The aryl hydrocarbon receptor repressor is a putative tumor suppressor gene in multiple human cancers

The aryl hydrocarbon receptor repressor (AHRR) is a bHLH/Per-ARNT-Sim transcription factor located in a region of chromosome 5 (5p15.3) that has been proposed to contain one or more tumor suppressor genes. We report here consistent downregulation of AHRR mRNA in human malignant tissue from different anatomical origins, including colon, breast, lung, stomach, cervix, and ovary, and demonstrate DNA hypermethylation as the regulatory mechanism of AHRR gene silencing. Knockdown of AHRR gene expression in a human lung cancer cell line using siRNA significantly enhanced in vitro anchorage-dependent and -independent cell growth as well as cell growth after transplantation into immunocompromised mice. In addition, knockdown of AHRR in non-clonable normal human mammary epithelial cells enabled them to grow in an anchorage-independent manner. Further, downregulation of AHRR expression in the human lung cancer cell line conferred resistance to apoptotic signals and enhanced motility and invasion in vitro and angiogenic potential in vivo. Ectopic expression of AHRR in tumor cells resulted in diminished anchorage-dependent and -independent cell growth and reduced angiogenic potential. These results therefore demonstrate that AHRR is a putative new tumor suppressor gene in multiple types of human cancers.

The aryl hydrocarbon receptor repressor (AHRR) is a newly discovered member of the growing superfamily of basic helix-loop-helix/Per-ARNT-Sim (bHLH/Per-ARNT-Sim) transcription factors, which includes the aryl hydrocarbon receptor (AHR) (1) and hypoxia inducible factor 1 (HIF1) (2), among others. AHRR represses the transcription activity of AHR by competing with this transcription factor for heterodimer formation with the AHR nuclear translocator (ARNT) (3) and subsequently binding to the xenobiotic response element (XRE) sequence (4), functioning as a naturally occurring dominant-negative factor. Furthermore, the expression of AHRR is induced by the AHR/ARNT heterodimer (5) through binding to XREs located in the 5′-flanking region of the AHRR gene. Therefore, AHRR and AHR constitute a regulatory loop in which the heterodimer AHR/ARNT activates expression of the AHRR gene, while the expressed AHRR inhibits the function of AHR (4).

Proof of the role of AHR as a ligand-activated transcription factor that mediates a pleiotropic response to environmental contaminants, such as benzo[a]pyrene, 2,3,7,8-tetrachlorodibenzo-p-dioxin, and tobacco smoke is extensive (6, 7). More importantly, data exist implicating AHR in the development of cancers from different anatomical origins (8–10). Compelling evidence of the tumorigenic potential of AHR comes from the fact that AHR-overexpressing transgenic mice spontaneously develop tumors (8, 9), whereas AHR-knockout mice have increased resistance to carcinogens (11). In this context, as an AHR regulator, AHRR is potentially involved in a vast array of normal and pathological processes ranging from xenobiotic response to tumor progression. In humans, AHRR maps to the short arm of chromosome 5 (5p15). This chromosomal region has been shown to be frequently deleted in a variety of tumor types, such as cervical and testicular germ tumors (12), colorectal cancer (13), early-stage ovarian tumors (14), bladder cancer (15), esophageal cancer (16), and lung tumors (17). Based on these loss-of-heterozygosity (LOH) studies, 5p15 has been often proposed as a chromosomal locus which likely harbors at least one tumor suppressor gene. The role of AHRR in carcinogenesis has not been extensively explored. Overexpression of AHRR in the breast cancer cell line MCF-7 results in growth inhibition (18) and genetic polymorphisms in AHRR are related to susceptibility to advanced endometriosis (19). Here we provide clinical, genetic, and functional evidence from in vitro and in vivo experiments supporting a critical role of AHRR as a putative tumor suppressor gene in several types of human cancer.

AHRR is downregulated in tumor cells. Our results show a consistent downregulation of AHRR throughout all the tumor types assessed in this study, including colon, breast, lung, stomach, cervical, and ovarian, when compared with normal tissues of the same anatomical origin (Figure 1). In preliminary experiments lung, breast, stomach, and colon primary tumors showed substantial downregulation of AHRR when compared with normal tissue from the same patient, adjacent to the tumor (Figure 1A). Interestingly, some of the hyperplastic or adenomatous nonmalignant colon polyps showed a smaller (up to 40%) downregulation of AHRR (Figure 1B), suggesting a direct relationship between levels of AHRR expression and degree of transformation in colon cancer. In a separate experiment, we found that 80% of cervical primary tumors and 100% of cervical cell lines showed significant reduction or complete abrogation of AHRR when compared with normal tissue (Figure 1C). Similar results were obtained with 14 ovarian carcinoma cell lines examined, which showed a strong downregulation of AHRR when compared with immortalized nonmalignant ovarian cell lines (Figure 1D). Additional proof of downregulation of AHRR mRNA in cancer comes from comparison of 11 lung tumor samples and normal tissue from the same patients (Figure 1E). All tumor samples showed significant silencing of AHRR mRNA when compared with their normal counterparts. Overall, these results show a consistent downregulation of AHRR mRNA in tumors of different histological origins (Supplemental data; supplemental material available online with this article; doi:10.1172/JCI30024DS1).

