Carcinogenesis vol.34 no.12 pp.2683–2693, 2013
doi:10.1093/carcin/bgt248
Advance Access publication July 10, 2013
The dioxin receptor has tumor suppressor activity in melanoma growth and metastasis
María Contador-Troca, Alberto Alvarez-Barrientos1,
Eva Barrasa, Eva M.Rico-Leo, Inmaculada CatalinaFernández2, Mauricio Menacho-Márquez3, Xosé
R.Bustelo3, José C.García-Borrón4, Aurea GómezDurán2, Javier Sáenz-Santamaría2 and Pedro
M.Fernández-Salguero*
Departamento de Bioquímica y Biología Molecular, Facultad de Ciencias
and 1Servicio de Técnicas Aplicadas a las Biociencias, Universidad de
Extremadura, 06071-Badajoz, Spain, 2Servicio de Anatomía Patológica,
Hospital Universitario Infanta Cristina, 06071-Badajoz, Spain,
3
Centro de Investigación del Cáncer, CSIC-Universidad de Salamanca,
37007-Salamanca, Spain and 4Departamento de Bioquímica y Biología
Molecular, Facultad de Medicina, Universidad de Murcia, 30003-Murcia,
Spain
*To whom correspondence should be addressed. Tel: +34 924 289422;
Fax: +34 924 289419;
Email: [email protected]
Melanoma is a highly metastatic and malignant skin cancer having poor rates of patient survival. Since the incidence of melanoma is steadily increasing in the population, finding prognostic
and therapeutic targets are crucial tasks in cancer. The dioxin
receptor (AhR) is required for xenobiotic-induced toxicity and
carcinogenesis and for cell physiology and organ homeostasis.
Yet, the mechanisms by which AhR affects tumor growth and
dissemination are largely uncharacterized. We report here that
AhR contributes to the tumor–stroma interaction, blocking melanoma growth and metastasis when expressed in the tumor cell
but supporting melanoma when expressed in the stroma. B16F10
cells engineered to lack AhR (small hairpin RNA for AhR) exacerbated melanoma primary tumorigenesis and lung metastasis
when injected in AhR+/+ recipient mice but not when injected in
AhR−/− mice or when co-injected with AhR−/− fibroblasts in an
AhR+/+ stroma. Contrary, B16F10 cells expressing a constitutively
active AhR had reduced tumorigenicity and invasiveness in either
AhR genetic background. The tumor suppressor role of AhR in
melanoma cells correlated with reduced migration and invasion,
with lower numbers of cancer stem-like cells and with altered levels of β1-integrin and caveolin1. Human melanoma cell lines with
highest AHR expression also had lowest migration and invasion.
Moreover, AHR expression was reduced in human melanomas
with respect to nevi lesions. We conclude that AhR knockdown
in melanoma cells requires stromal AhR for maximal tumor progression and metastasis. Thus, AhR can be a molecular marker in
melanoma and its activity in both tumor and stromal compartments should be considered.
Introduction
The dioxin receptor (AhR) is a transcription factor with important
roles in xenobiotic-induced toxicity and carcinogenesis and in tissue
homeostasis (1–3). Notably, AhR positively or negatively influences
proliferation and migration depending on the cellular phenotype. AhR
increases the growth of transformed cell lines (4) and induces liver
(5) and stomach tumors (6) in mice, but it also inhibits the proliferation of breast (7), prostate (8) and liver (9) cancer cells and intestinal
carcinogenesis in mice (10). In addition, we have shown that AhR
is down-modulated in human acute lymphoblastic leukemia cells
Abbreviations: AhR, dioxin receptor; CA-AhR, constitutively active AhR;
EMT, epithelial–mesenchymal transition; PBS, phosphate-buffered saline; shAhR, small hairpin RNA for AhR; SE, standard error; SMA, smooth muscle
actin; TMA, tissue microarray.
(11). These differential effects of AhR on tumor progression could
be partially dependent on its cell type-specific roles in cell migration.
