Received: 14 April 2020 Revised: 15 July 2020 Accepted: 15 July 2020

DOI: 10.1002/ptr.6832

RESEARCH ARTICLE

Plant cell culture extract of Cirsium eriophorum with skin pore

refiner activity by modulating sebum production and
inflammatory response

Sonia Laneri1 | Irene Dini1 | Annalisa Tito2 | Ritamaria Di Lorenzo1 |

Marida Bimonte | Assunta Tortora | Claudia Zappelli | Maura Angelillo3 |
2 2 3

Antonietta Bernardi1 | Antonia Sacchi1 | Maria Gabriella Colucci2,3 | Fabio Apone2,3

1
Department of Pharmacy, University of
Naples Federico II, Naples, Italy Facial pore enlargement is considered a significant esthetic and health concern in
2
Arterra Bioscience SpA, Naples, Italy skincare cosmetics. The pores fulfill the critical function of keeping the skin surface
3
Vitalab srl, Naples, Italy
hydrated and protected against microbial infections. The hyperseborrhea, the stress
Correspondence factors, and the hormonal triggers can cause pore size enlargement, causing higher
Dini Irene, Department of Pharmacy,
susceptibility of the skin to microbe aggressions and inflammatory reactions.
University of Naples Federico II, Via Domenico
Montesano 49, Naples 80131, Italy. Thus, reducing excessive sebum production and keeping functional pores are two of
Email: irdini@unina.it
the most requested activities in skincare cosmetics. A Cirsium eriophorum cell culture
Funding information extract was investigated for its role in sebum regulation, stratum corneum desquama-
PON (National Operational Program) “Imprese
tion, and anti-inflammation. The extract was able to regulate essential markers
e competitività” - Iniziativa, Grant/Award
Number: PMI2014-2020 associated with sebum secretion and pore enlargements, such as the enzyme
5α-reductase, which plays a central role in sebum production, and the trypsin-like
serine protease Kallikrein 5, which promotes skin exfoliation and antimicrobial
response. Moreover, the extract showed a sebum-normalizing and pore refining
activity in individuals having seborrheic or acne-prone skins, suggesting a role of the
C. eriophorum extract in rebalancing altered skin conditions responsible for pore
enlargement.

KEYWORDS

acne-prone, C. eriophorum cell cultures, dermo-cosmetic, mattifying effect, oily skin, sebum-
normalizing

1 | I N T RO DU CT I O N enlarged pores represent an easy way-in for harmful or potentially

Pores play a crucial role in facial skin, as they release the sebum pro-
duced by the sebaceous glands. However, when the production of
sebum is excessive, due to environmental stress, genetic predisposi-
tion or pathological skin conditions, the pore ducts can be clogged,
and the pores appear enlarged and deformed (Dong, Lanoue, &
Goldenberg, 2016). Enlarged facial pores represent one of the most
significant cosmetic concerns, as they became evident features, often
associated with comedones, pustules, papules, and cysts (Maia
Campos, Melo, & Mercurio, 2019). Besides the esthetic condition,
pathogenic aggressors, giving rise to local inflammatory reactions
(Ki & Rotstein, 2008). Therefore, minimize the appearance of the facial
pore size whilst decreasing the sebum secretion is one of the most
desired targets of skincare products. Two enzymes represent relevant
markers associated with sebum production and pore enlargements:
the 5α-reductase, associated with sebum formation and secretion
(Thiboutot et al., 2000), and the trypsin-like serine proteases Kallikrein
5 (KLK5), linked to physiological shedding and removal of pore-
clogging dead skin cells (Stefanini, Cunha, Henrique, & Tajara, 2015).
The intense pulsed light, the retinoic acid (RA) creams, the oral

