If
particles are too small, they may cause CO2 channelling through the
packed bed, reducing extraction efficiency.
3.2. Experimental results and statistical analysis
Table 1 shows results of extraction yield (Y1) from L. rivularis stalks
using supercritical CO2 as a function of modifier concentration and
extraction pressure. The highest extraction yield was 4.3-fold more than
the lowest yield. The highest value for total phenolics extraction yield
(Y2) was 5.1-fold more than the lowest value. For the antioxidant ac-
tivity (Y3, mmol TE/kg d.s.) and anti-inflammatory activity (Y4, mg/
mL) the differences were 4.4-fold and 2.3-fold respectively. Table 2
E. Uquiche, et al. The Journal of Supercritical Fluids 147 (2019) 1–8

Table 1
Total extract yield (Y1, g/kg d.s.), total phenolics extraction yield (Y2, mg GAE/kg d.s.), antioxidant activity (Y3, mmol TE/kg d.s.), anti-inflammatory activity (Y4,
mg/mL) and extraction rate parameter (k1, min−1) as a function of modifier concentration (C, wt.%) and pressure conditions (P, MPa).
C P X1 X2 Y1 Y2 Y3 Y4 k1 RMSD
wt.% MPa (―) (―) g/kg d.s. mg GAE/kg d.s. mmol TE/kg d.s. mg/mL min−1

0.5 20 ―1 ―1 6.29 ± 1.33 72.89 ± 15.03 0.0224 ± 0.0034 2.50 ± 0.01 0.154 0.04969
1.5 20 1 ―1 5.69 ± 0.42 79.75 ± 5.68 0.0130 ± 0.0015 1.72 ± 0.16 0.150 0.02742
0.5 40 ―1 1 21.86 ± 1.80 202.77 ± 17.30 0.0576 ± 0.0001 3.90 ± 0.12 0.163 0.03866
1.5 40 1 1 19.11 ± 1.73 148.70 ± 13.78 0.0450 ± 0.0093 2.90 ± 0.11 0.199 0.03440
1.0 30 0 0 22.70 ± 0.95 352.68 ± 1.68 0.0474 ± 0.0017 3.17 ± 0.27 0.231 0.05439
1.0 30 0 0 24.46 ± 1.13 371.37 ± 1.36 0.0572 ± 0.0049 3.27 ± 0.08 0.216 0.10466
1.0 30 0 0 22.61 ± 1.04 346.54 ± 2.51 0.0536 ± 0.0005 2.87 ± 0.17 0.203 0.02681
1.0 30 0 0 22.35 ± 0.71 352.81 ± 2.07 0.0567 ± 0.0090 3.31 ± 0.04 0.220 0.04561

summarizes the statistical indicators obtained from the ANOVA applied

to the models selected (Eqs. (4)–(7)), with significant coefficients only
(p ≤ 0.05). Models were significant at p < 0.001 and were considered
adequate given the high coefficient of determination (R2 > 0.95 and
adjusted-R2 > 0.92), the high signal-to-noise ratio (> 4) and low
coefficient of variation (< 10). The plot between predicted and ex-
perimental values showed good correlation for all responses (data not
shown), with significant correlation coefficients (r > 0.90, p < 0.01).
Analysis of variance was used to find significance for the selected
second-order model coefficients. A large regression coefficient and
small p-value would indicate a more significant effect on the response
variable.
For total extract yield (Y1), the quadratic coefficient of modifier
concentration ( A12 = ―9.793) had the largest effect, followed by the
linear term of pressure (A2= +7.248). For phenolic extraction yield
(Y2) the quadratic coefficient of modifier concentration ( A12 =
―229.824) was the factor with the largest effect, followed by the linear
coefficient of pressure (A2= +49.709). Quadratic coefficient of
modifier concentration ( A12 = ―0.01955) had the largest effect on Fig. 1. Surface plot of total extract yield (Y1, g/kg dry substrate) as a function
antioxidant activity, followed by the linear coefficient of pressure (A2= of modifier concentration (C, wt.%) and pressure (P, MPa).
+0.01682) and modifier concentration (A2= ―0.00549). For anti-in-
flammatory activity (Y3), the linear coefficient of pressure (A2= Y3 = 0.05406 − 0.00549X1 + 0.01682X2 − 0.01955X12 (6)
+0.648) was the most significant effect, followed by linear (A1=
―0.445) and quadratic ( A12 = ―0.401) coefficients of modifier con- Y4 = 3.155 − 0.445X1 + 0.648X2 − 0.401X12 (7)
centration. For variables Y1 and Y2 there was no significant effect of the
linear term of modifier concentration (p = 0.1671 and p = 0.2567, re-
spectively) but these coefficients were not removed in order to maintain 3.3. Total extract yield
model hierarchy. For all responses, there was no significant effect of the
interaction coefficient. Thus the second-order models established a Fig. 1 shows that total extract yield increased with rising pressure at
statistically significant relationship between response variables with the every modifier concentration value. Y1 increased from 15.78 to
modifier concentration (C) and pressure (P). To better visualize the 30.28 g/kg d.s. (˜2-fold) when P changes from 20 to 40 MPa (C = 1 wt
effect of C and P on response variables in the experimental region, %), which exemplified the positive linear effect of pressure
surface response graphs were generated using the following models, (p = 0.0001). This is due to the fact that increasing extraction pressure
at constant temperature thereby decreases intermolecular distance be-
Y1 = 23.03 − 0.838X1 + 7.248X2 − 9.793X12 (4)
tween CO2 molecules and increases CO2 density. Higher density in-
creases interactions between CO2 and organic molecules [20]. In the
Y2 = 355.85 − 11.80X1 + 49.71X2 − 229.82X12 (5)
literature, the positive effect of pressure on extraction yield from

