11-Cl-BBQ, a Select Modulator of AhR-regulated Transcription, Suppresses Lung Cancer Cell Growth via activation of p53 and p27^Kip1

Abstract

Aryl hydrocarbon receptor (AhR) is a ligand-activated transcription factor and functions as a tumor suppressor in different cancer models. In this study, we report detailed characterization of 11-chloro-7H-benzimidazo[2,1-a]benzo[de]iso-quinolin-7-one (11-Cl-BBQ) as a Select Modulator of AhR-regulated Transcription (SMAhRT) with anti-cancer actions. Treatment of lung cancer cells with 11-Cl-BBQ induced potent and sustained AhR-dependent anti-proliferative effects by promoting G1-phase cell cycle arrest. Investigation of 11-Cl-BBQ-induced transcription in H460 cells with or without the AhR expression by RNA-seq revealed activation of p53 signaling. In addition, 11-Cl-BBQ suppressed multiple pathways involved in DNA replication and increased expression of cyclin-dependent kinase inhibitors, including p27^Kip1, in an AhR-dependent manner. CRIPSR-cas9 knockout of individual genes revealed the requirement for both p53 and p27^Kip1 for the AhR-mediated anti-proliferative effects. Our results identify 11-Cl-BBQ as a potential lung cancer therapeutic, highlight the feasibility of targeting AhR and provide important mechanistic insights into AhR-mediated-anticancer actions.

Graphical abstract

Introduction

Aryl hydrocarbon receptor (AhR) is a ligand-activated transcription factor belonging to the bHLH (basis helix-loop-helix)-PER-ARNT-SIM (bHLH/PAS) subfamily of the bHLH family of transcription factors (1). The binding of a wide range of both endogenous and exogenous small molecules leads to the activation of AhR signaling that influences cellular outcomes in different developmental, immunological, and cancer-related contexts. Once activated, AhR translocates from the cytosol into the nucleus and heterodimerizes with AhR Nuclear Translocator (ARNT) (2). Binding of the AhR/ARNT complex to specific promoter sequences recruits other transcriptional regulators to modulate the expression of downstream target genes (3,4).

Accumulating evidence indicates that the AhR can have context-dependent pro- or anti-tumorigenic effects depending on the biological activity of the ligands. Analyzing genomic data of 947 human cancer cell lines revealed a subset of cancer cell lines with mutant NRAS that were dependent on AhR for their growth and survival (5). Constitutive activation of AhR by deleting a portion of its ligand binding domain promoted the development of stomach and liver tumor in mice (6–8). AhR has been shown to mediate and facilitate the pro-tumorigenic effects of environmental pollutants such as benzo[a]pyrene and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (9–11). Another evidence indicating that activating AhR by an endogenous ligand kynurene (a metabolite of tryptophan) facilitated the growth of the human LN-308 glioma cells in CD1^nu/nu mice (12). On the other hand, transgenic adenocarcinoma of the mouse prostate (TRAMP) AhR null mice had higher prostate cancer incidence than AhR wildtype mice (60% compared to 16%, respectively) (13). Recent studies showed that AhR exhibits tumor-suppressive effects against cancers of the intestinal system (14–16). Interestingly, the tumor incidence was significantly lower when the Apc^Min/-AhR^+/+, were fed with natural AhR ligands indole-3-carbinol and 3,3′-diindolylmethane (16). This effect was not observed in Apc^Min/-AhR^−/− mice.

As the fate of cancer cells can be influenced via AhR signaling, the AhR has been suggested as a molecular target for cancer drug development (17,18). Significant efforts have been focused on identifying AhR ligands with potent anticancer effects referred here as Select Modulators of AhR-regulated Transcription (SMAhRTs). Aminoflavone and its pro-drug (AFP464) are AhR agonists, exhibit anticancer effects on breast cancer cells and have progressed to phase I clinical trials for breast cancer treatment (19–21). Our laboratory initially focused on repurposing of FDA-approved drugs with well-characterized toxicity profiles in humans that could function through the AhR for treatment of cancer. Using this approach, we identified clinically used drugs leflunomide, flutamide, and raloxifene that suppressed the growth of cancer cells by inducing cell cycle arrest or cell death (22–25). We further conducted our search for SMAhRTs by screening more than 50,000 small molecules from the DIVERSet® library from ChemBridge. We recently reported the discovery of CGS-15953 as one such SMAhRT that promoted AhR tumor suppressive signaling and induction of cancer cell death (26). From this screening, we also identified a group of benzimidazoisoquinolines, including 11-Cl-BBQ and its analogs, as high affinity AhR ligands (27). These benzimidazoisoquinolines were shown to activate AhR at nanomolar concentrations, promote T-cell differentiation, and were non-toxic in mouse models (27–29). In the current study, we tested and characterized the anticancer effects of 11-Cl-BBQ in both small and non-small lung cancer cell lines. 11-Cl-BBQ induced potent anti-proliferative effects in lung cancer cells expressing high levels of the AhR, but not in cells with very low or no expression. Stable knockdown of AhR expression using CRISPR-cas9 in responsive cancer cell lines reversed the anti-cancer effects of 11-Cl-BBQ, indicating that these effects are AhR-dependent. In addition, 11-Cl-BBQ induced a stable G-1 phase cell cycle arrest, and the arrested cells were unable to grow in zebrafish tumor xenograft assays. Transcriptomic studies revealed that p53 signaling is induced and significantly contributed to the cell cycle arrest induced by 11-Cl-BBQ. In addition, these studies also revealed upregulation of three cyclin-dependent kinase inhibitors, p27^Kip1, p21^Cip1 and CABLES1. Knockdown experiments confirmed a role for p27^Kip1 in AhR-dependent inhibition of proliferation. Overall, these results increase our understanding of the mechanisms of growth suppression by the AhR and support development of 11-Cl-BBQ and other related benzimidazoisoquinolines as anti-lung cancer therapeutics.