Although AhR deficiency impairs migration and invasion of fibroblasts (12,13) and endothelial cells (14), it also promotes epithelial
cell migration (15). Altogether, these studies indicate that the pro- or
antitumoral roles of AhR are tissue specific and suggest that, in order
to understand how AhR influences cancer, we must consider both the
tumor itself and the surrounding stroma, which, in the skin, contains
among others, cells of the mesenchymal (fibroblasts), endothelial and
immune lineages.
Here, we have studied the role of AhR in the tumor–stroma interaction in melanoma, a highly aggressive tumor of melanocyte origin that
often develops from benign nevi (16). Melanoma is clinically relevant
because its high metastatic potential usually results in adverse prognosis
and poor patient survival (17,18). Despite that, few clinically relevant
molecular markers have been identified for melanoma, being tyrosinase
and Melan-A/MART1 (19,20) and B-RAF (21) the most commonly
used. Importantly, melanoma dissemination is controlled by the tumor
microenvironment, which either restricts or promotes the switch from
the radial (localized) to the vertical (metastatic) growth phases (22).
To address the tumor-intrinsic and the stromal-related AhR functions in melanoma, we have used two approaches. First, we have
molecularly engineered B16F10 mouse melanoma cell lines to either
down-modulate or to constitutively activate AhR in order to investigate how variations in AhR signaling affect melanoma growth and
metastasis. Second, we have transplanted those cell lines on AhR+/+ or
AhR−/− immunocompetent recipient mice to explore the contribution
of stromal AhR expression on melanoma tumorigenesis and metastasis. This scheme allowed us to investigate the interaction between
cell-autonomous (melanoma-dependent) and microenvironmental
(stromal-dependent) AhR functions in melanoma. We have found that
AhR plays antitumorigenic roles in the melanoma cell but pro-tumorigenic roles in the stroma, thus supporting cell type-specific AhR functions in cancer. These results may have clinical relevance because we
also observed that AhR expression was frequently reduced in human
melanomas relative to benign nevi. Taken together, we suggest that
AhR can be a novel molecular target with prognostic or therapeutic
value in melanoma.
Materials and methods
Cell lines and mice
B16F10 mouse melanoma cells were from the American Type Culture
Collection (Manassas, VA). Human C8161, A375, HBL, DOR and Hmel-1
melanoma cell lines were characterized by their growth pattern and morphology and B16F10 cells by the production of melanin. Human FF-C26
fibroblasts were γ-irradiated to halt cell division before use. AhR+/+ and
AhR−/− immortalized fibroblasts and aortic endothelial cells were produced
as described previously (13,14). Cells were tested to be mycoplasma free by
Hoechst 33342 staining. Cells were cultured in Dulbecco’s modified Eagle’s
medium containing 10% fetal bovine serum, 100 U/ml penicillin, 100 µg/ml
streptomycin and 2 mM l-glutamine at 37°C and 5% CO2 atmosphere. AhR+/+
and AhR−/− mice were produced by homologous recombination in embryonic
stem cells as described previously (23) and used at 8–10 weeks of age. Mice
experiments were approved by the Bioethics and Biosecurity Commission of
the University of Extremadura. Mice had free access to water and rodent chow.
Antibodies, vectors and reagents
The following antibodies were used: activated β1-integrin (9EG7), caveolin1
and β-catenin (Becton-Dickinson, Franklin Lakes, NJ); E-cadherin (Millipore,
Billerica, MA); AhR (Biomol, Plymouth, PA); total β1-integrin-fluorescein
isothiocyanate, CD133-PE, CD44-PerCP and CD29-fluorescein isothiocyanate
(BioLegend, San Diego, CA); fibronectin (Chemicon, Temecula, CA) and β-actin
(Sigma–Aldrich, St Louis, MO). Matrigel solution and matrigel-coated transwells
were from Becton-Dickinson. SuperScript II reverse transcriptase and SYBRGreen master mix were from Bio-Rad (Hercules, CA). The constitutively active
AhR (CA-AhR) was a generous gift from Dr Fujii-Kuriyama.
© The Author 2013. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
2683
M.Contador-Troca et al.