Table 2
Analysis of variance of regression coefficients and statistical appropriateness indicators of the second-order models selected.
Regression coefficients Y1 Y2 Y3 Y4

Estimate p-value Estimate p-value Estimate p-value Estimate p-value

A1 −0.838 0.1671 −11.801 0.2567 −0.00549 0.0397 −0.445 0.0076

A2 +7.248 0.0001 +49.709 0.0051 +0.01682 0.0008 +0.648 0.0019
A12 −9.793 0.0002 −229.824 < 0.0001 −0.01955 0.0016 −0.401 0.0338
A12
F 136.67 0.0002 121.41 0.0002 50.34 0.0012 29.1056 0.0035
R2 0.990 0.989 0.974 0.956
adjusted-R2 0.983 0.981 0.955 0.923
Signal-to-noise ratio 25.45 23.079 17.268 17.284
CV 5.48 7.409 8.252 6.05

4
E. Uquiche, et al.
herbaceous matrices has been extensively reported. Fig. 1 shows non-
linear effect (p = 0.0002) of modifier concentration on total extract
yield, regardless of pressure. At 40 MPa, Y1 increased when C increased
from 0.5 to 1 wt.% (from 21.3 to 30.3 g/kg d.s.) and then it decreased
until 19.7 g/kg d.s. when C changed to 1.5 wt.%. de Campos et al. [21]
have pointed out that the increase in ethanol concentration can affect
extraction yield by: (a) the saturation of CO2 with ethanol, which re-
sults in the formation of two phases, reducing the solvent capacity of
the CO2-ethanol mixture; (b) the formation of hydrogen bonds between
the hydrogen from one molecule with oxygen from another molecule
(ethanol–ethanol hydrogen bonding). The formation of ethanol–ethanol
hydrogen bonding reduces the availability of ethanol to interact with
solute molecules; and therefore less solute molecules are solubilized,
thereby decreasing the extraction yield [21,22]. Addition of ethanol as
a modifier affects the temperature (Tc) and pressure (Pc) critical for
CO2. Aidil et al. [23] reported the corresponding critical values for a
concentration of 5 wt.% of ethanol in CO2 as Tc = 42.5 °C and
Pc = 7.32 MPa. In the current work, temperature was at 60 °C and
pressure over 20 MPa, so that our working conditions were in the su-
percritical region for ethanol-modified CO2. Thus, the decrease in total
extract yield may be due to the prevalent formation of ethanol–ethanol
hydrogen bonding.
The increase and decrease of extraction yield with the modifier
addition in supercritical extraction from herbaceous matrices has been
reported in the literature. Daukšas et al. [7] studied supercritical ex-
traction from sage (Salvia officinalis L.) with ethanol-modified CO2 at
100 °C and 35 MPa. Extraction yield increases with 1 wt.% of modifier;
but decreases with an increase to 2 wt.%. Biscaia and Ferreira [24] used
ethanol-modified CO2 (2―7 wt.%) to extract propolis at 40 °C and
15 MPa. Higher recovery of extract was achieved with 5 wt.% modifier.
Similar behavior was informed by Kamali et al. [25] in supercritical
extraction from Biebersteinia multifida DC using ethanol-modified CO2.
Therefore, an increase and subsequent decrease of total extract yield
with an increase of modifier concentration, has been observed.
To illustrate the effect of modifier concentration and pressure on
extraction kinetics, cumulative extraction curves were constructed at
the extraction conditions studied (Fig. 2). The curves show the constant
extraction rate (CER) period, where the extraction mechanism is con-
trolled by solubility and the diffusion-controlled rate (DCR) period,
where the extraction mechanism is controlled by diffusion [20,26].
Differences between extraction curves in the CER period were observed.
Cumulative extraction curves were modeled using Peleg’s model (Eq.
(8)) [27] and the following kinetic parameters were estimated: k1 is the
Fig. 2. Effect of modifier concentration (wt.%) and pressure (MPa) on cumu-
lative extraction yield (Y1, g/kg d.s.) versus specific solvent consumption (kg
CO2/kg dry substrate) using supercritical CO2 at 60 °C: 0.5 wt.% ― 40 MPa
(square), 1.5 wt.% ― 40 MPa (circle), 0.5 wt.% ― 20 MPa (cross), 1.5 wt.% ―
20 MPa (diamond) and 1.0 wt.% ― 30 MPa (triangle).
5
The Journal of Supercritical Fluids 147 (2019) 1–8
parameter related to extraction rate at the very beginning of the process
(min−1); k2, the parameter related to maximum extraction yield
(min−1); q, amount of extract (g/kg d.s.); q0, maximum amount ex-
tracted (g/kg d.s.); and t, time (min). Parameters k1 and k2 were ob-
tained by adjusting experimental extraction data to the kinetic model
for each extraction condition. The difference between values observed
experimentally (Yexp) and values predicted by the kinetic model (Ypred)
were established by root mean square deviation (RMSD) (Eq. (9)).
q k1 t
=
q0 1 + k2 t (8)
n
∑i (Yexp − Ypred )2
RMSD =
n (9)
Initial extraction rate parameter (k1) ranged between 0.150 and
0.231 min−1 (Table 1). Value of k1 increased with pressure at a con-
stant temperature; and increased and decreased with the concentration
of modifier, similar to behaviour of Y1. In fact, there was a positive
correlation between Y1 and k1 (r = 0.804; p ≤ 0.05). This positive
correlation may be because the initial extraction rate depends on the
solute solubility in supercritical CO2. This condition does not necessa-
rily indicate that equilibrium was attained. Therefore, the differences
observed in the total extract yield are mainly explained by the effects of
extraction conditions on the solvation power of CO2 and consequently
on the initial extraction rate.
3.4. Total phenolics extraction yield
Fig. 3 shows an increase in total phenolics extraction yield with
pressure increase (p = 0.0051), at every modifier concentration. To
exemplify, Y2 increased from 88.12 to 187.54 mg GAE/kg d.s and from
64.52 to 163.93 mg GAE/kg d.s through increase from 20 to 40 MPa, at
0.5 and 1.5 wt.%, respectively. Fig. 3 shows significant quadratic effect
(p < 0.0001) of modifier concentration on phenolic extraction yield
(Y2) regardless of pressure, increasing and decreasing with the increase
of modifier concentration. To exemplify, at 40 MPa, Y2 increased from
187.54 to 405.56 mg GAE/kg d.s. by increase from 0.5 to 1 wt.% of
modifier. Then it decreased until 163.93 mg GAE/kg d.s. when 1.5 wt.%
was used. As has been argued before, the decrease in the phenolic ex-
traction yield at 1.5% may be due to the prevalence of the ethanol-
ethanol interaction, reducing the availability of ethanol to interact with
the solute molecules [21,22]. Machado et al. [28] studied conditions to
extract phenolic compounds from propolis using supercritical CO2
modified with 1 and 2 wt.% of ethanol at 50 °C and 25 MPa. Addition of
Fig. 3. Surface plot of total phenolics extraction yield (Y2, mg GAE/kg dry
substrate) as a function of pressure (P, MPa) and modifier concentration (C, wt
%).
E. Uquiche, et al. The Journal of Supercritical Fluids 147 (2019) 1–8