Results

11-Cl-BBQ inhibits lung cancer cell growth in an AhR-dependent manner

We tested the growth inhibition effect of 11-Cl-BBQ (Fig. 1A) on multiple lung cancer cell lines with varying levels of steady-state AhR expression (Fig. 1B). 11-Cl-BBQ decreased the growth of lung cancer cells in a dose-dependent manner from 10 nM to 10 μM concentrations (Fig. 1C). H460 and H69AR cell lines with relatively high expression of AhR were more sensitive to 11-Cl-BBQ than A549 and H1299 cells, each with relatively low expression of the AhR (Fig. 1B and 1C), suggesting a role for the AhR in the growth suppression induced by 11-Cl-BBQ. To determine whether the AhR was required for 11-Cl-BBQ-induced growth suppression, we generated stable AhR-deficient cells derived from H69AR lung cancer cells utilizing CRIPSR-cas9 (30,31) with three single guide RNAs (sgRNAs; CR-AHR1, CR-AHR2, and CR-AHR3) targeting the human AhR gene. Western blot analysis demonstrated reduction in steady-state levels of AhR expression in pooled cultures following expression of two AhR-specific sgRNAs (CR-AHR2 and CR-AHR3) compared to the control vector (CR-V2) (Fig. 1D). Suppression of AhR expression in H69AR cells significantly reversed the growth inhibitory effects of 11-Cl-BBQ (Fig. 1E). Thus, AhR is required for the growth inhibition induced by 11-Cl-BBQ in H69AR lung cancer cells.

Figure 1. 11-Cl-BBQ inhibits cell growth in an AhR-dependent manner.

(A) the structure of the small molecule 11-Cl-BBQ. (B) Steady-state AhR protein expression in lung cancer cell lines examined by western blot analysis of whole cell lysates. (C) Viability of lung cancer cells treated with 11-Cl-BBQ at indicated doses for 72 hours relative to vehicle control treated cells (set to 100%). (D) Western blot analysis of steady-state AhR protein expression in pooled cultures of H69AR cells transfected with three guide RNAs targeting human AhR (CR-AhR1, CR-AhR2, and CR-AhR3) or lentiCRISPRv2 control vector (CR-V2). (E) Relative cell viability of AhR-proficient (CR-V2, gray), and two different pools of AhR-deficient (CR-AhR2, orange bars and CR-AhR3, red bars) H69AR lung cancer cells treated with 11-Cl-BBQ at indicated doses for 72 hours. Statistical analyses were done with 3 biological replicates for each treatment with multiple comparison using Turkey – Kramer procedure; ***: adjusted p value < 0.001; error bars represent the standard errors.

To investigate the role of AhR in additional lung cancer cells, we generated AhR knockout H460 cells utilizing the CRIPSR-cas9 system. H460 AhR-knockout cells (CR-AHR3) were significantly less responsive to 11-Cl-BBQ treatment compared to the AhR-expressing CR-V2 control cells (Figs. 2A and 2B). In addition, continuous treatment with 11-Cl-BBQ for 14 days with 0.1–10 μM concentrations strongly inhibited colony formation of AhR-expressing H460 cells, but not the AhR-deficient H460 cells (Fig. 2C, and 2D). Based on these data we conclude that AhR mediates growth suppression of 11-Cl-BBQ in two distinct lung cancer cell lines.

Figure 2. AhR-mediate cell cycle arrest by 11-Cl-BBQ.

(A) Western blot analysis for the expression of AhR protein in clonal lines derived from H460 cells transfected with control vector (CR-V2) and or sgRNA targeting AhR (CR-AHR3). (B) Viability of H460 AhR-proficient (CR-V2, grey) or H460 AhR-deficient (CR-AHR3, red) cells treated with 11-Cl-BBQ at indicated doses for 72 hours. (C) Representative images of colonies formed by H460 AhR-proficient (CR-V2) or H460 AhR-deficient (CR-AHR3) treated with 11-Cl-BBQ at indicated doses for 14 days, with number of colonies were quantified (D) using OpenCFU software. (E) Annexin V staining of H460 AhR wildtype (CR-V2) cells or AhR knockout (CR-AHR3) treated with 11-Cl-BBQ (1 and 10 μM) or vehicle control for 72 hours, percentage of cell population positive with Annexin V staining (blue bar) was shown for each treatment group. (F) Cell cycle analysis of H460 AHR wildtype (CR-V2) cells or AHR knockout (CR-AHR3) treated with 11-Cl-BBQ (2.5 μM) or vehicle control for 24 hours. (G) Quantification of cell cycle distribution of data for four biological replicates. Turkey – Kramer’s multiple comparison test: * adjusted p value < 0.05, ** adjusted p value < 0.01, *** adjusted p value < 0.001, *** adjusted p value < 0.0001, ns: not statistically significant. Error bars in (B, D, and G) represent the standard error of three, three, and four biological replicates, respectively.

11-Cl-BBQ induces an irreversible, AhR-dependent cell cycle arrest in lung cancer cells

Visual inspection of the lung cancer cells using phase contrast light microscopy suggested that treatment with 11-Cl-BBQ did not induce cell death. In addition, H460 cells treated with 11-Cl-BBQ were negative for Annexin V staining (Fig. 2E), indicating that the growth inhibitory effect of 11-Cl-BBQ was not due to induction of cell death. Therefore, we tested the effect of 11-Cl-BBQ on H460 cell proliferation by analyzing the cell cycle distribution. There was a significant increase of cells in G1 phase (45% to >70%) and a reduction of cells in S-phase (35% to 15%) upon treatment with 2.5 μM 11-Cl-BBQ for 24 h (Fig. 2F and 2G). The G1 phase cell cycle arrest induced by 11-Cl-BBQ was AhR-dependent as this was not observed in H460 cells without the AhR expression (Fig. 2F and 2G). To determine whether the effect of 11-Cl-BBQ was reversible, AhR-expressing H460 cells were treated with 11-Cl-BBQ or DMSO (vehicle control) for 72 h, followed by a “washout” period for five days. Continuous exposure to 11-Cl-BBQ (2.5 μM) resulted in sustained growth suppression (Fig. 3A, red line), as expected. However, H460 cells didn’t recover from growth arrest even after 96 hours of 11-Cl-BBQ wash out period (Fig. 3A, black line). Thus, we investigated whether 11-Cl-BBQ-induced growth arrest was associated with an increase in the senescence-associated beta galactosidase (SA-β-gal), a well-characterized biomarker for an irreversible cell cycle arrest (32–35). Indeed, 11-Cl-BBQ treatment (2.5 μM for 120 h) resulted in significant increase in SA-β-gal expression and was AhR-dependent (Fig. 3B and 3C; compare AhR-proficient CR-V2 versus AhR-deficient CR-AHR3 cells). In contrast, treatment with doxorubicin, a topoisomerase II inhibitor, induced SA-β-gal expression independent of the AhR status in H460 cells (Fig. 3B and 3C). 11-Cl-BBQ treatment (1 μM for 120 h) also increased SA-β-gal in the H69AR lung cancer cell line (Fig. 3D and 3E).

