The test substances cyproconazole [CAS # 94361-06-5; Syngenta; Batch # CHF1E00042; purity 96.8 %] and epoxiconazole [Cas # 133855-98-8; BASF; Batch # 8563; purity 97.0 %] were obtained directly from the producing companies in technical quality (the same quality the substances are used in PPPs). The test substances fluxapyroxad [CAS # 907204-31-3; Batch # SZBF160XV; purity 99.9 %] and azoxystrobin [CAS # 131860-33-8; Batch # BCBT1118V; purity ≥ 98 %] were obtained from Sigma-Aldrich (Taufkirchen, Germany). The PPPs Adexar^® (registration no. 006958-00; BASF; Batch # 0013044538; 62.5 g/l epoxiconazole; 62.5 g/l fluxapyroxad) and Priori Xtra® (registration no. 005481-00; Syngenta; Batch # 139731; 200 g/l azoxystrobin; 80 g/l cyproconazole) were obtained commercially in Germany. The co-formulants 2-hydroxy-N,N-dimethylpropanamide (Lot# CDS021721), dipropylene glycol (Lot# BCBS8922V, CAS: 25265-71-8), propylene glycol (Lot# LRAB0286, CAS: 57-55-6), Eumulgin® B25 (Lot# BCBV7832, CAS: 68439-49-6), nanoclay, hydrophilic bentonite (Lot# MKCF4495, CAS: 1302-78-9), xanthan from xanthomonas campestris (Lot# BCBV5654, CAS: 1302-78-9), polymethacrylate, N,N-dimethyldecanamide (Lot# CDS001534) were obtained from Sigma-Aldrich (Taufkirchen, Germany). 2-ethylhexyl S lactate (Batch# 1702000034) was obtained from CorbionPurac (Amsterdam, Netherlands).

All substances or products were diluted in DMSO, resulting in a final DMSO medium concentration of 0.4 % in all exposed wells. All other chemicals were purchased from Merck (Darmstadt, Germany) or Sigma-Aldrich (Taufkirchen, Germany) in the highest available purity.

The HepaRG cells were obtained from Biopredic International (Saint Grégoire, France) and were grown for two weeks in Williams medium (Pan-Biotech GmbH, Aidenbach, Germany) containing 10 % fetal calf serum (FCS) (PAN-Biotech GmbH, Aidenbach, Germany), 100 U/ml penicillin, 100 µg/ml streptomycin, 0.05 % human insulin (PAA Laboratories GmbH, Pasching, Austria) and 50 µM hydrocortisone-hemisuccinate (Sigma-Aldrich, Taufkirchen, Germany). For the differentiation, cells were incubated for a further two weeks in differentiation medium. In addition to the already mentioned components, the differentiation medium contained 1.7 % DMSO. HepaRG cells were treated with phenol-red-free Williams medium (Pan-Biotech GmbH, Aidenbach, Germany) which contains all ingredients as the proliferation medium, but only 2 % FCS. Cells were incubated at 37° C in a 5 % CO₂ / 5 % humidity atmosphere in a Binder cell culture incubator.

NRU (neutral red assay) was used for cytotoxicity analysis of both PPPs and for the combination of the active substances according to the protocol by Repetto et al. (2008[30]). Before the neutral red assay was performed, HepaRG cells were seeded in 96-well plates (9x10³ cells per well). The cells were cultured for 4 weeks including 2 weeks of proliferation followed by 2 weeks of differentiation. Cells were then treated with the PPPs, the active substances' mix, and several individual co-formulants for each PPP (for each, eight different concentrations in culture medium with a final solvent concentration of 0.4 % DMSO) for 24 h. The detergent Triton X-100 (0.01 %) served as positive control. After 24 h, treatment medium was removed and cells were washed once with 100 μl PBS per well. Then, 100 μl neutral red medium was added per well and cells were incubated for 2 h at 37 °C. Afterwards, neutral red medium was removed and 150 μl of an acidic ethanol solution was added per well for extraction to remove the basic dye. Uptake of neutral red by endocytosis can be quantified by fluorescence measurement (Excision 530 nm, Emission 645 nm) using an Infinite M200pro plate reader (Tecan, Maennedorf, Switzerland).

