Journal of Chromatography A 1684 (2022) 463582

Contents lists available at ScienceDirect

Journal of Chromatography A
journal homepage: www.elsevier.com/locate/chroma

Planar chromatography-bioassays for the parallel and sensitive
detection of androgenicity, anti-androgenicity and cytotoxicity
Carolin Riegraf1,2 , Anna Maria Bell2 , Marina Ohlig, Georg Reifferscheid,
Sebastian Buchinger∗
Federal Institute of Hydrology, Am Mainzer Tor 1, 56068 Koblenz, Germany

a r t i c l e

i n f o

Article history:
Received 28 April 2022
Revised 13 October 2022
Accepted 17 October 2022
Available online 19 October 2022
Keywords:
Anti-androgenicity
Androgenicity
Cytotoxicity
High-performance thin-layer
chromatography
Effect-directed analysis

a b s t r a c t

Anti-androgens entering the aquatic environment, e.g., by effluents from wastewater treatment plants or
agricultural settings are contributing to endocrine disruption in wildlife and humans. Due to the simultaneous presence of agonistic compounds, common in vitro bioassays can underestimate the risk posed
by androgen antagonists. On the other hand, cytotoxic effects might lead to false positive assessments of
anti-androgenic effects in conventional bioassays. In the present study, a combination of normal phase
high-performance thin-layer chromatography (NP-HPTLC) with a yeast-based reporter gene assay is established for the detection of anti-androgenicity as a promising tool to reduce interferences of androgenic and anti-androgenic compounds present in the same sample. To avoid a misinterpretation of antiandrogenicity with cytotoxic effects, cell viability was assessed in parallel on the same plate using a
resazurin viability assay adapted to HPTLC plates. The method was characterized by establishing doseresponse curves for the model compounds flutamide and bisphenol A. Calculated effective doses at 10%
(ED10) were 27.9 ± 1.3 ng zone−1 for flutamide and 20.1 ± 5.1 ng zone−1 for bisphenol A. Successful distinction between anti-androgenicity and cytotoxicity was exemplarily demonstrated with 4-nitroquinoline
1-oxide. As a proof of concept, the detection and quantification of anti-androgenicity in an extract of a
landfill leachate is demonstrated. This study shows that the hyphenation of HPTLC with the yeast antiandrogen screen is a matrix-robust, cost-efficient and fast screening tool for the sensitive and simultaneous detection of anti-androgenic and cytotoxic effects in environmental samples. The method offers a
wide range of possible applications in environmental monitoring and contributes to the identification of
anti-androgenicity drivers in the course of an effect-directed analysis.
© 2022 The Author(s). Published by Elsevier B.V.
This is an open access article under the CC BY license ( />
1. Introduction
Many different environmental contaminants are present in the
aquatic environment continuously released by wastewater treatment plants (WWTPs), agricultural settings or aquacultures [1, 2].
Among these are compounds, which can have adverse effects on
the endocrine system, so called endocrine disrupting compounds
(EDCs). In particular, many previous studies were focused on EDCs
that trigger estrogenicity via estrogen receptor-mediated signaling
pathways associated with adverse effects such as feminization and
vitellogenin production in male fish [3, 4].
∗
Corresponding author at: Federal Institute of Hydrology, Department G3 - Biochemistry, Ecotoxicology, Am Mainzer Tor 1, 56068 Koblenz, Germany.
E-mail address: (S. Buchinger).
1
Present address: Swiss Centre for Applied Ecotoxicology, Überlandstrasse 133,
8600 Dübendorf, Switzerland
2
These authors contributed equally to this work.


Besides estrogenicity elicited by (xeno-)estrogens, several studies worldwide revealed the presence of compounds influencing the
androgen receptor, e.g., in effluents of paper and pulp industries
[5], rangeland grazing and beef cattle feedlot [6, 7], leather fabrication [8], and WWTPs [9, 10] or in river sediments [11]. These
compounds can act as receptor agonists or antagonists, which either activate (androgenic compounds) or inhibit (anti-androgenic
compounds) hormonal androgen receptors, respectively, and hence,
the endogenous hormonal activity. As consequence, masculinization and effects on the immune system of aquatic biota caused by
androgens were observed [12]. In contrast, anti-androgenic compounds induced inter alia feminization of non-mammalian vertebrate males, changes in gender ratio as well as inhibited oogenesis
and spermatogenesis [13].
Whereas estrogenic and androgenic activities were found to
be largely eliminated by biological and advanced treatment of
wastewater, antagonistic activities were only removed up to 50%

/>0021-9673/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license ( />

C. Riegraf, A.M. Bell, M. Ohlig et al.

Journal of Chromatography A 1684 (2022) 463582

by biological treatment and also not further by advanced treatment [14]. Thus, there is a need to further investigate antagonistic
effects such as anti-androgenicity, which could contribute besides
steroidal estrogens to endocrine disruption in wildlife [9, 15].
Several compounds are known for their anti-androgenic activity,
e.g., natural and synthetic steroids such as androstenone [16], nandrolone [16] or cyproterone acetate [13], but also industrial compounds such as bisphenol A and butyl benzyl phthalate [17], petrogenic naphthenic acids [18], the germicides chlorophene and triclosan [10], the fungicide vinclozolin [13], polycyclic musks [16] or
Coumarin 47 [19], and other consumer products [10]. Nevertheless,
the identification of anti-androgenic effect-inducing compounds
remains in many cases challenging [19].
Effect-based methods (EBMs) are tools to assess the overall effect potential of a sample taking into account the contribution of
unknown compounds. Several EBMs are available for the detection
of (anti-)androgenic effects. Recombinant cells such as yeast [17,

20] and mammalian cell lines [21, 22] are frequently used in respective receptor-based reporter gene assays. These in vitro assays
usually performed in microtiter well plates assess the overall effect potential of a sample including potential mixture effects. Depending on the underlying objective this can be a desired feature,
however, the presence of both agonistic and antagonistic could
mask their mutual activity potentially leading to false conclusions
[23]. Furthermore, cytotoxic effects might interfere with the detection of a specific effect in microwell-based assays [24]. Hashmi,
et al. [25] concluded in a recent study from 2020 that there is
a need “to better understand the occurrence of EDCs and masking compounds in different lipophilicity windows, to finally reduce
fractionation requirements for monitoring to a smart clean-up.” In
this respect the direct combination of high-performance thin-layer
chromatography (HPTLC) with in vitro-bioassays provides an efficient platform as a matrix-robust, cost-efficient and fast screening
approach that can guide subsequent in-depth EDA [26]. For this approach samples are separated on a HPTLC plate into different fractions prior to performing the EBM directly on the surface of the
HPTLC plate. After exposure of the cells, potential effect-inducing
fractions are observable and can be quantified. This approach was
successfully applied for the detection of compounds with specific
modes of action such as estrogenicity in personal care products
[27] or food contact materials [28], androgenicity in WWTP effluents [29] or photosystem II inhibition in surface water extracts
[30]. Recently, Klingelhöfer, et al. [31] published a reversed phase
HPTLC planar yeast anti-androgen screen (RP-HPTLC-p-YAAS) for
the detection of anti-androgenic compounds using lacZ as reporter
gene.
The objective of the present study was to develop a p-YAAS
on normal phase HPTLC plates as a complementary approach to
the reversed phase HPTLC to broaden the applicability, further increase the sensitivity of the bioassay and simultaneously enable a
quantification of antagonistic effects. In addition, a combination of
the p-YAAS with a test for cell viability was implemented to distinguish between anti-androgenic and cytotoxic compounds on the
same HPTLC plate. The developed methodology was examined and
optimized in terms of effective doses (ED) and repeatability using
mixtures of reference compounds. As proof of principle for an application to complex matrices, the methodology was used to assess
the anti-androgenic potential of a landfill leachate extract.


tained from Merck in the highest purity commercially available.
The solvents ethanol (≥99.8%), chloroform (99.0 - 99.4%), n-hexane
(99.0%), methanol (99.9%) and petroleum fraction (bp 65 – 100°C)
were purchased from Merck. Ethyl acetate (EtAc, 99.8%) was acquired from LGC Standards. Resazurin sodium salt (CAS: 62758-138) was obtained from Merck. The components for growth- and exposure medium were obtained from Merck in the highest grade
commercially available.