Figure 3. Induction of sustained growth suppression by 11-Cl-BBQ.

(A) Relative cell number (normalized to 0 h timepoint) of H460 cells treated with 11-Cl-BBQ (2.5 μM) or vehicle control for 72 hours, followed by a second round of 11-Cl-BBQ (2.5 μM) or vehicle treatment at indicated time. (B) Histogram of senescent-associated β-galactosidase (sen-β-gal) in H460 AHR wildtype (CR-V2) cells or AHR knockout (CR-AHR3) treated with 11-Cl-BBQ (2.5 μM), Doxorubicin (Dox, 50 nM) or vehicle control for 5 days. (C) Quantification of H460 cell populations positive with sen-β-gal staining in indicated treatment groups (six biological replicate per group, Turkey – Kramer’s multiple comparison test: **** adjusted p value < 0.0001. (D) Representative SA-β-gal staining versus side scatter (SSC) dot blots (with gating for SA-β-gal) of H69AR treated with vehicle or 11-Cl-BBQ (1μ) for 120 hours. (E) SA-β-gal positive cell populations were quantified from three biological replicates each treatment, p-value from a two-tailed ttest was shown. Error bars represent the standard error.

To investigate whether the stable cell cycle arrest induced by 11-Cl-BBQ sustains in an in vivo model, we monitored the growth of 11-Cl-BBQ-treated AhR-proficient and AhR-deficient lung cancer cells in a zebrafish tumor xenograft assay (36,37). AhR-expressing (CR-V2) and AhR-deficient (CR-AHR3) H460 cells were exposed to 11-Cl-BBQ (1 μM or 5 μM for 72 h), after which the compound was washed out. Cells were then injected into the yolk sac of live zebrafish embryos and the growth of the cancer cell xenografts was measured at days 1 and 4 of post-injection (Fig. 4A). As shown in Figure 4B and 4C, AhR-expressing 11-Cl-BBQ treated cells had significantly lower growth in this xenograft model. Taken together, we conclude that 11-Cl-BBQ treatment induces an AhR-dependent and irreversible G1-phase arrest in human lung cancer cells.

Figure 4. 11-Cl-BBQ induced arrested cells had lower survival in vivo.

(A) Schematic of zebrafish xenograft experiment: H460 AHR sufficient cells (CR-V2) or AHR deficient cells (CR-AHR3) treated with 11-Cl-BBQ (1 or 5 μM), or vehicle control (0.1% DMSO) for 72 hours were dyed with CMdil and injected into the yolk sac of zebrafish embryo; the growth of the cells was monitored at 1-day post-injection (1dpi) and again at 4dp using high content imager. (B) Relative tumor size of indicated treatment groups at 4dpi compared to 1dpi; statistical analyses using Dunnett’s multiple comparison test * adjusted p value < 0.05, ** adjusted p value < 0.01, ns: not statistically significant; error bars represent the standard error for 19 zebrafish embryo each treatment group, except H460 CR-V2 treated with 11-Cl-BBQ (5 μM) with 17 zebrafish embryo. (C) Representative images (10X magnification) of H460 zebrafish xenograft composed of zebrafish (bright field) and H460 cell (red channel) and overlayed images.

11-Cl-BBQ suppresses the expression of genes involved in DNA replication and cell cycle regulation

To identify AhR-mediated transcriptional programs relevant to inhibition of lung cancer cell proliferation, we measured changes in global gene expression. AhR-proficient and -deficient H460 cells were treated with 5 μM 11-Cl-BBQ or vehicle control for 4 h or 12 h and analyzed changes in gene expression by RNA-sequencing (RNA-seq). Treatment with 11-Cl-BBQ resulted in significant changes in expression of a number of genes in an AhR-dependent manner (Fig. 5A and 5B), including the elevation of well-known AhR targets: CYP1A1, CYP1B1, AhRR, and TIPARP (Fig. 5A). Expression changes of at least 2-fold for genes detected with seven or greater Transcripts Per kilobase Million (TPM) were considered significant. Some genes were differentially expressed both at 4 h and 12 h (33 elevated and 6 reduced), but several of differentially expressed genes were not the same between these two time points (85/67 elevated and 14/30 reduced at 4/12 h, respectively) (Fig. 5B). Changes in mRNA levels of previously known AhR-target genes (CYP1A1, IL1A, CDKN1B (p27^Kip1)) and several newly identified genes (CABLES1, CCNE2, and SHISA2) were confirmed by qPCR (Fig. 5C).

Figure 5. AhR-dependent changes in gene expression induced by 11-Cl-BBQ in lung cancer cells.

(A) Expression (TPM, transcripts per kilobase million) of well-known targets of AHR (CYP1A1, CYP1B1, AHRR and TIPARP) in H460 AHR wildtype (CR-V2) or AHR knockout (CR-AHR3) treated with 11-Cl-BBQ (5 μM), or vehicle control (0.1% DMSO) for 4 or 12 hours measured by RNA-seq method. (B) Venn diagram showed the number of genes that changed ≥ 2 fold with 11-Cl-BBQ treatment compared to vehicle treatment in an AHR-dependent manner at indicated times. (C) Expression of select genes in H460 AHR wildtype and AHR knockout treated with 11-Cl-BBQ (5 μM) or vehicle control for 4 hours, measured by RNA-seq and qPCR methods; standard errors of three biological replicates (CDKN1B, E2F7, and CYP1A1) or two technical replicates (other genes) were added as error bars for qPCR data. (D) All canonical pathways that enriched upon 11-Cl-BBQ treatment for 12 hours in AHR wildtype (FDR < 25%, blue bars) and in AHR knockout cells (red bars) by gene set enrichment analysis (GSEA). (E) Top 9 gene ontology terms of a cluster related to cell cycle regulation (adj p-val < 0.01) revealed by Gene Ontology analysis of differentially expressed transcripts upon 11-Cl-BBQ treatment for 12 hours.