Phytotherapy Research. 2020;1–11. wileyonlinelibrary.com/journal/ptr © 2020 John Wiley & Sons Ltd 1
2 LANERI ET AL.
isotretinoin uptake, the iontophoresis, and the glycolic acid peeling
are currently the most used treatments to reduce pore size. However,
only partial success has been reported due to a lack of efficacy or
occurrence of irritative reactions (Roh, Han, Kim, & Chung, 2006;
Thielitz, Abdel-Naser, Fluhr, Zouboulis, & Gollnick, 2008). The growing
demand for plant-derived ingredients in cosmetics is indeed driven by
a significant reduction of the side effects and the chance to use them
in unlimited amounts (Juliano & Magrini, 2018). Plants of the genus
Cirsium (family of Asteraceae) have been extensively used in the food
and biomedical applications (Das, Behera, & Pramanik, 2017), due to
their content of active phytochemicals, such as flavonoid glycosides,
a et al., 2015), with
phenolic acids, sterols, triterpenes and lignans (Bog
known antimicrobial, anti-inflammatory and antiphlogistic activities
(Dini & Laneri, 2019; Inui, Aoshima, Nishiyama, & Itami, 2011; Jeong,
Jung, & Choi, 2008). In this study, a new potential dermo-cosmetic
ingredient was obtained by Cirsium eriophorum cell suspension cul-
tures and, after a chemical characterization, it was tested in vitro on
skin cell cultures and in vivo on oily and acne-prone skins. Plant cell
cultures are becoming an always more accessible source of active
ingredients for skin care applications, due to their bio-sustainable and
standardized characteristics in industrial production (Apone, Tito,
Arciello, Carotenuto, & Colucci, 2020; Eibl et al., 2018; Karg &
Kallio, 2009). C. eriophorum cell culture extract reduced the expression
and the activity of essential markers associated with sebum produc-
tion and pore enlargements, such as 5α-Reductase and Kallikrein
5, and inhibited the expression of pro-inflammatory mediators in skin
cells. Moreover, it improved epidermal barrier functionality, restored
an average sebum level, and reduced pore size, suggesting a potential
effect as a dermo-cosmetic ingredient in alleviating complications
related to acne-prone and seborrheic skins.
2 | MATERIALS AND METHODS
2.1 | Plant tissue cultures and extract preparation
Cell cultures were obtained by C. eriophorum plants (Plant World
Seeds, Devon, UK) by inducing the proliferation of meristematic leaf
cells on solid agar plates until obtaining calluses. The cells were trans-
ferred to the liquid growth medium (Gamborg B5 medium, sup-
plemented with 500 mg/L myo-inositol, 30 g/L sucrose, and
phytohormones) and grown as suspension cultures at 27
C in the dark
under orbital shaking. Once the cultures of about 300 g/L were
obtained, the cells were collected and lysed in a phosphate buffer at
pH 7.4 to prepare a water-soluble extract, which was lyophilized
(CeCCE). The powder was dissolved in water or cell culture media at
the appropriate concentrations for testing.
2.2 | Chemical analyses
Samples of CeCCE were dissolved in a water solution of 1%
trifluoroacetic acid. The solution was analyzed by Liquid
Chromatography–Mass Spectrometry (LC–MS/MS). A UHPLC (Ultra-
High-Performance Liquid Chromatography; UltiMate 3000 UHPLC,
ThermoFisher Dionex, Bartlesville) was interfaced with a high-
resolution mass spectrometer based on Orbitrap technology (Q-
Exactive, ThermoFisher, Bartlesville) and connected to an EASY-Spray
source. The samples were desalted by a 0.1% formic acid solution on
a C18 PepMap 100 precolumn (Thermo Scientific, Bartlesville) and
then loaded onto the EASY-Spray analytical column (15 cm x 75 μm
ID PepMap RSLC C18, 3 μm, 100 Angstrom; Thermo Scientific,
Bartlesville) using 0.1% formic acid in water (solvent A) at a flow of
300 ml/min. The elution from the column was performed by using an
increasing concentration of a solvent B (acetonitrile/0.1% formic acid).
Mass data (MS) and MS/MS were acquired from the mass spectrome-
ter operating in positive ion mode with a vaporization temperature of
350
C, a capillary temperature of 280
C, and a capillary voltage of
1.9 kV. The spectrometer was set in data-dependent acquisition mode
by selecting the five most intense ions (isolation window 1.0 m/z,
stepped collision energy 20, 40) with a dynamic exclusion range of
15 s, selecting the ions with a single and double charge. The MS/MS
data from the compound fragmentation was analyzed in triplicate and
were processed using the Compound Discoverer 2.1 Software
(ThermoFisher, Bartlesville) for the identification of the metabolites.
2.3 | Cell cultures
Human Dermal Fibroblasts (HDF, Cell Applications, Inc., San Diego,
CA) derived from neonatal foreskin and immortalized human
keratinocytes HaCaT (AddexBio Technologies, San Diego, CA) were
maintained in DMEM supplemented with 10% of fetal bovine serum
in a 95% air, 5% CO2 and humidified atmosphere at 37
C. Human
Epidermal Keratinocytes (HEK, Gibco/Life Technologies, Carlsbad,
CA) were maintained in EpiLife medium supplemented with Human
Keratinocyte Growth Supplement in a 95% air, 5% CO2 and humidi-
fied atmosphere at 37
C.
2.4 | 5-alpha reductase activity
8 x 103 HDF were seeded in 96-well plates and treated with CeCCE
plus 100 nM testosterone for 24 hr. As a positive control, the syn-
thetic aza-steroid Finasteride was used at a concentration of 10 nM.
After the treatment, the supernatants were collected and stored at
4
C for the analysis. Multiwell plates were coated with 20 ng/ml of
BSA-conjugated dihydrotestosterone (BSA-DHT) in Na2CO3 50 mM,
pH 9.0, at 4
C. After washing in phosphate buffer (PBS), the plates
were incubated with 50 μl of cell supernatant and 50 μl of anti-DHT
primary antibody (Cloud-Clone Corp. CCC), conjugated with biotin,
dissolved in PBS containing 1% Bovine Serum Albumin (BSA). After
2 hr, the plates were washed three times and incubated with 5 μg/ml
of streptavidin-conjugated Horse Radish Peroxidase in PBS containing
1% BSA. The color, developed by a 0.5 mg/ml solution of O-Phenylen
Diamine (OPD) in 0.012% H2O2, was measured by the absorbance at
LANERI ET AL.
490 nm (Seo, Yumnam, Jeong, & Kim, 2018). The reported values,
expressed as percentages, were the means of three independent
experiments. The observed differences were statistically significant as
calculated by the paired-samples t test.
2.5 | Gene expression analyses of cytokine genes
1.5 x 105 HaCaT were seeded in 6-well plates and treated for 16 hr
with CeCCE or RA 1 μM. Parallely, cultures of Staphylococcus
epidermidis cells (NCTC 11047 - Sigma Aldrich, Germany) were grown
overnight until OD = 0.45. About 1 x 108 Staphylococcus epidermidis
cells (1:20 dilution in the cell culture medium) were then added to the
keratinocyte cultures and incubated for additional 3 hr. After the
treatments, total RNA was extracted with the GenElute Mammalian
Total RNA Purification Kit (Sigma-Aldrich, Milano, Italy) and treated
with deoxyribonuclease (DNAse) I (Thermo Fisher Scientific, Dallas,
TX) at 37
C for 30 min. Reverse transcription was performed using
the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scien-
tific, Dallas, TX). Semi-quantitative PCR was performed using the
Quantum RNA kit (Thermo Fisher Scientific) containing primers to
amplify 18S ribosomal RNA (18S rRNA) along with competimers, that
reduced the amplified 18S rRNA product within the range to be used
as endogenous standard. The amplification reactions were made by
using specific oligonucleotides (IL-1αFor: ATGGCCAAAGTTCCAG
ACAT; IL-1αRev: TGTAATGCAGCAGCCGTGAG; IL-1βFor: ATGGCA
GAAGTACCTGAGCT; IL-1βRev: AAGGACATGGAGAACACCAC; IL-8
For: GCCACCGGAGCACTCCATAA; IL-8Rev: CTCTTCAAAAACTTCT
CCACAACC; TNFαFor: ATGAGCACTGAAAGCATGATCC; TNFαRev:
TCATACCAGGGCTTGGCCTCA) in the Mastercycler ProS (Eppendorf,
0