2.2. Media and solutions
Stock solutions of model compounds were prepared in ethanol.
The stock solution of resazurin was prepared in double distilled
water and stored at 4°C in the dark.
The growth medium contained 6.7 g L−1 yeast nitrogen base
without amino acids, 20 g L−1 glucose, and the appropriate amino
acids, i.e. adenine (20 mg L−1 ), arginine (20 mg L−1 ), aspartic acid
(100 mg L−1 ), glutamic acid (100 mg L−1 ), histidine (20 mg L−1 ),
isoleucine (30 mg L−1 ), leucine (100 mg L−1 ), lysine (30 mg L−1 ),
methionine (20 mg L−1 ), phenylalanine (50 mg L−1 ), serine (400
mg L−1 ), threonine (200 mg L−1 ), tyrosine (30 mg L−1 ) and valine
(150 mg L−1 ), in double-distilled water. The exposure medium was
five-times higher concentrated than the growth medium and additionally supplemented with 8 μL mL−1 CuSO4 -solution (2.5 g L−1
in double distilled water).
The lacZ reaction mixture consisted of 10 mL lacZ-buffer (10.67
g L−1 Na2 HPO4 · 2 H2 O, 0.75 g L−1 KCl, 0.25 g L−1 MgSO4 · 7 H2 0,
5.5 g L−1 NaH2 PO4 · H2 O, and 1 g L−1 sodium dodecyl sulfate)
and 0.5 mg mL−1 of 4-methylumbelliferyl-β -D-galactopyranoside
(MUG, CAS: 6160-78-7, dissolved in dimethyl sulfoxide (DMSO,
CAS: 67-68-5)).

2.3. Landfill leachate extract
Leachate of a mixed deposition site was collected as a grab
sample at a landfill site in Germany prior to leachate water treatment. Until 2005, this landfill was also used for deposition of domestic and bulky waste, from 2006 only commercial and construction waste as well as sewage and industrial sludges were disposed

of there. The sample was collected, filtered and 200-fold enriched
by solid phase extraction as described in detail in Riegraf, et al.
[32]. Sample extract (1 mL) in methanol was stored at – 20°C in
1.5 mL amber glass vials until use.

2.4. Chromatographic separation
HPTLC was performed on 20×10 cm silica gel 60 F254 HPTLC
plates (Merck, Germany). HPTLC plates were pre-washed by chromatographic development with methanol to 5 mm below the rim
in a TLC-developing-chamber (CAMAG) and subsequently activated
at 110°C for 30 min in an oven prior to use [33]. Dilutions of
model compounds or sample extracts were applied as 5 mm band
and 5×3 mm area, respectively, at 8 mm distance from the lower
edge of the pre-washed HPTLC plates using an Automatic TLC Sampler 4 (ATS 4, CAMAG). A two-step chromatographic separation
was performed using an Automated Multiple Development System (AMD 2, CAMAG) with 1.) methanol up to 20 mm and 2a.)
chloroform:EtAc:petroleum fraction (55:20:25, V:V:V, modified after Cimpoiu, et al. [34]) for separating flutamide and BPA or sample
extracts or 2b.) EtAc:n-hexane (50:50, V:V) for the separation of
flutamide and testosterone up to 90 mm [32]. All automated CAMAG devices were operated under the software visionCATS (version 2.5. SP1, CAMAG).

2. Materials and methods
2.1. Chemicals
The model compounds testosterone (CAS: 58-22-0), flutamide
(CAS: 13311-84-7), bisphenol A (BPA, CAS: 80-05-7, ≥99%) and
4-nitroquinoline 1-oxide (4-NQO, CAS: 56-57-5, ≥98%) were ob2


C. Riegraf, A.M. Bell, M. Ohlig et al.

Journal of Chromatography A 1684 (2022) 463582

2.5. HPTLC-based planar yeast (anti-)androgen screen


captured by scanning densitometry using the TLC Scanner 4 (CAMAG) in absorption mode at λ = 575 nm under light emitted by a
deuterium and a halogen-tungsten lamp without applying a filter.

Effects on the androgen receptor were evaluated using the androgenic test strain BJ1991 derived from Saccharomyces cerevisiae
(MATa pep4-3, prbl-1122, ura3-52, leu2, frpl, GAL) [17, 35]. The
preparation of cells for the planar androgen bioassay was described elsewhere [29]. Briefly, cells were cultivated overnight (20
mL growth medium inoculated with 1 mL cryogenic BJ1991 yeast
suspension) at 30°C on a shaker (IKA® KS 30 0 0 i control, orbital
shaking at 200 rpm). Subsequently, cells were pelleted by centrifugation (Hettich® Universal 320R, 2500 g for 10 min) and resuspended in fresh exposure medium. The cell density was adjusted
to 1500 ± 50 FAU for automatic spray application, calibrated according to ISO 7027-1 [36] at a wavelength of 600 nm using a plate
reader (Tecan Infinite® 200 PRO).
The planar yeast androgen screen (p-YAS) was performed as described in Riegraf, et al. [29]. In brief, the adjusted cell suspension
was sprayed on the HPTLC plate using a HPTLC derivatizer (CAMAG) operated in a closed system (application volume: 3 mL, nozzle: yellow, spraying level: 5). For exposure, the HPTLC plates were
placed in a plastic box containing a paper towel soaked with 5 mL
double-distilled water and incubated at 30°C and 90% relative humidity for 20 h (NuAire CO2 -incubator with humidity control, NU5820E). After incubation, HPTLC plates were dried with cold air for
5 min using a fan. Meanwhile, the lacZ reaction mixture was prepared and sprayed on the dried HPTLC plate using a HPTLC derivatizer (CAMAG, application volume: 2.5 mL, nozzle: green, spraying
level: 5). Subsequently, the HPTLC plate was placed in a plastic box
without lid, which in turn was placed in an incubator at 37°C for
15 min to give time to the enzymatic reaction.
For the detection of anti-androgenicity, a planar yeast antiandrogen screen (p-YAAS) was developed by adapting the pYAS procedure described as follows. A testosterone stock solution
in ethanol (2 mg mL−1 ) was diluted 1:200 with 10x exposure
medium. Just before the exposure of the cells to the chromatographically separated sample components on the HPTLC plate, the
adjusted yeast suspension was spiked with testosterone to a final
concentration of 50 ng mL−1 . This yeast suspension was immediately sprayed on the HPTLC plate using a HPTLC derivatizer (CAMAG) operated in a closed system (application volume: 3mL, nozzle: yellow, spraying level: 5). Then, the procedure of the p-YAS as
described above was pursued.
Agonistic and antagonistic potentials of the separated compounds were detected qualitatively under UV light at a wavelength
of λ = 366 nm and exposure times of 550 ms, 20 0 0 ms as well
as automatic settings using the TLC Visualizer 2 (CAMAG). Furthermore, signals were documented using a TLC Scanner 4 (CAMAG) at
λex = 320 nm (deuterium lamp) and a cut-off filter of 400 nm.