Figure 1

AHRR mRNA expression levels in tumors and normal controls as assessed by real-time PCR. (A) Significant downregulation in AHRR mRNA (between 30% and 90%) was observed in tumors from several origins. Data are presented as a ratio of mRNA levels in tumor versus normal control. (B) Primary colon tumors showed a very strong downregulation of AHRR mRNA when compared with normal controls. Interestingly, nonmalignant colon polyps exhibited a more moderate downregulation of AHRR mRNA when compared with colon tumors. Statistically significant differences were achieved when comparing levels of AHRR in normal tissue with levels in polyps. (C) AHRR expression levels in normal cervical tissue, cervical tumor cell lines, and primary cervical tumors. Strong downregulation or complete ablation of AHRR mRNA was observed in 100% of cell lines and 80% of the primary tumors when compared with normal controls. Little variation was observed in the levels of AHRR mRNA among the normal controls. (D) All ovarian tumor cell lines included in this study showed downregulation of AHRR mRNA levels when compared with normal ovarian cell lines. (E) Box-and-whisker plot showing significant downregulation AHRR mRNA levels observed in lung tumors and normal tissues from the same patients. ***P < 0.001.

AHRR satisfies Knudson’s “2-hit” hypothesis. Because of the downregulated expression of the AHRR gene in tumors from different tissue origins, and the frequent LOH at the 5p15 chromosomal region (12), we reasoned that AHRR might be a potential tumor suppressor gene. Knudson’s “two hits” hypothesis considers tumor suppressor genes by their functional inactivation achieved by a combination of LOH, hypermethylation of the gene’s promoter, and mutations in the gene’s coding region (20). Although, as expected from previous studies (12, 21), a number of polymorphisms and evidence for frequent LOH of this region was obtained in this study (Supplemental Table 2), no evidence of pathogenic mutations in the coding region of the AHRR gene was found, pointing to hypermethylation as a possible mechanism responsible for its steady downregulation in tumors. Promoter hypermethylation was found in over 78% of the tumors examined (Table 1 and Supplemental Figure 2), while no methylation was found in normal tissue. Remarkably, almost all ovarian, cervical, and colon tumors exhibited promoter hypermethylation. The lowest hypermethylation rate (13%) was found in seminoma testicular tumors (it has been reported that seminomatous germ cell tumors rarely exhibit promoter hypermethylation; ref. 22). Interestingly, levels of promoter hypermethylation were closely correlated with cancer grade in cervical and esophageal specimens: low-grade precancerous lesions showed the lowest levels of promoter methylation (4.2% and 0%, respectively), higher levels were observed in high-grade lesions (41% and 28%, respectively), and virtually all established cervical tumors (95.5%) and 52% of esophageal tumors were methylated. Further supporting evidence that hypermethylation is a main regulatory mechanism of AHRR silencing in tumors comes from experiments in which tumor cells are exposed to demethylating agents. Treatment of tumor cells with 5-aza-dC and/or trichostatin A significantly reactivated AHRR expression (more than 1.5-fold increase) in 71.5% of the tumor cell lines included in the experiment (Supplemental Table 3). These data thus show that epigenetic alterations (i.e., hypermethylation) are largely responsible for downregulation of AHRR in cancer cells. Analysis of the 5′ regulatory region of AHRR showed several binding motif sites were found in the same region affected by hypermethylation including AP-2, c–Est-1, and Elk-1 (Supplemental Table 4). Especially interesting was a short sequence, located close to the star codon (–55 to –41 bp from the start codon), which contained potential binding sites for several members of the bHLH family of transcription factors including AHR and HIF.

Table 1

AHRR promoter hypermethylation in cancer

The collective data shown above strongly suggest that AHRR may function as a tumor suppressor gene in cancers from different tissue origins. It is assumed that the loss of tumor suppressor gene function correlates with an aggressive tumorigenic phenotype (23). siRNA technology in combination with in vitro cell-based growth assays (such as anchorage-independent growth assay) has been recently recognized as a successful strategy for identification of new tumor suppressor genes (24). In this context, and to better understand the functional relevance of AHRR as a tumor suppressor gene, we artificially suppressed its expression by siRNA technology in the lung tumor A549 and normal breast MCF10A cell lines (Supplemental Figure 3). Transfected cells were tested for a number of established hallmark traits of the tumorigenic phenotype such as growth and colony formation both in vivo and in vitro and resistance to apoptosis, migration, and angiogenic potential (25).