Milano, Italy) with the following general scheme: 2 at 94 C followed
by 35 cycles of 94
C for the 30 s, 50
C for 30 s, and 72
C for 30 s,
with a 10 min final extension at 72
C. The PCR products were loaded
on 1.5% agarose gel, and the amplification bands were visualized and
quantified with the Geliance 200 Imaging system (Perkin Elmer,
Woodbridge, Ontario, Canada). The amplification band corresponding
to the analyzed gene was normalized to the amplification band
corresponding to the 18S and reported as a percentage of bacteria
treated cells (set as 100%; Wang, Bell, Keeney, & Strobel, 2010). The
reported values represented the means of three independent experi-
ments and the observed differences were statistically significant as
calculated by the paired-samples t test.
2.6 | Gene expression and activity of Kallikrein 5
5
For KLK5 gene expression analysis, 1.5 x 10 HEK were seeded in
6-well plates and treated for 6 hr with CeCCE or RA 1 μM. After the
treatment, the cells were collected and total RNA was extracted with
the GenElute Mammalian Total RNA Purification Kit (Sigma-Aldrich).
The RNA was then processed as described above for the cytokine
gene expression analysis. The oligos used in the amplification reac-
tions were KLK5For: ATCACAGCCTTGCTTCTG and KLK5Rev:
3
GACCTTAGGGAAGTGCACT. For KLK5 activity evaluation, the cells
were seeded in 6-well plates and at confluency 1.2 mM CaCl2 was
added to induce cell differentiation in the presence or not of the
extract or 1 μM RA. After 48 hr the cell culture media were collected
and incubated with 1 mM Na-Benzoyl-D,L-arginine 4-nitroanilide
hydrochloride (DL-BAPA, Sigma-Aldrich) in 50 mM Tris–HCl + 50 mM
CaCl2, at pH 8.0. After 72 hr of incubation at 37
C the absorbance at
410 nm was measured. The values reported in the graph were
expressed as percentages respect to CaCl2 treated cells and represen-
ted the means of three independent experiments.
2.7 | ELISA assay for collagen measure
To determine Pro-Collagen I production, 8 x 104 HDF were seeded
in 96-well plates and treated for 24 hr with the extract or (trans-
forming growth factor beta) TGFβ 2.5 ng/ml as positive control.
After washing in Phosphate-buffered saline (PBS; Thermo Fisher
Scientific), cells were fixed in 4% paraformaldehyde for 10 min,
washed three times with the “ELISA buffer” (PBS 1X, 0.5 mM CaCl2,
1 mM MgCl2, 0.1% Triton X100) and treated with 3% of BSA for
30 min in ELISA buffer. After 2 hr of incubation with the primary
antibody (sc-166,572, Santa Cruz Biotechnology, Dallas, TX) at room
temperature, the cells were washed, incubated with the anti-mouse
secondary antibody (Biorad, CA) for 1 hr and then the amount of
Procollagen I measured by a colorimetric reaction, using 0.35 mg/ml
solution of OPD (Sigma-Aldrich, Milano, Italy) and 0.012% H2O2 in
50 mM citrate buffer (di Martino et al., 2017). The values were
reported as percentages to the untreated control (set as 100%) and
represented the means of three independent experiments. The
observed differences were statistically significant as calculated by
the paired-samples t test.
2.8 | In vivo studies
The test was made according to the principles of the Helsinki Declara-
tion (1964) and subsequent revisions endorsed by the European Com-
munity (fourth revision, called Somerset West, South Africa, 1996;
Saunders & Wainwright, 2003) and according to the Colipa Guidelines
for the evaluation of the efficiency of cosmetic products (May 2008).
2.8.1 | Apparatus
Trans Epidermal Water Loss and skin hydration were tested by a
Tewameter Probe and a Corneometer probe DUAL MPA 580 equipped
to a Software CK-MPA-Multi-Probe-AdapterFB (Courage+Khazaka
electronic GmbH Mathias-Brüggen-Str.9150829 Köln, Germany). Sebum
secretion was obtained by Sebumeter SM 815, while for pore counting
and type, we used a VisioFace 1000D connected to the Software CSI
(Complete Skin Investigation). These instruments are Courage+Khazaka
electronic GmbH, Köln, Germany.
4 LANERI ET AL.
2.8.2 | Cream composition
The cream with Cirsium extract contained: Phase O (Triticum vulgare
germ oil 5%); Emulsifiers (Cetyl alcohol 2%, Polyglyceryl-3methylglucose
distearate 5.0%, Cetearyl alcohol 3%); Phase W Aqua (q.b to 100);
Cirsium extract (0.5%); Preservants (Potassium sorbate, sodium benzoate
0.5%) and Parfum (0.1%). The cream indicated as placebo contained all
the components without Cirsium extract. All the components were
bought from ACEF (Fiorenzuola D'Arda, Italy) and Parfum by Farotti
Essenze (Rimini, Italy).
The creams (active and placebo) were prepared shaking vigor-
ously the oil phase (O) at 70
C with the aqueous phase, heated up to
the same temperature, by a Silverson L5M-A Laboratory Mixer.
Subsequently the cream was cooled in an ice bath and the remaining
components were added at room temperature. The pH of the creams
was measured by a pHmeter Crison GPL20 and they were 5.5–5.4;
the viscosity of both the creams (with the ingredient and placebo) was
29.254–30.156 mPa (L4, 20 rpm) and was calculated by a rheometer
Visco Basic Plus, Fungilab.
2.8.3 | Test conditions
Skin biophysical parameters were detected in a chamber (T = 25 ± 2
C;
relative humidity 40 ± 5%), after an acclimation period of approxi-
mately 30 min, at the beginning (D0), after 2 weeks (D14) and after
4 weeks (D28) of twice-daily treatment with the test product
(2 mg/cm2).
2.8.4 | Volunteer selection
A panel of 40 volunteers, aged from 20 to 40, with combination or
oily skin, applied a dermo-cosmetic cream containing 0.5% p/p
plant extract or the corresponding placebo twice daily for 4 weeks
on faces. Biophysical parameters were monitored over 2 and
4 weeks of product use. Volunteers were selected based on inclu-
sion and non-inclusion specific criteria and on their ability to
respect the constraints required by the protocol. All volunteers
signed an informed consent sheet, according to the Helsinki Decla-
ration of ethical principles for medical research and claimed not to
be allergic to the dermo-cosmetic cream components. Each volun-
teer had a card with the instruction for dermo-cosmetic cream use
and, at the end of the study, answered a questionnaire to determine
product compliance.
2.8.5 | The inclusion and exclusion criteria
The subjects included in the study were female and male aged
between 20 and 40 with phototype I-IV (Fitzpatrick Scale), with
combination or oily skin, impure and acne-prone skin, in functional
health status, without systemic pathologies. The volunteers do not
apply other products than the one studied on the test sites within
7 days before the test and during the study period. In this study,
pregnant or lactating women, subjects with a history of skin hyper-
reactivity or reactions of intolerance to cosmetic products/ingredi-
ents, affected by skin diseases (eczema, psoriasis, lesions), subjects
with photosensitivity, overexposed to intensive sunlight (natural or
artificial) the month preceding the study or expected to be exposed
to UV rays during the study period, were excluded.
2.9 | Statistical analyses
All the in vitro experiments were conducted in triplicates and
repeated three times. The results were presented as means ± standard
deviations (SD) of three independent experiments. A paired-samples
t test was conducted using the GraphPad QuickCalcs Website:
https://www.graphpad.com/quickcalcs/ttest1.cfm; a p value lower
than .05 was considered statistically significant. For the in vivo tests,
statistical tests were performed using SPSS software 15.0 for
Windows (SPSS Science, Chicago, IL). Student t-test was assessed.
A p value <.05 was considered significant.
3 | RE SU LT S
3.1 | Chemical characterization of C. eriophorum
cell culture extract and phenolic quantification
The most abundant phenols present in the CeCCE extract
were identified (Table 1) and quantified by LC-HRMS (Orbitrap)
analysis. Thirteen phenolics, including nine phenolic acids, three
caffeoylquinic acids derivatives, and two dicaffeoylquinic acids,
were identified.
Validation parameters for the quantification of the phenolics by
UPLC were explained in Table 2.
The caffeoylquinic acid derivatives were the most representative
class of phenols and the chlorogenic acid, the most concentrated phy-
tochemicals in the extract (Table 3).
3.2 | In vitro studies
3.2.1 | 5α-reductase activity
Pore dilation in oily and acne-prone skins is often associated with
sebum overproduction, which is, in turn, regulated by the activity of
the enzyme 5α-Reductase (5α-R). To investigate on the capacity of
CeCCE to modulate sebum production in skin cells, the 5α-R activity
was first induced in human dermal fibroblasts by testosterone
100 nM. Then the cells were treated with CeCCE at two concentra-
tions and with the compound (5α,17β)-(1,1-Dimethylethyl)-3-oxo-
4-azaandrost-1-ene-17-carboxamide, a 5α-R inhibitor, known as
Finasteride (Rattanachitthawat, Pinkhien, Opanasopit, Ngawhirunpat, &
LANERI ET AL. 5