Retroviral transduction
B16F10 melanoma cells were transduced with expression vectors containing a
small hairpin RNA for AhR (sh-AhR) or a CA-AhR receptor as described previously (24). Mouse aortic endothelial cells were transduced with the sh-AhR
expression vector following the same protocol.
Sodium dodecyl sulfate–polyacrylamide gel electrophoresis and
immunoblotting
Sodium dodecyl sulfate–polyacrylamide gel electrophoresis and immunoblotting were performed using sh-AhR, wild-type (WT)-AhR and CA-AhR
B16F10 total cell extracts as indicated (13).
Clonogenic assays
Clonogenic assays in 2D were done by plating 5 × 102 or 103 sh-AhR, WTAhR or CA-AhR cells in plain tissue culture dishes. After 5 days, the medium
was removed and clones were washed in phosphate-buffered saline (PBS),
stained for 10 min with 0.5% (wt/vol) crystal violet and counted. For 3D clonogenic assays, cells were grown in matrigel as described previously (14).
Briefly, 5 × 103 sh-AhR, WT-AhR or CA-AhR B16F10 cells were seeded on
matrigel plugs for 3 days in complete medium. Clones formed were counted
and analyzed for cell spreading using the ImageJ software (version 1.45S).
Melanoma cell injection, lung metastasis and extravasation assays
Aliquots of 105 exponentially growing cells were injected subcutaneously in
both flanks of AhR+/+ and AhR−/− recipient mice and tumors were allowed
to grow for 15 days. For co-injection experiments, 105 sh-AhR, WT-AhR
or CA-AhR cells were mixed with 4 × 104 immortalized AhR+/+ or AhR−/−
fibroblasts (13) and the cellular mixture was injected subcutaneously in WT
mice. In these last set of experiments, each melanoma cell line was injected
with AhR+/+ fibroblasts on one flank and with AhR−/− fibroblasts on the other
flank. Mice were killed and tumors were recovered, weighed and measured
with a caliper. Tumor volume was calculated as length × width2 × 0.4. For
lung metastasis, 105 cells of each line were resuspended in 100 µl of PBS
and injected through the tail vein of AhR+/+ and AhR−/− recipient mice. After
21 days, mice were killed and the lungs were extracted and observed for the
presence of melanoma-derived metastasis. Tumor lesions were recorded and
photographed. Lung extravasation was determined after 48 h of cells injection
through the tail vein. One hour before killing, mice were injected with 100
µl of rhodamine-conjugated lectin to stain capillaries. Lungs were recovered,
fixed in paraformaldehyde, sectioned at 10 µm and the number of melanoma
cells found outside the blood vessels were quantified by confocal microscopy.
Cell migration and invasion
Cell migration was analyzed in wound healing experiments as reported in refs
13 and 15 using the Cell-R system (Olympus, Tokyo, Japan). Invasion through
matrigel-coated transwells was analyzed using an Olympus FV1000 confocal
microscope as described previously (14).
Flow cytometry
The analysis of side population cells in sh-AhR, WT-AhR and CA-AhR
B16F10 cultures was done in a MoFlo-XDP equipment (Beckman-Coulter,
Indianapolis, IN) using published criteria (25). Cells were stained for 90 min
at 37°C with 5 µg/ml Hoechst 33342 (Sigma–Aldrich), centrifuged and resuspended in N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid–Hanks’
balanced salt solution buffer containing 2% fetal bovine serum. Aliquots of
cells were treated with the ABCG2 extrusion bomb inhibitor fumitremorgin C
(50 µM; Sigma–Aldrich). Propidium iodide (10 nM) was used to discriminate
dead cells. To quantify side population subpopulations, Hoechst 33342-labeled
cells were stained for CD133+ (CD133-PE), CD44+ (CD44-PerCP) and CD29+
(CD29-fluorescein isothiocyanate) for 30 min at 10°C. To-Pro (0.1 µM) was
used to discriminate dead cells from these analyses.
