Research Article
Received: 9 June 2010;
Revised: 1 October 2010;
Accepted: 4 October 2010
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/pca.1305
Fast Ultrasound‐assisted Extraction of Polar
(phenols) and Nonpolar (lipids) Fractions in
Heterotheca inuloides Cass.
O. F. Mijangos Ricárdez,† J. Ruiz‐Jiménez, L. Lagunez‐Rivera‡ and
M. D. Luque de Castro*
ABSTRACT:
Introduction – Heterotheca inuloides Cass., also known as “arnica”, is used in traditional medicine in Mexico.
Objective – Development of fast methods for the extraction of lipidic and phenolic fractions from arnica plants and their
subsequent characterization.
Methodology – Ultrasound was applied to accelerate extraction of the target compounds from this plant and reduce the use of
organic solvents as compared with conventional methods. Gas chromatography–ion trap mass spectrometry and liquid
chromatography with diode‐array detection were used for the characterization of the lipidic and phenolic fractions, respectively.
Results – Under optimal extraction conditions, 9 and 55 min were necessary to complete extraction of the lipidic and
phenolic fractions, respectively. The fatty acids present at the highest concentrations in H. inuloides were eicosatetraenoic n3
(24.6 μg/g), cis‐9‐hexadecenoic n7 (23.1 μg/g), exacosanoic (22.7 μg/g) and cis‐9‐octadecenoic acid (21.3 μg/g), while the rest
were in the range 7.6–1.3 μg/g. The most concentrated phenols were guaiacol (41.5 μg/g), catechin (38.7 μg/g), ellagic acid
(35.9 μg/g), carbolic acid (24.2 μg/g) and p‐coumaric acid (19.5 μg/g), while the rest were in the range 5.1–0.4 μg/g.
Conclusion – Ultrasound reduces the time necessary to complete the extraction 160 and 26 times, the extraction volume 2.5
and 4 times, and increases the extraction efficiency 5 and 3 times for lipidic and phenolic fractions, respectively, in
comparison with conventional extraction methods. In addition, the characterization of the lipidic and phenolic fractions
constitutes a first approach to the H. inuloides metabolome. Copyright © 2011 John Wiley & Sons, Ltd.
Keywords: Heterotheca inuloides Cass.; ultrasound‐assisted extraction; fatty acids; phenols; GC‐MS; HPLC–DAD
Introduction
Plants commonly known in Mexico as “arnica” include, among
others, the following species: Mentezlia conzatti, Zexmenia
pringlei, Haplopappus spp. and Heterotheca inuloides (Gené
et al., 1998; Maldonado‐López et al., 2008), the last being the
most abundant in Mexico, where it also has the names acahual,
cuauteteco and xochihuepal. It is used in traditional medicine
for treatment of contusions, pain and injuries of the skin, similar
to the use in Europe of mountain arnica (Arnica montana L.), its
European counterpart (Gené et al., 1998).
Heterotheca inuloides is a pilar plant with yellow flowers, a
graduated involucre and narrow bracts, long petioles in the
basal leaves, and average height 50–70 cm, reaching 1 m under
optimum growing conditions (Delgado et al., 2001). Its turgid
stem, generally not very graft, underneaths the inflorescence. It
is a perennial plant, and its habitat is ruderal and weedy,
disturbed pastures, and cleared forests (Sagrero‐Nieves, 1996).
Previous research on H. inuloides flowers has reported the
presence of polar compounds, mainly flavonoids, sesquiterpenoids
(Kubo et al., 1995), triterpenoids and sterols (Haraguchi et al., 1997;
Kubo et al., 2000). None of the previous studies has involved both
polar and nonpolar fractions or that of the lipidic fraction alone.
The growing trends towards the development of green
chemistry and automation have also reached extraction steps,
thus originating a high demand for shorter extraction methods,
Phytochem. Anal. 2011
with reduced consumption of organic solvents and less human
intervention. Auxiliary energy, such as microwaves, ultrasound
or the use of superheated liquids (Wang and Weller, 2006;
Priego‐Capote and Luque de Castro, 2007), in continuous and
discontinuous approaches, have been used with the aim of
meeting this demand.
Ultrasound has demonstrated its usefulness for improving
extraction efficiency and shortening the time required for this
step. The effects of ultrasound on leaching are primarily related
to cavitation. Thus, the implosion of bubbles formed during
ultrasound application produces rapid adiabatic compression of
gases and vapours within the bubbles or cavities and a highly
efficient temperature and pressure as a result (Luque de Castro
* Correspondence to: M. D. Luque de Castro, Department of Analytical
Chemistry, Annex Marie Curie Building, Campus of Rabanales, University of
Córdoba, E‐14071, Córdoba, Spain. E-mail: [email protected]
†
‡
Permanent address: Instituto Politécnico Nacional CIIDIR Oaxaca, Hornos
No. 1003, Col. Noche Buena, Santa Cruz Xoxocotlán, Oaxaca, Mexico.
Author of collection, classification and sending of the target samples.
Instituto Politécnico Nacional CIIDIR Oaxaca, Hornos No. 1003. Noche
Buena, Santa Cruz, Xoxocotlán, Oaxaca, Mexico.
Department of Analytical Chemistry, Annex Marie Curie Building, Campus of
Rabanales, University of Córdoba, E‐14071, Córdoba, Spain
Copyright © 2011 John Wiley & Sons, Ltd.