Biological processes often involve multiple genes or a network of genes working in a coordinated manner in the same or closely related pathways. We performed gene set enrichment analysis (GSEA) using single sample GSEA projection (38). Several negative regulators of cell cycle progression (CDK inhibitors p27^Kip1 and CABLES1 (Fig. 5C) and p21^Cip1 (Fig. 6A) were elevated in an AhR-dependent manner after 4 h of 11-Cl-BBQ treatment. GSEA further revealed enrichment of low expressed genes (FDR <25%) in canonical pathways associated with DNA replication and cell cycle progression in AhR-expressing (but not in AhR-deficient) H460 lung cancer cells after 12 h exposure to 11-Cl-BBQ (Fig. 5D). These results are consistent with the induction of G1 cell cycle arrest by 11-Cl-BBQ (Fig. 2).

Figure 6. Induction of p21 and p53 11-Cl-BBQ.

(A) Expression of p21 (CDKN1A) mRNA in H460 AHR wildtype (CR-V2) or AHR knockout (CR-AHR3) cells treated with 11-Cl-BBQ (5 μM) or vehicle control for 4 and 12 hours measured by RNA-seq. (B) Western blot showed the expression of p21 protein in H460 AHR wildtype or AHR knockout cells treated with 11-Cl-BBQ (2.5 μM), TCDD (30 nM) or vehicle control. (C and D) Western blot analysis of p53 protein levels in H460 cell treated with (C) 11-Cl-BBQ (5 μM) or (D) TCDD (30 nM) at the indicated time.

The RNA-seq data set includes paired-end, long reads (150 bps) which allows for transcript-level quantification (39,40). Therefore, we generated a transcript-level quantification of the RNA-seq data as an alternate approach to identify changes in gene expression induced by 11-Cl-BBQ in lung cancer cells. This led to the identification of 777 transcripts altered in an AhR-dependent manner by 2-fold or more after 12 h of 11-Cl-BBQ treatment. We next analyzed gene ontology (GO) enrichment of the transcripts using DAVID functional annotation clustering tools (41,42). The results revealed a cluster of functions related to cell cycle regulation (Cluster 4, enrichment score 5.04; Fig. 5E) as one of the top enriched clusters. Consistent with the gene-level GSEA results (Fig .5D), the top 9 most enriched functional annotation terms in the cluster 4 were associated with DNA replication, G1/S phase cell cycle progression, and signal transduction associated with cellular responses to DNA damage for both the G1 and M phase checkpoints (Fig. 5E). Among the enriched GO terms was signal transduction by p53 resulting in cell cycle arrest, suggesting the involvement of the tumor suppressor p53 in 11-Cl-BBQ-induced cell cycle arrest. Together, our analyses of cell cycle progression (Fig. 2) and gene expression (Fig. 5) demonstrate that activation of AhR by 11-Cl-BBQ results in an irreversible G1-phase cell cycle arrest in lung cancer cells associated with the activation of signal transduction pathways related to cellular checkpoint responses and the suppression of the expression of genes necessary for DNA replication. This analysis identified additional potential key mediators of 11-Cl-BBQ-induced growth suppression downstream of AhR activation.

Tumor suppressor p53 is required for 11-Cl-BBQ-induced G1 phase cell cycle arrest in lung cancer cells

The analysis of gene ontology terms associated with changes of transcripts suggested the involvement of p53-dependent signal transduction in the 11-Cl-BBQ-induced cell cycle arrest in lung cancer cells (Fig 5E). In addition, transcription of the CDK inhibitor p21^Cip1, a well-known downstream target of p53 (43–47) was up-regulated by 11-Cl-BBQ in H460 lung cancer cells in an AhR-dependent manner (Fig. 6A and 6B). This prompted us to investigate the role of p53 in the AhR-mediated cell cycle arrest. H460 lung cancer cells express wild type p53 (48,49) and treatment with 11-Cl-BBQ (5 μM) resulted in increased p53 protein levels starting from 4 h, with further increases after 6 and 9 h (Fig. 6C). However, expression of p53 protein was not altered (Fig. 6D) by the treatment with TCDD, a potent ligand of AhR.

To test if p53 has a role in AhR-mediated growth suppression, we generated p53-deficient H460 cells using the CRISPR-cas9 (Fig. 7A). Knockout of p53 lowered the basal expression of p21^Cip1 and its induction, both at mRNA and protein levels (Fig. 7A and 7B). H460 p53-proficient cells were arrested in G1 phase upon 11-Cl-BBQ treatment as expected, however, the G1 cell cycle arrest was not detected in p53 knockout H460 cells (Fig. 7C and 7D). Although it was not statistically significant, we noticed an increase in G2 phase cells in p53 knockout cells upon treatment with 11-Cl-BBQ. Knockout of p53 also significantly reduced SA-β-gal, a biomarker for long-term cell cycle arrest (Fig. 7E) upon 11-Cl-BBQ treatment. These data indicate an important role of p53 in AhR-mediated cell cycle arrest.

Figure 7. Tumor suppressor p53 is required for G1 phase cell cycle arrest induced by 11-Cl-BBQ in H460 lung cancer cells.