TABLE 1 Validation parameter for phenolics analysis by UHPLC–MS/MS

Calculated
RT Theoretical Experimental errors
Phenolic compounds (min) Formula m/z m/z Δppm LC/MSn data (% base peak)
Phenolic acid
Caffeic acid 4.65 C9H7O4 179.0338 179.0332 −3.2 MS2[179] 135.0451 (100)
Caffeic acid glucosidic 1.55 C15H18O9 341.0866 341.0877 0.40 179.0345 (100), 135.0400 (10)
derivatives
p-Coumaric acid 9.71 C9H10O5 163.04007 163.04028 1.29 119.05023
Sinapinic acid hexoside 6.29 C17H22O10 385.1128 385.1139 0.11 223.0610 (100), 208.0400 (20),
179.0698 (10)
Ferulic acid hexoside 6.42 C16H20O9 355.123 355.1034 0.10 193.0507 (100), 178.0280 (10)
Protocatechuic acid 2.45 C7H6O4 153.183 153.0193 0.06 109.0292 (100)
Hydroxybenzoic acid 2.88 C7H6O3 137.02442 137.02458 1.17 93.03431
Gallic acid 7.55 C7H7O5 169.01425 169.01639 3.81
Caffeoylquinic acids
5-O-caffeoylquinic acid 4.94 C16H17O9 353.0867 353.0875 0.7 MS2[353] 191.0557(100),
179.0343 (7), 173. 0450 (5)
1-O-caffeoylquinic acid 4.27 C16H18O9 353.0867 353.0877 0.2 191.0555 (100), 179.0343 (10)
3-O-caffeoylquinic acid 2.68 C16H17O9 353.0867 353.0877 0.2 MS2[353] 191.0548(100),
(chlorogenic acid) 179.0338(46), 135.0443(7))
Dicaffeoylquinic acids
1.5-O-dicaffeoylquinic acid 15.53 C25H23O12 515.1184 515.1176 −1.54 MS2 [515]: 353.0853(100),
(cynarin) 191.0552(5), 179.0342(5
4.5-O-dicaffeoylquinic acids 15.81 C25H23O12 515.1184 515.1182 −0.3 MS2[515] 353.0865(100)

TABLE 2 Validation parameter for phenolic analysis by UHPLC–MS/MS

Phenolic compounds Linearity (mg/g) R2 LOD (mg/g) LOQ (mg/g) RSD % (n = 3), (50 mg/g)
Phenolic acid
Caffeic acid 0.5–2 0.998 0.542 2.50 1.19
Caffeic acid exoses 3–4 0.995 3.147 3.781 1.24
Coumaric acid 0.1–1 0.998 0.207 0.822 0.9
Sinapinic acid hexoside 0.1–3 0.991 0.105 0.458 1.17
Ferulic acid hexoside 0.01–0.1 1.000 0.012 0.092 1.27
Protocatechuic acid 1–5 0.912 3.5 5.236 1.24
Hydroxybenzoic acid 1–5 0.991 0.100 2 1.8
Gallic acid 1–5 0.991 0.994 1.526 1.52
Caffeoylquinic acids
1-O-caffeoylquinic acid 5–13 0.992 11.983 12.578 1.22
3-O-caffeoylquinic acid (chlorogenic acid) 1–120 0.991 102.321 104.567 1.19
Dicaffeoylquinic acids
1.5-O-dicaffeoylquinic acid (cynarin) 1–3 0.991 0.45 3.567 1.23
4.5-O-dicaffeoylquinic acids 3–5 0.991 4.231 5.012 1.21

Abbreviation: RSD, relative standard deviation, LOD, Limit of detection, LOQ, Limits of quantification.