2.7. Data processing and statistical analysis
Excel® and R 3.5.2 [37], in particular the ‘drc‘ [38] and the ‘ggplot2‘ [39] packages, were used for data processing and statistical
analysis. Peak areas of signals were extracted from the scan chromatograms and expressed as arbitrary units (AU). Dose-response
curves for model compounds were established by regression analysis. A four-parameter log-logistic function [40] was used to fit
averaged obtained data from up to five replicates to a sigmoidal
curve. This dose-response curve served as basis for the calculation
of ED10 and ED50.
The anti-androgenic effect of the landfill leachate was quantified by the calculation of biological equivalence concentrations
(BEQs) by relating the observed effects caused by the antagonistic
fractions to the dose-response data of the model compounds BPA
and flutamide. The resulting values reflect the amount of BPA and
flutamide producing the same effect as the sample or their fractions of unknown composition considering the enrichment factor,
pre-dilution and application volumes.
3. Results
3.1. Development of p-YAAS on normal phase silica gel HPTLC plates
The YAAS conducted in 96-well plates served as starting point
for the development of the p-YAAS. Similarly to the YAAS, a high
background signal is created induced by an agonist spiked to the
applied yeast suspension. Antagonistic compounds can then be detected based on fluorescence signal suppression. In a first step,
the engineered yeast cells were exposed on a HPTLC plate to various amounts of the androgen receptor (AR) agonist testosterone
to adjust the background level for an optimal detection of antagonistic effects visible as a suppression of the fluorescence signal.
Concentrations of 50, 100, 150, 200 and 250 ng mL−1 testosterone
spiked to the applied yeast suspension were tested. Dose-response
curves of the anti-androgenic model compounds flutamide and BPA
were established by spot application without chromatographic development using the different testosterone concentrations. The detectable anti-androgenic effects decreased with increasing spike
concentration (see Fig. S1) while the intensity of cytotoxic signals
was not affected by the background agonist concentration (see Fig.
S2). A testosterone spike of 50 ng mL−1 led to the lowest ED50values and was thus chosen as the final agonist spike concentration.
After the successful development of the p-YAAS procedure

without chromatographic development (Fig. 1a.)), mixtures of
model compounds were applied and chromatographically separated. The mobile phase composition was adapted to reach a complete separation of the model compounds. The final solvent composition for separating flutamide and BPA consisted of a focusing step using 100% methanol followed by a separation step using chloroform:EtAc:petroleum fraction (55:20:25, V:V:V, modified
after Cimpoiu, et al. [34]).
A signal suppression in a dose responding manner with the
separated model compounds flutamide and BPA were observed
(Fig. S3). The respective dose-response curves are shown in
Fig. 1a.) without and Fig. 1b.) with chromatographic development.
The obtained dose response curves were used to calculate effective
doses (Table 1). Established ED10 values were lower with chromatographic development compared to tests without chromatographic development. The ED10 value for BPA with 35.1 ± 3.5
ng spot−1 was lower compared to the ED10 of flutamide with an

2.6. HPTLC-based planar resazurin cell viability assay for the
detection of yeast cytotoxicity
The planar assay for cell viability can be either performed on
the same HPTLC plate as the p-YAAS after the application of the
lacZ reaction mixture or on a separate HPTLC plate treated identically but without applying lacZ reaction mixture. After drying the
plates with cold air for 5 min, 3 mL of resazurin solution (0.1 g
L−1 in double distilled water) was sprayed on the HPTLC plate using a HPTLC derivatizer (CAMAG, nozzle: green, spraying level: 5).
The HPTLC plate was incubated at 30°C and 90% relative humidity for 30 min. After the incubation, signals visualized as dark rose
spots on a colorless background were captured at white light (direct light, transmitted light and a combination of the both) using
the TLC Visualizer 2 (CAMAG). Subsequently, HPTLC plates were
dried again with cold air for 5 min, followed by a second documentation of the signals at white light. Moreover, signals were
3


C. Riegraf, A.M. Bell, M. Ohlig et al.

Journal of Chromatography A 1684 (2022) 463582

Fig. 1. Planar yeast anti-androgen screen (p-YAAS) a.) without and b.) with chromatographic separation of the model compounds bisphenol A and flutamide. A two-step

chromatographic development was performed using methanol and chloroform:EtAc:petroleum fraction (55:20:25, V:V:V, modified after Cimpoiu, et al. [31]). In the graphs,
mean signal intensities determined by the peak area are plotted against the applied amount of model compound. Error bars show the respective standard deviation with a.)
n=4 and b.) n=3.
Table 1
Effective doses 10% (ED10) and 50% (ED50) of flutamide and bisphenol A in the
planar yeast anti-androgen screen (p-YAAS) with and without chromatographic development. Shown are mean values ± standard deviation. The number of replicates
was nwithout =4 and nwith =3.
Compound

Effective dose

Chromatographic development
without (ng spot−1 )
with (ng spot−1 )

Flutamide

ED10
ED50
ED10
ED50

46.3 ± 4.7
167 ± 13
35.1 ± 3.5
235 ± 18

Bisphenol A

3.3. Distinction between antagonistic and cytotoxic effects by means

of a resazurin cell viability assay
For a correct interpretation of antagonistic effects in reporter
gene assays, it is of high importance to distinguish specific inhibitory effects from a general cytotoxicity that also results in signal suppression. In 96-well plate assays, this is often done by assessing cellular growth, which is not possible on a HPTLC plate.
Therefore, a HPTLC-based assay for cell viability was integrated
into the overall procedure. This assay is based on the characteristic of viable yeast cells to irreversibly reduce the redox dye resazurin (orange at pH < 6.5 and dark violet at pH > 6.5) into
the pink, fluorescent product (resorufin) and further to the colorless dihydroresorufin. NAD(P)H generated by catabolic reactions in
the cytoplasm and the mitochondria of living cells serves as the
reducing agent for this reaction. Thus, the viability of the yeast
cells on the HPTLC plate can be assessed visually after the application of resazurin by a discoloration of the HPTLC plate at positions with metabolically active cells. In contrast, depending on
the pH value and moisture content of the HPTLC plate, cytotoxic
effects become visible as orange, rose, violet or blue spots. This
general principle had been previously adapted to the application
on HPTLC plates to detect the antimicrobial activity of peptides
[41] and plant extracts [42]. Furthermore, resazurin was used to
confirm the results of a newly developed planar cytotox CALUX
bioassay [43].
In line with the current study, several parameters for the onplate resazurin assay were optimized. First, the application of resazurin was improved in terms of application volume, nozzle size
for spraying and concentration. For this purpose, six different resazurin concentrations spanning a range of 0.1 – 5 g L−1 were
applied on a HPTLC plate on which different amounts of the cytotoxic 4-NQO were applied. With increasing resazurin concentration, the non-toxic area turned orange to brownish while the cytotoxic signals stand out as rose or violet spots (Fig. S5). A concentration of 0.1 g L−1 resazurin was chosen as final concentration
for the spraying because this resulted in a complete conversion of
the resazurin to the colorless dihydroresorufin and thus provided
the highest contrast to the background. Second, cytotoxic signals
of different amounts of 4-NQO and BPA were scanned at five different wavelengths (λ = 350 nm, 360 nm, 520 nm, 540 and 575
nm) (Fig. S6). A wavelength of 575 nm resulted in the highest sig-