Silencing of AHRR enhances tumor growth in vitro and in vivo through deregulation in cell cycle control. The hallmark of all neoplastic development is deregulated cell proliferation (25). Downregulation of AHRR resulted in enhanced cell growth potential (Figure 2A). After 5 days of transfection, a difference of 4.3- and 2.6-fold in growth was observed between A549E and A549G/F, respectively. Cell cycle analysis on these cells showed that the cell number on G₂/M phase of A549F/G was significantly higher than that of A549E (Figure 2, B–D). Consistently, G₀/G₁ phase was significantly decreased in these cells. In addition, we determined that several prooncogenic factors were significantly elevated on A549G, including Ras, PKCa, and cyclin A, while tumor suppressor factors such as PTEN and Ku80 were downregulated (Table 2). Further strengthening these results, downregulation of AHRR enhanced the ability of A549F and A549G to form colonies on anchorage-independent clonogenic assays (162% and 306% over A549E, respectively) (Figure 2E). Of note, the enhanced clonogenic capability was accompanied by differences in the morphology of the colonies. A549E typically grew in solid spheroids with defined colony contours (Figure 2G), whereas some of the colonies formed by A549F/G showed irregular contours and cells detached from the main core of the colony (Figure 2H). To further explore the link between AHRR silencing and clonogenic potential, we artificially silenced AHRR in the immortalized nonmalignant breast epithelial cell line MCF10A. In agreement with a previous report (26), MCF10A-E failed to form colonies in soft agar (Figure 2F). In contrast, MCF10A-F/G were able to grow in colonies displaying similar morphology to the one observed in A549F/G.

Figure 2

Effects of siRNA-induced silencing of AHRR on growth. (A) On a time-course MTT assay, A549G (squares) grew 4.3-fold and A549F (triangles) grew 2.6-fold faster than the empty nonsilenced A549 cells (circles, A549E; diamonds, A549SR). No differences were observed between A549E and A549SR. (B–D) Cells deficient in AHRR (A549F [C] and A549G [D]) showed a significant shift in the number of cells towards S and G₂/M phases as compared with the control A549E (B). Consistently, a reduction in the number of cells in G₀/G₁ phase was observed in A549F/G. (E and F) Artificial downregulation of AHRR enhanced A549 (E) and MCF10A (F) colony formation (60% and 300% increase in average colonies for A549F/G and A549E, respectively). MCF10A-F and MCF10A-G were able to clone in soft agar, while, as expected, MCF10A-E did not form colonies. All clonogenic assays were run in triplicate. (G and H) Comparison of the morphology of the colonies formed by A549E (G) and A549G (H). A549E grew in compact spheroids with defined contours. A549G colonies showed irregular morphology and cells detached and partially separated from the core of the colony. Original magnification, ×20. (I) In an in vivo experiment, A549 cells transfected with a siRNA for AHRR showed a significant increase in xenograft tumor volume 8 weeks after injection when compared with empty plasmid–transfected A549 cells (3-fold increase for A549F [squares] compared with A549E [triangles] and 7-fold increase for A549G [diamonds] compared with A549E). n = 10 animals/group. *P < 0.05; **P < 0.01; ***P < 0.001.

Table 2

Comparison of the protein expression levels between A549G and A549E

Overall, the data described above clearly demonstrate a link between AHRR downregulation and tumor cell growth in vitro. To further address this finding in an in vivo model of tumorigenesis, we injected A549E/F/G cells into nude mice and followed the development of xenograft tumors over time. A549F- and A549G-induced tumors showed a 3- and 7-fold increase in volume, when compared with tumors generated by A549E (Figure 2I). Smaller differences were found between A549E and A549F (3-fold increase). These data further suggest that downregulation of AHRR is a contributing factor to enhanced tumorigenicity.

AHRR silencing protects against apoptosis and enhances angiogenic potential, migration, and invasion in tumor cells. In addition to growth deregulation and clonal expansion, resistance to apoptosis (27), enhanced angiogenic potential (28), and motility (25) lie at the heart of all tumor development. Here we show that silencing of AHRR confers resistance to apoptosis induced by overnight exposure to various proapoptotic signals (Figure 3A). Proteomics expression profile of A549G cells showed a significant elevation of the antiapoptotic factor Bcl-x (Table 2), suggesting potential explanatory mechanisms. We also studied the angiogenic potential of transfected A549 cells in vivo using DIVAA. An 8-fold increase in neovascularization-related fluorescence was observed in A549G when compared with A549E (Figure 3B). Similarly, A549F exhibited higher angiogenic potential (5-fold increase) compared with A549E. Although sprouting of capillaries from quiescent vasculature on the tumor is a prerequisite for metastasis (28), enhanced motility and invasiveness are also required. The morphological alterations observed in A549F/G in clonogenic assays (Figure 2, F and G) suggest loss of cell-cell contact within the colonies and possibly higher migratory potential. Additionally, A549F/G showed increased (7- and 4.5-fold increase, respectively) migratory potential over A549E (Figure 3C). Similar results were obtained when testing the invasive potential of A549-transfected cells. Silencing of AHRR correlated with increased invasion, as observed through Matrigel staining (Figure 3C). These results are consistent with upregulation of prometastatic factors such as vimentin or phosphorylated focal adhesion kinase (FAK) and loss of E-cadherin, which is associated with loss of cell-cell interactions and enhanced invasive phenotype, in cells with artificially silenced AHRR (Table 2).