(A) Western blot analysis of p53 and p21 protein levels in H460 p53 wildtype (WT) and p53 knockout (KO) treated with 11-Cl-BBQ (2.5 μM), TCDD (30 nM), or vehicle control for 9 hours. (B) Expression of p21 mRNA in H460 p53 wildtype (WT) or p53 knockout (KO) cells treated with 11-Cl-BBQ (2.5 μM) or vehicle control for 4 hours measured by qPCR; error bar showed the standard deviation of two technical replicates. (C and D) Representative histogram (C) and quantification (D) of cell cycle distribution of H460 p53 WT or p53 KO cells treated with 11-Cl-BBQ (2.5 μM) or vehicle control for 24 hours. (E) Representative histogram and quantification of sen-β-gal in H460 p53 WT and p53 KO treated with 11-Cl-BBQ (2.5 μM) or vehicle control for 5 days. Media on cells was changed on day 3 and were treated with 11-Cl-BBQ or vehicle. **: adjusted p value < 0.01, ****: adjusted p value < 0.0001, ns: not statistically significant for Turkey – Kramer’s multiple comparison test. Error bars show the standard error of four and six biological replicates in (D and E), respectively.

p27^Kip1 is an important downstream target gene of the AhR for suppression of lung cancer cell proliferation.

We identified an early induction (4 h) of CDKN1B (p27^Kip1) by 11-CL-BBQ in an AhR-dependent manner during RNA-seq analysis (Fig. 5C). We reconfirmed the induction of mRNA in a time-course experiment (0.5 h to 8 h) upon treatment with 11-Cl-BBQ (2.5 μM) by qPCR (Fig. 8A). To investigate whether the AhR regulated p27^Kip1 directly, we performed chromatin immunoprecipitation (ChIP) using an AhR-specific polyclonal antibody to detect association of the AhR with CDKN1B (p27^Kip1) gene promoter. Results of the ChIP assay confirmed recruitment of the AhR to CYP1A1 gene promoter, upon treatment with both 11-Cl-BBQ and TCDD (Fig. 8B). Similarly, the AhR was recruited to the CDKN1B (p27^Kip1) gene promoter by both 11-Cl-BBQ and TCDD (Fig. 8C). In addition, a time-course analysis (Fig. 8A) demonstrated that CDKN1B (p27^Kip1) is induced by 11-Cl-BBQ as early as 1 h, consistent with the notion that p27^Kip1 is a direct target of the AhR (Fig. 8C). The increased p27^Kip1 protein levels were seen even after 72 h of treatment (Fig. 8D).

Figure 8. Cyclin inhibitor p27^Kip1 is an important downstream target of AhR in mediating cell cycle arrest by 11-Cl-BBQ treatment.

(A) Expression of p27 (CDKN1B) mRNA measured by qPCR in H460 cells treated with 11-Cl-BBQ (2.5 μM) or TCDD (30 nM) relative to vehicle control treatment (0.1% DMSO) at indicated times. (B and C) Enrichment of AHR at CYP1A1 (B) and p27 (C) promoters in H460 cells after treatment with 11-Cl-BBQ (2.5 μM), TCDD (30 nM), or vehicle control for one hour measured by qPCR after chromatin immunoprecipitated (IPed) with AHR specific antibody (AHR) or immunoglobulin G control (IgG); y-axis is percentage of IPed (%IPed) relative to input control of corresponding treatment. (D) Western blot analysis of p27 protein in H460 AHR proficient (CR-V2) and AHR deficient (CR-AHR3) cells treated with 11-Cl-BBQ at indicated doses for 72 hours. (E) Western blot analysis of p27 protein levels in H460 cells transfected with CRIPSR-cas9 with sgRNA specific for p27 gene (CR-p27) or CRISPR-cas9 control vector (CR-V2) treated with 11-Cl-BBQ (2.5 μM) for 24 hours. (F) Cell cycle distribution of control H460 cells expressing p27 (CR-V2) or not expressing p27 (CR-p27) treated with 11-Cl-BBQ (1 μM) or vehicle control for 24 hours. (G) Quantification of sen-β-gal in H460 p27 expressing (CR-V2) and p27 non-expressing (CR-p27) cells treated with 11-Cl-BBQ. Error bars show the standard deviation of two technical replicates (A, B, and C) or four and three biological replicates (F and G, respectively). Statical analysis using Turkey – Kramer’s multiple comparison test: ** adjusted p value < 0.01, **** adjusted p value < 0.0001, ns: not statistically significant.

To determine whether p27^Kip1 was required for the 11-Cl-BBQ-induced cell cycle arrest, we used the CRISPR-cas9 to generate stable p27^Kip1-deficient cells from the H460 lung cancer cell line. The knockout of p27^Kip1 was clearly confirmed at the protein level by Western blot analysis (Fig. 8E). There were higher percentage of G1 phase cells in p27^Kip1-deficient H460 cells and excitingly they did not respond to 11-Cl-BBQ treatment (Fig. 8F). Furthermore, SA-β-gal induction after 11-Cl-BBQ treatment was also reduced in the absence of p27^Kip1 expression (Fig. 8G). These data reveal critical roles of both p53 and p27^Kip1 in AhR-mediated suppression of lung cancer cell proliferation by 11-Cl-BBQ.

Discussion

Lung cancer is the second-most-common cancer, but it is the leading cause of cancer death - more than colon, breast, and prostate cancers combined (50). Much progress has been made in personalized medicine (51), however chemotherapy and radiation are still the primary treatment options for advanced lung cancer. Lack of effective treatment is reflected by the low 5-year survival rate of 19%, which decreases to 6% in small cell lung cancer cases (52). Therefore, expanding targeted anticancer drugs is critical to improve the outcomes for lung cancer patients.

Aryl hydrocarbon receptor (AhR) is a ligand-activated transcription factor and regulates cell cycle progression and cell death in a ligand-dependent manner. Significant efforts have been ongoing in our laboratory to exploit the anti-proliferative and pro-apoptotic functions of the AhR for cancer therapy. AhR is highly expressed in lung cancer, in TCGA pan-cancer gene expression data accessed through UALCAN portal (53) as well as reported in a previous study (54), representing an attractive target to develop therapeutics for lung cancer treatment. Our laboratory has identified some clinically approved drugs that are AhR agonists with strong anticancer effects (22–25,55) suggesting activation of AhR might be beneficial for cancer patients. We discovered benzimidazoisoquinoline 11-Cl-BBQ and its analogs as high-affinity AhR ligands that are well tolerated in mouse models (27,28). The current study demonstrates that 11-Cl-BBQ strongly suppresses growth of lung cancer cells via an AhR-mediated induction of permanent G1 cell cycle arrest.