27.9 ± 1.3
205 ± 11
20.1 ± 5.1
139 ± 29


ED10 of 46.3 ± 4.7 ng spot−1 (Table 1). However, the ED50 values showed a different picture with lower values determined for
flutamide in comparison to BPA without chromatographic development.
3.2. Separation of anti-androgenic and androgenic compounds for
interference reduction
One main advantage of performing the YAAS on HPTLC plates
is the possibility to separate anti-androgenic and androgenic compounds to reduce interferences of agonistic and antagonistic effects
that might result in a mutual masking of these specific effects. A
proof of principle for the analysis of androgenic affects in the presence of an anti-androgen and vice versa was done by the analysis
of a mixture of testosterone and flutamide in a concentration ratio
of 1 to 1500. This mixture was first analyzed by the classic 96-well
based YAS and YAAS. A significant reduction of agonistic and vice
versa antagonistic effects in the mixture was observed in the 96well based assays compared to the application of testosterone (Fig.
S4a.)) and flutamide (Fig. S4b.)) alone.
Aliquots of the same mixture were applied in different volumes
on split HPTLC plates that were directed to the p-YAS and the pYAAS, respectively. In Fig. 2, the results of the p-YAS are shown on
the left side and the results of the p-YAAS are shown on the right
side. By the comparison of signals detected in the mixture and
respective positive controls, i.e. testosterone and flutamide alone,
it becomes evident that a distinction of agonistic and antagonistic effects is possible by the HPTLC-based versions of the YAS and
the YAAS. In case of the p-YAAS slightly increased signal intensities
are visible at the testosterone-specific migration distance when the
mixture is analyzed.
4


C. Riegraf, A.M. Bell, M. Ohlig et al.

Journal of Chromatography A 1684 (2022) 463582

Fig. 2. Comparison of planar yeast androgen screen (p-YAS, left) and planar yeast anti-androgen screen (p-YAAS, right) shown with the model compounds testosterone and

flutamide applied as a mix (3.33 μg L−1 testosterone and 5 mg L−1 flutamide) or individually. A two-step chromatographic development was performed using 1.) methanol
and 2.) ethyl acetate:n-hexane 50:50 (V:V). The image shows the signal detection with fluorescence-imaging at λexcitation = 366 nm without (left) and with (right) testosterone
spike (50 ng mL−1 ).

nal intensity and was thus selected as final measuring wavelength.
Finally, different incubation times after resazurin application were
investigated (t = 5 min, 15 min, 30 min and 60 min) (Fig. S7). An
incubation time of 30 min resulted in the clearest signals and the
best contrast to the background and was therefore chosen as final
incubation time.

The distinction between anti-androgenic and cytotoxic effects
on the same HPTLC plate was tested using this optimized procedure. Therefore, the anti-androgenic model compounds BPA and
flutamide and the cytotoxic 4-NQO were applied in different
amounts on a HPTLC plate. First, the p-YAAS was performed to detect anti-androgenic activity (Fig. 3a.)). After signal detection, re-

Fig. 3. Distinction between cytotoxic and anti-androgenic effects using the planar yeast anti-androgen screen (p-YAAS) in combination with a resazurin cell viability assay
without chromatographic separation. The model compounds bisphenol A (anti-androgenic), flutamide (anti-androgenic) and 4-nitroquinoline 1-oxide (cytotoxic) were applied
on a HPTLC plate in amounts indicated above the images. The two images show the same HPTLC plate: a.) Anti-androgenic effect detection by testosterone spike (50 ng
mL−1 ) with fluorescence-imaging at λexcitation = 366 nm, b.) cytotoxic effect detection with resazurin under white light.
5


C. Riegraf, A.M. Bell, M. Ohlig et al.

Journal of Chromatography A 1684 (2022) 463582

Fig. 4. Analysis of a.) anti-androgenic and b.) cytotoxic effects in different dilutions of a landfill leachate extract using the planar yeast anti-androgen screen (p-YAAS) in
combination with a resazurin cell viability assay. a.) fluorescence-imaging at λexcitation = 366 nm, b.) signal detection under white light. Landfill leachate extracts were applied
in different volumes on a HPTLC plate as indicated at the top of the figure and subsequently separated in a two-step chromatographic development with 1.) methanol and

2.) chloroform:EtAc:petroleum fraction (55:20:25, V:V:V, modified after Cimpoiu, et al. [31]). 4-Nitroquinoline 1-oxide (4-NQO), bisphenol A (BPA) and flutamide served as
positive controls applied on the three rightmost tracks.

sazurin was sprayed on the same HPTLC plate to visualize possible
cytotoxic effects (Fig. 3b.)). Based on the results shown in Fig. 3,
the signal suppression of the p-YAAS caused by 4-NQO can be attributed clearly to a cytotoxic effect whereas the absence of cytotoxic effects indicates the specific antagonistic effects of BPA and
flutamide in the amounts applied to the HPTLC plate.

sazurin cell viability assay performed directly after the p-YAAS
on the same HPTLC plate (Fig. 4b.)). These areas of cytotoxicity
are located in the center of the respective signals detected by
the p-YAAS. After dilution of the sample, anti-androgenic effects
were detected in each sample dilution as two separated signals of
comparable intensity and with dose-dependent variation of intensity (Fig. 4a.)). The upper signals shared the same Rf-value as the
model compound BPA. In contrast, the lower signals could not be
assigned to a candidate compound. The androgenic activity of the
same sample had been investigated by p-YAS in the course of an
earlier study conducted by Riegraf, et al. [32]. In contrast to the
anti-androgenic and cytotoxic effects, androgenicity was not observed in the tested concentrations.
For the quantification of antagonistic effects, the diluted sample
extract was investigated in three different volumes on the HPTLC
plate in parallel to a calibration spanning the amounts of 10 to
500 ng BPA and flutamide (Fig. S8). The equivalent concentrations
were calculated as arithmetic mean of three independent experiments resulting in a total of nine replicates. The upper signal
corresponded to 0.45 ± 0.08 mg BPA-EQ L−1 or 1.26 ± 0.19 mg
flutamide-EQ L−1 in the original leachate. Furthermore, 1.10 ± 0.20

3.4. Analysis of antagonistic and cytotoxic effects in a landfill
leachate extract
Finally, the applicability of the developed method for environmental samples was investigated. Environmental samples often

contain complex matrices composed of several compound classes
which potentially can affect the performance of bioassays. For a
proof of concept, the p-YAAS and the planar cytotoxicity assay
were applied to an extracted landfill leachate. Despite the overload of the stationary phase as evident by insufficiently separated and elongated and/or broadened signals, the application of
the undiluted sample extract led to the detection of three antiandrogenic spots. Two of this three anti-androgenic fractions additionally showed cytotoxic effects as demonstrated with the re6


C. Riegraf, A.M. Bell, M. Ohlig et al.

Journal of Chromatography A 1684 (2022) 463582

mg flutamide-EQ L−1 were calculated for the lower signal of antiandrogenicity. In total, 2.36 ± 0.38 mg flutamide-EQ L−1 were determined in the landfill leachate under the assumption of a quantitative extraction of the compounds.