Melanosphere formation
sh-AhR, WT-AhR and CA-AhR cells were plated in Nunc ultra-low adherence
24-well plates (Thermo Fisher, Waltham, MA) at 3 × 104 cells per well and melanospheres were allowed to form for 24 h. Images were acquired with an Olympus
Cell-R microscope and sphere number and size were quantified using the ImageJ
software (version 1.45S). Spheres were gently dissociated, centrifuged and
diluted in PBS containing 0.1% of bovine serum albumin. Then, CD133-PE and
CD44-PerCP antibodies were added for 30 min at 10°C and labeled cells were
analyzed in a Cytomics FC-500 flow cytometer (Beckman-Coulter).
Human melanoma tissue microarrays and immunohistochemistry
Human nevi and melanoma tissues were obtained from the tumor collection of
the Infanta Cristina University Clinical Hospital. Each biopsy was diagnosed
by two independent pathologists. Human tissue microarrays (TMAs) were
assembled from individual biopsies using a Beecher MTA-I equipment (Sun
Prairie, WI). One millimeter cylinders were taken from paraffin-embedded
2684
tissues and inserted in triplicate in TMAs containing 17 melanomas and 6 nevi
(69 cores). TMAs were sectioned at 3–5 µm and analyzed for AhR and caveolin1 expression by immunohistochemistry.
Immunofluorescence
Immunofluorescence was done in an Olympus FV1000 confocal microscopy
on cultures fixed for 20 min at room temperature in 4% paraformaldehyde
(Polysciences Inc, Warrington, PA). Fluorochromes used were Alexa 488
(E-cadherin) and Alexa 633 [β-catenin and α-smooth muscle actin (SMA)].
4′,6-Diamidino-2-phenylindole was used to stain cell nuclei. Fluorescence distribution analysis was done using the FV10 software (Olympus).
β1-integrin activation measurement
sh-AhR, WT-AhR and CA-AhR B16F10 cells were detached at 4°C in PBS
containing 2 mM of ethylenediaminetetraacetic acid, centrifuged, resuspended
and incubated at 4°C for 10 min in PBS containing 1% bovine serum albumin
to block unspecific antibody binding. The levels of total and active β1-integrin
were quantified as described previously (26).
Statistical analyses
Data are shown as mean ± standard error (SE). Statistical comparison between
experimental conditions was done using GraphPad Prism 4.0 software
(GraphPad, San Diego, CA). The Student’s t-test and the analysis of variance
analyses were applied. Experiments were done in duplicate in at least three
cultures of each cell line.
Results
AhR down-modulation in the tumor cell increased melanoma growth
and metastasis
To analyze the role of AhR in the tumor–stroma interaction in melanoma, we have used B16F10 mouse cells. They were chosen for
their ability to grow tumors in non-immunocompromised AhR+/+ and
AhR−/− mice, which are essential to understand how AhR expression
in the stroma affects tumor development. To our knowledge, no equivalent cell line is available from human melanomas. B16F10 cells were
transduced with retroviruses encoding either a sh-AhR or a CA-AhR
receptor to explore the effects of loss or gain of AhR function, respectively (Supplementary Figure S1A, available at Carcinogenesis
Online). B16F10 cells were also transduced with an empty retrovirus
as control (Supplementary Figure S1A, available at Carcinogenesis
Online). These cell lines will be referred to hereafter as sh-AhR,
CA-AhR and WT-AhR, respectively. Immunoblot analysis demonstrated proper knockdown of the endogenous AhR and overexpression
of the ectopically expressed receptor in sh-AhR and CA-AhR cells,
respectively (Supplementary Figure S1B, available at Carcinogenesis
Online). Constitutive AhR activation in CA-AhR cells was demonstrated by the increase in messenger RNA for the Cyp1a1 target gene
(Supplementary Figure S1C, available at Carcinogenesis Online).
Silencing or constitutive activation of the AhR pathway did not significantly affect proliferation rates (Supplementary Figure S2A, available
at Carcinogenesis Online) or cell cycle distribution of B16F10 cells
(Supplementary Figure S2B, available at Carcinogenesis Online).