HepaRG cells were seeded in 6 well plates and allowed to proliferate and differentiate for 4 weeks. Total RNA was isolated from cells treated for 24 h with the two PPPs, their active substance mix and some of their co-formulants whilst ensuring a final concentration of 0.4 % DMSO in each well. The isolation was carried out using peqGOLD TriFast™ (peqlab, Erlangen, Germany) reagent according to the manufacturer's instructions. Quantity and quality of the RNA samples were measured using a Nanodrop spectrophotometer (NanoDrop 2000; Thermo Fisher Scientific, Waltham, MA, USA). For a ratio absorbance at 260/280 and 260/230 nm, values were accepted within the range of 1.8-2.2 and 1.6-2.2 nm respectively. As recommended by the supplier, all samples were purified prior to the RT2 Profiler PCR Array PAHS-3401Z procedures following the supplier's manual using the RNeasy® MiniElute Cleanup kit. RNA samples were then transcribed to cDNA according to the supplier's manual using the RT2 First Strand kit. Gene expression analysis was subsequently performed with RT2 Profiler PCR Array PAHS-3401Z kit in combination with RT² SYBR® Green qPCR mastermix (Qiagen®; Hilden, Germany) on an ABI 7900HT instrument (Applied Biosystems, Darmstadt, Germany). The cycling condition for PCR reaction was programmed according to the table provided by the supplier's manual. A table of CT values was created in an excel sheet and then uploaded to the data analysis web portal (http://www.qiagen.com/geneglobe). CT values were normalized based on a manual selection procedure. The data analysis web selected and displayed the standard HKG genes by default. Afterwards, normalization was done if the selected genes had only a small change in their expression across different sample groups. The selected HKG were ACTB, B2M, GAPDH, HPRT1, and RLP0. The data analysis web portal tool calculated fold change/regulation using the CT method, where delta CT was calculated between gene of interest (GOI) and an average of reference genes (HKG), followed by delta-delta CT calculations (delta CT (test group)-delta CT (control group)). Fold Change was then calculated using 2^ (-delta delta CT) formula. The data was downloaded as Excel sheets and PDF files. To decrease false-positives due to random variability, p-values were adjusted by FDR-method (Benjamini and Hochberg, 1995[4]). In addition, gene ontology (GO) term enrichment analysis based on R software was applied to identify the pathways in which differentially expressed genes (DEG) significantly enriched.

HepaRG cells were cultured, allowing for their proliferation and differentiation for 4 consecutive weeks in 6-well plates (2x10⁵ cells per well). Cells were subsequently treated (with the formulations, their active substances in combination and some of their co-formulants) and incubated for 24 h at 37 °C. Cells were then lysed following the protocol provided by Signatope. The quantification of the Cyp enzymes and ABC transporters was carried out by Signatope with the use of quantitative mass spectrometry-based immunoassays. There, the amount of protein was determined, and the tryptic digestion of the lysates was carried out, followed by the addition of the isotope labeled signature peptides from the sequences of CYP450 enzymes and transporters. In a further step, these synthetic peptides and the endogenous peptides (originating from the sample) were enriched and precipitated with the aid of peptide group-specific antibodies. These peptides were eluted from the antibodies after precipitation and detected by mass spectrometry (LC-MS-MS QExactive, Thermo). The proteins were quantified indirectly by referencing the signal from the endogenous peptide to the signal from the synthetic peptide.

The extraction of the products and their active substances from HepaRG cells, PBS and the medium was done using the Quick, Easy, Cheap, Effective, Rugged and Safe (QuEChERS) technique with a few deviations from the original proposal by Anastassiades et al. (2003[1]). Sample preparation was as follows: Cell samples were dispersed in 10 ml of acetonitrile and Milli-Q water mix (4:1 ratio respectively) and placed in an ultrasonic bath for 15 min. Medium and PBS samples were directly sonicated for 15 min. For each sample 1 ml of the mixture was added to a 50 ml centrifuge tubes to which 4 ml of Milli-Q water was added; 10 ml of precooled acetonitrile (ACN) was added, and samples were vigorously shaken by a vortex mixer for 10 min. One Supel QuE Citrate Extraction Tube 55227-U (Sigma-Aldrich, Taufkirchen, Germany) was added to the samples and the centrifuge tubes were shaken immediately for one minute by hand. Centrifuge tubes were vigorously shaken by a vortex mixer for another 10 min. Centrifuge tubes were subject to centrifugation for 5 min at 3000 g and 4 °C. 7 ml of the supernatant was transferred to 15 ml centrifuge tubes from which the ACN was evaporated with a nitrogen stream EVA 1 Vis (VLM GmbH, Bielefeld, Germany). Residues were redissolved in 0.20 ml of the mobile phase (start conditions), mixed for 5 min and sonicated for 5 min, and then transferred to a vial for chromatographic analysis.