[31] using reverse-phase HPTLC plates was not designed for quantification purposes, a direct comparison of sensitivities is not possible. Only the higher spike-level of testosterone in the RP-HPTLC
(800 pg testosterone/mm−2 instead of 7.5 pg testosterone/mm−2 )
might indicate a higher sensitivity of the assay based on normal
phase HPTLC. The proposed method allows the detection of antiandrogenicity in the range of suggested effect-based trigger values of 3.3 to 14.4 μg flutamide-EQ L−1 [53]. When compared with
chemical target analyses for flutamide, the ED10 of the method
[52] presented is in the same order of magnitude as the LOD of an
approach using HPLC-UV [54] and even lower than that of an electrochemical sensor recently developed for the trace-level recognition in biofluids [55].
By the application of the p-YAAS (and p-YAS), androgenic and
anti-androgenic compounds can be spatially separated prior to the
application of the test organisms (Fig. 2). This is an advantage for
the investigation of complex environmental samples. For example,
Pannekens, et al. [56] showed that counter-acting substances, i.e.
receptor agonists and antagonists, concurrently occur in wastewaters from municipal and hospital WWTPs resulting in suppressed
biological signals in reporter gene assays. Agonist effects can even
be completely masked by antagonistic compounds as shown by
Weiss, et al. [11] for androgenic effects in river sediments. Sample
fractionation could be done in a higher resolution by HPLC, however, an increased sample throughput would require investments

in lab-automation. Furthermore, used mobile phases for separation in HPLC might interfere with subsequent bioassays if not completely removed by evaporation [57]. The proposed method allows
a simultaneous analysis of ten samples per HPTLC plate without
the need to prepare and test dilution series of the sample, due to
the inherent dilution of the sample on the surface of the HPTLC
plate by diffusion processes. Thus, specific anti-androgenic effects
are detectable even at high sample concentrations leading to cytotoxic effects as shown for the undiluted extract of the landfill
leachate. The developed planar cytotoxicity assay was performed
on the same HPTLC plate subsequent to the p-YAAS reducing cost
and time-need considerably. Cytotoxic effects caused by the extracted landfill leachate were detectable unambiguously as colored
spots (Fig. 4b) - specific anti-androgenic effects are detectable in
parallel in the areas surrounding the cytotoxic center of the signal
(Fig. 4a).
The total concentration of anti-androgenic substances in the
landfill leachate equals to 2.36 ± 0.38 mg flutamide-EQ L−1 and
vastly exceeded the levels of 11.7 to 56.4 μg flutamide-EQ L−1
found in municipal and hospital wastewaters [56]. Furthermore, 6
to 32 μg flutamide-EQ L−1 were reported in the context of investigations on effluents of 12 WWTPs in Danube river basin [58] and
levels of up to 90 μg flutamide-EQ L−1 were detected during a
two-year survey at three Dutch surface waters [59]. Thus, the antiandrogenic effect in the landfill leachate was around 25 to 400
times higher than that previously reported in waste- and river waters. Escher, et al. [53] suggested an effect-based trigger value for
anti-androgenicity in the range of 3.3 to 14.4 μg flutamide-EQ L−1 .
This threshold was exceeded by a factor of about 160 to 700 by
the landfill leachate. However, the investigated leachate is not a direct threat to the environment since it is treated onsite before it is
discharged into a municipal WWTP for further treatment. Though
a leakage of the draining system might lead to a contamination
of surrounding soils and water bodies with anti-androgenic compounds.
The calculated BPA-EQ of 0.45 ± 0.08 mg L−1 for the upper antagonistic signal with the same Rf-value as the model compound
BPA is in the same range as the previously reporter BPA concentration of 2.9 mg L−1 determined by GC-MS/MS [32]. This study of
Riegraf, et al. additionally revealed the presence of nonylphenols
and 4-tert-octylphenol in the respective landfill leachate. Although


4. Discussion
In the presented study, a procedure based on the direct coupling of normal phase HPTLC with the yeast anti-androgen screen
for the detection of anti-androgenic effects was successfully established using the model compounds flutamide and BPA. The detectability of specific antagonistic effects on the androgen-receptor
in contrast to general inhibitory effects such as cytotoxicity is
clearly underlined by the observation that the antagonistic effects mediated by flutamide and BPA can be masked by increasing
spike-levels of testosterone (Fig. S1). In contrast, the signal suppression of the fluorescent background by the cytotoxic 4-NQO was
not affected by the spike-level of the receptor agonist (Fig. S2) indicating that the inhibitory effect is not mediated by the androgen receptor but by its toxicity to the yeast cells. In general, all effects of compounds reducing the activity of the androgen receptor,
e.g., via a competition for the ligand binding domain or allosteric
regulation would be detectable by the proposed assay. Itzel, et al.
[44] provides an overview of 89 compounds whose anti-androgenic
action was verified in different bioassays. However, as for all cellbased in vitro-assays the definite outcome depends on the cellular
context and modes of action. For example, interferences with the
binding of the androgen receptor to individual, cell-specific transcriptional cofactors might escape the detection by the proposed
assay [45, 46].
The relative potency of BPA to flutamide was found to be
0.68 without and 1.47 with chromatographic development. Similar results were reported by Rostkowski, et al. [10] who determined a relative potency for BPA of 0.60 by a YAAS in microtiter
plate format. These findings also correlate well with that of Fang,
et al. [47] who detected comparable binding affinities for flutamide and BPA to the androgen receptor. The shift in the relative potency caused by the chromatographic development in the
proposed method might be explained by a different diffusion of
BPA compared to flutamide during the development of the HPTLC
plate which lead to small shifts in the dose response relationship.
The resulting changes of relative potencies have to be considered,
e.g., for the calculation of effect contributions of compounds in a
mixture-based on data of chemical analysis.
Due to the possibility to apply high sample volumes to the
HPTLC plate, the presented method has a higher effective sensitivity compared to its equivalent in the 96-well format. Assuming a
common 10 0 0-fold enrichment of environmental samples, the determined ED10 of 27.9 ng flutamide per spot translates to a LOD
of 5.58 μg L−1 (0.02 μM) in case of an application volume of 5 μL
and even only 0.28 μg L−1 (0.001 μM) after the application of 100

μL sample extract. In comparison, the inhibitory concentration associated with 10% and 50% effect (IC10 and IC50) of flutamide using the same yeast strain in the classic 96-well plate approach was
1.53 ± 0.19 mg L−1 (5.56 μM) and 4.29 ± 0.32 mg L−1 (15.54 μM),
respectively (Fig. S4). These results are in line with the sensitivity
of other recombinant yeast strains expressing the human androgen receptor, which e. g. obtained an IC50 of 6.14 μM or 20.3 μM
of flutamide in microtiter plates [48, 49]. Assays based on mammalian cell lines are considered to detect androgen receptor mediated effects more sensitive than yeast-based assays [50]. For example, Hu, et al. [51] recently reported an IC50 of 2.3 μM flutamide
using an assay based on MDA-kb2 cells. In the context of an international ring trial, the IC50 detected by the anti-AR-CALUX method
actually ranged between 0.11 and 1.1 μM flutamide [52]. Since the
experimental set-up of the study published by Klingelhöfer, et al.
7


C. Riegraf, A.M. Bell, M. Ohlig et al.

Journal of Chromatography A 1684 (2022) 463582

both substances showed anti-androgenic activities in prior reporter
gene assays [10, 60], the observed antagonistic effects cannot be
assigned to nonylphenols and 4-tert-octylphenol as their retardation factor did not correspond to the main signals detected in the
sample. However, the third anti-androgenic signal only detectable
in the undiluted sample extract could be caused by nonylphenols
and 4-tert-octylphenol as they show a similar migration distance
under identical separation conditions [32]. In contrast to BPA, these
two compounds have not been quantified by chemical analysis, so
that their expected anti-androgenic effect cannot be estimated. As
the complete explanation of anti-androgenic effects in the landfill
leachate was out of the scope of our study, the identification of
causing compounds was not pursued any further. The extraction
of the stationary phase at relevant positions and the subsequent
analysis by mass-spectrometry could support the identification of
bioactive compounds in terms of an effect-directed analysis as proposed by Weiss, et al. [61].