O. F. Mijangos Ricárdez et al.
and Priego‐Capote, 2006). The increased temperature enhances
the solubility of the analytes in the leachant and facilitates their
diffusion from the sample matrix to the outer region. On the
other hand, the increased pressure facilitates penetration of the
leachant into the sample matrix and transfer from the matrix to
the liquid phase at the interface. Various types of analytes,
including polar and nonpolar organic compounds, organometals,
and metallic elements, have been thus efficiently leached from
solid samples (Luque de Castro and Priego‐Capote, 2007a, b).
The aim of the present research was to develop methods for
the extraction of the two main fractions of compounds – polar
and nonpolar – in H. inuloides, based on the use of ultrasound to
accelerate removal and drastically reduce organic extractants
(green method), thus avoiding the use of conventional time‐
consuming and far‐from‐green methods, such as Soxhlet
extraction. There are no methods in the literature for these
compounds in H. inuloides as the majority of the research on this
plant has been focused on the volatile fraction of its flowers.
in 10 mL methanol. The standard solutions, which contained the 19
phenols, were prepared by dilution of the appropriate volume of each
stock solution in methanol. All the above solutions were stored at −20°C in
glass vials and kept in the dark until use.
Potassium methylate (0.5 M) in methanol (Panreac, Barcelona, Spain)
was used as a derivatization reagent in order to hydrolyse and transform
the fat into FAMEs. All safety precautions (gloves, mask, hood‐fume, etc.)
were adopted. Methyl esters of acids nonanoic (9:0), dodecanoic (12:0),
hexadecanoic (16:0), cis‐9‐hexadecenoic n7 (16:1 n7), heptadecanoic
(17:0), 10 heptadecenoic (17:1 n10), octadecanoic (18:0), trans‐9
octadecenoic (18:1 n9t), cis‐9 octadecenoic (18:1 n9), cis‐7‐octadecenoic
n7 (18:1 n7), trans, trans‐octadecadienoic (18:2 t9, t12), cis, trans‐
octadecadienoic (18:2 c, t), trans, cis‐octadecadienoic (18:2 t, c), cis, cis‐
octadecadienoic (18:2 c, c), trans, trans‐cis octadecadienoic (18:3 t, t, c ), cis,
trans, trans‐octadecadienoic (18:3 c, t, t ), cis, trans‐cis octadecadienoic
(18:3 c, t, c), trans‐cis, cis‐octadecadienoic (18:3 t, c, c), cis,cis,
cis‐octadecadienoic (18:3 c, c, c), eicosatrienoic (20:3 n6), eicosatetraenoic
n3 (20:4 n3), eicosapentaenoic n3 (20:5 n3) and exacosanoic (26:0) from
Sigma‐Aldrich were used as standards for calibration. Nonadecanoate
acid methyl ester (19:0) from Fluka (Steinheim, Germany) was used as
internal standard in the determination step.
Experimental
Samples
Apparatus
Ultrasonic irradiation was applied by means of a Branson 450 digital
sonifier (20 kHz, 450 W) equipped with a cylindrical titanium alloy probe
(12.7 mm diameter), which was immersed in a stainless steel container
with eight compartments for the test tubes. A centrifuge (Selecta,
Barcelona, Spain) was used to separate suspension particles. A rotary
evaporator (R‐200, Büchi, Switzerland) was used to release the hexane
after each extraction. An analytical balance (Explorer Analytical Balance,
Ohaus, USA) was used for gravimetric determination of the extracted oil.
A Varian CP 3900 gas chromatograph coupled to a Saturn 2100 ion trap
mass spectrometer (Sugar Land, TX, USA), furnished with a SP‐2380 fused
silica capillary column (60 m × 0.25 mm i.d., 0.2 μm film thickness)
provided by Supelco (Bellefonte, PA, USA) was used for the analysis of
fatty acid profiles (esterified fatty acids, EFAs, and nonesterified fatty acids,
NEFAs) after conversion into their respective fatty acid methyl esters
(FAMEs). The absorbance of the phenolic extract, after Folin–Ciocalteu
(F‐C) derivatization, was measured by a Thermo Spectronic, Helios gamma
(UK), UV–vis spectrometer. The liquid chromatograph was a ProStar
system (Varian, Palo Alto, CA, USA), consisting of a pump (ProStar 240,
solvent delivery mixture, a diode array detector (ProStar 330) and an
autosampler (ProStar 410) furnished with a column oven and a 0.6 mL
sample loop.
The operational variables were optimized using the software
Statgraphics plus version 5.1. for Windows (Stat Point Inc., Herndon,
VA, USA).
Reagents
Ethanol was from Panreac (Barcelona, Spain) and n‐hexane was provided
by Scharlab (Barcelona, Spain). High‐quality water (resistivity 18.2 mΩ
obtained from a Milli‐Q system from Millipore, Bedford, MA, USA) was
used to prepare both water–ethanol extractant mixtures and mobile
chromatographic phases. HPLC‐grade ethanol, methanol, acetonitrile,
sodium carbonate, F‐C reagent and ortophosphoric acid were also from
Panreac.