Active substances were chromatographically separated on an Agilent 1260 series liquid chromatography system (Agilent Technologies, Heilbronn, Germany) consisting of a binary pump system (G1312B), degasser (G4225A), column oven (G1316A TCC), auto sampler with thermostat (G1367E HiP ALS + G1330B) and an Instant Pilot controller (G4208A). The separation was performed at 30 °C with a reversed-phase Luna^® C₈ LC Column (150 x 2 mm, 5 particle size) Phenomenex, Aschaffenburg, Germany). The mobile phase had a flow of 0.3 ml/min and the injection volume was 2 µl. Gradient conditions for the analysis of the active substances of both PPPs can be found in the supplementary information (Supplementary Tables 1, 2).

Mass analysis in positive ESI mode was carried out using an AB Sciex 6500 QTRAP system (Applied Biosystems, Toronto, Canada) operated in the scheduled multiple-reaction-monitoring mode. The following parameters were used: nebulizing gas/ion source gas 1 (GS1) 20 psi and drying gas/ion source gas 2 (GS2) 50 psi, ionspray voltage (IS) 5500 V, entrance potential (EP) 10 V, collision gas (CAD) medium, curtain gas (CUR) 40 psi, and source temperature (TEM) 400 °C. Isolation and fragmentation of the parent ions were done, for each active substance (azoxystrobin, cyproconazole, epoxiconazole and fluxapyroxad). Optimization of declustering potential (DP), collision energy (CE) and collision cell exit potential (CXP) was done for each m/z transition, in order to maximize obtainable intensities. Target scan time was set to 2.3251 s and scheduled MRM detection window was set to 60 s. Detection parameters are displayed in Table 2(Tab. 2). Using the ion ratio as confirmatory parameter, two m/z transitions with the highest intensity were obtained for each analyte. Each sample was injected twice. Analyst software was used for data acquisition and processing, and Multiquant Software was used for data analysis.

Data were graphically visualized using the Graphpad Prism software (version 9.3.1).

Statistical analyses were performed with SigmaPlot (version 14) and the R software (version). Error bars represent the standard deviation of the mean.

For cell viability, concentration data were transformed using X=log(X) and then normalized to the solvent control. Error bars indicate standard deviation, n=2 biological replicates each performed with six technical replicates. For gene expression analysis p-values were adjusted by the FDR-method (Benjamini and Hochberg, 1995[4]).

Cell viability results showed a dose-response relationship with increasing concentrations of the substances in the case of Adexar^®, Priori Xtra^® and their active substances mix in both the NRU and WST-1 assays on HepaRG cells after 24 h. Figure 1(Fig. 1) and 2(Fig. 2) summarize the results from the NRU assays. The WST-1 results are provided in the supplementary data. EC₅₀ values obtained from the best fit values of the dose response curves generated from the results of the NRU assay are summarized in Table 3(Tab. 3).

All available co-formulants of Adexar® were tested, but only one co-formulant which will be referred to as 'co-formulant A' not listed in MSDS, resulted in a dose-response curve with increasing concentration. The other co-formulants of Adexar®, 2-ethylhexyl-S-lactate and two other co-formulants not listed in the MSDS did not show observable cytotoxicity at the concentrations tested. Ethoxylated alcohol listed as a co-formulant in the formulation of Priori Xtra® also resulted in a dose-response relationship with increasing concentration on HepaRG cells. This co-formulant will be referred to as 'co-formulant 1'. Two other co-formulants of Priori Xtra^® did not decrease cell viability at the concentrations used.

Gene expression analysis shows a similar gene expression profile between Adexar^® and its active substances' mix. For cells treated with Adexar^® significant changes after p-value correction observed were 4 out of 374 transcripts and the magnitudes of the differences were up to 32 fold regulation. CYP1A1 and CYP3A4 were upregulated, with the CYP1A1 showing the largest upregulation of over 32-fold. Cells treated with Adexar^® and its active substances' mix similarly upregulated CYP1A1, CYP3A4, and MSMO1 significantly, whereas CYP1A2 and CYP2B6 expression were significantly affected only by the active substances' mix. In addition, the expression of ACAT2 was upregulated by Adexar^® and not by the mix. Figures 3(Fig. 3) and 4(Fig. 4) summarize the results.