Figure 3

Effects of siRNA-induced silencing of AHRR. (A) A549E (black bars), A549F (white bars), and A549G (gray bars) were incubated in serum-free media (R0) or the appropriate media containing α-Fas or MK886. In all experiments (n = 3), A549F/G showed enhanced resistant to proapoptotic signals when compared with A549E. (B) DIVAA analysis showed that A549F and A549G have enhanced angiogenic potential when compared with A549E, and this effect was more prominent for A549G than for A549F (n = 8). (C) Silencing of AHRR (A549F/G) enhances the migratory (black bars) and invasive (white bars) potential of the A549 tumor cell line (n = 6). Detail photos of the porous membranes used to determine migratory potential of A549E and A549G are also shown. Original magnification, ×10. *P < 0.05; **P < 0.01; ***P < 0.001.

Ectopic expression of AHRR opposes the effect of siRNA for AHRR. Overexpression of AHRR was achieved in a transient fashion in A549 lung adenocarcinoma cells (Supplemental Figure 3C). AHRR was initially described as a repressor of the AHR, and our hypothesis contemplates a suppressor activity of AHRR linked to the AHR tumorigenic potential. To confirm biological activity of the overexpressed AHRR, we measured levels of the AHR-induced cancer-related gene CYP1A1. As expected, transfection of AHR-induced CYP1A1 expression and cotransfection of AHRR counteracted the effect of AHR on reducing the levels of CYP1A1 (Figure 4A), therefore confirming biological activity of the overexpressed AHRR as a repressor of AHR. Interestingly, A549-AHRR showed reduced growth rate over 3 days compared with A549-empty, as assessed by Electric Cell-Substrate Impedance Sensing (Figure 4B). Consistently, AHRR transfer resulted in inhibition of anchorage-independent growth in clonogenic assays (Figure 4C). Colonies were morphologically similar and reached the same size (Figure 4B), although significantly fewer colonies were found in A549-AHRR samples. Also, using a modified tube formation assay in which PAE-GFP were cocultured (in separate Matrigel compartments) with transfected A549 cells (Figure 4D), we demonstrated reduced tube formation after exposure to A549-AHRR compared with cells exposed to A549-empty, showing that overexpression of AHRR reduced the angiogenic potential of A549.

Figure 4

Effect of ectopic expression of AHRR in A549. (A) Overexpression of AHRR partially blocked expression of AHR-induced genes such as CYP1A1 (plasmid amounts expressed in μg). (B) Increased levels of AHRR in A549 resulted in reduced anchorage-dependent growth. The blue line represents A549-AHRR, and the red line represents A549-empty. Dotted lines denote standard deviation (n = 4). The maximum difference was reached 56 hours after inoculation of cells in the wells. (C) Ectopic expression of AHRR resulted in reduced anchorage-independent growth (adjacent images). Original magnification, ×10 (C); ×2 (D). (D) Representative images of tube formation assays using cocultures of A549-empty or A549-AHRR and PAE-GFP cells. A549-empty induced tube formation of PAE-GFP cells comparable to that of the internal positive control (10% FBS). Reduced tube formation was observed when PAE-GFP cells were cocultured with A549-AHRR. **P < 0.01; ***P < 0.001.

Tumor formation arises as a consequence of the acquisition of unique cellular capabilities, which involve, among others, deregulation of cell proliferation, resistance to apoptosis, enhanced cell motility, augmented angiogenic potential, and anomalies in cell-cell interaction and the microenvironment, resulting in invasion and metastasis (25). Over the last several years, we have learned that this process is regulated by a relatively small subset of genes that act by either enhancing (oncogenes) or diminishing (tumor suppressor genes) the final malignant outcome (29). Finding tumor suppressor genes is challenging, and although approaches based on analysis of LOH have been proven to be successful (29), we now know that multiple lines of evidence, including functional implications, are needed to confirm the identification of newly defined tumor suppressor genes (30). Here we provide clinical, genetic, and functional evidence supporting the hypothesis that silencing of the AHRR occurs in tumors from different tissue origins and that this mute expression results in an aggressive tumorigenic phenotype, thus suggesting that AHRR plays an important role in suppressing tumor formation in humans.