Chanvorachote, 2019), as a positive control. Significant inhibition of the 3.2.2 | Kallikrein 5 activity
testosterone-induced enzymatic activity was produced by CeCCE
0.002% (52.2%) and 0.01% (44.5%), even stronger than those obtained We analyzed the involvement of CeCCE in the regulation of
with finasteride (Figure 1). kallikrein activity, as kallikreins are important enzymes modulating
6 LANERI ET AL.

hyper-keratinisation, which is often associated with pore dilation (An et al., 2017), mimicking a skin bacteria dysbiosis in vitro, and the
(Igawa et al., 2017). Skin keratinocytes were treated with the extract or expression levels of the inflammatory cytokines interleukin-1α (IL-1α),
with RA as positive control and the expression level and the enzymatic interleukin-1β (IL-1β) interleukin-8 (IL-8) and Tumor Necrosis Factor-α
activity of Kallikrein 5 (KLK5), the main protease involved in desmo- (TNF-α) were measured by RT-PCR. The results, shown in Figure 3,
some degradation, were measured. The CeCCE extract at 0.002% and indicated that the treatment with CeCCE, at both the concentrations,
0.01 increased both expression and enzymatic activity at significant led to significant inhibitions of all the analyzed cytokines.
way similar to those obtained with RA, indicating a potential effect of
the extract in promoting skin desquamation (Figure 2a,b).
3.2.3 | Collagen I production
Antiinflammatory activity
The capacity of CeCCE to inhibit the release of inflammatory We analyzed CeCCE ability to induce collagen I production in skin
mediators by skin cells was evaluated. Epidermal keratinocytes were fibroblasts, since this essential protein is the main responsible for
challenged with an excess of the bacteria Staphilococcus epidermidis keeping the skin firm around the pores. The results of the analysis
indicated that CeCCE significantly stimulated Pro-Collagen I produc-
tion in ELISA assay, analogously to the positive control TGFβ (Lijnen &
TABLE 3 Polyphenols composition of CeCCE
Petrov, 2002), suggesting a positive role of the extract in maintaining
Concentration in a correct dermis tone (Figure 4).
Compound CeCCE (mg/g of powder)
Phenolic acids
Caffeic acid 1.81 3.3 | In vivo studies
Caffeic acid exoses 3.41
Coumaric acid 0.61 Twenty volunteers (female and male healthy individuals with oily or
Sinapinic acid hexoside 0.26 acne prone skin) were treated on the face with an cream containing

Ferulic acid hexoside
Protocatechuic acid
Hydroxybenzoic acid
Gallic acid
Caffeoylquinic acids
3-O-caffeoylquinic acid
(chlorogenic acid)
1-O-caffeoylquinic acid
Dicaffeoylquinic acids
1.5-O-dicaffeoylquinic acid
(cynarin)
4.5-O-dicaffeoylquinic acid
0.08 0.5% p/p of CeCCE, twice a day for 28 days, and the effects on
sebum production, pore size and skin hydration were compared to
4.70
those of 20 volunteers who were treated with a placebo cream.
1.55
1.23
3.3.1 | Activity on sebum production
103.34
The results of the sebum production test showed that CeCCE con-
12.39
taining cream reduced the sebum production by 34.92% after
2 weeks, and 46.53% after 4 weeks, significantly more than the pla-
2.16 cebo cream, which inhibited sebum production of 10.49 and 10.11%
after 2 and 4 weeks, respectively (Figure 5).
4.77

120

100

80
*
60
**
**
40
***
20

0 F I G U R E 1 5α-Reductase activity. The error

Control CeCCE 0,002% CeCCE 0,01% Finasteride 10nM
bars represent standard deviations, all the
Untreated
measures are statistically significant (*p < .05;
Testosterone 100nM 100 52.2 44.55 68.82
**p < .01; ***p < .001)
LANERI ET AL. 7

F I G U R E 2 (a and b) Kallikrein 5 expression (A) 200

level and activity. The error bars represent
standard deviations, all the measures are 180
***
statistically significant (*p < .05; 160 *** **
**p < .01; ***p < .001)
140

120

100

80

60

40

20

0
Control CeCCE 0,002% CeCCE 0,01% RA 1uM
Series1 100 137.23 140.26 161.64

(B)
200
*
*
KLK5 activity (% to control = 100)

150 *

100

50

0
CeCCE
Control CeCCE 0.01% RA 1uM
0.002%
Series1 100 128.7 166.3 152.3

3.3.2 | Activity on pore size
The second parameter measured was the reduction in the number of visible
pores, indicating a decrease of the pore orifice size. The CeCCE containing
cream reduced the number of visible pores by 40.0 and 37.91% after 2 and
4 weeks of treatment, respectively, differently from the placebo cream,
whose effect on pore size was not significant (Figure 6a). The treatment with
the CeCCE containing cream produced an effect also on the forehead, where
it reduced the number of visible pores by 44.80% (2 weeks) and 43.01%
(4 weeks; data not shown). Photographs of the cheeks, taken before and
after 4 weeks of treatment, are reported (Figure 6b).
the trans epidermal water loss (TEWL) produced by the extract con-
taining cream and that of the placebo. The cream containing CeCCE
reduced TEWL values by 20.34% after 2 weeks, and by 18.05% after
4 weeks, while the placebo cream reduced it only by 12.18 and
12.51%, after 2 and 4 weeks, respectively (Figure 7). Moreover, the
cream containing CeCCE enhanced epidermal moisture level, mea-
sured by corneometer, by 19.5 and 20.0% after 2 and 4 weeks of
treatment, while the placebo cream produced only a modest effect of
4.5% after 4 weeks (Figure 7).
4 | DI SCU SSION

3.3.3 | Activity on skin hydration In the present article, we showed that C. eriophorum cell cultures were
promising sources of bioactive molecules, including phenols, typically
As pore enlargement is often associated with a decrease in the skin present in plants of the genus Cirsium and the genus Cynara (family of
hydration level (Katsuta, Iida, Inomata, & Yoshida, 2004), we measured Asteraceae; Bo
ga et al., 2015; D'Antuono et al., 2018). For this
8 LANERI ET AL.