The phenolic compounds were purchased from Extrasynthese (Genay,
France) in the case of p‐hydroxybenzoic acid, guaiacol, protocatechuic
acid, pyrogallol, catechin, syringaldehyde, coniferaldehyde, acetovanillone,
acetosyringone and pyrocatechol; carbolic acid, vanillin, vanillic
acid, ellagic acid, sinapaldehyde, p‐coumaric acid, ferulic acid, gallic acid,
syringic acid and the internal standard 4‐methylphenol were from Sigma‐
Aldrich (St Louis, MO, USA). The stock standard solution of each phenolic
compound was prepared at 1000 μg/mL by dissolving 10 mg each phenol
View this article online at wileyonlinelibrary.com
Heterotheca inuloides was collected in the area called “Cerro de Monte
Albán” in Santa Cruz Xoxocotlán, Oaxaca, México (17°3′ 43″ N, 96°43′ 18″ W)
within the months July to September 2008 and transported to the food
laboratory of the Interdisciplinary Center of Investigation for the
Regional Integral Development (CIIDIR) Unit Oaxaca of the Instituto
Politécnico Nacional, México. Plants were identified and authenticated
by the taxonomist Dr. R. Solano, and a voucher specimen was deposited
in the herbarium of the CIIDIR. The plants were dehydrated by
preservation from the sun for 14 days at room temperature (25–26°C)
and transported to the University of Córdoba, Spain, where they were
milled and sized (60 μm), then kept at 4°C in dark until use.
Methods
Conventional extraction methods were used to compare the performance of those here proposed.
Conventional extraction of the lipidic fraction. One gram of sample
was placed in a cellulose thimble (25 × 88 mm, Albert, Spain). The overall
Soxhlet glassware was fitted to a tared distillation flask containing 80 mL
hexane and two‐to‐three boiling glass regulators. After extraction for
24 h, the hexane in the extract was evaporated to dryness in a Büchi
rotary‐evaporator at 35°C, with recovery of the solvent over 82%.
Conventional extraction of the phenolic fraction. Maceration by
continuous agitation was selected as the reference to compare the
efficiency of the proposed extraction method for phenols. For this
purpose, 1 g dried and milled plant and 100 mL ethanol were placed in a
flask and agitated vigorously for 24 h, after which the liquid phase was
stored until analysis.
Proposed ultrasound‐assisted extraction of the lipidic fraction.
The experimental set‐up used for static ultrasound‐assisted extraction is
shown in Fig. 1. The powdered plant (0.5 g portions) was placed in test
tubes to which portions of 10 mL hexane had been added. A total of
eight tubes were placed in the stainless steel container, which was
immersed in the water bath, where ultrasonic irradiation under the
optimal working conditions (duty cycle 50%, applied power 225 W,
probe position 1 cm from the bath bottom, water bath temperature 35°C,
and irradiation time 3 min) was applied. After extraction, the extract
(lipidic fraction) was centrifuged for 5 min at 3000 g to deposit potential
in‐suspension particles and the supernantant stored in an amber flask at
−20°C. The cycle was repeated twice more and the total extraction time
was 9 min. The hexane extract was evaporated to dryness at 35°C.
Copyright © 2011 John Wiley & Sons, Ltd.
Phytochem. Anal. 2011
Ultrasound‐assisted Extraction of Lipids and Phenols from H. inuloides
Figure 1. Device for the simultaneous leaching of eight samples with the assistance of an ultrasonic probe. (A) Sample tube; (B) upper view of the
device for tube positioning; (C) controller of the ultrasonic probe; (D) probe; (E) water bath.
Proposed ultrasound‐assisted extraction of the phenolic fraction.
The powdered plant (0.5 g portions) was placed in test tubes to which
portions of 5 mL 40:60 ethanol–water were added. Eight tubes were
placed in the stainless steel container, which was immersed in the water
bath, where ultrasonic irradiation under the optimal working conditions
(duty cycle 50%, output amplitude 50% of the converter, applied power
225 W, probe position 3 cm from the bath bottom, bath temperature
35°C, and irradiation time 11 min) was applied for each cycle. After
extraction, the extract (phenolic fraction) was centrifuged for 5 min at
3000 g in order to deposit potential in‐suspension particles and the
clean extract collected in a 25 mL volumetric flask. The cycle was
repeated four times (total extraction time 55 min) and the sum of the
extracts was made up to 25 mL in the volumetric flask using the
extractant mixture.
Overall determination of the lipidic fraction. The dry residue from
the hexane extract was kept overnight in a heater at 120°C to remove
water traces; then it was weighed for determination of the overall lipidic
fraction.
Derivatization method of the lipidic fraction. Conversion of fatty
acids into their more volatile FAMEs is a necessary step prior to gas
chromatrography individual separation, for which the lipidic extract was
diluted to 2 mL with hexane and homogeneized in a vial for 30 s; then,
0.5 mL potassium methylate in methanol was added and the mixture
shaken vigorously in the vial for 3 min and centrifuged at 3000 g for
5 min. The supernatant hexane fraction, which contained the FAMEs
from EFAs, was transferred to a test tube. The remaining methanolic
fraction was subjected to the following steps to obtain the FAMEs from
NEFAs: addition of anhydride sodium sulphate and homogenization for
30 s; subsequent addition of 0.5 mL 5% sulphuric acid in methanol,
heating in a water bath at 50°C for 30 min with vigorous shaken of the
mixture. After cooling, 1 mL n‐hexane was added, mixed for 2 min in a
vortex and, after phase separation, and the top n‐hexane phase
containing the derivatized NEFA was transferred to a test tube. This
operation was repeated with the remaining fraction. Thus, the NEFA oil
fraction was obtained. This fraction was dried under a stream of N2. The
residue was dissolved in 500 μL n‐hexane and shaken for 2 min. Finally,
dilutions were made for FAMEs from EFA (1:50) and from NEFA (1:10) in
n‐hexane containing 5 μg/mL internal standard.