In this study we demonstrate downregulation of AHRR throughout all the tumor types assessed, including colon, breast, lung, stomach, cervical, and ovarian, when compared with normal tissues of the same origin. Functional inactivation of tumor suppressor genes is achieved by a combination of LOH, hypermethylation of the gene’s promoter, and mutations (20, 31). We have found high rates of LOH (together with several new polymorphisms) and hypermethylation in the promoter region of the AHRR gene in cervical, testicular, and ovarian tumors. The majority of tumors, particularly ovarian cancer and testicular germ cell tumors, exhibited only methylated alleles, which can result from LOH of one allele and methylation of the other or methylation of both alleles. The first mechanism may certainly account for a proportion of cases, since 5p15 LOH is a common event in cervical cancer and testicular germ cell tumors (12, 21, 32–34). These data, together with the reactivation of AHRR expression after exposure to demethylating agents, are consistent with Knudson’s 2-hit hypothesis (20, 31) and indicate that LOH and hypermethylation constitute the main mechanism for the silencing of AHRR in tumors. Inactivation of tumor suppressor genes can represent an early event and a prerequisite for clonal expansion. In support of this idea we, have found a direct relationship between hypermethylation of the AHRR promoter region and tumor grade in cervical and esophageal cancer. Further supporting this idea, a small set of colon polyps (which we can consider as a naturally occurring example of premalignant lesions) showed intermediate downregulation of AHRR compared with colon tissues from healthy patients and frank colon tumors. This leads us to postulate that a reduced dosage of AHRR in the polyps, rather than its absolute absence, may contribute to cancer susceptibility in initial stages of the carcinogenic process.

Hypermethylation in the neighborhood of the start codon has been shown to suppress gene expression in tumor cells by either interfering with RNA polymerase II initiation or transcription factor binding (35). This region is therefore more likely to contain binding sequences for transcription factors relevant in AHRR regulation in tumor cells. Several regions of the promoter of the AHRR gene have been shown to be potentially relevant for the regulation of its expression (36, 37). We have found that the sequence contained between –55 bp and –41 bp from the start codon of the AHRR gene (which we have shown to be hypermethylated in tumors) is rich in potential binding sites for several members of the bHLH family of transcription factors, some of which are known to be upregulated in tumor cells (AHR, c-myc/MAX, and HIF-1). Upregulation of these transcription factors in cancer cells would potentially correlate with an increase in the levels of AHRR and an enhanced cancer suppressive action. However, hypermethylation of this promoter region and LOH prevent AHRR expression with concomitant loss of its repressor function and enhancement of the tumorigenic phenotype (Figure 5). Due to the significant length of the regulatory region on the AHRR gene, additional studies targeting this area are guaranteed to deepen our understanding of the role that methylation plays on AHRR expression in tumor cells.

Figure 5

Working model of the tumor suppression mechanism for AHRR integrated in a gene regulation feedback loop with AHR. (A) In normal cells, activation of AHR causes induction of AHRR through XREs present in its promoter. AHRR acts as a negative regulator of AHR by competing for binding to ARNT and XREs present in the promoter regulatory region of variety of genes. The balanced relationship between the positive and negative transactivation signals of AHR and AHRR results in cellular homeostasis. (B) In cancer cells, methylation of the AHRR promoter and LOH cause blockade of its expression despite the higher levels of AHR. Thus absence of AHRR eliminates competition for binding to ARNT and XREs, which results in an imbalance between positive and negative transactivation signals, thereby causing the induction of a battery of genes related to tumorigenesis and cancer progression.

Functional evidence of the tumor suppressive nature of AHRR arises from the study of the malignant phenotype in cells with artificially altered AHRR expression. Independent of the etiology of different tumors, deregulated cell proliferation together with suppressed apoptosis constitute a common platform upon which all neoplastic evolution occurs. Here we demonstrate that downregulation of AHRR enhances growth potential through deregulation in cell cycle control and resistance to apoptosis (potentially through deregulation of cyclin A and Bcl-x, respectively). AHRR inactivation is also linked to enhancement in anchorage-independent growth of tumor cells. Artificial upregulation of AHRR results in impaired anchorage-dependent and -independent growth. In agreement with these data, a recent report shows that overexpression of AHRR results in blockade of growth in the breast cancer cell line MCF-7 (18). Interestingly, partial silencing of AHRR in nontumoral MCF10A (which lack the ability of anchorage-independent growth; ref. 26) enables them to clone in soft agar, further supporting the important role of AHRR as a negative regulator of clonogenicity. Recently it has been shown that immortalized mouse mammary fibroblasts lacking AHR have impaired tumorigenicity in a subcutaneous mouse xenograft model (38) and that its overexpression promotes progression through the cell cycle (39), triggering a malignant phenotype in vivo (8, 9). Tumors cells showing inactivation of the AHR negative regulator AHRR showed enhanced tumorigenic potential, while AHRR activation resulted in diminished tumorigenic potential, in agreement with the logical hypothesis that inactivation of AHRR should lead to similar physiological consequences. The proteomic profile of cells silenced for AHRR shows that several prooncogenic factors such as Ras, HES-1, and casein kinase IIa and antiapoptotic factors such as Bcl-x are elevated, while tumor suppressor molecules such as PTEN are diminished. This could explain the enhanced growth and apoptosis resistance in these cells. These collective data implicate AHRR as an important regulator of the overall tumor development, the expression of which inversely correlated with cell growth and protected against apoptosis in cancer cells.