We then analyzed the effect of the loss- and gain-of-function of
AhR in melanoma growth and metastasis. sh-AhR and CA-AhR
cells were injected in AhR+/+ recipient mice and tumor growth
was evaluated 15 days later. An inverse correlation between AhR
expression and melanoma primary tumorigenesis was found since
sh-AhR and CA-AhR cells promoted larger and smaller tumors than
control cells, respectively (Figure 1A and Supplementary Figure S3A,
available at Carcinogenesis Online). To analyze the effect of AhR in
melanoma lung metastasis, the aforementioned cells were injected
intravenously in WT mice and the number of metastatic nodules in
the lungs counted de visu 21 days later. AhR knocked-down cells
induced 2-fold more and larger metastatic nodules than control cells
(Figure 1B and C). In this case, however, no statistically significant
differences between control and CA-AhR cells were detected
(Figure 1B and C), suggesting that, unlike primary tumorigenesis
(Figure 1A), endogenous AhR functions appear enough to activate the
pro-metastatic program toward the lung. The increased number and
larger size of the metastasis induced by sh-AhR cells are probably
related to an enhanced fitness to colonize the lung parenchyma
AhR expression suppresses melanoma progression
Fig. 1. AhR expression suppresses melanoma growth and metastasis in a murine model. (A) sh-AhR, WT-AhR and CA-AhR cells were injected subcutaneously in
AhR+/+ or AhR−/− recipient mice and tumors were harvested, weighed and measured 15 days later. The following number of tumors were analyzed: AhR+/+
mice–10, 13 and 9 for sh-AhR, WT-AhR and CA-AhR cells, respectively; AhR−/− mice–11, 10 and 11 for sh-AhR, WT-AhR and CA-AhR cells, respectively. (B and
C) Each cell line was injected intravenously through the tail vein in AhR+/+ (n = 9) or AhR−/− (n = 9) recipient mice and lung metastases counted after 21 days. (D) Cell
lines were analyzed for their extravasation potential to the lungs as described in Materials and methods. Data are shown as mean ± SE. Bar 1 cm. n.s., nonsignificant.
because they did not show any significant extravasation advantage
when compared with control and CA-AhR cells (Figure 1D).
AhR knockdown increased migration and invasion of mouse
melanoma cells
Examination of sh-AhR, WT-AhR and CA-AhR B16F10 cells in 2D
and 3D cultures indicated that AhR was linked to significant changes
in cell morphology. In 2D cultures, the sh-AhR clones usually grew
in spread cell colonies with reduced intercellular contacts (Figure 2A,
left top). These cells also formed highly migratory colonies in 3D
cultures that were very different in morphology from the rounded
colonies produced by control cells (Figure 2A, left bottom) and that
were reminiscent of cells undergoing an epithelial–mesenchymal
transition (EMT). This phenotype was a direct effect of AhR since the
CA-AhR clones showed an exacerbated, ‘epithelial-like’ morphology
under both culture conditions (Figure 2A, right panels). Consistent
with an EMT-like process, the sh-AhR and CA-AhR cells had
increased or reduced migration in wound healing (Figure 2B) and
2685
M.Contador-Troca et al.
Fig. 2. AhR knockdown alters melanoma cell morphology, migration and invasion. (A) sh-AhR, WT-AhR and CA-AhR B16F10 cells were grown in 2D (upper left) of 3D
matrigel cultures (lower left) and clone morphology was analyzed. The number of clones formed in 2D cultures (middle) and the number of spreading cells in 3D matrigel
cultures were counted (right). Arrows mark the contour of the clones. n.s., nonsignificant. (B) Cell migration was analyzed in wound healing assays as indicated (15).
(C) Invasion of sh-AhR, WT-AhR and CA-AhR B16F10 cell lines was quantified by confocal microscopy in matrigel-coated transwells. (D) sh-AhR, WT-AhR and CA-AhR
melanoma cells were analyzed by immunofluorescence for the expression of E-cadherin, β-catenin and α-SMA. 4′,6-Diamidino-2-phenylindole staining was used to locate
cell nuclei. (E and F) Protein levels of epithelial and mesenchymal markers (E) and of caveolin1 (F) were determined in each B16F10 cell line by immunoblotting. Numbers
under the figure indicate relative protein levels. (G) β1-integrin expression and activation were analyzed by flow cytometry as indicated in Materials and methods. β1t and β1a
stand for total and activated β1-integrin, respectively. Data are shown as mean ± SE from triplicate experiments in at least two cultures of each line. Bar 50 µm.