3.5. Antioxidant activity

Fig. 4 shows that antioxidant activity increased with rising pressure

Fig. 4. Surface plot of antioxidant activity (Y3, mmol TE/kg dry substrate) as a
function of pressure (P, MPa) and modifier concentration (C, wt.%).
regardless of modifier concentration value. Y3 increases from 0.0232 to
0.0568 mg TE/kg d.s. (˜2.4-fold) at C = 0.5 wt.% and from 0.0122 to
0.0458 mg TE/kg d.s. (˜3.8-fold) at C = 1.5 wt.%, when P changes from
20 to 40 MPa, demonstrating the linear effect of pressure (p = 0.0008).
Fig. 4 shows significant non-linear effect (p = 0.0016) of modifier
concentration on antioxidant activity, regardless of pressure. To ex-
emplify, at 40 MPa, Y3 increased when C increased from 0.5 to 1 wt.%
(from 0.0568 to 0.0709 g/kg d.s.) and then it decreased until 0.0458 g/
kg d.s. when C changed to 1.5 wt.%. The effect of the modifier addition
and the pressure on Y3 followed a behavior similar to the extraction of
phenolic compounds. In fact, Y2 and Y3 correlated positively (p < 0.1).
Phenolic compounds are known for their antioxidant and anti-in-
flammatory properties [30]. Phenolic compounds exert their anti-
oxidant activity by neutralizing free radicals or by chelating metal ions
[31]. In this study, the behavior of the antioxidant activity with the
variables modifier concentration and pressure, can be explained by the
phenolic compounds extracted. A positive correlation has been in-
formed between antioxidant activity and phenolic compounds, in-
dicating that these compounds are mostly responsible antioxidant ca-
pacity [32,33].