Method for determination of fatty acid profile. The appropriate
separation of FAMEs was carried out by GC, then detected and quantified
by MS using the GC‐MS method developed by Sánchez‐Ávila et al. (2007),
modified in the injection step (injection volume 1.0 μL and isothermal
operation of injector in splitless mode at 250°C during the run).
The samples were analysed using the following oven temperature
program: initial temperature 70°C (held for 1.2 min), increased at
Phytochem. Anal. 2011
25°C/min to 120°C and followed by a second gradient of 2°C/min to
243°C and, finally, increased by 40°C/min to 270°C and held at this
temperature for 5 min.
The mass spectrometer was operated in the EI mode and the ion
preparation mode was μ_Selected Ion Storage (μ_SIS, similar to Selected Ion
Monitoring). The manifold, trap and transfer line temperatures were set at
60, 170 and 200°C, respectively. Helium at a constant flow‐rate of 1 mL/min
was used as carrier gas for the GC‐MS analysis of the FAME extracts.
Method for overall determination of phenols. The amount of total
phenolics was measured by a modified version of the F‐C method (Girón
et al., 2009) using gallic acid as standard. A calibration curve was run
using solutions of 100, 200, 300, 400, 500 and 600 mg/L of this
compound (y = 0.002x + 0.075, R2 = 0.9931). A 0.5 mL aliquot of dilute
extract (all extracts were diluted with distilled water to adjust the
absorbance within the calibration limits), 10 mL distilled water, 1 mL F‐C
reagent and 3 mL sodium carbonate (20% w/v) were mixed in this order,
made up to 25 mL with distilled water and heated at 50°C for 5 min.
After 30 min, the absorbance was measured at 765 nm against a blank
similarly prepared, but containing distilled water instead of extract. The
spectrophotometric measure was repeated three times for each extract
and the average datum interpolated within the gallic acid calibration
curve; the total concentration of phenols was expressed as micrograms
of gallic acid per gram of arnica as calculated by the F‐C method.
Method for determination of the phenolics profile. The applied
method was that proposed by Ferreiro‐Vera et al. (personal communication) for the individual determination of phenolic compounds in vine‐
shoots, but adapted to extracts from H. inuloides.
The analytical column used was a reversed‐phase Inertsil ODS‐2
(250 × 4.6 mm i.d., 5 μm), the injection volume was 20 μL and the binary
gradient involved solvents A (water, 0.2% phosphoric acid) and B (1:1
acetonitrile–methanol). The gradient was as follows: 0–20 min, 96–82%
A and 4–18% B, flow‐rate 1 mL/min; 20–40 min, 82–0% A and 18–100%
B, flow‐rate 1 mL/min; 40–84 min, 82–74% A and 18–26% B, flow‐rate
1 mL/min; 84–93 min, 74–50% A and 26–50% B, flow‐rate 1 mL/min. The
extracts were injected directly into the chromatograph and the eluated
monitored at 260, 280, 320 and 360 nm (elution time, 93 min).
Results and Discussion
The use of the experimental set‐up shown in Fig. 1 for
ultrasound‐assisted extraction allows expedited optimization
of the extraction methods (as eight tubes are simultaneously
irradiated), and apply homogeneous irradiation to all tubes in it.
Therefore, the device can be very useful in routine analysis.
Copyright © 2011 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com
O. F. Mijangos Ricárdez et al.
Optimization of the extraction method
Lipidic fraction. The number of variables to be optimized in
the extraction step of the lipidic fraction was seven, namely:
probe position, ultrasound radiation amplitude, percentage of
ultrasound exposure (duty cycle), irradiation time, water‐bath
temperature, volume of hexane as extractant and cycle number.
The response variable was expressed as overall weight of the
lipidic fraction determined by gravimetry.
A Plackett‐Burman design 27 × 3/32 type III resolution
allowing four degrees of fredom and involving 12 randomized
runs plus three centre points was built for a screening study of
the behaviour of the seven factors affecting the extraction
process. The upper and lower values given to each factor were
selected from the available data and experience gathered in the
preliminary experiments. The tested and optimum values
obtained for each variable are shown in Table 1.
The conclusion of this screening study was that the maximum
lipidic fraction can be extracted under the following optimal
conditions: probe position 1 cm from the water‐bath bottom,
50% ultrasound radiation amplitude, 50% ultrasound exposure
duty cycle, 3 min irradiation time, 35°C water‐bath temperature,
10 mL extractant volume and three cycles.
Finally, a kinetics study under the optimal working conditions
was performed to assess the number of cycles necessary for
Table 1. Optimization of the ultrasound‐assisted extraction
of the lipidic fraction of H. inuloides
Variable
Range
tested
Probe position (cm)a
1–3
Ultrasound radiation amplitude (%)
10–50
Percentage of ultrasound exposure 10–50
(duty cycle) (%)
Irradiation time (min)
3–7
Water‐bath temperature (°C)
15–35
Extractant volume (mL)
5–10
Cycle number
1–4
a
From the bath bottom.
