All-trans retinoic acid (at-RA) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) were purchased from Sigma-Aldrich (Taufkirchen, Germany). The RAR agonist Am580 (CD336; [4-(5,6,7,8-tetrahydro-5,5,8,8,-tetramethyl-2-naphthalenylcarboxamido)benzoic acid]) and the RXR agonist CD2608 (LG1069; (4-[1-(4,5,5,8,8-pentamethyl-5,6, 7,8-tetrahydro-2-naphthyl)-ethyl] benzoic acid) were kindly provided by Dr. U. Reichert (CIRD-Galderma, Sophia-Antipolis, France). All other chemicals were purchased from Merck (Darmstadt, Germany) or Sigma-Aldrich in the highest available purity.

The human colon adenocarcinoma cell line Caco-2 (ECACC, Porton Down, UK) was cultured in Dulbecco's modified Eagle's medium (DMEM; PAA Laboratories GmbH, Coelbe, Germany) supplemented with 10 % fetal calf serum, 100 IU/ml penicillin, and 100 µg/ml streptomycin at 37 °C in a humidified atmosphere of 5 % CO₂. Cells from passages 29-40 were used for all experiments. Caco-2 cells were cultured in six-well plates until complete confluency and were further cultivated for 14 to 18 days in order to allow differentiation. For a recent detailed description of Caco-2 cell culture, please refer to Lichtenstein et al. (2017[19]). Verification of successful differentiation into enterocyte-like cells, which was also used as a verification of cell line identity for Caco-2 cells, has also been described in detail in the aforementioned previous publication (Lichtenstein et al., 2017[19]).

Cells were treated with test compounds in concentrations as indicated in the figures. Compounds were always freshly dissolved in culture medium prior to treatment, from stock solutions prepared in dimethyl sulfoxide (DMSO). The final DMSO concentration in the treatment medium was 0.1 % (v/v). After 24 h of incubation, the medium was removed, the cell monolayer was washed with ice-cold phosphate-buffered saline (PBS), and cells were harvested for subsequent RNA isolation. All compounds were used at non-cytotoxic concentrations, as determined by routine cytotoxicity testing, to ensure the absence of unspecific effects caused by cellular stress.

Total RNA was isolated using the RNeasy kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol including a DNase digestion step. The concentration and quality of the RNA was determined with a Nano Drop spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA). 2 µg of RNA were reverse transcribed by 200 U Moloney murine leukemia virus reverse transcriptase (Promega, Heidelberg, Germany) using 100 pmol of oligo (dT)₁₅ as primers (2 h at 42 °C). Real-time quantitative PCR was performed on a 7900HT real-time PCR cycler (Applied Biosystems, Darmstadt, Germany). Experiments were set up in triplicates containing SYBR Green Mastermix (AbGene, Hamburg, Germany) and the respective primers (3 pmol per reaction). The primer sequences are available from Table 1(Tab. 1). The thermal cycling protocol comprised an initial denaturation step at 95 °C for 15 min, followed by 40 cycles of denaturation at 95 °C for 15 s, annealing and extension at 60 °C for 1 min, and a final extension at 60 °C for 15 min. The mRNA amount of CYP1A1 in each sample was normalized to its ACTB (encoding β-actin) content and then referred to vehicle controls treated with the solvent DMSO only to obtain fold inductions.

For chromatin immunoprecipitation (ChIP) experiments differentiated Caco-2 cells grown in 175 cm² cell culture flasks were used. Cells were treated for 1.5 h with TCDD, at-RA, or vehicle (DMSO), respectively. ChIP analyses were performed as described in detail by Ehlers et al. (2008[6]), where a more detailed description of the experimental protocol can be found. Immunoprecipitation was carried out with antibodies directed against acetyl-histone H3 (H3-K9/K14, 06-599; Upstate, Temecula, CA, USA), RXRα (D 20), RXRβ (C 20), and AHR (H 211) (all from Santa Cruz Biotechnology, Santa Cruz, CA, USA).

A genomic region around 3,000 base pairs upstream of the transcription starting point of CYP1A1 was chosen for amplification following ChIP. This region corresponds to the XRE cluster (i.e., a region containing multiple AHR-binding so-called xenobiotic response elements) in the CYP1A1 promoter (Figure 1(Fig. 1)). Subsequent to the digestion of RNA and proteins, the enriched fragmented DNA was analyzed for the abundance of the promoter fragments of interest, in comparison to the input control as well as control samples from solvent-treated cells. Successful enrichment of immunoprecipitated DNA was demonstrated by real-time quantitative PCR via comparison with the amount of DNA in the input control. In addition, the second exon of the PAX5 gene located more than 13,000 base pairs downstream of its start codon was amplified as a negative control to demonstrate specificity of DNA enrichment. The DNA content of each sample was normalized to its ACTB content and then referred to input DNA to obtain fold enrichment. Primer sequences are shown in Table 1(Tab. 1).

Statistical analyses were performed with SigmaPlot software. Statistical significance of differences between mean values were determined by Student's t-test or, for concentration-dependent studies, by one-way ANOVA followed by Dunnett's post-hoc test. Statistically significant differences were assumed at p < 0.05.

First, Caco-2 cells were exposed to agonists of different nuclear receptors to characterize the impact of activation of these signaling pathways on CYP1A1 mRNA expression (Figure 2(Fig. 2); for detailed data from individual replicates, please refer to the Supplementary Data Table). The real-time PCR data show that the model AHR agonist TCDD, as expected, drastically stimulated CYP1A1 gene expression. Exposure of cells to 50 nM TCDD induced CYP1A1 gene expression by approximately 425-fold (Figure 2(Fig. 2)). Activation of RAR was initially accomplished by incubation of cells with increasing concentrations of at-RA. Only at concentrations of 10 µM or higher, a slight, but statistically significant approximately 3.5-fold increase in CYP1A1 expression was recorded, while lower concentrations of at-RA did not exert remarkable effects (Figure 2(Fig. 2)). Activation of RAR with 100 nM of the agonist Am580 did not significantly affect CYP1A1 expression, whereas activation of RXR by its ligand CD2608 led to a very slight, but significant approximately 2-fold increase in CYP1A1 mRNA (Figure 2(Fig. 2)).