Acknowledgement
This work was supported by the German Federal Ministry for
the Environment, Nature Conservation, Nuclear Safety and Consumer Protection in line of the BMUV-project ‘General and specific
ecotoxicology’. The authors thank Ramona Pfänder for the excellent technical assistance with the yeast assays in microtiter plate
format.
Supplementary materials
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.chroma.2022.463582.
References
[1] A. Pal, K.Y.-H. Gin, A.Y.-C. Lin, M. Reinhard, Impacts of emerging organic contaminants on freshwater resources: review of recent occurrences, sources, fate
and effects, Sci. Total Environ. 408 (2010) 6062–6069, doi:10.1016/j.scitotenv.
2010.09.026.
[2] D.J. Lapworth, N. Baran, M.E. Stuart, R.S. Ward, Emerging organic contaminants
in groundwater: a review of sources, fate and occurrence, Environ. Pollut. 163
(2012) 287–303, doi:10.1016/j.envpol.2011.12.034.
[3] S. Jobling, N. Beresford, M. Nolan, T. Rodgers-Gray, G.C. Brighty, J.P. Sumpter,
C.R. Tyler, Altered sexual maturation and gamete production in wild roach
(Rutilus rutilus) living in rivers that receive treated sewage effluents1, Biol. Reprod. 66 (2002) 272–281, doi:10.1095/biolreprod66.2.272.
[4] K.A. Kidd, M.J. Paterson, M.D. Rennie, C.L. Podemski, D.L. Findlay, P.J. Blanchfield, K. Liber, Direct and indirect responses of a freshwater food web to
a potent synthetic oestrogen, Philos. Trans. R. Soc. B 369 (2014) 20130578,
doi:10.1098/rstb.2013.0578.
[5] R.J. Ellis, M.R. van den Heuvel, E. Bandelj, M.A. Smith, L.H. McCarthy,
T.R. Stuthridge, D.R. Dietrich, In vivo and in vitro assessment of the androgenic
potential of a pulp and paper mill effluent, Environ. Toxicol. Chem. 22 (2003)
1448–1456, doi:10.1002/etc.5620220705.
[6] E.P. Kolodziej, D.L. Sedlak, Rangeland grazing as a source of steroid hormones
to surface waters, Environ. Sci. Technol. 41 (2007) 3514–3520, doi:10.1021/
es063050y.
[7] J. Durhan Elizabeth, S. Lambright Christy, A. Makynen Elizabeth, J. Lazorchak,

C. Hartig Phillip, S. Wilson Vickie, L.E. Gray, T. Ankley Gerald, Identification of
metabolites of Trenbolone acetate in androgenic runoff from a beef feedlot,
Environ. Health Perspect. 114 (2006) 65–68, doi:10.1289/ehp.8055.
[8] V. Kumar, C. Majumdar, P. Roy, Effects of endocrine disrupting chemicals from
leather industry effluents on male reproductive system, J. Steroid Biochem.
Mol. Biol. 111 (2008) 208–216, doi:10.1016/j.jsbmb.20 08.06.0 05.
[9] E.M. Hill, K.L. Evans, J. Horwood, P. Rostkowski, F.O. Oladapo, R. Gibson, J.A. Shears, C.R. Tyler, Profiles and some initial identifications of
(Anti)Androgenic compounds in fish exposed to wastewater treatment works
effluents, Environ. Sci. Technol. 44 (2010) 1137–1143, doi:10.1021/es901837n.
[10] P. Rostkowski, J. Horwood, J.A. Shears, A. Lange, F.O. Oladapo, H.T. Besselink,
C.R. Tyler, E.M. Hill, Bioassay-directed identification of novel antiandrogenic
compounds in bile of fish exposed to wastewater effluents, Environ. Sci. Technol. 45 (2011) 10660–10667, doi:10.1021/es202966c.
[11] J.M. Weiss, T. Hamers, K.V. Thomas, S. van der Linden, P.E.G. Leonards,
M.H. Lamoree, Masking effect of anti-androgens on androgenic activity in European river sediment unveiled by effect-directed analysis, Anal. Bioanal.Chem.
394 (2009) 1385–1397, doi:10.1007/s00216- 009- 2807- 8.
[12] S. Milla, S. Depiereux, P. Kestemont, The effects of estrogenic and androgenic
endocrine disruptors on the immune system of fish: a review, Ecotoxicol. 20
(2011) 305–319, doi:10.1007/s10646-010- 0588- 7.
[13] Y. Kiparissis, T.L. Metcalfe, G.C. Balch, C.D. Metcalfe, Effects of the antiandrogens, vinclozolin and cyproterone acetate on gonadal development in the
Japanese medaka (Oryzias latipes), Aquat. Toxicol. 63 (2003) 391–403, doi:10.
1016/S0166-445X(02)00189-3.
[14] L. Gehrmann, H. Bielak, M. Behr, F. Itzel, S. Lyko, A. Simon, G. Kunze, E. Dopp,
M. Wagner, J. Tuerk, Anti-)estrogenic and (anti-)androgenic effects in wastewater during advanced treatment: comparison of three in vitro bioassays, Environ. Sci. Pollut. Res. Int. 25 (2018) 4094–4104, doi:10.1007/s11356- 016- 7165- 4.
[15] T.J. Runnalls, L. Margiotta-Casaluci, S. Kugathas, J.P. Sumpter, Pharmaceuticals in the aquatic environment: steroids and anti-steroids as high
priorities for research, Hum. Ecol. Risk Assess., 16 (2010) 1318-1338.
/>[16] J.M. Weiss, E. Simon, G.J. Stroomberg, R. de Boer, J. de Boer, S.C. van der Linden, P.E.G. Leonards, M.H. Lamoree, Identification strategy for unknown pollutants using high-resolution mass spectrometry: androgen-disrupting compounds identified through effect-directed analysis, Anal. Bioanal.Chem. 400
(2011) 3141–3149, doi:10.10 07/s0 0216- 011- 4939- x.
[17] P. Sohoni, J. Sumpter, Several environmental oestrogens are also antiandrogens, J. Endocrinol. 158 (1998) 327, doi:10.1677/joe.0.1580327.
[18] K.V. Thomas, K. Langford, K. Petersen, A.J. Smith, K.E. Tollefsen, Effect-directed
identification of naphthenic acids as important in Vitro Xeno-estrogens and