Optimum
value
1
50
50
3
35
10
3
complete extraction. Figure 2 shows that three cycles are
sufficient for exhaustive extraction of the lipidic fraction in
arnica by the proposed ultrasound‐assisted method.
Ultrasonic application allows extraction of the total fat
content in a time much shorter than that required by the
Soxhlet extraction method. The time was shortened from 24 h
to 9 min and, in addition, the yield was increased about 4 times.
The volume of the organic solvent used in the proposed
method was also reduced (30 mL of hexane vs. 100 mL required
for the conventional method). Therefore, this is a very simple
method, the results of which are similar to (but cheaper than)
those provided by an SFE method (Roldán‐Gutiérrez, et al.,
2008), and it does not suffer from significant degradation, as
happens when microwave irradiation is used to accelerate
extraction (Ruiz‐Jiménez and Luque de Castro, 2004).
Phenolic fraction. The variables were the same as for
extraction of the lipidic fraction, including in this case the
extractant composition (ethanol–water mixtures). The response
variable was expresed as total concentration of phenols
expressed as microgram of gallic acid per gram of arnica as
calculated by the F‐C method.
A Plackett‐Burman design 27 × 3/32 type III resolution allowing
four degrees of fredom and involving 12 randomized runs plus
three centre points was built for the screening study of the
behaviour of the height factors affecting the extraction process.
The conclusions of this screening study were that the probe
position, water bath temperature, radiation amplitude, extractant
volume and duty cycle were not statistically influential factors
within the ranges under study. However, the results showed
higher extraction efficiencies with the lowest value tested for the
extractant volume (5 mL) and the highest values tested for probe
position (3 cm), ultrasound radiation amplitude 50%, duty cycle
50% and water bath temperature 35°C. Irradiation time, cycle
number and ethanol percentage in the extractant mixture were
influential factors within the range under study. The first and
second factors had a positive influence on the process and the
influence of the third was negative. The second screening
involving higher values for the irradiation time and the cycle
number and lower values for the percentage of ethanol in the
extraction mixture was carried out using the optimal conditions
for the rest of the variables. The tested and optimum values
obtained for each variable are shown in Table 2.
Figure 2. Study of the extraction kinetics of the lipidic fraction by the proposed ultrasound‐assisted method.
View this article online at wileyonlinelibrary.com
Copyright © 2011 John Wiley & Sons, Ltd.
Phytochem. Anal. 2011
Ultrasound‐assisted Extraction of Lipids and Phenols from H. inuloides
Table 2. Optimization of the ultrasound‐assisted extraction
of the phenolic fraction of H. inuloides
Variable
Range tested
Probe position (cm)a
Ultrasound irradiation
amplitude (%)
Percentage of ultrasound
exposure (duty cycle) (%)
Irradiation timeb
Water‐bath temperature (°C)
Ethanol in the extractant
mixture (%)
Extractant volume (mL)
Cycles number
a
Optimum
value
First
design
1–3
10–50
Second
design
3
50
3
50
10–50
50
50
3–7
15–35
50–80
7–11
35
20–50
11
35
40
5–10
1–3
5
3–6
5
5
From the bath bottom.
The results of the second experimental design (see Table 2)
show that only the number of cycles was an influential factor;
therefore its influence was studied by a univariate approach by
fixing the other variables at their optimum values (ethanol
percentage in the extractant 40% and irradiation time 11 min).
Between one and six extraction cycles were performed. The
results obtained showed that the extraction efficiency increased
with the number of cycles up to five cycles, and levelled off for
higher numbers (see Fig. 3).
Ultrasonic application allows extraction of the total phenols
content in a time much shorter than that required by the
maceration method. The time was shortened from 24 h to 55 min
and, in addition, the yield was increased about 3 times. The
volume of the organic solvent used in the proposed method was
also reduced (25 mL ethanol–water was required for complete
extraction, vs. 100 mL required in the conventional method).
Characterization of the extraction method
In order to determine the precision of the method, within‐
laboratory reproducibility and repeatability were evaluated in a
single experimental set‐up with duplicates. The experiments
were carried out under the optimal conditions for the phenolic
and lipidic fractions. Two measurements of the sample per day
were carried out in 7 days.
To determine the variance due to the between‐day effect,
equation (1) was used:
2
sbetween ¼ðMSbetween −MSwithin Þ=nj
(1)
where MS is the mean square (residual sum of squares rated by
the freedom degrees) and nj is the number of replicates per day.
The within‐laboratory reproducibility, sWR2 , was calculated by
equation (2):
s2WR ¼ sr2 þ sbetween2
(2)
sr2
where is the residual mean squares within‐days and sbetween2
is the variance due to the between‐day effect. The results
obtained are listed in Table 3. The repeatability, expressed as
relative standard deviation (RSD), was 5.9%, while the within
laboratory reproducibility, also expressed as RSD, was 10.4% for
the lipidic fraction. For the phenolic fraction, the repeatability
(as RSD) was 3.6% and laboratory reproducibility 5.9%.
Characterization of the lipidic and phenolic fractions
Lipidic fraction. The fatty acids in the lipidic fraction were
analysed by GC‐MS. The experimental GC‐MS variables were
optimized. The optimal working conditions were those described in the Experimental section. Appropriate separation was
achieved within 78 min. Methyl nonadecenoate acid (C19:1) was
used as internal standard due to its physical and chemical
behaviour being similar to that of the derivatized analytes and
its absence in the analysed samples.
