| Sažetak (engleski) | Introduction
Over time, due to the action of internal and external adverse factors, a number of
structural changes occur in the skin. Those changes include changes in pigmentation, elasticity
or hydration, which all manifest themselves as skin aging. To combat those changes, numerous
dermatopharmaceutical products are being developed with the aim of displaying positive effects
on the skin’s health and appearance. The bioactive components of such products are often
secondary metabolites of medicinal plants, such as polyphenols and alkaloids. They are known
for their beneficial effect on the skin including antioxidant, anti-inflammatory, and antihyperpigmentation
effects.
Besides their skin-related beneficial effects, contemporary cosmetic and dermatological
products are often prepared so that they meet the ethical and environmental demands for a
sustainable production. One of the pivotal aspects of producing eco-friendly plant extracts is
selection of eco-friendly extraction techniques and solvents, and ultrasound-assisted extraction
is one of the most used green extraction techniques. This technique achieves high yields with
lower energy and solvent use by applying sound waves to break cell walls, expediting
compound release. Furthermore, green solvents, characterized by their biodegradability, nontoxicity
and low flammability are an integral part of green extraction methods. Glycerol, a nontoxic
and biodegradable solvent of natural origin, is characterized by all the characteristics that
a green solvent should display. It also has an additional advantage which makes it highly
suitable for cosmetic applications due to its influence on viscosity and stability of the cosmetic
product and the moisture-retaining properties on the skin. Therefore, glycerol offers a valuable
eco-friendly option for plant-based extract production.
Numerous scientific studies show that herbal drugs Berberidis radicis cortex,
Echinaceae purpureae herba, Liquiritiae radix, and Silybi mariani fructus have a significant
dermatopharmaceutical potential. Therefore, in this dissertation, the optimization of glycerolbased
ultrasound-assisted extraction of the phenolic and other active components from these
drugs was performed using chemometric methods. The antioxidant activity of optimized
extracts was determined and their ability to inhibit the enzymes collagenase, elastase, tyrosinase
and lipoxygenase was examined. Additionally, their effect on hyaluronidase, biocompatibility
with HaCaT cells as well as their effect on wound healing in the "scratch" test was assessed.
Methods
The identity of the herbal drugs was confirmed by the procedures described in the
relevant pharmacopoeial monographs and other appropriate scientific literature. Prior to the
extraction, plant material was pulverized and sieved to obtain uniform particle size. In the
preliminary extractions, comparison of macerations using either glycerol/water or
ethanol/water mixtures was performed in order to select the optimal solvent. In addition to that,
the effectiveness of ultrasound assisted extraction with glycerol/water mixtures was compared
with the effectiveness of maceration performed using the same solvent. In order to obtain
extracts with desired chemical and biological characteristics ultrasound-assisted extraction was
optimized using response surface methodology. Box-Behnken design and two-level factorial
design were used for design of these experiments. Influences of several factors (independent
variables; glycerol concentration, extraction duration, ultrasound power, temperature, addition
of ascorbic acid to the extraction mixtures and the weight of herbal drug used for the extraction)
on the extraction outcomes (dependent variables) were assessed.
Concentrations of the selected bioactive ingredients in the extracts, berberine from
Berberidis radices cortex, cichoric and caftaric acid from Echinaceae purpureae herba, glabridin
and isoliquiritigenin from Liquiritiae radix, and silymarin (flavonolignans silibinin A and B,
isosilibinin A and B, silicristin and silidianin concentrations) from Silybi mariani fructus, were
determined by high-performance liquid chromatography. UV-VIS spectrophotometry was used
for determination of total phenolic compounds in Liquiritiae radix. Total phenolic acids in
Echinaceae purpureae herba was calculated as the sum of cichoric and caftaric acid
concentrations. Relative extraction efficiency of total phenols from Liquiritiae radix was
calculated as the ratio of total phenol content and the weight of the plant material used for the
extraction.