One of the essential capabilities that enhances cancer cells’ tumorigenicity is their ability to migrate and become invasive. Here we show that silencing of AHRR tracks with increased motility and invasive potential in tumor cells and is accompanied by loss of E-cadherin, augmentation in vimentin, and phosphorylated FAK expression. Deregulation of these proteins has been previously related to an increase in migratory potential and a more invasive phenotype in tumors (40–42). Consistent with our results, a recent report has shown that cells lacking AHR expression show downregulation of FAK and impaired migratory capabilities (38). The increased motility and the defined proteomic profile in AHRR-deficient cells could also account for the abnormal colony morphology observed in clonogenic assays in which cells appeared to detach and migrate away from the main core of the colony. Tumor growth and metastasis rely on angiogenesis, the induction of new sprouting capillaries from quiescent vasculature (28). AHRR expression in tumor cells inversely correlates with their angiogenic potential. Tumor cells ectopically overexpressing AHRR showed lower angiogenic potential, while tumor cells in which AHRR expression was blocked showed high angiogenic potential. In summary, our data support that enhanced growth, clonal expansion, motility, and angiogenic potential, together with increased resistance to apoptotic signals, are cellular capabilities gained as a result of downregulation of AHRR expression and account for the xenograft growth potential of AHRR low-expressing cells in vivo.

The clinical, genetic, and functional data presented here implicate AHRR as a key regulator in the tumorigenic process whose downregulation (a consequence of methylation in tumor cells) induces pleiotropic effects, leading to tumor promotion and progression. Our data are consistent with the hypothesis that AHRR acts as a tumor suppressor gene in cancers from different tissue origins. The broad spectrum of signaling pathways regulated by AHRR renders this molecule an attractive biological target for developing new therapeutic intervention approaches against human cancers.

View Supplemental data

The study was supported by the University of Colorado Cancer Center Lung_SPORE grant 5P50 CA 058187-13 and by NCI Intramural funding project Z01SC000173.

Address correspondence to: Enrique Zudaire, NCI Angiogenesis Core Facility, NCI, NIH, Advanced Technology Center, Suite 115, Bethesda, Maryland 20892-4605, USA. Phone: (301) 496-8050; Fax: (301) 435-8036; E-mail: zudairee@mail.nih.gov.

Nonstandard abbreviations used: AHR, aryl hydrocarbon receptor; AHRR, AHR repressor; ARNT, AHR nuclear translocator; bHLH, basic helix-loop-helix; FAK, focal adhesion kinase; LOH, loss of heterozygosity; NCI, National Cancer Institute; RPMI-10, RPMI supplemented with 10% FBS; XRE, xenobiotic response element.

Conflict of interest: The authors have declared that no conflict of interest exists.