2686
AhR expression suppresses melanoma progression
matrigel-based invasion assays (Figure 2C) with respect to control
WT-AhR cells.
Given the observations previously described, we examined whether
the morphology of the sh-AhR and CA-AhR cells could correlate with
an EMT and mesenchymal–epithelial transition process, respectively.
We combined immunofluorescence and immunoblotting to analyze
changes in the classical markers E-cadherin, β-catenin, fibronectin
and α-SMA, as well as in the EMT inducer Snail. In agreement with
an EMT-like process, the more spread and migratory sh-AhR cells
partially delocalized E-cadherin and β-catenin from the plasma membrane and increased their α-SMA content by more than 2-fold when
compared with both WT-AhR and CA-AhR cells (Figure 2D and E).
Additionally, sh-AhR cells had a moderate increase in Snail and in the
mesenchymal marker fibronectin with respect to the less tumorigenic
CA-AhR cells (Figure 2E).
We have recently observed in murine fibroblasts that AhR exerts
a dual opposite role activating caveolin1 and inhibiting β1-integrin
(26). Considering these results, and since caveolin1 is involved in
melanoma malignancy and its expression is E-cadherin dependent
(27), we analyzed caveolin1 and β1-integrin in our B16F10 cell lines.
Immunoblot analyses revealed that caveolin1 was heavily upregulated
in CA-AhR cells but down-modulated in sh-AhR knockdown cells
(Figure 2F). Using flow cytometry, we also found that AhR knockdown upregulated the expression (Figure 2G, left) and the activation
(Figure 2G, right) of β1-integrin, a caveolin1-inhibited integrin that
regulates melanoma motility and invasion (28,29).
Stromal AhR modulates the tumorigenic and metastatic abilities of
melanoma cells
To investigate how stromal AhR affects melanoma progression, we
inoculated each B16F10 cell line in AhR−/− recipient mice. For primary tumorigenesis, the AhR−/− microenvironment did not significantly affect the growth of either CA-AhR or WT cells (Figure 1A
and Supplementary Figure S3A, available at Carcinogenesis Online).
In contrast, it severely compromised the growth advantage of sh-AhR
cells previously seen in WT mice (Figure 1A and Supplementary
Figure S3A, available at Carcinogenesis Online), reducing tumors
to the sizes observed with control cells (Figure 1A). Regarding lung
metastasis, the AhR−/− stroma reduced the number of metastatic foci
induced by each of the three cell lines under study (Figure 1C). The
AhR-expressing stroma likely produces secretable factors that promote melanoma progression since the invasive potential of sh-AhR
B16F10 clones was higher in AhR+/+ fibroblast-conditioned medium
than in that of AhR−/− fibroblasts (Figure 3A). CA-AhR and WT-AhR
cells, on the contrary, did not significantly differ in invasion in either
conditioned media (Figure 3A). Although fibroblasts are a prototypical cell type in maintaining a functional stroma, we also analyzed
the contribution of the endothelium on the invasive potential of our
B16F10 cells. Mouse aortic endothelial cells from AhR+/+ mice were
transduced with a sh-AhR vector to reduce their basal receptor levels
(Supplementary Figure S3B, available at Carcinogenesis Online). The
invasion of sh-AhR melanoma clones increased in AhR+/+ endothelium-conditioned medium with respect to that of sh-AhR endothelium
Fig. 3. Stromal AhR modulates melanoma cell migration. (A and B) sh-AhR, WT-AhR and CA-AhR cells were cultured in medium conditioned by AhR+/+ or
AhR−/− mouse fibroblasts (A) or by AhR+/+ or sh-AhR endothelial cells (B) and clone morphology and cell scattering were analyzed by microscopy. (C) sh-AhR,
WT-AhR or CA-AhR melanoma cells were co-injected with immortalized fibroblasts from AhR+/+ or AhR−/− mice in WT-AhR mice. Injections were set such
as each melanoma cell line was injected with one fibroblast cell line on one flank of the mouse and with the other fibroblast cell line on the other flank. The
following number of tumors were analyzed: AhR+/+ fibroblasts–5, 5 and 6 for sh-AhR, WT-AhR and CA-AhR cells, respectively; AhR−/− fibroblasts–6, 6 and 8
for sh-AhR, WT-AhR and CA-AhR cells, respectively. (D) Tumors produced by each cell line in AhR+/+ recipient mice were processed to quantify their blood
vessel content (see Supplementary Figure S4A, available at Carcinogenesis Online). Data are shown as mean ± SE from triplicate cultures. Three tumors were
analyzed for each experimental condition. Bar 50 µm.