Antioxidant potential of the extracts was determined using several methods. Antiradical
activity of the extracts was tested spectrophotometrically with 2,2-diphenyl-1-picrylhydrazyl
radical. The inhibition of oxidative degradation of unsaturated fatty acids was tested by
monitoring the kinetics of b-carotene degradation in the presence of linoleic acid at elevated
temperature, whereas the ability to chelate Fe2+ ions was determined in the reaction with
ferrozine. Butylated hydroxyanisole (BHA) and EDTA were used as positive controls,
respectively.
Appropriate spectroscopic methods were used to determine the ability of plant extracts,
as well as glycerol in the concentrations used in the extracts, to inhibit specific enzymes that
can affect the health and appearance of the skin. Tyrosinase inhibition was determined by
monitoring formation of dopaquinone from 3,4-dihydroxy-L-phenylalanine. Kojic acid was
used as a reference inhibitor. Elastase activity was tested with the substrate N-succinyl-Ala-
Ala-Ala-p-nitroanilide and oleanolic acid as a positive control. Collagenase activity was
determined by using N- 3-2-furyl-acryloyl-Leu-Gly-Pro-Ala as the substrate and gallic acid as
a positive control. Anti-hyaluronidase activity was performed in the assay with pdimethylaminobenzaldehide,
while tannic acid was used as positive control. The antiinflammatory
activity of plant extracts was tested by measuring the concentration of conjugated
diene resulting from the oxidation of linoleic acid by 5-lipoxygenase and in the reaction with
ovalbumin. Nordihydroguaiaretic acid and diclofenac were used as positive controls,
respectively. Evaluation of biocompatibility and wound healing effects were evaluated on
HaCaT cells, immortal keratinocyte cell line from human skin. Cell viability was assessed using
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT test), while wound
healing effects were determined using the "scratch" test method.
Measurements were performed in triplicate and expressed as arithmetic mean ± standard
deviation. The activities in the performed assays were calculated by regression analysis and
expressed as an IC50 values. Statistical differences were examined using ANOVA and
corresponding post-hoc tests (Tukey's and Dunnett's post-hoc tests for comparison among the
extracts and the control, respectively). Validity of the response surface methodology models
was assessed using ANOVA.
Results
Preliminary experiments were focused on the comparison of the two extraction solvent
systems (glycerol/water mixtures vs. ethanol/water mixtures) and the two extraction techniques
(maceration vs. ultrasound assisted extraction). They were performed using the extraction of
phenolic acids from Echinaceae purpureae herba and silymarin from Silybi mariani fructus as
the models for further development of extraction procedures. Maceration results showed that
glycerol/water mixtures were equally effective solvents for the extraction of phenolic acids
from Echinaceae purpureae herba as ethanol/water mixtures. However, the efficiency of
glycerol/water for extraction of silymarin from Silybi mariani fructus was inferior to
ethanol/water mixtures. On the other hand, ultrasound-assisted extraction of target compounds
was superior to maceration in both cases. It achieved significantly higher yields of target
phenolics despite much shorter extraction time. In addition, the efficiency of glycerol/water for
ultrasound-assisted extraction of silymarin from Silybi mariani fructus reached the efficiency
of maceration using ethanol/water mixtures. Taking this into account, glycerolic ultrasoundassisted
extraction was proposed as the method of choice in further investigations performed
within this work.
Preliminary ultrasound-assisted extraction of Echinaceae purpureae herba, performed
according to the two-level factorial design, was used for initial selection of independent
variables for further extraction optimization. It was found that the extraction outcome was most
influenced by glycerol concentration, extraction time, temperature and ultrasound power. The
addition of ascorbic acid had a negative, time-dependent, influence on the concentration of all
the Echinaceae purpureae herba phenolic acids analyzed in this work.
Based on the results obtained in the preliminary experiments, further optimization of
the glycerol extraction of bioactive compounds from selected herbal drugs was performed.
Glycerol content and the temperature were used as independent variables in all the performed
extraction optimizations, while the other variables were selected on case-to-case basis.