Calibrations plots were run for all analytes using the peak area
as a function of the standard concentration of each compound.
The calibration curves are shown in Table 4. The limit of
detection (LOD) for each analyte was expressed as the mass of
analyte giving a signal 3σ above the mean blank signal (where σ
is the standard deviation of the blank signal). The LODs
obtained ranged between 2 and 95 ng/g. The limits of
quantitation, expressed as the mass of analyte giving a signal
10σ above the mean blank signal, ranged from 6.6 to 561.0 ng/g.
LODs and LOQs were estimated for the target analytes in both
Figure 3. Study of the extraction kinetics of the phenolic fraction by the proposed ultrasound‐assisted method.
Phytochem. Anal. 2011
Copyright © 2011 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com
O. F. Mijangos Ricárdez et al.
Table 3. Study of the precision of the proposed method
Day
Replicate 1
Replicate 2
Average concentration
Experiments for the determination of the within‐laboratory reproducibility and repeatability, lipidic fractiona
1
47.0
45.5
2
44.9
51.5
3
43.1
44.7
4
37.1
40.1
5
42.1
47.6
6
42.8
42.6
7
37.8
35.7
Repeatability 5.9
Reproducibility 10.4
Experiments for the determination of the within‐laboratory reproducibility and repeatability, phenolic fractionb
1
244.7
245.9
2
237.1
223.6
3
203.9
210.2
4
225.7
227.6
5
221.3
221.2
6
224.6
215.2
7
247.8
222.5
Repeatability 3.6
Reproducibility 5.9
a
b
46.2
48.2
43.9
38.6
44.8
42.7
36.7
245.3
230.3
207.1
226.6
221.2
219.9
235.1
Values obtained by gravimetry, expressed as mg fat/g sample.
Concentration obtained by the F‐C method, expressed as μg phenols/g sample.
Table 4. Calibration curve, regression coefficient, detection and quantification limits (LOD and LOQ), and concentration of each
analyte by GC–MS for the lipidic fraction of H. inuloides
Compound
Calibration curve
r2
LODa
LOQa
Concentrationa
C9
C12
C16:0
C16:1 n7
C17
C17:1 n10
C18
C18:1n9 t
C18:1n9
C18:1n7
C18:2 t9,t12
C18:2 c, t
C18:2 t, c
C18:2 c, c
C18:3 t, t, c
C18:3 c, t, t
C18:3 c, t, c
C18:3 t, c, c
C18:3 c, c, c
C20:3 n6
C20:4n3
C20:5 n3
C26:0
y = 0.068x + 0.003
y = 0.025x + 0.003
y = 0.049x + 0.005
y = 0.023x − 0.001
y = 0.087x − 0.001
y = 0.029x − 0.009
y = 0.087x − 0.007
y = 0.022x + 0.005
y = 0.028x + 0.001
y = 0.036x − 0.006
y = 0.015x − 0.001
y = 0.060x − 0.008
y = 0.055x − 0.003
y = 0.088x − 0.007
y = 0.012x − 0.002
y = 0.012x − 0.007
y = 0.014x − 0.001
y = 0.027x − 0.004
y = 0.011x − 0.004
y = 0.083x − 0.006
y = 0.042x + 0.002
y = 0.014x − 0.002
y = 0.024x − 0.001
0.995
0.997
0.991
0.994
0.998
0.993
0.997
0.999
0.995
0.995
0.998
0.997
0.999
0.997
0.998
0.986
0.993
0.992
0.965
0.999
0.965
0.989
0.998
200.0
10.0
170.0
8.0
2.0
5.0
10.0
4.0
4.0
4.0
90.0
95.0
70.0
10.0
80.0
60.0
6.0
4.0
6.0
30.0
5.0
6.0
2.0
660.0
33.0
561.0
26.4
6.6
16.5
33.0
13.2
13.2
13.2
297.0
313.5
231.0
33.0
264.0
198.0
19.8
13.2
19.8
99.0
16.5
19.8
6.6
3.3 ± 0.4
5.3 ± 0.8
2.4 ± 0.3
23.1 ± 2.5
4.7 ± 0.4
6.6 ± 0.6
5.7 ± 0.5
7.7 ± 0.6
21.3 ± 2.4
5.7 ± 0.6
3.4 ± 0.9
1.3 ± 0.2
1.8 ± 0.2
4.5 ± 0.4
1.6 ± 0.2
5.8 ± 0.8
2.4 ± 0.2
4.1 ± 0.3
3.6 ± 0.3
5.6 ± 0.9
24.6 ± 2.6
5.7 ± 0.9
22.6 ± 2.3
a
As ng/g.
extracts and standard solutions and are shown in Table 4. The
highest concentrations in H. inuloides were acids eicosatetraenoic
n3 (24.6 μg/g), cis‐9‐hexadecenoic n7 (23.1 μg/g), exacosanoic
View this article online at wileyonlinelibrary.com
(22.6 μg/g) and cis‐9‐octadecenoic (21.3 μg/g); the rest were in
the range 7.6 μg/g (trans‐9 octadecenoic acid) to 1.3 μg/g
(cis, trans‐octadecadienoic acid). It is worth emphasizing the
Copyright © 2011 John Wiley & Sons, Ltd.