2687
M.Contador-Troca et al.
and no significant differences were found for WT-AhR and CA-AhR
melanoma cells in either conditioned media (Figure 3B). The sh-AhR
endothelium-conditioned medium also reduced the clone formation
potential of sh-AhR cells (Supplementary Figure S3C, available at
Carcinogenesis Online), again suggesting that AhR is needed to produce stroma-secreted factors required for the growth of melanoma
cells depleted of AhR expression. To further sustain this possibility,
we decided to co-inject each B16F10 cell line with immortalized
AhR+/+ or AhR−/− fibroblasts (13) in WT recipient mice. The presence
of AhR−/− fibroblasts reduced the growth (Figure 3C) and the volume
(Supplementary Figure S3D, available at Carcinogenesis Online)
of the tumors formed by sh-AhR melanoma cells in WT mice, thus
supporting a tumor promoting effect for stromal AhR (see Figure 1A
and Supplementary Figure S3A, available at Carcinogenesis Online).
AhR−/− fibroblasts did not significantly affect the growth of WT-AhRor CA-AhR-derived tumors with respect to AhR+/+ fibroblasts (Figures
3C and Supplementary Figure S3D, available at Carcinogenesis
Online). The effect of stromal AhR on the growth of sh-AhR-derived
tumors possibly involves an enhanced angiogenic response since the
number of blood vessels were higher in sh-AhR than in WT-AhR or
CA-AhR tumors (Figures 3D and Supplementary Figure S4A, available at Carcinogenesis Online). Altogether, these results indicate that
(i) AhR contributes to the control of melanoma cell migration and
invasion, (ii) AhR can repress pro-migratory molecules such as β1integrin, (iii) AhR hyperactivation leads to the further enhancement
of some (caveolin1) but not all (β1-integrin) pro-migratory molecules
and (iv) AhR plays intrinsic tumor suppressor-like roles in melanoma
cells but pro-tumorigenic roles in the stroma.
AhR knockdown increased the pool of cancer stem-like cells
Next, we explored one potential cause for the tumor suppressor-like
role of AhR in melanoma cells. Since EMT can be associated with a
stem-like cell phenotype in cancer cells, we analyzed if the sh-AhR
B16F10 line contained an increased number of stem-like cells. We
monitored by flow cytometry the percentage of ABCB5+/CD29+
cells (30) that coexpressed CD133 and CD44, two surface markers
associated to the stem-like phenotype (31). sh-AhR cells, but not
their CA-AhR counterparts, displayed larger numbers of CD133+/
CD44+/CD29+ cells when compared with control cells (Figure 4A).
To confirm these results, melanosphere formation assays in nonadherent conditions in vitro (32) were performed. We found that the
sh-AhR cells generated more and larger melanospheres than either
WT-AhR or CA-AhR cells (Figure 4B). The stem-like status of the
melanospheres was confirmed by flow cytometry (Figure 4C). These
results indicate that one possible mechanism for the tumor suppressor activity of endogenous AhR is to reduce the stem-like cell pool of
melanoma cells.
Human melanoma cells and melanomas from human patients had
reduced AHR expression
The results above suggest an inverse correlation between melanoma
malignancy and the tumor suppressor-like role of AhR. We thus investigated whether AhR could serve as a surrogate marker for melanoma
progression and malignancy. We analyzed a battery of five human
melanoma cell lines including A375, C8161, HBL, Hmel-1 and DOR.