3.6. Anti-inflammatory activity

Fig. 5 shows that Y4 (IC50 values) increased with increasing pressure

regardless of modifier concentration value, demonstrating the linear
effect of pressure (p = 0.0019). The IC50 value represents the extract
concentration required to inhibit the activity of lipoxygenase by 50%.
Therefore, the anti-inflammatory activity of the extract is higher at low
IC50 values. This means that the anti-inflammatory activity decreased
with the increase in pressure. Lipoxygenase are oxidative enzymes in-
volved in inflammatory disorders [8]. Phenolic compounds inhibit li-
poxygenase, by chelating or reducing its non-heme iron atom in their
active site [9]. Zhang et al. [33] and Talhouk et al. [34] point out that
phenolic compounds can act as anti-inflammatory agents. A positive
correlation between anti-inflammatory activity and total polyphenols
Fig. 5. Surface plot of anti-inflammatory activity (Y4, mg/mL) as a function of
has been reported [35]. However in this study, the extraction of phe-
modifier concentration (C, wt.%) and pressure (P, MPa). nolic compounds did not explain the behavior of the anti-inflammatory
activity. A plausible explanation would be that in a low pressure range,
the process could co-extract compounds with higher anti-inflammatory
Table 3
activity than phenolic compounds. For instance, compounds such as
Characteristics of extract from L. rivularis stalks obtained with ethanol-modified
terpenoids have been shown to have anti-inflammatory activity [36].
CO2 at 1 wt.% and 40 MPa (60 °C).
Regarding the effect of the modifier, Fig. 5 shows that Y4 decreased
Characteristics Experimental from 2.63 to 1.66 mg/mL and from 3.93 to 2.96 mg/mL, when C
changed from 0.7 to 1.5 wt.% at 20 and 40 MPa, respectively. This
Total extract yield (g/kg d.s.) 34.15 ± 0.89
Total phenolics extraction yield (mg GAE/kg d.s.) 658.44 ± 10.85 means that the anti-inflammatory activity improved with increasing
TEAC (mmol TE/kg d.s.) 0.103 ± 0.003 modifier concentration, in the upper range, which correspond to the
FRAP (mmol Fe+2/kg d.s.) 16.69 ± 0.52 significant linear effect (p = 0.0019). While, in the lower range of C
Characteristics of extract (0―0.7 wt.%), the quadratic effect becomes important, observing a
Total phenolics content (mg GAE/g) 19.28 ± 0.19 plateau.
FRAP (μmol Fe+2/g) 488.57 ± 2.45 In the upper range of C (0.7―1.5 wt.%), anti-inflammatory activity
TEAC (mmol TE/kg) 3.018 ± 0.162
Lipoxygenase inhibition assay (IC50, mg/mL) 3.38
continued to improve (decreased Y4); this contrasts with the decreased
phenolic extraction yield, suggesting that in this region the anti-in-
Bioactive compounds
flammatory properties may be given by other constituents of the ex-
Caryophyllene oxide (mg/g) 7.28
Catechin (mg/100 g) 32.46 tract. Ashraf-Khorassani et al. [37] pointed out that the addition of
Quercetin (mg/100 g) 16.40 small amounts of modifier can enhance the efficiency of trace metal
Resveratrol (mg/100 g) 36.59 extraction. In this respect, Ravipati et al. [38] reported that the anti-
inflammatory activity of extracts from some Chinese medicinal plants
correlated well with the trace metal contents, pointing to metals such as
ethanol significantly improved extraction of phenolic compounds, with Zn, Mg and Se. Alam et al. [39] pointed out that metal complexes such
1 wt.% ethanol being significantly better than 2 wt.% or without as Cu and Zn showed excellent anti-inflammatory activity. Thus one
modifier. Similar behavior for supercritical extraction of phenolic possible explanation for the increase in anti-inflammatory activity may
compounds from Citrus depressa Hayata using etanol-modified CO2, was be the presence of a higher concentration of trace metals in the upper
informed [29]. range of C. This hypothesis deserves further exploration in future