The gene expression of co-formulant A was investigated at the highest sub-cytotoxic concentration investigated (Figure 4(Fig. 4)). The gene expression analysis showed significant downregulation of the following gene transcripts: UGT2B4, GPX2, CTSE, FABP1, SERPINA3, CYP2C19 and ACOT12. ACACA was significantly upregulated.

There were no significant gene expression changes observed in the treatments with Priori Xtra^® or with its active substances' mix (cyproconazole and azoxystrobin) at the concentrations selected.

GO term enrichment analysis suggested that upregulated DEGs significantly enriched in steroid biosynthesic processes in the case of Adexar^® (1.25 mg/L). Whereas, the epoxiconazole and fluxapyroxad mix (1.25 mg/L each) at the same concentration showed an influence.

Protein quantification data reveals that the concentrations of several enzymes were influenced by the treatments. Figure 5(Fig. 5) summarizes the results. Specifically, CYP1A2, CYP3A4, and CYP2C8 proteins increased as a result of treatment with Adexar^® as compared to the solvent control (0.4 % DMSO). Co-formulant A induces the production of ABCB1 and ABCC2 transporters and CYP2C8 but not CYP1A2 and CYP3A4.

The intracellular concentrations of the active substances were obtained in order to compare the concentrations of the active substances within the cells treated with the product, with the concentration of those treated with active substances' mix. Cells treated with Adexar^® and the active substance mix of Adexar^® were analyzed for epoxiconazole and fluxapyroxad. Similarly, cells treated with Priori Xtra^® and its active substances' mix were analyzed for cyproconazole and azoxystrobin. The percentage concentration of each active substance present in the medium, the cells, and PBS was calculated. The intracellular concentration values obtained are presented in Figure 6(Fig. 6). A higher percentage concentration of each active substance was detected in the HepaRG cells treated with the products as well as with the associated medium and PBS, compared to those treated with the mixtures of the respective active substances, except in the case of cyproconazole 2. The concentration of epoxiconazole detected in cells treated with Adexar® and cells treated with its respective active substances' mix (epoxiconazole and fluxapyroxad), was 1.77 % and 1.05 % respectively. A similar result was observed for fluxapyroxad which was detected at 1.43 % in the cells treated with Adexar^® and at 0.55 % with those treated with the active substances' mix. For the cells treated with Priori Xtra^® and its active substances mix azoxystrobin was detected at a 15 times higher concentration in the product (0.15 %) than in the mix (0.01 %). Cyproconazole is isomeric and was detected in two peaks as mentioned in the method. For the peak designated as cypro 1, values obtained were 0.46 % for the product and 0.28 % for the mix, whereas cypro 2 had similar values for the product and the mix at 0.30 % and 0.29 % respectively.

A higher percentage concentration of each active substance was detected in the HepaRG cells treated with the products as well as with the associated medium and PBS, compared to those treated with the mixtures of the respective active substances, except in the case of cyproconazole 2. The concentration of epoxiconazole detected in cells treated with Adexar® and cells treated with its respective active substances' mix (epoxiconazole and fluxapyroxad), was 1.77 % and 1.05 % respectively. A similar result was observed for fluxapyroxad which was detected at 1.43 % in the cells treated with Adexar^® and at 0.55 % with those treated with the active substances' mix. For the cells treated with Priori Xtra^® and its active substances mix azoxystrobin was detected at a 15 times higher concentration in the product (0.15 %) compared to in the mix (0.01 %). Cyproconazole is isomeric and was detected in two peaks as mentioned in the method. For the peak designated as cypro 1, values obtained were 0.46 % for the product and 0.28 % for the mix, whereas cypro 2 had similar values for the product and the mix at 0.30 % and 0.29 % respectively.

This work was funded and supported by the German Federal Institute for Risk Assessment.

The authors declare to have no conflict of interest.

The authors would like to express gratitude to Dr. Corinna Kuerbis for allowing us to use the LC-MS/MS. We would also like to thank Annemarie Richter for technical support.