5. Conclusion
Coupling HPTLC with bioassays to detect anti-androgenic activity and cytotoxicity in parallel allows a matrix-robust, costefficient, fast and sensitive elucidation of effects and a reduction of
interferences from agonists of the androgen receptor as well as cytotoxic effects that might lead to false positive test results. The pYAAS allows the detection of a group of toxic substances with high
environmental relevance and is thus a valuable addition to the existing methods that combine HPTLC and various specific effectbased bioassays. The proposed method helps to reduce the number of dilution steps, e.g., needed in liquid culture assays, while at
the same time increasing the possibility to detect very low concentrations of compounds of interest offering a wide range of possible
applications in environmental monitoring. So far, only a limited set
of model compounds and sample extracts was investigated in the
present study. The routine investigation of anti-androgenic effects
using the proposed method can improve the correlation between
results of chemical analyses and the overall anti-androgenic activity, because an underestimation of effects due to a mutual masking
of agonistic and antagonistic compounds but as well false positive
test results can be avoided.
Data availability statement
The data that support the findings of this study are available
from the corresponding author upon reasonable request.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.
CRediT authorship contribution statement
Carolin Riegraf: Conceptualization, Formal analysis, Validation,
Visualization, Writing – original draft. Anna Maria Bell: Formal
analysis, Validation, Visualization, Writing – original draft. Marina
Ohlig: Data curation, Investigation, Writing – review & editing. Georg Reifferscheid: Writing – review & editing. Sebastian
Buchinger: Conceptualization, Supervision, Writing – review &
editing.
Data availability
Data will be made available on request.
8



C. Riegraf, A.M. Bell, M. Ohlig et al.

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]
[27]

[28]

[29]

[30]

[31]

[32]


[33]

[34]

[35]

[36]
[37]

[38]
[39]
[40]

Journal of Chromatography A 1684 (2022) 463582

anti-androgens in north sea offshore produced water discharges, Environ. Sci.
Technol. 43 (2009) 8066–8071, doi:10.1021/es9014212.
M. Muschket, C. Di Paolo, A.J. Tindall, G. Touak, A. Phan, M. Krauss, K. Kirchner, T.-B. Seiler, H. Hollert, W. Brack, Identification of unknown antiandrogenic compounds in surface waters by effect-directed analysis (EDA) using
a parallel fractionation approach, Environ. Sci. Technol. 52 (2018) 288–297,
doi:10.1021/acs.est.7b04994.
M.L. Eldridge, J. Sanseverino, A.C. Layton, J.P. Easter, T.W. Schultz, G.S. Sayler,
Saccharomyces cerevisiae BLYAS, a new Bioluminescent bioreporter for detection of androgenic compounds, Appl. Environ. Microbiol. 73 (2007) 6012–6018,
doi:10.1128/aem.00589-07.
B. van der Burg, R. Winter, H.-y. Man, C. Vangenechten, P. Berckmans,
M. Weimer, H. Witters, S. van der Linden, Optimization and prevalidation of
the in vitro AR CALUX method to test androgenic and antiandrogenic activity
of compounds, Reprod. Toxicol. 30 (2010) 18–24, doi:10.1016/j.reprotox.2010.
04.012.
V.S. Wilson, K. Bobseine, C.R. Lambright, L.E. Gray Jr, A novel cell line, MDAkb2, that stably expresses an androgen- and glucocorticoid-responsive reporter
for the detection of hormone receptor agonists and antagonists, Toxicol. Sci. 66

(2002) 69–81, doi:10.1093/toxsci/66.1.69.
C.J. Houtman, R. ten Broek, A. Brouwer, Steroid hormonal bioactivities, culprit natural and synthetic hormones and other emerging contaminants in
waste water measured using bioassays and UPLC-tQ-MS, Sci. Total Environ. 630
(2018) 1492–1501, doi:10.1016/j.scitotenv.2018.02.273.
A. Abbas, I. Schneider, A. Bollmann, J. Funke, J. Oehlmann, C. Prasse, U. SchulteOehlmann, W. Seitz, T. Ternes, M. Weber, H. Wesely, M. Wagner, What you
extract is what you see: Optimising the preparation of water and wastewater samples for in vitro bioassays, Water Res. 152 (2019) 47–60, doi:10.1016/j.
watres.2018.12.049.
M.A.K. Hashmi, M. Krauss, B.I. Escher, I. Teodorovic, W. Brack, Effect-directed
analysis of progestogens and glucocorticoids at trace concentrations in river
water, Environ. Toxicol. Chem. 39 (2020) 189–199, doi:10.1002/etc.4609.
G. Morlock, W. Schwack, Hyphenations in planar chromatography, J. Chromatogr. A 1217 (2010) 6600–6609, doi:10.1016/j.chroma.2010.04.058.
S. Buchinger, D. Spira, K. Bröder, M. Schlüsener, T. Ternes, G. Reifferscheid, Direct coupling of thin-layer chromatography with a bioassay for the detection
of estrogenic compounds: applications for effect-directed analysis, Anal. Chem.
85 (2013) 7248–7256, doi:10.1021/ac4010925.
A.J. Bergmann, E. Simon, A. Schifferli, A. Schönborn, E.L.M. Vermeirssen, Estrogenic activity of food contact materials—evaluation of 20 chemicals using
a yeast estrogen screen on HPTLC or 96-well plates, Anal. Bioanal.Chem. 412
(2020) 4527–4536, doi:10.10 07/s0 0216- 020- 02701- w.
C. Riegraf, G. Reifferscheid, S. Belkin, L. Moscovici, D. Shakibai, H. Hollert,
S. Buchinger, Combination of yeast-based in vitro screens with highperformance thin-layer chromatography as a novel tool for the detection of
hormonal and dioxin-like compounds, Anal. Chim. Acta 1081 (2019) 218–230,
doi:10.1016/j.aca.2019.07.018.
C. Riegraf, G. Reifferscheid, B. Becker, S. Belkin, H. Hollert, U. Feiler,
S. Buchinger, Detection and quantification of photosystem II inhibitors using the freshwater Alga Desmodesmus subspicatus in combination with
high-performance thin-layer chromatography, Environ. Sci. Technol. 53 (2019)
13458–13467, doi:10.1021/acs.est.9b04634.
I. Klingelhöfer, N. Hockamp, G.E. Morlock, Non-targeted detection and differentiation of agonists versus antagonists, directly in bioprofiles of everyday products, Anal. Chim. Acta 1125 (2020) 288–298, doi:10.1016/j.aca.2020.05.057.
C. Riegraf, G. Reifferscheid, L. Moscovici, D. Shakibai, H. Hollert, S. Belkin,
S. Buchinger, Coupling high-performance thin-layer chromatography with a
battery of cell-based assays reveals bioactive components in wastewater and
landfill leachates, Ecotoxicol. Environ. Saf. 214 (2021) 112092, doi:10.1016/j.

ecoenv.2021.112092.
A. Schoenborn, P. Schmid, S. Bräm, G. Reifferscheid, M. Ohlig, S. Buchinger, Unprecedented sensitivity of the planar yeast estrogen screen by using a spray-on
technology, J. Chromatogr. A 1530 (2017) 185–191, doi:10.1016/j.chroma.2017.
11.009.
C. Cimpoiu, A. Hosu, S. Hodisan, Analysis of some steroids by thin-layer chromatography using optimum mobile phases, J. Pharm. Biomed. Anal. 41 (2006)
633–637, doi:10.1016/j.jpba.20 05.12.0 04.
I.J. Purvis, D. Chotai, C.W. Dykes, D.B. Lubahn, F.S. French, E.M. Wilson,
A.N. Hobden, An androgen-inducible expression system for Saccharomyces
cerevisiae, Gene 106 (1991) 35–42, doi:10.1016/0378-1119(91)90563-Q.
ISO 7027-1, International organization for standardization. Water quality — Determination of turbidity — Part 1: quantitative methods., in, 2016.
R Development Core TeamR: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria, 2008 http:
//www.R-project.org.
C. Ritz, F. Baty, J.C. Streibig, D. Gerhard, Dose-response analysis using R, PLoS
One 10 (2016) e0146021, doi:10.1371/journal.pone.0146021.
H. Wickham, ggplot2: Elegant Graphics for Data Analysis, Springer, New York,
2016, doi:10.1007/978- 3- 319- 24277- 4.
C. Ritz, Toward a unified approach to dose–response modeling in ecotoxicology, Environ. Toxicol. Chem. 29 (2010) 220–229, doi:10.1002/etc.7.