6
E. Uquiche, et al.
Table 4
Percent inhibition of α-amylase and α-glucosidase at varying concentrations of extract obtained with ethanol-modified CO2 at 1 wt.% and 40 MPa (60 °C).
Enzyme α-amylase
Concentration Inhibition IC50
(mg/mL) (%) mg/mL
0.1 24.6 ± 1.2 15.1 ± 1.1
0.5 28.6 ± 1.9
2 32.5 ± 2.0
6 34.4 ± 1.7
10 43.9 ± 2.6
14 48.8 ± 2.1
20 56.8 ± 4.4
studies.
3.7. Characteristics of the selected extract
Supercritical extraction at 1 wt.% ethanol addition and 40 MPa was
selected because it produced higher total extract yield (30.28 g/kg d.s.),
total phenolics extraction yield (405 mg GAE/kg d.s.) and antioxidant
activity (0.071 mmol TE/kg d.s.). To validate the analysis, experimental
extractions were carried out at the selected supercritical condition.
Table 3 shows the characteristics of the extract obtained at the selected
condition. The second-order models showed good coefficients of de-
termination (R2≥0.92, Table 2) and were adequate for predicting re-
sponse variables. In addition, the antioxidant activity of the selected
extract was measured by FRAP assay, obtaining 16.69 mmol Fe+2/kg
d.s. FRAP represents the reducing capacity of antioxidant compounds in
a redox reaction, in which one reactive species is reduced at the expense
of the oxidation of another [14].
There is no information in the literature for L. rivularis stalks.
Uquiche and Martinez [4] reported antioxidant activity (0.099 mmol
TE/kg d.s.) and anti-inflammatory activity (IC50 = 1.12 mg/mL) for
extract of L. rivularis leaves obtained at 40 °C and 10 MPa. Uquiche and
Garcés [5] reported a total phenolic content of 1001 mg GAE/kg d.s. for
extract of the same substrate obtained at 40 °C and 15 MPa. Although
the differences between these values and those presented in Table 3 are
clear, and are attributable to differences in substrate type, extraction
condition and solvent, the values are of the same order of magnitude
and demonstrate important properties in L. rivularis stalk extract. To
complement the characteristics of the selected extract, a quantitative
determination was made of bioactive compounds (Table 3), such as
caryophyllene oxide (7.28 mg/g), catechin (32.46 mg/100 g), quercetin
(16.40 mg/100 g), resveratrol (36.59 mg/100 g). For caryophyllene
oxide, the value is of similar magnitude to values reported in previous
works on supercritical extraction from L. rivularis leaves, of 7.49 mg/g
[4] and 10 mg/g [4,5]. Bimakr et al. [40] reported between 8.1 and
13.5 mg/100 g of catechin in supercritical extract from spearmint
(Mentha spicata L.) leaves. Sesquiterpenes (e.g. caryophyllene) and
flavonoids (e.g. catechin, quercetin, resveratrol) have important
bioactive properties. Quercetin exhibits antioxidant and anti-in-
flammatory properties [41,42], as does resveratrol [43]. Caryophyllene
oxide possesses several biological effects such as anti-inflammatory
activity [44].
The selected supercritical extract exhibited inhibitory properties
against α-amylase and α-glucosidase, two hydrolytic enzymes asso-
ciated with diabetes mellitus type 2. Table 4 shows the capacity of the
selected extract to inhibit (%) α-amylase and α-glucosidase as a func-
tion of its concentration. According to the IC50 value, the selected ex-
tract was more effective in inhibiting α-glucosidase (2.7 mg/mL) than
α-amylase (15.1 mg/mL). There are reports in the literature of the
properties of flavonoids as inhibitors of α-glucosidase [45] and α-
amylase [46]. On the inhibitory effect of sesquiterpenes, it is known
that these compounds are important constituents in essential oils. An
inhibitor effect on α-amylase and α-glucosidase has been reported for
7
The Journal of Supercritical Fluids 147 (2019) 1–8
Enzyme α-glucosidase
Concentration Inhibition IC50
(mg/mL) (%) mg/mL
0.1 38.4 ± 3.4 2.7 ± 0.1
0.5 46.6 ± 1.4
0.8 48.6 ± 2.6
1.0 53.6 ± 0.3
1.5 72.4 ± 0.3
3.0 78.5 ± 1.2
essential oils from Cymbopogon nardus (citronella grass) [18], Cymbo-
pogon citratus (lemongrass) [18] and Ocimum basilicum (sweet basil)
[47]. Thus sesquiterpenes may have antidiabetic properties.
4. Conclusions
Modifier (ethanol) concentration and extraction pressure sig-
nificantly affected the total extract yield, total phenolics extraction
yield, and antioxidant and anti-inflammatory activities of extracts from
L. rivularis stalks. The appropriate condition to obtain supercritical ex-