Both, the AHR and the retinoid receptors, act as ligand-dependent transcription factors which bind to the promoters of target genes and thereby regulate their transcription. This is generally associated with an increased histone acetylation of the respective promoter region, and with increased binding of the activated transcription factors to their recognition sequences within the gene regulatory region of the DNA. The histone H3 acetylation status of the promoter of the CYP1A1 gene was therefore analyzed by chromatin immunoprecipitation (ChIP) assays (see also scheme in Figure 1(Fig. 1)). Differentiated Caco-2 cells were exposed to 50 nM TCDD for AHR activation or to 10 µM at-RA for RAR activation. Immunoprecipitations from nuclear cellular Caco-2 extracts were carried out with an antibody against acetylated histone H3 and a DNA region approximately 3,000 base pairs upstream of the transcription start was amplified from the precipitates by PCR.

Figure 3A(Fig. 3) demonstrates the histone H3 acetylation status in the promoter region of CYP1A1 (-3,000 bp upstream of the start codon) in Caco-2 cells (for detailed data from individual replicates, please refer to the Supplementary Data Table). In control cells only treated with the solvent, histone H3 showed some constitutive acetylation in the promoter region of CYP1A1 even without xenobiotic treatment, while a substantial enhancement in the amplified DNA amount, corresponding to an increased acetylation of histone H3 in the respective genetic region, was recorded following treatment with both, at-RA and TCDD (Figure 3A(Fig. 3)). Histone H3 acetylation was not detected in the negative control, where a DNA region outside of any promoter region (i.e., the coding region of the gene PAX5; Figure 3A(Fig. 3)) was amplified; this demonstrates the specificity of our assay.

Increased histone H3 acetylation in the promoter region of CYP1A1 following xenobiotic treatment was observed (Figure 3A(Fig. 3)). This was plausible since histone acetylation is generally associated with gene activation. We now analyzed whether also the recruitment of the respective relevant transcription factors, namely AHR and retinoic acid receptors, to the same regions of the gene promoters would be detectable by ChIP assays. Activation of the AHR again was achieved by treatment with TCDD, while RAR activation was performed using at-RA. ChIP analyses were carried out with antibodies directed against the AHR, RXRα and RXRβ. In control cells treated with the solvent only, AHR and RXRβ were barely associated with the promoter of the CYP1A1 gene, whereas basal binding significantly differing from non-antibody controls was detected for RXRα (Figure 3(Fig. 3)). The known interaction between the AHR and the CYP1A1 promoter (-3,000 base pairs) was verified in samples from cells exposed to TCDD which showed massive binding of the AHR to this genetic region (Figure 3B(Fig. 3)). In addition, an increase in RXRα binding to the CYP1A1 promoter was also observed following treatment of Caco-2 cells with TCDD (Figure 3C(Fig. 3)). No major impact of TCDD treatment was observed with regard to the binding of RXRβ (Figure 2D(Fig. 2)). Cell treatment with at-RA resulted in a statistically significant enrichment of the RXRα-CYP1A1 promoter interaction (Figure 3C(Fig. 3)), without affecting RXRβ binding to the gene promoter (Figure 3D(Fig. 3)). Interestingly, an increase in AHR binding to the CYP1A1 promoter was also visible following cell treatment with at-RA (Figure 3B(Fig. 3)). Taken together these results indicate that individual treatment with either at-RA or TCDD might, to a certain degree, be able to assist the recruitment of both receptors, AHR and RXRα, to the CYP1A1 gene promoter, even without addition of an exogenous ligand of the second receptor.

Experiments with at-RA concentrations in a range from 0.1 µM to 25 µM verified the generated data for the histone H3 acetylation at the CYP1A1 promoter and demonstrated the increased binding of AHR and RXRα (Table 2(Tab. 2); for detailed data from individual replicates, please refer to the Supplementary Data Table). Moreover, substantial enrichment was now also seen for RXRβ, but only at the highest tested concentration of at-RA (Table 2(Tab. 2)). For histone H3 acetylation and AHR binding, lower concentrations of at-RA caused higher enrichment at the CYP1A1 promoter (Table 2(Tab. 2)). Enrichment was again not detected in the negative control in these experiments (data not shown).

The data presented above indicate a cross-talk between the different receptors. Recently it has been shown that the cross-talk of the AHR with Wnt/β-catenin signaling at the CYP1A1 promoter synergistically enhances the responses to AHR agonists (Schulthess et al., 2015[29]). We were now interested whether similar effects might also be observed with AHR agonists and retinoids activating their respective receptors. Thus, in subsequent mixture experiments, the RXR agonists CD2608 and at-RA were combined with TCDD (Figure 2(Fig. 2); right bars). The stimulation of both pathways, namely activation of the AHR with TCDD combined with an activation of the RAR/RXR-dependent signaling pathways by at-RA or CD2608, enhanced CYP1A1 mRNA expression by up to 870-fold, which is significantly more than expected from the same concentration of the AHR agonist alone, and also significantly more than expected when additive effects of stimulation of the two individual signaling pathways is assumed. Thus a combined stimulation of AHR- as well as RAR/RXR-dependent transcription enhances the cellular levels of CYP1A1 mRNA in an over-additive manner (Figure 2(Fig. 2)).