Additional independent variable for optimization of berberine yield and antiradical activity in
Berberidis radicis cortex extraction was the ultrasonication power, while plant material to
solvent mass ratio was the additional variable for the extraction of total phenolic compounds,
glabridin and isoliquiritigenin from Liquiritiae radix. Extraction of Echinaceae purpureae herba
was optimized for the yield of caftaric acid, chicoric acid, total phenolic acids, and for
antiradical activity. Ultrasonication power and time were used as additional independent
variables, while time was the additional independent variable for preparation of Silybi mariani
fructus extracts optimized for the yield of silymarin and antiradical activity.
Further experiments were conducted based on the protocols established by Box-
Behnken design. Upon the preparation of the extracts and the determination of the target
dependent variables, the independent variables that significantly affected the extraction
efficiency were determined. The most important independent variables, the extraction
temperature and glycerol concentration, significantly influenced the efficiency in all the
performed experiments. In all the extractions conducted within this research, the extraction
outcome was proportional to the negative square value of the glycerol concentration. Also, a
negative linear effect of glycerol concentration was observed during the extraction of
Echinaceae purpureae herba, Berberidis radicis cortex and Liquiritiae radix. On the other hand,
the extraction temperature had a positive linear influence on the outcome of all the phenolic
compounds extractions performed within this work. Even though the other independent
variables were not employed in all the extractions performed herein, in individual cases they
also influenced extraction efficiency. For example, ultrasonication power influenced the
extraction of berberine from Berberidis radicis cortex, and the extraction of phenolic acids from
Echinaceae purpureae herba, while mass ratio of plant material to solvent influenced the
extraction of all the dependent variables from Liquiritiae radix. Extraction time significantly
influenced the content of target phenolic compounds from Echinaceae purpureae herba and
Silybi mariani fructus.
Based on the values of dependent variables in the prepared extracts (concentrations of
the target compounds and antiradical activity) response surface methodology was used to
propose the equations for extraction models and to calculate the values of independent variables
needed to achieve the extraction goals. Two extracts were prepared from Berberidis radicis
cortex. The extract optimized for berberine content contained 145.5 μg/mL of the alkaloid, and
it was prepared at 80 °C using 50% (w/w) glycerol and 144 W ultrasonication power. The extract
with maximum antiradical activity (IC50 = 58.88 μL extract/mL) was prepared at the same
temperature, but using lower glycerol content (30 %, w/w) and stronger ultrasonication (720
W). The conditions necessary for preparation of the Echinaceae purpureae herba extracts
optimized for caftaric acid, cichoric acid and total phenolic acid contents were essentially the
same. Therefore, only one extract was prepared using the calculated conditions: 70% glycerol
(w/w), temperature of 60 °C and 360 W of ultrasonication power. The extraction lasted for 60
min. Contents of caftaric acid, cichoric acid and total phenolic acid the extracts were 31.82
μg/mL, 113.11 μg/mL and 144.93 μg/mL, respectively. The Echinaceae purpureae herba extract
displaying the strongest antiradical activity was prepared at 60 °C and using 177 W
ultrasonication power. The extraction was performed with 50%, glycerol (w/w) and lasted for
55 min.
Liquiritiae radix extract optimized for total phenolics, obtained using 20% glycerol
(w/w), 70 °C and using 0.93 g of plant material, contained 854.6 μg/mL of the target compounds.
The extract with the highest relative extraction efficiency of total phenols contained 791.6
μg/mL of phenols per g of plant material. It was prepared using 30% glycerol (w/w), at 70 °C
and using 0.7 g of plant material. The conditions required for preparation of the extracts
optimized for glabridin and isoliquiritigenin contents coincided and only one extract was
prepared. The most appropriate extraction solvent for its preparation was 85% glycerol (w/w).
The extract was prepared at 70 °C using 1 g of plant material. Concentrations of glabridin and
isoliquiritigenin in thus prepared extract were 21.89 μg/mL and 6.23 μg/mL, respectively. The
Silybi mariani fructus extract with strongest antiradical activity (IC50 = 192.30 μL extract/mL)
was prepared at 60 °C, but using a rather low glycerol content (20%, w/w). The extraction was
performed for 60 min. The best conditions for the extraction of silymarin were 60 min at 80 °C
using 40% (w/w) glycerol. Silymarin content in the extract optimized for the highest
concentration of active substances, reached 99.6 μg/mL (expressed as silibinin).