tract from L. rivularis stalks was considered to be 1 wt.% modifier and
40 MPa, according to the higher total extract yield, total phenolics ex-
traction yield, and antioxidant activity. A higher modifier concentration
does not necessarily result in an improvement in the total extract yield
and total phenolics extraction yield. Higher anti-inflammatory ability of
the extract was obtained with higher modifier concentration. Bioactive
compounds were quantified in the selected extract, such as car-
yophyllene oxide (sesquiterpene), catechin, quercetin and resveratrol
(flavonoids). The selected extract presented the capacity to inhibit en-
zymes associated with diabetes (α-amylase and α-glucosidase), showing
its potential benefits for health. This is the first report on extraction
from L. rivularis stalks using supercritical fluid technology.
Author disclosure statement
The authors declare no conflict of interest.
Declarations of interest
None.
Acknowledgments
This research was funded by Chilean agency Fondecyt (project
1170841).
References
[1] H.M. Niemeyer, Composition of essential oils from five aromatic species of
Asteraceae, J. Essent. Oil Res. 21 (2009) 350–353, https://doi.org/10.1080/
10412905.2009.9700189.
[2] C. Bosio, G. Tomasoni, R. Martínez, A.F. Olea, H. Carrasco, J. Villena, Cytotoxic and
apoptotic effects of leptocarpin, a plant-derived sesquiterpene lactone, on human
cancer cell lines, Chem. Biol. Interact. 242 (2015) 415–421, https://doi.org/10.
1016/j.cbi.2015.11.006.
[3] E.M.Z. Michielin, A.A. Salvador, C.A.S. Riehl, A. Smânia Jr, E.F.A. Smânia,
S.R.S. Ferreira, Chemical composition and antibacterial activity of Cordia verbe-
nacea extracts obtained by different methods, Bioresour. Technol. 100 (2009)
6615–6623, https://doi.org/10.1016/j.biortech.2009.07.061.
[4] E. Uquiche, M. Martínez, Glandular trichome disruption by rapid CO2 depressur-
ization as pretreatment for recovery of essential oil from Leptocarpha rivularis
leaves, Ind. Crops Prod. 83 (2016) 522–528, https://doi.org/10.1016/j.indcrop.
2015.12.075.
[5] E. Uquiche, F. Garcés, Recovery and antioxidant activity of extracts from
Leptocarpha rivularis by supercritical carbon dioxide extraction, J. Supercrit. Fluids
110 (2016) 257–264, https://doi.org/10.1016/j.supflu.2015.12.003.
E. Uquiche, et al.
[6] J.M. del Valle, E. Uquiche, L. Valenzuela, Supercritical fluid extraction of poly-
phenols, in: J. Cuevas-Valenzuela, J.R. Vergara-Salinas, J.R. Pérez-Correa (Eds.),
Advances in Technologies for Producing Food-Relevant Polyphenols, Editorial CRC
Press, Taylor & Francis Group, Boca Raton, FL, 2016, pp. 125–192.
[7] E. Daukšas, P.R. Venskutonis, V. Povilaityte, B. Sivik, Rapid screening of anti-
oxidant activity of sage (Salvia officinalis L.) extracts obtained by supercritical
carbon dioxide at different extraction conditions, NahrungFood 45 (2001) 338–341,
https://doi.org/10.1002/1521-3803(20011001)45:5<338::aid-food338>3.0.
co;2-t.
[8] R. Wisastra, F.J. Dekker, Inflammation, cancer and oxidative lipoxygenase activity
are intimately linked, Cancers 6 (2014) 1500–1521, https://doi.org/10.3390/
cancers6031500.
[9] M.J. Laughton, P.J. Evans, M.A. Moroney, J.R.S. Hoult, B. Halliwell, Inhibition of
mammalian 5-lipoxygenase and cyclo-oxygenase by flavonoids and phenolic dietary
additives: relationship to antioxidant activity and to iron ion-reducing ability,
Biochem. Pharmacol. 42 (1991) 1673–1681, https://doi.org/10.1016/0006-
2952(91)90501-U.
[10] American Society of Agricultural Engineers Standards, ANSI/ASAE S319.3: Method
of Determining and Expressing Fineness Of Feed Materials By Sieving, St. Joseph,
MI, 2010.
[11] E. Uquiche, E. Huerta, A. Sandoval, J.M. del Valle, Effect of boldo (Peumus boldus
M.) pretreatment on kinetics of supercritical CO2 extraction of essential oil, J. Food
Eng. 109 (2012) 230–237, https://doi.org/10.1016/j.jfoodeng.2011.10.013.
[12] V.L. Singleton, R. Orthofer, R.M. Lamuela-Raventós, Analysis of total phenols and
other oxidation substrates and antioxidants by means of folin-ciocalteu reagent,
Methods Enzymol. 299 (1999) 152–178, https://doi.org/10.1016/S0076-6879(99)
99017-1.
[13] W. Brand-Williams, M.E. Cuvelier, C. Berset, Use of free radical method to evaluate
antioxidant activity, Lebensm, Wiss. Technol. 28 (1995) 25–30, https://doi.org/10.
1016/S0023-6438(95)80008-5.