120

100 **
*** * *
* * * *
80 * *
* *** *
60 *
*
*
40

20

0
IL-1a IL-1b IL-8 TNF-a
control 68.31 39.94 60.5 64.58
S.epidermidis 100 100 100 100
S.epidermidis+CeCCe 0,002% 85.89 60 72.8 74
S.epidermidis+CeCCe 0,01% 84.44 50.69 71.5 73.46
S.epidermidis+T0901317 10uM 77.84 54.31 26.83 82.26

F I G U R E 3 Cytokine expression level modulation. The values ± standard deviations are expressed as relative % to the S. epidermidis treated
samples, set as 100% (*p < .05; **p < .01; ***p < .001)

* D14 D28
160 0
*
140
Sebum production (%)
Pro-Col I (% versus control=100)

*** -10

120
-20

100
-30
80
-40
60
-50
40
-60
20 D14 D28
CeCCE cream -34.92 -46.53
0 Placebo -10.49 -10.11
Control CeCCE 0.002% CeCCE 0.01% TGFβ 2.5ng/mL
Series1 100 126.175172 131.2393164 142.8185126
F I G U R E 5 Sebum level measure. D14: 14 days of treatment;
D28: 28 days of treatment. All the measures are statistically
F I G U R E 4 Pro-Collagen I production measure. The error bars
significant (p < .05)
represent standard deviations, all the measures are statistically
significant (*p < .05; ***p < .001)

and intra-day repeatability was validated by the relative standard

purpose, the phenolic profile and dosage were investigated by using
an UHPLC-Orbitrap platform in MS and MS/MS levels. The MS
methods were validated in terms of linearity, precision, and sensitivity.
The linearity of the calibration curve was established by correlation
factor of the calibration curve, the method sensitivity in the range of
the LODs and the LOQs, confirmed method sensitivity, and the inter
deviation (RSD) values <15%. Phenolic identification was obtained
comparing the retention time, mass spectra, accurate mass measure-
ments, and MS2 analyses with standards and with literature data
(Li, Zhang, Zeng, Yang, & Deng, 2011; Regueiro et al., 2014; Troise,
Ferracane, Palermo, & Fogliano, 2014). Caffeoylquinic acid derivatives
were the most representative class of phenolics in the CeCCE, and
LANERI ET AL.
A. Number of visible pores on the cheek
10 D14
Number of poresvariation (%)
-10
-20
-30
-40
-50
D14
CeCCE cream -40
Placebo -1.6
F I G U R E 6 Measures of visible pores on the cheek. D0: the day after treatment; D14: 14 days of treatment; D28: 28 days of treatment. All
the measures are statistically significant (p < .05) [Colour figure can be viewed at wileyonlinelibrary.com]
among these, the chlorogenic acid was the most concentrated. In this
study we demonstrated that a water-soluble extract derived from
C. eriophorum cell cultures was able to inhibit 5α-Reductase activity
and increase Kallikrein 5 expression when tested on skin cells in vitro,
suggesting its potential role in sebum regulation and skin stratum cor-
neum exfoliation. Both 5α-Reductase inhibitors and kallikrein
5 inducers have recently drawn much interest in Cosmetics and Der-
matology, due to their activity in helping anti-acne therapeutic drugs
and reducing the desquamation impairment, respectively (Yin,
Hwang, & Lee, 2019; Zhang et al., 2016; Zouboulis, Dessinioti,
Tsatsou, & Gollnick, 2017). Indeed, hyper-keratinization of the infun-
dibulum contributes to pore enlargement, together with alterations in
the thickness of the epidermis and a loss of dermis structural compo-
nents, thus the role of kallikrein 5 in reducing skin desquamation rep-
resents a key issue when treating this type of dysfunctions
(Murakami, Kawamura, Jie, & Tanno, 2006).
In this study, CeCCE negatively modulated the over-expression of
proinflammatory cytokines IL-1α, IL-1β,
keratinocytes, and increased collagen synthesis. The effect on colla-
gen I stimulation indicated that CeCCE might improve the dermal skin
properties, which get compromised in the skin areas around enlarged
pores (Dong et al., 2016). Indeed, a dermis relaxation and collapse of
the pore duct is also associated with an alteration of the collagen pro-
duction and stability in the aged skin, resulting in an enlargement of
the pore orifice diameter (Schütz et al., 2019). This effect becomes
even more critical in the presence of skin lesions that are conse-
quences of significant inflammatory reactions such as those leading to
pimples and pustules, where a fast and correct healing process is
9
B. Photographs of the faces before and after
treatment.
(The red and green dots indicate the number of
visible large and small pores respectively)
D0
D28
D28
D28
-37.91
3.52
critical (Fabbrocini et al., 2010). Our study provides results on the
effect of the phenolic compounds contained in a Cirsium cell culture
extract on the regulation of four main skin activities: sebum produc-
tion, desquamation, inflammatory response, and collagen synthesis,
which all translate into a pore refining and moisturizing effect, con-
firmed by the clinical studies.
Indeed, the study on human skin in vivo demonstrated that CeCCE
effectively reduced sebum secretion and the number of enlarged pores,
confirming the ability of the extract to attenuate the visible signs of
seborrheic and acne-prone skins. When tested on the cheeks and fore-
head, the CeCCE containing cream, differently from a placebo cream,
significantly reduced the number of enlarged pores, confirming that the
observed activities of the extract in vitro resulted in a pore-refining
effect on the skin in vivo. Moreover, a significant reduction in pore size
was observed in volunteers treated with CeCCE containing cream. As
previously shown, volunteers with dysfunctional pores exhibit a higher
TEWL value and a lower conductance within the stratum corneum,
IL-8, and TNF-α in which leads to dry skin conditions and more frequent irritations
(Logger, Olydam, Wietsker-van der Weg, & van Erp, 2019; Yamamoto,
Takenouchi, & Ito, 1995). Application of CeCCE containing cream
counteracted pore enlargement and inhibited TEWL, suggesting its
capacity to restore and maintain the skin barrier. The increased conduc-
tance values observed within the stratum corneum (Corneometry test),
clearly indicated an ameliorated permeability of the upper skin layers in
volunteers treated with CeCCE extract. The CeCCE extract studied in
this work could effectively improve the skin appearance by modulating
sebum production, decrease pore size, and ameliorate skin barrier prop-
erties without inducing any short-term and long-term irritation.
10 LANERI ET AL.