The difference between the experimental results and the values calculated by the
theoretical models was less than 8% indicating the validity of the models. The prepared extracts
were used for further investigation of biological activities.
The optimized extracts displayed notable antioxidant activities in the performed
antioxidant assays. For example, Silybi mariani fructus extract optimized for antiradical activity
had a statistically equal effect to the employed positive control BHA, while the extract
optimized for silymarin content had a better effect than BHA in the test with b-carotene and
linoleic acid. Both Echinaceae purpureae herba extracts showed a statistically better effect than
BHA in the test with b-carotene. The Liquiritiae radix extract optimized for glabridin and
isoliquiritigenin content showed a better radical scavenging activity than BHA, while the
extract optimized for total phenolic content was as effective an iron chelator as EDTA.
The optimized extracts effectively inhibited enzymes and processes that negatively
affected the appearance and health of the skin, indicating the potential for positive effects on
inflammatory processes, pigmentation, skin firmness, hydration and elasticity. The activity of
several extracts in the performed assays was equal to or statistically higher than the activity of
standard inhibitors. By far the most effective among the tested optimized extracts were the
extracts prepared from Liquiritiae radix. The extract optimized for glabridin and
isoliqiuritigenin content was a stronger inhibitor of protein coagulation, as well as elastase and
tyrosinase activity than the employed positive controls diclofenac, ursolic and kojic acid.
Compared to the standard inhibitors, the extract optimized for total phenolic content was an
equally effective elastase inhibitor and a better tyrosinase inhibitor, while the extract optimized
for extraction efficiency of total phenols was an equally effective elastase and tyrosinase
inhibitor. Echinaceae purpureae herba extracts also displayed notable efficacy in the performed
assays. For example, the extract optimized for the highest content of phenolic acids and the
extract optimized for the most pronounced antiradical activity were more effective than tannic
acid in the hyaluronidase inhibition test, while the extract optimized for total phenolic acid
content was a statistically equally effective tyrosinase inhibitor as kojic acid. Glycerol, at the
concentrations in which it was present in the extracts, inhibited heat-induced protein
coagulation, actively contributing to the effect of the extracts. This ability of glycerol further
confirms that the advantages of using glycerol extracts in cosmetics go beyond its use as a green
solvent.
Biocompatibility and wound healing activity was investigated using Echinaceae
purpureae herba glycerol extracts. When tested for biocompatibility, neither the extracts nor
glycerol, did significantly adversely affect the viability of HaCaT cells in concentration of up
to 25 μL/mL. When used for in vitro investigation of wound healing using the in vitro "scratch"
model wound, the extracts used in these concentrations significantly increased wound healing
rate. After 48 h, the scratch surface in the cell monolayer treated with the extract optimized for
antiradical activity in concentration of 2.5 μL/mL reduced by approximately 60 %. Similar
results were observed in the extract optimized for phenolic acid content, indicating excellent
wound healing activity.
Conclusion
The performed research confirmed that glycerol is an effective solvent for the extraction
of the active components of herbal drugs: Berberidis radicis cortex, Echinaceae purpureae
herba, Liquiritiae radix, and Silybi mariani fructus. Ultrasound-assisted extraction was the
extraction method of choice due to shorter extraction time and superior extraction results over
maceration. Application of glycerol resulted in eco-friendly extracts with pronounced
antioxidant properties that inhibited selected skin-related enzymes. In addition to its many
advantages over conventional solvents, such as humectant properties and biocompatibility, that
eliminates the need to remove the solvent from the extract, glycerol was also active in the
protein coagulation assay thus displaying anti-inflammatory properties. The observed
biocompatibility with HaCaT cells indicated that the prepared glycerol extracts could be used
in dermatopharmaceutical preparations without prior removal of the solvent, thereby reducing
energy consumption and negative environmental impact. The observed biological effects of the
tested plant extracts indicate that they have a promising role in development of innovative
dermatopharmaceutical products intended for diminishing the consequences of skin aging.
Additional research is needed to determine the method of incorporation of the extracts into
dermopharmaceutical cosmetics, as well as their exact dosage.
Keywords: |