[14] I.F. Benzie, J.J. Strain, The ferric reducing ability of plasma (FRAP) as a measure of
antioxidant power: the FRAP assay, Anal. Biochem. 239 (1996) 70–76, https://doi.
org/10.1006/abio.1996.0292.
[15] A.S. Ahmed, E.E. Elgorashi, N. Moodley, L.J. McGaw, V. Naidoo, J.N. Eloff, The
antimicrobial, antioxidative, anti-inflammatory activity and cytotoxicity of dif-
ferent fractions of four South African Bauhinia species used traditionally to treat
diarrhoea, J. Ethnopharmacol. 143 (2012) 826–839, https://doi.org/10.1016/j.jep.
2012.08.004.
[16] J.S. Kim, Investigation of phenolic, flavonoid, and vitamin contents in different
parts of korean ginseng (Panax ginseng C.A. Meyer), Prev. Nutr. Food Sci. 21 (2016)
263–270, https://doi.org/10.3746/pnf.2016.21.3.263.
[17] F. Rouessac, A. Rouessac, Chemical Analysis: Modern Instrumentation Methods and
Techniques, 2nd ed., John Wiley & Sons, Chichester, UK, 2007.
[18] N. Rahali, S. Mehdi, F. Younsi, M. Boussaid, C. Messaoud, Antioxidant, α-amylase,
and acetylcholinesterase inhibitory activities of Hertia cheirifolia essential oils: in-
fluence of plant organs and seasonal variation, Int. J. Food Prop. 20 (2017)
1637–1651, https://doi.org/10.1080/10942912.2017.1352597.
[19] V. Muccilli, N. Cardullo, C. Spatafora, V. Cunsolo, C. Tringali, α-Glucosidase in-
hibition and antioxidant activity of an oenological commercial tannin. Extraction,
fractionation and analysis by HPLC/ESI-MS/MS and 1H NMR, Food Chem. 215
(2017) 50–60, https://doi.org/10.1016/j.foodchem.2016.07.136.
[20] G. Brunner, Gas Extraction. An Introduction to Fundamentals of Supercritical Fluids
and the Applications to Separation Processes, Steinkopff Darmstadt Springer, New
York, 1994.
[21] L.M.A.S. de Campos, F.V. Leimann, R.C. Pedrosa, S.R.S. Ferreira, Free radical
scavenging of grape pomace extracts from Cabernet sauvingnon (Vitis vinifera),
Bioresour. Technol. 99 (2008) 8413–8420, https://doi.org/10.1016/j.biortech.
2008.02.058.
[22] H.I. Castro-Vargas, P. Benelli, S.R.S. Ferreira, F. Parada-Alfonso, Supercritical fluid
extracts from tamarillo (Solanum betaceum Sendtn) epicarp and its application as
protectors against lipid oxidation of cooked beef meat, J. Supercrit. Fluids 76
(2013) 17–23, https://doi.org/10.1016/j.supflu.2012.10.006.
[23] I.H. Aidil, H.I. Cetin, M.E. Yener, A. Bayındırlı, Subcritical (carbon dioxide
+ethanol) extraction of polyphenols from apple and peach pomaces, and de-
termination of the antioxidant activities of the extracts, J. Supercrit. Fluids 43
(2007) 55–63, https://doi.org/10.1016/j.supflu.2007.04.012.
[24] D. Biscaia, S.R.S. Ferreira, Propolis extracts obtained by low pressure methods and
supercritical fluid extraction, J. Supercrit. Fluids 51 (2009) 17–23, https://doi.org/
10.1016/j.supflu.2009.07.011.
[25] H. Kamali, E. Golmakani, A. Golshan, A. Mohammadi, T.A. Sani, Optimization of
ethanol modified supercritical carbon dioxide on the extract yield and antioxidant
activity from Biebersteinia multifida DC, J. Supercrit. Fluids 91 (2014) 46–52,
https://doi.org/10.1016/j.supflu.2014.04.004.
[26] J.M. del Valle, J.C. de la Fuente, D.A. Cardarelli, Contributions to supercritical
extraction of vegetable substrates in Latin America, J. Food Eng. 67 (2005) 35–57,
The Journal of Supercritical Fluids 147 (2019) 1–8
https://doi.org/10.1016/j.jfoodeng.2004.05.051.
[27] S. Kitanović, D. Milenović, V.B. Veljković, Empirical kinetic models for the resinoid
extraction from aerial parts of St. John’s wort (Hypericum perforatum L.), Biochem.
Eng. J. 41 (2008) 1–11, https://doi.org/10.1016/j.bej.2008.02.010.
[28] B.A.S. Machado, G. de Abreu Barreto, A.S. Costa, S.S. Costa, R.P.D. Silva, D.F. da
Silva, H.N. Brandão, J.L.C. da Rocha, S.B. Nunes, M.A. Umsza-Guez, F.F. Padilha,
Determination of parameters for the supercritical extraction of antioxidant com-
pounds from green propolis using carbon dioxide and ethanol as co-solvent, PLoS
One 10 (8) (2015) e0134489, , https://doi.org/10.1371/journal.pone.0134489 doi:
10.1371/journal.pone.0134489.
[29] Y.H. Lee, A.L. Charles, H.F. Kung, C.T. Ho, T.C. Huang, Extraction of nobiletin and
tangeretin from Citrus depressa Hayata by supercritical carbon dioxide with ethanol
as modifier, Ind. Crops Prod. 31 (2010) 59–64, https://doi.org/10.1016/j.indcrop.
2009.09.003.
[30] F. Ghiringhelli, C. Rebe, A. Hichami, D. Delmas, Immunomodulation and anti-in-