Phytochem. Anal. 2011
Ultrasound‐assisted Extraction of Lipids and Phenols from H. inuloides
percentage of the relevant families of fatty acids in the overall
lipidic fraction, namely: monosaturated fatty acids 37.2% and
insaturated fatty acids 71.8% (among the latter, trans fatty acids
16.3%, omega3 fatty acids 17.5%, and omega‐6 fatty acids
27.6%). The presence of fatty acids in H. inuloides has so far not
been studied, so the study of these compounds constitutes
one of the first steps to determine the metabolome of this
plant.
Phenolic fraction. Owing to the complexity of the matrix of
H. inuloides, quantitation of the target analytes was based on
the standard‐addition method (see Table 5). The results show a
high content of phenolic compounds, specifically, 190.1 μg/g.
This value was lower than that obtained by the F‐C
spectrophotometric method, which provided a concentration
of 226.4 μg/g (absorbance = 2.0 × 10−3x + 75.0 10−3, x being the
total concentration of phenols expressed as μg/g, with
R2 = 0.9931). In addition to the fact that the molecular weight
of the phenols actually in the extract did not coincide with that
of the gallic acid used as standard, interferences were detected
in the F‐C method caused by sulphur dioxide and nitrogen‐
containining compounds such as amino acids and proteins,
causing an error by excess.
Table 5 shows that the most concentrated phenols in this
fraction were guaiacol (41.5 μg/g), catechin (38.7 μg/g), ellagic
acid (35.9 μg/g), carbolic acid (24.2 μg/g) and coumaric acid
(19.5 μg/g), while the rest were in the range 5.1 μg/g
(pyrocatechol) to 0.5 μg/g (vanillic acid). This fraction contained
significant amounts of antioxidants, the action of which in
medicine has been widely demonstrated. Thus, ellagic acid has
antiproliferative and antioxidant properties (Roki et al., 2001),
catechin is a flavonoid which acts as an antioxidant in plant
metabolism (Hurst, 2008), syringaldehyde is an aromatic
aldehyde in plants (Hurst, 2008), phenol has antiseptic
properties (Dewick, 2002), guaiacol is an expectorant, antiseptic
and anaesthetic (Dewick, 2002) and coumaric acid has
antioxidant and anti‐carcinogenic properties (Morimoto et al.,
2008). All these compounds are present in the extract obtained
by the proposed method; the concentrations of them and
their individual quantitation had not been calculated in
previous studies on H. inuloides, mainly devoted to studying
the volatile fraction (Sagrero‐Nieves, 1996; Gené et al., 1998;
Maldonado‐López et al., 2008).
Comparison of the efficiency of the proposed methods with
that of conventional methods. Using the conventional
methods based on Soxhlet and maceration extraction, both
applied for 24 h, their efficiency was calculated taking as 100%
efficiency that achieved by the proposed methods. In this way
22 and 36% were provided by these methods for the lipidic and
phenolic fractions, respectively. In addition to the highest yield
of the extraction process, the reported methods did not
produce degradation of the target compounds as demonstrated
by the plateaux obtained in Figs. 2 and 3 by increasing the
number of extraction cycles and thus the subjection of the
extracted compounds to longer ultrasonic irradiation times.
The information provided by the present research constitutes
a first approach to the “arnica metabolome” for an in‐depth
knowledge of this plant in order to justify and enlarge its use in
traditional medicine in Mexico and other countries with high
production of this plant.
Acknowledgements
The Spanish Ministerio de Ciencia e Innovación (MICINN) is
thanked for financial support (project no. CTQ2009‐07430). O.F.
M.R. expresses his gratitude to Consejo Nacional de Ciencia y
Tecnología (CONACyT), Mexico, for a scholarship.
Table 5. Calibration curve, regression coefficient, detection and quantification limits (LOD and LOQ), and concentration of each
analyte by HPLC‐DAD for the phenolic fraction of H. inuloides
Compounds
Calibration curve
R2
Catechin
Vanillin
Syringaldehyde
Coniferaldehyde
Sinapaldehyde
Acetovanillone
Acetosyringone
Pyrogallol
Pyrocatechol
Carbolic acid
Guaicol
Gallic acid
Protocatechuic acid
p‐Hydroxybenzoic acid
Syringic acid
p‐Coumaric acid
Ferulic acid
Ellagic acid
Vanillic acid
y = 234.5x + 31.2
y = 255.1x + 35.7
y = 87.34x + 23.3
y = 345.8x + 43.9
y = 556.7x + 87.5
y = 44.6x + 28.9
y = 467.3x + 79.7
y = 212.5x + 12.9
y = 346.9x + 65.4
y = 457.7x + 78.9
y = 292.4x + 55.4
y = 77.9x + 33.5
y = 377.6x + 93.4
y = 586.3x + 88.6
y = 422.7x + 75.4
y = 341.3x + 58.6
y = 76.6x + 32.1
y = 51.2x + 26.2
y = 88.7x + 44.3
0.997
0.998
0.999
0.994
0.987
0.998
0.999
0.995
0.997
0.990
0.990
0.990
0.986
0.990
0.996
0.997
0.990
0.990
0.990
a
LODa
LOQa
6.5
3.2
5.6
4.3
3.3
6.4
2.6
4.2
6,6
4.4
4.4
3.4
3.9
2.7
2.1
3.3
3.3
5.6
3.5
21.5
10.8
18.7
14.3
10.8
21.3
8.7
13.9
21.6
19.4
22.6
11.2
13.1
8,9
6.8
17.7
11.1
16.9
11.5
Wavelength
Concentrationa
280
320
320
320
360
320
360
360
360
280
320
320
320
360
360
280
280
320
320
38.7 ± 2.3
1.2 ± 0.2
1.3 ± 0.3
0.5 ± 0.1
0.8 ± 0.1
3.4 ± 0.9
2.7 ± 0.5
1.1 ± 0.3
5.1 ± 0.8
24.2 ± 0.9
41.5 ± 2.6
4.9 ± 1.9
3.1 ± 1.2
0.7 ± 0.1
2.8 ± 0.3
19.5 ± 0.9
2.7 ± 0.6
35.9 ± 0.9
0.47 ± 0.1
As μg/g.