RNA-seq analysis indicate up-regulation of CDK inhibitors such as p21^Cip1, p27^Kip1 and CABLES1 (Fig. 5C and 6A). The up-regulation of p21^Cip1 by 11-Cl-BBQ treatment is in agreement with previous studies showing that p21^Cip1 is a direct target of AhR (57). The data presented here further supports that p27^Kip1 is a direct transcriptional target of AhR and plays an important role in 11-Cl-BBQ-induced G1-phase cell cycle arrest (Fig. 8). We previously reported p27Kip1 as a target gene of the AhR in hepatoma cells (58). Furthermore, recently we identified a new regulatory cross-talk between p27Kip1 and the AhR (59). CABLES1 is a newly identified CDK inhibitor which has been demonstrated to play important roles in cell cycle regulation and suppression of tumor growth (60–63). However, the regulation of CABLES1 mRNA expression remains poorly understood. Up-regulation of CABLES1 mRNA has been detected in transcriptomic studies using AhR ligands (64–66). Our preliminary data indicate that the 11-Cl-BBQ upregulates CABLES1 mRNA as early as one hour (data no shown). In addition, the expression of CABLES1 mRNA was also lower in AhR knockout cells compared to AhR-expressing H460 cells (Fig. 5C). Collectively, these data suggest a role for AhR in regulating CABLES1 expression. AhR exhibits strong tumor suppressor roles in colon cancer (14,16). Interestingly, knockout of CABLES1 in a colon cancer mouse model increased oncogenic Wnt/B-catenin signaling and increased cancer progression (60). Therefore, the role of CABLES1 in cell cycle regulation and tumor suppression by the AhR needs further investigation.

Gene set enrichment analysis showed that 11-Cl-BBQ suppressed genes in several pathways that are important for cell G1-S phase cell cycle progression including, G1-S phase specific transcription, DNA strand elongation, lagging strand synthesis, and E2F-mediated regulation of DNA replication (Fig. 5). Importantly, the suppression of these pathways was AhR-dependent. Our results are consistent with a previous study showing that treatment of Cl-BBQ (40:60 mixture of 10- and 11-Cl-BBQ) also suppressed genes enriched in cell cycle regulation pathways resulting in inhibiting the proliferation of CD4+ T cells in mice (29). Similarly, ligand-activated AhR has been shown to interact with the tumor suppressor retinoblastoma (RB) protein leading to suppression of E2F signaling and resulting in a G1 cell cycle arrest (67–69).

Gene ontology analysis of RNA-seq data in the current study at the transcript-level identified involvement of the tumor suppressor p53 signaling in AhR-regulated cell cycle arrest (Fig. 5E). Further investigation revealed that 11-Cl-BBQ induces p53 signaling, and that p53 was required for 11-Cl-BBQ-induced growth arrest (Figs. 6 and 7). Although it was not statically significant after applying correction for multiple comparisons, we observed a slight increase in G2/M phase in p53 knockout cells treated with 11-Cl-BBQ. A shift from G1 to G2 arrest in cells lacking p53 has been reported by others (43). A plausible explanation for this effect is that the inducing event of p53 signaling (e.g., a small molecule or ionizing radiation) may cause both G1 arrest (p53-dependent) and a G2 arrest (p53-independent). In the presence of p53, the p53-dependency dominates the effect and results in G1 arrest, whereas G2 arrest becomes more obvious in the absence of p53.

Interestingly, activation of the AhR by TCDD, the most potent and well-studied AhR ligand, does not activate p53 and has minimal effects on lung cancer cell proliferation (data not shown). The mechanism by which 11-Cl-BBQ activates p53 via AhR is yet to be investigated. H460 cells express wildtype p53 and are highly responsive to 11-Cl-BBQ. However, H69AR multidrug resistant cells harboring a p53 c.551G>T (p.E171*) mutation are still sensitive. In addition, we have recently reported that the AhR has a prominent role as a tumor suppressor in the absence of p53 in a mouse model (70). We reported significantly increased tumor incidences and reduced survival in p53-deficient mice associated with the loss of AhR. These data indicate an interplay between these pathways that needs to be investigated to have a better understanding for the role of AhR in cancer.

In conclusion, AhR mediates the growth suppression of 11-Cl-BBQ in lung cancer cells. Up-regulation of cell cycle inhibitor p27^Kip1 and activation of p53 tumor suppressor signaling are important contributors to the potent anti-proliferative effects of 11-Cl-BBQ (Fig. 9). The present study opens up new opportunities to investigate SMAhRTs, and in particular 11-Cl-BBQ and their analogs, as cancer therapeutics.

Figure 9:

11-chloro-7H-benzimidazo[2,1-a]benzo[de]iso-quinolin-7-one (11-Cl-BBQ) has been identified as a Select Modulator of AhR-regulated Transcription (SMAhRT) with anti-cancer actions. 11-Cl-BBQ promoted potent and sustained AhR-dependent anti-proliferative effects in cancer cells by activation of p53 signaling. 11-Cl-BBQ induced AhR- and p53-dependent increase in the senescence-associated beta galactosidase (SA-β-gal), a biomarker for an irreversible cell cycle arrest.

Materials and Methods

Cell culture

Lung cancer cell lines A549, H460, H146, H1299 and H69AR were purchased from ATCC and maintained in RPMI 1640 medium, except the A549 cells were maintained in F12K medium, supplemented with 10% fetal bovine serum (FBS). HEK293T cells were cultured in DMEM medium supplemented with 10% FBS. Cell cultures were split twice or three times a week when cells reached 70–80% confluent. Cell were cultured at 37°C and 5% CO₂ in humidified cell incubator.