flammatory roles of polyphenols as anticancer agents, Anticancer Agents Med.
Chem. 12 (2012) 852–873 https://doi.org/ 10.2174/187152012802650048.
[31] F. Shahidi, P.K. Janitha, P.D. Wanasundara, Phenolic antioxidants, Crit. Rev. Food
Sci. Nutr. 32 (1992) 67–103, https://doi.org/10.1080/10408399209527581.
[32] S.C. Liu, J.T. Lin, C.K. Wang, H.Y. Chen, D.J. Yang, Antioxidant properties of var-
ious solvent extracts from lychee (Litchi chinenesis Sonn.) flowers, Food Chem. 114
(2009) 577–581, https://doi.org/10.1016/j.foodchem.2008.09.088.
[33] L. Zhang, A.S. Ravipati, S.R. Koyyalamudi, S.C. Jeong, N. Reddy, P.T. Smith,
J. Bartlett, K. Shanmugam, D.G. Münch, M.J. Wu, Antioxidant and anti-in-
flammatory activities of selected medicinal plants containing phenolic and flavo-
noid compounds, J. Agric. Food Chem. 59 (2011) 12361–12367 https://doi:
10.1021/jf203146e.
[34] R.S. Talhouk, C. Karam, S. Fostok, W. El-Jouni, E.K. Barbour, Anti-inflammatory
bioactivities in plant extracts, J. Med. Food 10 (2007) 1–10 https://doi: 10.1089/
jmf.2005.055.
[35] O. Radmark, O. Werz, D. Steinhilber, B. Samuelsson, 5-Lipoxygenase: regulation of
expression and enzyme activity, Trends Biotechnol. 32 (2007) 332–341 https://doi:
10.1016/j.tibs.2007.06.002.
[36] Q.Q. Zhao, S.F. Wang, Y. Li, Q.Y. Song, K. Gao, Terpenoids with anti-inflammatory
activity from Abies chensiensis, Fitoterapia 111 (2016) 87–94 https://doi: 10.1016/
j.fitote.2016.04.002.
[37] M. Ashraf-Khorassani, M.T. Combs, L.T. Taylor, Supercritical fluid extraction of
metal ions and metal chelates from different environments, J. Chromatogr. A 774
(1997) 37–49, https://doi.org/10.1016/S0021-9673(97)00358-0.
[38] A.S. Ravipati, L. Zhang, S.R. Koyyalamudi, S.C. Jeong, N. Reddy, J. Bartlett,
P.T. Smith, K. Shanmugam, G. Münch, M.J. Wu, M. Satyanarayanan, B. Vysetti,
Antioxidant and anti-inflammatory activities of selected Chinese medicinal plants
and their relation with antioxidant content, BMC Complement. Altern. Med. 12
(2012) 173–186, https://doi.org/10.1186/1472-6882-12-173.
[39] M.S. Alam, J. Choi, D. Lee, Synthesis of novel Schiff base analogues of 4-amino-1, 5-
dimethyl-2- phenylpyrazol-3-one and their evaluation for antioxidant and anti-in-
flammatory activity, Bioorg. Med. Chem. 20 (2012) 4103–4108, https://doi.org/10.
1016/j.bmc.2012.04.058.
[40] M. Bimakr, R.A. Rahman, A. Ganjloo, F.S. Taip, L.M. Salleh, M.Z.I. Sarker,
Optimization of supercritical carbon dioxide extraction of bioactive flavonoid
compounds from spearmint (Mentha spicata L.) leaves by using response surface
methodology, Food Bioprocess Technol. 5 (2012) 912–920, https://doi.org/10.
1007/s11947-010-0504-4.
[41] A.W. Boots, G.R.M.M. Haenen, A. Bast, Health effects of quercetin: from antioxidant
to nutraceutical, Eur. J. Pharmacol. 585 (2008) 325–337, https://doi.org/10.1016/
j.ejphar.2008.03.008.
[42] A.T. Jan, M.R. Kamli, I. Murtaza, J.B. Singh, A. Ali, Q.M.R. Haq, Dietary flavonoid
quercetin and associated health benefits—an overview, Food Res. Int. 26 (2010)
302–317, https://doi.org/10.1080/87559129.2010.484285.
[43] W. Yu, Y.C. Fu, W. Wang, Cellular and molecular effects of resveratrol in health and
disease, J. Cell. Biochem. 113 (2012) 752–759, https://doi.org/10.1002/jcb.23431.
[44] S. Sain, P.K. Naoghare, S.S. Devi, A. Daiwile, K. Krishnamurthi, P. Arrigo,
T. Chakrabarti, Beta caryophyllene and caryophyllene oxide, isolated from Aegle
marmelos, as the potent anti-inflammatory agents against lymphoma and neuro-
blastoma cells, Antiinflamm. Antiallergy Agents Med. Chem. 13 (2014) 45–55,
https://doi.org/10.2174/18715230113129990016.
[45] Y.Q. Li, F.C. Zhou, F. Gao, J.S. Bian, F. Shan, Comparative evaluation of quercetin,
isoquercetin and rutin as inhibitors of α-glucosidase, J. Agric. Food Chem. 57
(2009) 11463–11468, https://doi.org/10.1021/jf903083h.
[46] X.S. Shu, J.H. Lv, J. Tao, G.M. Li, H.D. Li, N. Ma, Antihyperglycemic effects of total
flavonoids from Polygonatum odoratum in STZ and alloxan-induced diabetic rats, J.
Ethnopharmacol. 124 (2009) 539–543, https://doi.org/10.1016/j.jep.2009.05.006.
[47] A.O. Ademiluyi, S.I. Oyeleye, G. Oboh, Biological activities, antioxidant properties
and phytoconstituents of essential oil from sweet basil leaves, Comp. Clin. Pathol.
25 (2016), https://doi.org/10.1007/s00580-015-2163-3 169 176.