Reference information: J. Clin. Invest.118:640–650 (2008). doi:10.1172/JCI30024

1. Burbach, K.M., Poland, A., Bradfield, C.A. 1992. Cloning of the Ah-receptor cDNA reveals a distinctive ligand-activated transcription factor. Proc. Natl. Acad. Sci. U. S. A. 89:8185-8189.
   View this article via: CrossRef PubMed Google Scholar
2. Wang, G.L., Jiang, B.H., Rue, E.A., Semenza, G.L. 1995. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc. Natl. Acad. Sci. U. S. A. 92:5510-5514.
   View this article via: CrossRef PubMed Google Scholar
3. Hoffman, E.C., et al. 1991. Cloning of a factor required for activity of the Ah (dioxin) receptor. Science. 252:954-958.
   View this article via: PubMed Google Scholar
4. Mimura, J., Ema, M., Sogawa, K., Fujii-Kuriyama, Y. 1999. Identification of a novel mechanism of regulation of Ah (dioxin) receptor function. Genes Dev. 13:20-25.
   View this article via: PubMed Google Scholar
5. Reyes, H., Reisz-Porszasz, S., Hankinson, O. 1992. Identification of the Ah receptor nuclear translocator protein (Arnt) as a component of the DNA binding form of the Ah receptor. Science. 256:1193-1195.
   View this article via: CrossRef PubMed Google Scholar
6. Denison, M.S., Nagy, S.R. 2003. Activation of the aryl hydrocarbon receptor by structurally diverse exogenous and endogenous chemicals. Annu. Rev. Pharmacol. Toxicol. 43:309-334.
   View this article via: CrossRef PubMed Google Scholar
7. Kasai, A., et al. 2006. High levels of dioxin-like potential in cigarette smoke evidenced by in vitro and in vivo biosensing. Cancer Res. 66:7143-7150.
   View this article via: CrossRef PubMed Google Scholar
8. Moennikes, O., et al. 2004. A constitutively active dioxin/aryl hydrocarbon receptor promotes hepatocarcinogenesis in mice. Cancer Res. 64:4707-4710.
   View this article via: CrossRef PubMed Google Scholar
9. Andersson, P., et al. 2002. A constitutively active dioxin/aryl hydrocarbon receptor induces stomach tumors. Proc. Natl. Acad. Sci. U. S. A. 99:9990-9995.
   View this article via: CrossRef PubMed Google Scholar
10. Koliopanos, A., et al. 2002. Increased arylhydrocarbon receptor expression offers a potential therapeutic target for pancreatic cancer. Oncogene. 21:6059-6070.
    View this article via: CrossRef PubMed Google Scholar
11. Shimada, T., et al. 2002. Arylhydrocarbon receptor-dependent induction of liver and lung cytochromes P450 1A1, 1A2, and 1B1 by polycyclic aromatic hydrocarbons and polychlorinated biphenyls in genetically engineered C57BL/6J mice. Carcinogenesis. 23:1199-1207.
    View this article via: CrossRef PubMed Google Scholar
12. Mitra, A.B., et al. 1995. Genetic alterations at 5p15: a potential marker for progression of precancerous lesions of the uterine cervix. J. Natl. Cancer Inst. 87:742-745.
    View this article via: CrossRef PubMed Google Scholar
13. Xu, S.F., Peng, Z.H., Li, D.P., Qiu, G.Q., Zhang, F. 2003. Refinement of heterozygosity loss on chromosome 5p15 in sporadic colorectal cancer. World J. Gastroenterol. 9:1713-1718.
    View this article via: PubMed Google Scholar
14. Wang, V.W., Bell, D.A., Berkowitz, R.S., Mok, S.C. 2001. Whole genome amplification and high-throughput allelotyping identified five distinct deletion regions on chromosomes 5 and 6 in microdissected early-stage ovarian tumors. Cancer Res. 61:4169-4174.
    View this article via: PubMed Google Scholar
15. Bohm, M., Kleine-Besten, R., Wieland, I. 2000. Loss of heterozygosity analysis on chromosome 5p defines 5p13-12 as the critical region involved in tumor progression of bladder carcinomas. Int. J. Cancer. 89:194-197.
    View this article via: CrossRef PubMed Google Scholar
16. Peralta, R.C., et al. 1998. Distinct regions of frequent loss of heterozygosity of chromosome 5p and 5q in human esophageal cancer. Int. J. Cancer. 78:600-605.
    View this article via: CrossRef PubMed Google Scholar
17. Wieland, I., et al. 1996. Allelic deletion mapping on chromosome 5 in human carcinomas. Oncogene. 12:97-102.
    View this article via: PubMed Google Scholar
18. Kanno, Y., Takane, Y., Izawa, T., Nakahama, T., Inouye, Y. 2006. The inhibitory effect of aryl hydrocarbon receptor repressor (AhRR) on the growth of human breast cancer MCF-7 cells. Biol. Pharm. Bull. 29:1254-1257.
    View this article via: CrossRef PubMed Google Scholar
19. Kim, S.H., et al. 2007. Association between susceptibility to advanced stage endometriosis and the genetic polymorphisms of aryl hydrocarbon receptor repressor and glutathione-S-transferase T1 genes. Hum. Reprod. 22:1866-1870.
    View this article via: CrossRef PubMed Google Scholar
20. Knudson, A.G. 2001. Two genetic hits (more or less) to cancer. Nat. Rev. Cancer. 1:157-162.
    View this article via: CrossRef PubMed Google Scholar
21. Arias-Pulido, H., Narayan, G., Vargas, H., Mansukhani, M., Murty, V.V. 2002. Mapping common deleted regions on 5p15 in cervical carcinoma and their occurrence in precancerous lesions. Mol. Cancer. 1:3.
    View this article via: PubMed Google Scholar
22. Smiraglia, D.J., et al. 2002. Distinct epigenetic phenotypes in seminomatous and nonseminomatous testicular germ cell tumors. Oncogene. 21:3909-3916.
    View this article via: CrossRef PubMed Google Scholar
23. Donehower, L.A., et al. 1992. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature. 356:215-221.
    View this article via: CrossRef PubMed Google Scholar