[41] M. Jaskiewicz, M. Orlowska, G. Olizarowicz, D. Migon, D. Grzywacz, W. Kamysz,
Rapid screening of antimicrobial synthetic peptides, Int. J. Pept. Res. Ther. 22
(2016) 155–161, doi:10.1007/s10989-015-9494-4.
[42] H. Dehghan, P. Rezaee, A. Aliahmadi, Bioassay screening of 12 Iranian plants
and detection of antibacterial compounds from Heracleum persicum using a
TLC bioautography method, J. Liq. Chromatogr. Relat. Technol. 43 (2020) 381–
387, doi:10.1080/10826076.2020.1725557.
[43] I. Klingelhofer, L.P. Ngoc, B. van der Burg, G.E. Morlock, A bioimaging system
combining human cultured reporter cells and planar chromatography to identify novel bioactive molecules, Anal. Chim. Acta 1183 (2021) 11, doi:10.1016/j.
aca.2021.338956.
[44] F. Itzel, L. Gehrmann, T. Teutenberg, T.C. Schmidt, J. Tuerk, Recent developments and concepts of effect-based methods for the detection of endocrine
activity and the importance of antagonistic effects, TrAC, Trends Anal. Chem.

118 (2019) 699–708, doi:10.1016/j.trac.2019.06.030.
[45] Y. Hikichi, M. Yamaoka, M. Kusaka, T. Hara, Selective androgen receptor modulator activity of a steroidal antiandrogen TSAA-291 and its cofactor recruitment
profile, Eur. J. Pharmacol. 765 (2015) 322–331, doi:10.1016/j.ejphar.2015.08.052.
[46] H. Dotzlaw, U. Moehren, S. Mink, A.C.B. Cato, J.A.I. Lluhi, A. Baniahmad, The
amino terminus of the human AR is target for corepressor action and antihormone agonism, Mol. Endocrinol. 16 (2002) 661–673, doi:10.1210/me.16.4.661.
[47] H. Fang, W.D. Tong, W.S. Branham, C.L. Moland, S.L. Dial, H.X. Hong, Q. Xie,
R. Perkins, W. Owens, D.M. Sheehan, Study of 202 natural, synthetic, and environmental chemicals for binding to the androgen receptor, Chem. Res. Toxicol.
16 (2003) 1338–1358, doi:10.1021/tx030011g.
[48] J. Stragierowicz, S. Stypula-Trebas, L. Radko, A. Posyniak, M. Nasiadek, M. Klimczak, A. Kilanowicz, An assessment of the estrogenic and androgenic properties
of tetra- and hexachloronaphthalene by YES/YAS in vitro assays, Chemosphere
263 (2021) 10, doi:10.1016/j.chemosphere.2020.128006.
[49] D.H. Kim, C.G. Park, S.H. Kim, Y.J. Kim, The Effects of Mono-(2-Ethylhexyl)
Phthalate (MEHP) on human estrogen receptor (hER) and androgen receptor (hAR) by YES/YAS In Vitro assay, Molecules 24 (2019) 10, doi:10.3390/
molecules24081558.
[50] G. Bagchi Bhattacharjee, S.M. Paul Khurana, In vitro reporter assays for screening of chemicals that disrupt androgen signaling, J. Toxicol. 2014 (2014)
701752, doi:10.1155/2014/701752.
[51] X.X. Hu, W. Shi, S. Wei, X.W. Zhang, H.X. Yu, Identification of (anti-)androgenic
activities and risks of sludges from industrial and domestic wastewater treatment plants, Environ. Pollut. 268 (2021) 9, doi:10.1016/j.envpol.2020.115716.
[52] A. Milcamps, R. Liska, I. Langezaal, W. Casey, M. Dent, J. Odum, Reliability of the AR-CALUX (R) In Vitro method used to detect chemicals with
(Anti)Androgen activity: results of an international ring trial, Toxicol. Sci. 184
(2021) 170–182, doi:10.1093/toxsci/kfab078.
[53] B.I. Escher, S. Aїt-Aїssa, P.A. Behnisch, W. Brack, F. Brion, A. Brouwer,
S. Buchinger, S.E. Crawford, D. Du Pasquier, T.J.S.o.t.T.E. Hamers, Effectbased trigger values for in vitro and in vivo bioassays performed on
surface water extracts supporting the environmental quality standards
(EQS) of the European water framework directive, 628 (2018) 748-765.
/>[54] H. Bhatia, A. Kumar, Y. Ogino, J. Du, A. Gregg, J. Chapman, M.J. McLaughlin,
T. Iguchi, Effects of the commercial antiandrogen flutamide on the biomarkers
of reproduction in male Murray rainbowfish (Melanotaenia fluviatilis), Environ.
Toxicol. Chem. 33 (2014) 1098–1107, doi:10.1002/etc.2524.
[55] L. Durai, A. Gopalakrishnan, S. Badhulika, A low-cost and facile electrochemical

sensor for the trace-level recognition of flutamide in biofluids using large-area
bimetallic NiCo2O4 micro flowers, New J. Chem. 46 (2022) 3383–3391, doi:10.
1039/d1nj05246b.
[56] H. Pannekens, A. Gottschlich, H. Hollert, E. Dopp, Evaluation of mixture effects
of endocrine active substances in wastewater using CALUX reporter-gene assays, Int. J. Hyg. Environ. Health 222 (2019) 670–677, doi:10.1016/j.ijheh.2019.
04.008.
[57] M.G. Weller, A unifying review of bioassay-guided fractionation, effect-directed
analysis and related techniques, Sensors 12 (2012) 9181–9209, doi:10.3390/
s120709181.
[58] N.A. Alygizakis, H. Besselink, G.K. Paulus, P. Oswald, L.M. Hornstra, M. Oswaldova, G. Medema, N.S. Thomaidis, P.A. Behnisch, J. Slobodnik, Characterization
of wastewater effluents in the Danube River Basin with chemical screening, in
vitro bioassays and antibiotic resistant genes analysis, Environ. Int. 127 (2019)
420–429, doi:10.1016/j.envint.2019.03.060.
[59] C.J. Houtman, K. Brewster, R. ten Broek, B. Duijve, Y. van Oorschot, M. Rosielle,
M.H. Lamoree, R. Steen, Characterisation of (anti-)progestogenic and (anti)androgenic activities in surface and wastewater using high resolution effectdirected analysis, Environ. Int. 153 (2021) 13, doi:10.1016/j.envint.2021.106536.
[60] J. Li, M. Ma, Z.J. Wang, In vitro profiling of endocrine disrupting effects of phenols, Toxicol. Vitro 24 (2010) 201–207, doi:10.1016/j.tiv.20 09.09.0 08.
[61] S.C. Weiss, N. Egetenmeyer, W. Schulz, Coupling of in vitro bioassays
with planar chromatography in effect-directed analysis, in: G. Reifferscheid,
S. Buchinger (Eds.), Vitro Environmental Toxicology - Concepts, Application and
Assessment, Springer International Publishing Ag, Cham, 2017, pp. 187–224,
doi:10.1007/10_2016_16.

9