D14 D28 OR CID

0
Irene Dini https://orcid.org/0000-0003-1418-1431
Fabio Apone https://orcid.org/0000-0002-3693-2761
-5
TEWL variation (%)

RE FE RE NCE S
-10 An, Q., Sun, M., Qi, R. Q., Zhang, L., Zhai, J. L., Hong, Y. X., … Gao, X. H.
(2017). High Staphylococcus epidermidis colonization and impaired per-
meability barrier in facial Seborrheic dermatitis. Chinese Medical Jour-
-15
nal, 130(14), 1662–1669.
Apone, F., Tito, A., Arciello, S., Carotenuto, G., & Colucci, M. G. (2020).
-20 Plant tissue cultures as sources of ingredients for skin care applica-
tions. Annual Plant Reviews, 3, 135–150.
Bo ga, M., Yılmaz, P., Cebe, D., Fatima, M., Siddiqui, B. S., & Kolak, U.
-25 (2015). Chemical constituents and biological activities of Cirsium
D14 D28
leucopsis, C. sipyleum, and C. eriophorum. Zeitschrift für
CeCCE cream -20.34 -18.05
Naturforschung C, 69(9–10), 381–390.
Placebo -12.18 -12.51 D'Antuono, I., Carola, A., Sena, L. M., Linsalata, V., Cardinali, A.,
Logrieco, A. F., … Apone, F. (2018). Artichoke polyphenols produce
skin anti-age effects by improving endothelial cell integrity and func-
25 D14 D28
tionality. Molecules, 23(11), E2729. https://doi.org/10.3390/
Skin moisture variation