Viability assay

Cancer cells were seeded in 96-well plates at a density of 3,000 cells per well in their optimal culture medium supplemented with 1% FBS overnight, then compounds were added at indicated concentrations and incubated for 72 hours, otherwise, different treatment times were specified. Cell viability was measured using the ATP-based CellTiter Glo assay (Promega - Madison, WI) following the manufacturer’s protocol. Luminescent signals were recorded using the Tropix TR717 microplate luminometer. The viability of cultures following each treatment was normalized to vehicle treatment (0.1% DMSO (v/v)).

Colony forming assay

H460 cells were seeded in 6-well plates at a density of 200 cell per well in complete growth media. 11-Cl-BBQ or DMSO were added the following day as indicated concentrations for two weeks in 1%FBS RMPI media, media and treatments were replenished twice a week. Cell colonies were fixed and stained with methylene blue in 50% ethanol solution overnight. Pictures of cell colonies were captured using the ChemiDoc imaging instrument, and quantified using OpenCFU software (71).

Primary antibodies and Western blot

Whole cell lysate was prepared in RIPA lysis buffer with 1X Laemmli buffer, then denatured by heating at 100°C for 5 minutes. Denatured proteins were separated in SDS-PAGE gel and transferred onto PVDF membrane. The PVDF membrane was incubated in TBST with 5% nonfat dry milk for one hour at room temperature prior to incubating with primary antibody of interest overnight at 4°C with gentle rotation. Appropriate horse radish peroxidase conjugated secondary antibody was applied for one hour at room temperature. Signal was developed by applying SuperSignal West Pico chemiluminescence substrate and captured using the ChemiDoc imaging instrument. Primary antibodies used in this study: anti-AHR antibody (BML-SA210 – Enzo Biochem, Farmingdale, NY), anti-p27Kip1 antibody (610242 – BD Biosciences), anti-p53 antibody (DO-1 sc-126 – Santa Cruz Biotechnology, Dallas, TX), anti-p21 antibody (2947S – Cell Signaling Technology, Danvers, MA), and anti-GAPDH antibody (sc-36502, Santa Cruz).

Generation of stable knockout cell lines using CRISPR-cas9

Guide RNAs against human AhR, CDKN1B, and TP53 genes in one vector CRIPSR-cas9 lentiCRISPRv2 plasmid have been described (31) and were purchased from GeneScript (Piscataway, NJ). Stable AhR, CDKN1B (p27) and TP53 (p53) knockout lines were generated using CRISPR-cas9 system as previously described in detail (72). Briefly, for each gRNA, viral particles were generated by co-transfecting CRISPR-cas9 sgRNA plasmid with packaging plasmids psPAX2 and pMD2.G (Addgene, Watertown, MA) into HEK293T cells using lipofectamine 2000 (ThermoFisher, Waltham, MA). The viral particles were collected twice on the next two days for a total 4 ml of supernatant containing viral particles, frozen in −80°C overnight, filtered through 0.2 μm filter, divided into 0.5 ml aliquots and kept in −80°C. H460 or H69AR cells were transduced by reverse transduction with the viral particles in RPMI 1640 medium supplemented with 1% FBS and 10 μM protamine sulfate. Infected cells were selected for using 0.5 μg/ml puromycin for one week. Monoclonal cell lines were generated by limited dilution in 96-well plates. After 10 days in culture, cells were checked under a microscope and wells containing only one clone were selected for further expansion. Knockout phenotype was confirmed by Western Blot. Guide RNAs against AhR have been described in our previous study (55); guide RNAs targeting human p27 and p53 genes: p27 gRNA-1 ATT GCT CCG CTA ACC CCG TC; p27 gRNA-2 GGG TTA GCG GAG CAA TGC GC; p27 gRNA-3 TTC CCC AAA TGC CGG TTC TG; p53 gRNA-1 CCG GTT CAT GCC CAT GC; p53 gRNA-2 CGC TAT CTG AGC GCT CA; p53 gRNA-3 CCC CGG ACG ATA TTG AAC AA.

Cell cycle analysis

Cells were harvested by applying trypsin-EDTA then washed twice with phosphate-buffered saline (PBS) solution, fixed with 70% ethanol for 15 min at room temperature and kept in −20°C overnight. Fixed samples were washed twice with PBS and stained with Hoechst 33258 (Invitrogen, Carlsbad, CA) at the final concentration of 1 μg/ml in the 0.1% Triton X-100/PBS for 20 minutes at room temperature. DNA content was analyzed by flow cytometer for 10,000 cells per sample, singlets population was selected for further analysis and cell cycle was assigned by fitting the DNA content into Dean-Jett-Fox model using FlowJo software.

Senescence-associated beta-galactosidase (SA-β-gal) staining

H460 cells were seeded at 2×10⁵ cells/well in 6-well plate overnight and treated with indicated compounds for 5 days, media and treatments were refreshed on the day 3^rd. Cells were stained for senescence-associated β-galactosidase (SA-βgal) using 5-dodecanoylaminofluorescein di-β-D-galactopyranoside (C₁₂FDG) as previously described in detail (32). Briefly, media were aspirated and replaced with fresh media containing 0.1 μM bafilomycin A1 (Calbiochem) and incubated for 1 hour in cell culture incubator. C₁₂FDG (Setare Biotech) was added with a final concentration of 33 μM and incubated for another hour. Fluorescent signal was recorded using CytoFlex S cytometer for 10,000 cells per sample.