Phytochem. Anal. 2011
Copyright © 2011 John Wiley & Sons, Ltd.
View this article online at wileyonlinelibrary.com
O. F. Mijangos Ricárdez et al.
References
Delgado G, Olivares MS, Chávez MI, Ramírez‐Apan T, Linares E, Bye R,
Espinoza‐García FJ. 2001. Antiinflammatory constituents from
Heterotheca inuloides. J Nat Prod 64: 861–864.
Dewick PM. 2002. Medicinal Natural Products: a Biosynthetic Approach
(2nd edn). John Wiley and Sons: New York; 316.
Gené RM, Segura L, Adzet T, Marin E, Iglesias J. 1998. Heterotheca inuloides:
anti‐inflammatory and analgesic effect. J Ethnopharmacol 60: 157–162.
Girón MV, Ruiz‐Jiménez J, Luque de Castro MD. 2009. Dependence of
fatty‐acid composition of edible oil on their enrichment in olive
phenols. J Agric Food Chem 57: 2797–2802.
Haraguchi H, Ishikawa H, Sanchez Y, Ogura T, Kubo Y, Kubo I. 1997.
Antioxidative constituents in Heterotheca inuloides. Bioorg Med Chem
5: 865–871.
Hurst WJ. 2008. Methods of Analysis for Functional Foods and
Nutraceuticals (2nd edn). CRC Press: Boca Raton, FL; 280.
Kubo I, Ishiguro K, Chaudhuri SK, Kubo Y, Sanchez Y, Ogura T. 1995. A
plant growth inhibitory sesquiterpenoid from Heterotheca inuloides.
Phytochem Anal 38: 553–554.
Kubo I, Kinst‐Hori I, Chaudhuri SK, Kubo Y, Sánchez Y, Ogura T. 2000.
Flavonols from Heterotheca inuloides: tyrosinase inhibitory activity
and structural criteria. Bioorg Med Chem 8: 1749–1755.
Luque de Castro MD, Priego‐Capote F. 2006. Analytical Applications of
Ultrasound. Elsevier: Amsterdam; 393.
Luque de Castro MD, Priego‐Capote F. 2007a. Ultrasound assistance to
liquid–liquid extraction: a debatable analytical tool. Anal Chim Acta
583: 2–9.
Luque de Castro MD, Priego‐Capote F. 2007b. Ultrasound‐assisted
preparation of liquid samples. Talanta 72: 321–334.
View this article online at wileyonlinelibrary.com
Maldonado‐López Y, Linares‐Mazari E, Bye R, Delgado G, Espinoza‐García FJ.
2008. Mexican arnica anti‐inflammatory action: plant age is correlated
with the concentration of anti‐inflammatory sesquiterpenes in the
medicinal plant Heterotheca inuloides Cass. (Asteraceae). Econ Bot 62:
161–170.
Morimoto M, Cantrell CL, Libous‐Bailey L., Duke, SO. 2008. Phytotoxicity
of constituents of glandular trichomes and the leaf surface of
camphorweed, Heterotheca subaxillaris. Phytochem Anal 70:
69–74.
Priego‐Capote F, Luque de Castro MD. 2007. Ultrasound in analytical
chemistry. Anal Bioanal Chem 387: 249–257.
Roldán‐Gutiérrez JM, Ruiz‐Jiménez J, Luque de Castro MD. 2008.
Ultrasound‐assisted dynamic extraction of valuable compounds
from aromatic plants and flowers as compared with steam distillation
and superheated liquid extraction, Talanta 75: 1369–1375.
Roki D, Menkovic N, Savikin‐Fodulovic K, Krivokuca‐Dokic D, Ristic M,
Grubisic D. 2001. Flavonoids and essential oil in flower heads of
introduced Arnica chamissonis. J Herbs Spices Med Plants 8:
19–27.
Ruiz‐Jiménez J, Luque de Castro MD. 2004. Forward‐and‐back dynamic
ultrasound‐assisted extraction of fat from bakery products. Anal Chim
Acta 502: 75–82.
Sagrero‐Nieves L. 1996. Volatile components from the leaves of
Heterotheca inuloides Cass. Flav Fragr J 11: 49–51.
Sánchez‐Ávila N, Priego‐Capote F, Luque de Castro MD. 2007. Ultrasound‐
assisted extraction and silylation prior to gas chromatography–mass
spectrometry for the characterization of the triterpenic fraction in
olive leaves. J Chromatogr A 1165: 158–165.
Wang L, Weller CL. 2006. Recent advances in extraction of nutraceutical
from plants. Trends Food Sci Technol 17: 300–312.
Copyright © 2011 John Wiley & Sons, Ltd.
Phytochem. Anal. 2011