molecules23112729
20
Das, U., Behera, S. S., & Pramanik, K. (2017). Ethno-herbal-medico in
15 wound repair: An incisive review. Phototherapy Research, 31, 579–590.
di Martino, O., Tito, A., De Lucia, A., Cimmino, F., Cicotti, F., Apone, F., …
10
Calabrò, V. (2017). Hibiscus syriacus extract from an established cell
5 culture stimulates skin wound healing. Biomedical Research Interna-
tional, 2017, 7932019. https://doi.org/10.1155/2017/7932019
0 Dini, I., & Laneri, S. (2019). Nutricosmetics: A brief overview. Phototherapy
-5 Research, 33(12), 3054–3063.
D14 D28 Dong, J. B. A., Lanoue, J. M. D., & Goldenberg, G. M. D. (2016). Enlarged
CeCCE cream 19.5 20 facial pores: An update on treatments. Cutis, 98(1), 33–36.
Eibl, R., Meier, P., Stutz, I., Schildberger, D., Hühn, T., & Eibl, D. (2018).
Placebo 0.6 4.53
Plant cell culture technology in the cosmetics and food industries: Cur-
rent state and future trends. Applied Microbiology and Biotechnology,
F I G U R E 7 Effect on skin hydration. D14: 14 days of treatment; 102, 8661–8675.
D28: 28 days of treatment. All the measures are statistically Fabbrocini, G., Annunziata, M. C., D'Arco, V., De Vita, V., Lodi, G.,
significant (p < .05) Mauriello, M. C., … Monfrecola, G. (2010). Acne scars: Pathogenesis,
classification and treatment. Dermatology Research and Practice, 13,
1–13. https://doi.org/10.1155/2010/893080
Igawa, S., Kishibe, M., Minami-Hori, M., Honma, M., Tsujimura, H.,
5 | C O N CL U S I O N Ishikawa, J., … Ishida-Yamamoto, A. (2017). Incomplete
KLK7 secretion and upregulated LEKTI expression underlie hyperkera-
C. eriophorum cell culture extract reduced the expression and the totic stratum corneum in atopic dermatitis. Journal of Investigative Der-
matology, 137(2), 449–456.
activity of essential markers associated with sebum production and
Inui, S., Aoshima, H., Nishiyama, A., & Itami, S. (2011). Improvement of
pore enlargements, such as 5α-Reductase and Kallikrein 5, and
acne vulgaris by topical fullerene application: Unique impact on skin
inhibited the expression of pro-inflammatory cytokines in skin cells. care. Nanomedicine Nanotechnology Biology and Medicine, 7, 238–241.
Moreover, it restored an average sebum level, reduced pore size, and Jeong, D. M., Jung, H. A., & Choi, J. S. (2008). Comparative antioxidant
ameliorated epidermal barrier functionality, suggesting a potential activity and HPLC profiles of some selected Korean thistles. Archives
of Pharmacol Research, 31, 28–33.
effect as a dermo-cosmetic ingredient in alleviating critical conditions
Juliano, C., & Magrini, G. (2018). Cosmetic, functional ingredients from
related to acne-prone and seborrheic skins. botanical sources for anti-pollution skincare products. Cosmetics, 5, 19.
https://doi.org/10.3390/cosmetics5010019
ACKNOWLEDGMENTS Karg, S., & Kallio, P. T. (2009). The production of biopharmaceuticals in
plant systems. Biotechnology Advances, 27, 879–894.
PON (National Operational Program) “Imprese e competitività” -
Katsuta, Y., Iida, T., Inomata, S., & Yoshida, S. (2004). Improving the
Iniziativa PMI2014-2020. The funders had no role in the design of the appearance of facial pores. Cosmetics & Toiletries, 119(10), 59–64.
study; in the collection, analyses, or interpretation of data; in the writ- Ki, V., & Rotstein, C. (2008). Bacterial skin and soft tissue infections in
ing of the manuscript, or in the decision to publish the results. PON adults: A review of their epidemiology, pathogenesis, diagnosis, treat-
ment, and site of care. The Canadian Journal of Infectious Diseases &
(National Operational Program).
Medical Microbiology, 19(2), 173–184.
Li, Y., Zhang, X., Zeng, L., Yang, Y., & Deng, Y. (2011). Chemical profiling of
CONF LICT OF IN TE RE ST bioactive constituents in Sarcandra glabra and its preparations using
The authors declare no conflict of interest. ultra-high-pressure liquid chromatography coupled with LTQ Orbitrap
LANERI ET AL.
mass spectrometry. Rapid Communication in Mass Spectrometry, 25(17),
2439–2447. https://doi.org/10.1002/rcm.5123
Lijnen, P., & Petrov, V. (2002). Transforming growth factor-beta 1-induced
collagen production in cultures of cardiac fibroblasts is the result of
the appearance of myofibroblasts. Methods and Findings in Experimen-
tal and Clinical Pharmacology, 24, 333–344.
Logger, J. G. M., Olydam, J. I., Wietsker-van der Weg, W., & van
Erp, P. E. J. (2019). Noninvasive skin barrier assessment: Multi-
parametric approach and pilot study. Cosmetics, 6(1), 20. https://doi.
org/10.3390/cosmetics6010020
Maia Campos, P. M. B. G., Melo, M. O., & Mercurio, D. G. (2019). Use of
advanced imaging techniques for the characterization of oily skin.
Frontiers in Physiology, 10, 254–263.
Murakami, M., Kawamura, A., Jie, G., & Tanno, O. (2006). Morphological
analysis of facial pores and factors related to changes with aging in pore
shape. New York, NY: IFSCC Congress.
Rattanachitthawat, N., Pinkhien, T., Opanasopit, P., Ngawhirunpat, T., &
Chanvorachote, P. (2019). Finasteride enhances stem cell signals of
human dermal papilla cells. In Vivo, 33(4), 1209–1220.
Regueiro, J., Sánchez-González, C., Vallverdú-Queralt, A., Simal-
Gándara, J., Lamuela-Raventós, R., & Izquierdo-Pulido, M. (2014).
Comprehensive identification of walnut polyphenols by liquid chroma-
tography coupled to linear ion trap–Orbitrap mass spectrometry. Food
Chemistry, 152, 340–348.
Roh, M., Han, M., Kim, D., & Chung, K. (2006). Sebum output as a factor
contributing to the size of facial pores. British Journal of Dermatology,
155, 890–894.
Saunders, J., & Wainwright, P. (2003). Risk, Helsinki 2000 and the use of
placebo in medical research. Clin Med, 3, 435–9.
Schütz, R., Rawlings, A. V., Wandeler, E., Jackson, E., Trevisan, S.,
Monneuse, J. M., … Imfeld, D. (2019). Bio-derived hydroxystearic acid
ameliorates skin age spots and conspicuous pores. International Journal
of Cosmetic Science, 41(3), 240–256.
Seo, J. O., Yumnam, S., Jeong, K. W., & Kim, S. Y. (2018). Finasteride
inhibits melanogenesis through regulation of the adenylate cyclase in
melanocytes and melanoma cells. Archives of Pharmacal Research, 41
(3), 324–332. https://doi.org/10.1007/s12272-018-1002-x
Stefanini, A. C., da Cunha, B. R., Henrique, T., & Tajara, E. H. (2015). Involve-
ment of Kallikrein-related peptidases in normal and pathologic processes.
Dis Markers, 2015, 946572. https://doi.org/10.1155/2015/946572
11
Thiboutot, D., Bayne, E., Thorne, J., Gilliland, K., Flanagan, J., Shao, Q., …
Helm, K. (2000). Immunolocalization of 5alpha-reductase isozymes in
acne lesions and normal skin. Archives of Dermatology, 136(9),
1125–1129.
Thielitz, A., Abdel-Naser, M. B., Fluhr, J. W., Zouboulis, C. C., & Gollnick, H.
(2008). Topical retinoids in acne—An evidence-based overview. Journal
der Deutschen Dermatologischen Gesellschaft, 6(12), 1023–1031.
Troise, A. D., Ferracane, R., Palermo, M., & Fogliano, V. (2014). Targeted
metabolite profile of food bioactive compounds by Orbitrap high reso-
lution mass spectrometry: The “FancyTiles” approach. Food Research
International, 63B, 139–146.
Wang, Y., Bell, J. C., Keeney, D. S., & Strobel, H. W. (2010). Gene regula-
tion of CYP4F11 in human keratinocyte HaCaT cells. Drug Metabolism
and Disposition: The Biological Fate of Chemicals, 38(1), 100–107.
https://doi.org/10.1124/dmd.109.029025
Yamamoto, A., Takenouchi, K., & Ito, M. (1995). Impaired water barrier
function in acne vulgaris. Archives of Dermatological Research, 287(2),
214–218.
Yin, J., Hwang, I. H., & Lee, M. W. (2019). Anti-acne vulgaris effect includ-
ing skin barrier improvement and 5α-reductase inhibition by
tellimagrandin I from Carpinus tschonoskii. BMC Complementary Medi-
cine Research, 19, 323. https://doi.org/10.1186/s12906-019-2734-y
Zhang, L., Ferreyros, M., Feng, W., Hupe, M., Crumrine, D. A., Chen, J., …
Roop, D. R. (2016). Defects in stratum Corneum desquamation are the
predominant effect of impaired ABCA12 function in a novel mouse
model of harlequin Ichthyosis. PLoS One, 11(8), 0161465.
Zouboulis, C. C., Dessinioti, C., Tsatsou, F., & Gollnick, H. P. M. (2017).
Anti-acne drugs in phase 1 and 2 clinical trials. Expert Opinion on Inves-
tigational Drugs, 26, 813–823.
How to cite this article: Laneri S, Dini I, Tito A, et al. Plant cell
culture extract of Cirsium eriophorum with skin pore refiner
activity by modulating sebum production and inflammatory
response. Phytotherapy Research. 2020;1–11. https://doi.org/
10.1002/ptr.6832