RNA isolation and quantitative reverse transcription PCR (qPCR)

Total RNA samples were isolated using the E.Z.N.A. Total RNA Kit (Omega Bio-tek, Norcross, GA) following the manufacturer’s protocol. One microgram of total RNA was used to make cDNA using the Transcriptor First Strand cDNA Synthesis Kit (Roche, Basel, Switzerland) in a total volume of 20 μl. The cDNA (0.5 μl) was used for each qPCR reaction with the FastStart Universal SYBR Green Master (Rox) (Roche) in a total volume of 20 μl using standard thermal cycling on 7500 Fast real time PCR system (ABI Applied Biosystems). qPCR primers specific for human CYP1A1, CDKN1A (p21), CDKN1B (p27^Kip1), and GAPDH as described in our previous study (57). qPCR primers for other genes were designed and checked for target specificity using Primer-BLAST tool (73); these primer sequences were: CYP1B1 forward 5’-AAG TTC TTG AGG CAC TGC GAA-3’; reverse 5’-CCG GTA CGT TCT CCA AAT CC-3’, AHRR forward 5’-ATC GTG GAC TAT CTG GGC TTC-3’; 5’-TTC TGG TGC ATTA CAT CCG TC-3’, CABLES1 forward 5’-TCA AGA ACA TGC GGC AAC AC-3’; reverse 5’-GGG TAC TGT CGC GAT ACG G-3’, SHISA2 forward 5’-CGG CTG CGA CAA TGA CCG-3’; reverse 5’-CAA TGA GGA ACG GCA CGT AG-3’,IL1A forward 5’-AGT AGC AAC CAA CGG GAA GG-3’; reverse 5’-AAG GTG CTG ACC TAG GCT TG-3’, CCNE2 forward 5’-TCA CTG ATG GTG CTT GCA GT-3’ ; reverse 5’-GCC AGG AGA TGA TTG TTA CAG GA-3’, E2F7 forward 5’-TCG CTC TCC CTT CCC GAT G-3’; reverse 5’-ACC TCC ATC CCT GCT TTC CTA AG-3’.

AhR chromatin immunoprecipitation (ChIP)

H460 cells were seeded overnight in RPMI media supplemented with 10% FBS and were treated with 11-Cl-BBQ (2.5 μM), TCDD (30 nM), or 0.1% DMSO as vehicle control for one hour. Chromatin immunoprecipitation (ChIP) was carried out using the MAGnify™ Chromatin Immunoprecipitation System (Invitrogen) following the manufacturer protocol. Rabbit polyclonal antibody anti-AhR (BML-SA210 – Enzo Biochem) and rabbit IgG provided with the ChIP kit was used as control antibody. The enrichment of AHR at CYP1A1 and p27 (CDKN1B) promoters was measured using qPCR method as described earlier (74). qPCR primers for p27 promoter and CYP1A1 promoter using qPCR with primers p27forward 5’-GGC CGT TTG GCT AGT TTG TT-3’; reverse 5’-GAG ATT GGC TGG TCG CGT-3’, CYP1A1 forward 5’-TGC CCA GGC GTT GCG TGA GAA G-3’; reverse 5’-ACC CGC CAC CCT TCG ACA GTT C-3’. P27 primers were designed to cover a putative AHR binding site in the promoter region of p27 that has been described in a previous study (56), CYP1A1 primers have been described from a previous study (75) that are specific for the AHR binding site at CYP1A1 enhancer region.

Global transcriptomic study using RNA-seq

Total RNA samples were isolated at the indicated time using the E.Z.N.A. Total RNA kit as mentioned above. The samples were quantified using nanodrop method, and the three biological samples were combined with the same molarity and used as a representative sample for each treatment. The mixed samples were quantified using the Bioanalyzer 2100 instrument with the RNA Nano chip. All the samples had RIN numbers of 10, except one sample with a RIN of 9.9. For each sample, one microgram of total RNA was used to select for messenger RNAs using the Takara PrepX PolyA mRNA isolation kit. Libraries of mRNA were prepared using the Takara PrepX RNA-seq for Illumina Library Prep kit. The final libraries were quantified using the HS-D5000 tape on the Tapestation 4200 instrument then mixed together with an equal molarity and sequenced using one lane of a Hi-Seq 3000 flow cell for 150-bp paired-end reads. RNA libraries were made and sequenced at the Center for Genome research and Bioinformatic.

RNA-seq reads were pseudo-aligned to the reference human transcriptome (release 92 from Ensembl (78)) and quantified using kallisto with default settings (77). Gene set enrichment analyses (GSEA) were performed using the GSEA java application for desktop (58) using the normalized transcript per million (TPM) values for each treatment, in AhR wild type and AhR knock out cells separately, using the single sample GSEA projection with curated canonical pathway gene sets (version v6.2) (38,79).

Zebrafish xenograft experiments:

Zebrafish were housed at the Sinnhuber Aquatic Research Laboratory at Oregon State University (SARL; Corvallis, OR, USA), and experiments were conducted according to the IACUC-approved Animal Care and Use Protocol ACUP 5113. H460 AHR wildtype (CR-V2) and knockout (CR-AHR3) cells were treated with 11-Cl-BBQ (1 and 5 μM) or 0.1% DMSO as vehicle control for 72 hours in RPMI media supplemented with 1% FBS. Cells were trypsinized, dyed with cell viable dye CM-Dil (Thermofisher), and micro-injected into zebrafish embryos. The growth cancer xenografts were monitored at one day and four days post injection as described in detailed in our previous studies (36,37). Images were taken with a 10x air objective using a Zeiss LSM 780 confocal microscope.

Statement of Data Availability:

The data that supports the key findings of this study are in Figures 2A, 2C, 3A, 3C, 4C, 7A, 7E, 8E, and 8G.

Abbreviations:

AhR

Aryl Hydrocarbon Receptor

ARNT

Aryl hydrocarbon Receptor Nuclear Translocator protein

11-Cl-BBQ

11-chloro-7H-benzimidazo[2,1-a]benzo[de]iso-quinolin-7-one

bHLH

basic-helix-loop-helix

ChIP

Chromatin immunoprecipitation

CR-AHR

AhR-specific single guide RNA

CRISPR

Clustered regularly interspaced short palindromic repeats

CR-V2

Control vector

FDR

False discovery rate

GO

Gene ontology

GSEA

Gene set enrichment analysis

mRNA

Messenger RNA

PAS

Per-AhR/Arnt-Sim (Per: Period gene, Sim: Simple Minded gene)

qPCR

Quantitative real-time polymerase chain reaction

SA-β-gal

Senescence-associated beta galactosidase

sgRNA

Single guide RNA

SMAhRT

Selective Modulator of AhR-regulated Transcription

TCDD

2,3,7,8-tetrachlorodibenzo-p-dioxin

TCGA

The Cancer Genome Atlas

TRAMP

Transgenic adenocarcinoma of the mouse prostate