24. Downward, J. 2005. RNA interference libraries prove their worth in hunt for tumor suppressor genes. Cell. 121:813-815.
    View this article via: CrossRef PubMed Google Scholar
25. Hanahan, D., Weinberg, R.A. 2000. The hallmarks of cancer. Cell. 100:57-70.
    View this article via: CrossRef PubMed Google Scholar
26. Soule, H.D., et al. 1990. Isolation and characterization of a spontaneously immortalized human breast epithelial cell line, MCF-10. Cancer Res. 50:6075-6086.
    View this article via: PubMed Google Scholar
27. Evan, G.I., Vousden, K.H. 2001. Proliferation, cell cycle and apoptosis in cancer. Nature. 411:342-348.
    View this article via: CrossRef PubMed Google Scholar
28. Hanahan, D., Folkman, J. 1996. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell. 86:353-364.
    View this article via: CrossRef PubMed Google Scholar
29. Stanbridge, E.J. 1990. Identifying tumor suppressor genes in human colorectal cancer. Science. 247:12-13.
    View this article via: CrossRef PubMed Google Scholar
30. White, R.L. 1998. Tumor suppressing pathways. Cell. 92:591-592.
    View this article via: CrossRef PubMed Google Scholar
31. Jones, P.A., Laird, P.W. 1999. Cancer epigenetics comes of age. Nat. Genet. 21:163-167.
    View this article via: CrossRef PubMed Google Scholar
32. Murty, V.V., et al. 1994. Allelic loss and somatic differentiation in human male germ cell tumors. Oncogene. 9:2245-2251.
    View this article via: PubMed Google Scholar
33. Ku, W.H., et al. 1997. Genomic deletion and p53 inactivation in cervical carcinoma. Int. J. Cancer. 72:270-276.
    View this article via: CrossRef PubMed Google Scholar
34. Luft, F., Gebert, J., Schneider, A., Melsheimer, P., von Knebel Doeberitz, M. 1999. Frequent allelic imbalance of tumor suppressor gene loci in cervical dysplasia. Int. J. Gynecol. Pathol. 18:374-380.
    View this article via: PubMed Google Scholar
35. Tate, P.H., Bird, A.P. 1993. Effects of DNA methylation on DNA-binding proteins and gene expression. Curr. Opin. Genet. Dev. 3:226-231.
    View this article via: CrossRef PubMed Google Scholar
36. Baba, T., et al. 2001. Structure and expression of the Ah receptor repressor gene. J. Biol. Chem. 276:33101-33110.
    View this article via: CrossRef PubMed Google Scholar
37. Haarmann-Stemmann, T., et al. 2007. Analysis of the transcriptional regulation and molecular function of the aryl hydrocarbon receptor repressor in human cell lines. Drug Metab. Dispos. 35:2262-2269.
    View this article via: CrossRef PubMed Google Scholar
38. Mulero-Navarro, S., et al. 2005. Immortalized mouse mammary fibroblasts lacking dioxin receptor have impaired tumorigenicity in a subcutaneous mouse xenograft model. J. Biol. Chem. 280:28731-28741.
    View this article via: CrossRef PubMed Google Scholar
39. Elferink, C.J. 2003. Aryl hydrocarbon receptor-mediated cell cycle control. Prog. Cell Cycle Res. 5:261-267.
    View this article via: PubMed Google Scholar
40. Hsia, D.A., et al. 2003. Differential regulation of cell motility and invasion by FAK. J. Cell Biol. 160:753-767.
    View this article via: CrossRef PubMed Google Scholar
41. Cavallaro, U., Christofori, G. 2004. Cell adhesion and signalling by cadherins and Ig-CAMs in cancer. Nat. Rev. Cancer. 4:118-132.
    View this article via: PubMed Google Scholar
42. Hu, L., et al. 2004. Association of Vimentin overexpression and hepatocellular carcinoma metastasis. Oncogene. 23:298-302.
    View this article via: PubMed Google Scholar
43. Narayan, G., et al. 2003. Frequent promoter methylation of CDH1, DAPK, RARB, and HIC1 genes in carcinoma of cervix uteri: its relationship to clinical outcome. Mol. Cancer. 2:24.
    View this article via: PubMed Google Scholar
44. Koul, S., et al. 2002. Characteristic promoter hypermethylation signatures in male germ cell tumors. Mol. Cancer. 1:8.
    View this article via: PubMed Google Scholar
45. Maines-Bandiera, S.L., Kruk, P.A., Auersperg, N. 1992. Simian virus 40-transformed human ovarian surface epithelial cells escape normal growth controls but retain morphogenetic responses to extracellular matrix. Am. J. Obstet. Gynecol. 167:729-735.
    View this article via: PubMed Google Scholar
46. Manzano, R.G., et al. 2002. CL100 expression is down-regulated in advanced epithelial ovarian cancer and its re-expression decreases its malignant potential. Oncogene. 21:4435-4447.
    View this article via: CrossRef PubMed Google Scholar
47. Hendricks, D.T., Taylor, R., Reed, M., Birrer, M.J. 1997. FHIT gene expression in human ovarian, endometrial, and cervical cancer cell lines. Cancer Res. 57:2112-2115.
    View this article via: PubMed Google Scholar
48. Waltenberger, J., et al. 1994. Different signal transduction properties of KDR and Flt1, two receptors for vascular endothelial growth factor. J. Biol. Chem. 269:26988-26995.
    View this article via: PubMed Google Scholar
49. Quandt, K., Frech, K., Karas, H., Wingender, E., Werner, T. 1995. MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 23:4878-4884.
    View this article via: CrossRef PubMed Google Scholar
50. Giaever, I., Keese, C.R. 1993. A morphological biosensor for mammalian cells. Nature. 366:591-592.
    View this article via: CrossRef PubMed Google Scholar
51. Guedez, L., et al. 2003. Quantitative assessment of angiogenic responses by the directed in vivo angiogenesis assay. Am. J. Pathol. 162:1431-1439.
    View this article via: PubMed CrossRef Google Scholar