Clone formation assays in 2D showed that C8161 cells produced more
spread and scattered clones than did the remaining cell lines analyzed
(Figure 5A). C8161 cells also had a more migratory phenotype than
A375, Hmel-1 and DOR cells, whereas HBL cells had an intermediate migration potential (Figure 5B). When cultured in matrigel, a
similar pattern was obtained being C8161, the most invasive among
the cell lines tested (Figure 5C). Interestingly, immunoblot analyses
revealed that AHR levels were lower in the highly migratory and invasive C8161 cells than in the less motile A375, Hmel-1 and DOR cells,
with HBL cells having an intermediate amount of AHR expression
(Figure 6A). Thus, as for the murine B16F10 model, human melanoma cells with reduced AHR expression had increased migration and
invasion potentials. In agreement, rescue of AHR expression in C8161
2688
cells (Figure 6B) reduced their migration as compared with the parental cell line (Figure 6C). Moreover, C8161 cells formed more invasive
and scattered colonies than A375 cells when grown on FF-C26 human
fibroblasts (Figure 6D), further suggesting a negative role for AHR on
the invasion of melanoma cells.
We then examined the expression of AHR in human nevi and
in melanomas from human patients using tissue arrays and standard immunohistochemistry. Immunofluorescence was also used
to confirm AHR expression in human samples (Supplementary
Figure S5, available at Carcinogenesis Online). As summarized
in Supplementary Table SI, available at Carcinogenesis Online,
AHR protein was detected in five out of six nevi lesions analyzed
(83%; Figure 6E). In contrast, 14 human melanomas were negative for AHR, whereas 2 biopsies displayed weak immunoreactivity (Supplementary Table SI, available at Carcinogenesis Online,
and Figure 6E; n = 16). The negative and weak immunoreactive melanomas correlated with advanced Clark III-IV-V or early
Clark <III stages, respectively (Supplementary Table SI, available
at Carcinogenesis Online). Since the pro-tumorigenic and prometastatic sh-AhR B16F10 cells had reduced caveolin1 expression
(Figure 2F), we analyzed caveolin1 levels in the same tissue arrays.
Caveolin1 was detected more frequently in nevi lesions (80%) than
in malignant melanomas (6%; Supplementary Table SI, available at
Carcinogenesis Online, and Figure 6E). AHR and caveolin1 were
coexpressed in 67% of all nevi analyzed (Supplementary Table SI,
available at Carcinogenesis Online). Some melanomas had detectable AHR expression in the stroma (Supplementary Figure S4B,
available at Carcinogenesis Online), which could be suggestive
of its effects in that compartment. Thus, AHR and caveolin1 seem
inversely correlated with human melanoma progression.
Discussion
AhR promotes or inhibits cell proliferation and migration depending
on the phenotype of the target cell (33). However, only few studies
have analyzed the role of AhR in tumor development in vivo, and
it is mostly unknown if differences in AhR expression contribute
to tumor progression and dissemination. Moreover, no studies have
yet addressed whether AhR is relevant in the interaction between
the tumor cell and the stroma. This encouraged us to investigate the
tumorigenic and metastatic response of melanoma cells with varying levels of AhR expression in an AhR-expressing or AhR-lacking
microenvironments. Melanoma was studied for its increasing clinical
impact and for its metastatic potential that often results in adverse
clinical outcomes.
An important finding of this study is that as AhR expression/activation decreases in the tumor cell, melanoma tumorigenesis and
metastasis became markedly stimulated. The intrinsic suppressive
role of AhR in melanoma cells may be related to its ability to modulate cell migration and invasion, thus strengthening the existence of
cell-autonomous AhR functions in cancer (12,14,15). Accordingly,
although AhR knockdown favored a mesenchymal-like phenotype
with increased migration and tumorigenicity, constitutive AhR activation reduced cell migration and impaired melanoma growth and
dissemination. The correlation found in the murine melanoma model
(e.g. increased migratory and invasive potentials are associated to
reduced AhR levels) was also observed in human melanoma cells,
which not only sustains the tumor suppressor function of AhR but
also validates the murine model. The change from an anti-invasive
(epithelial) to a pro-invasive (mesenchymal) phenotype caused by
AhR depletion may reflect an EMT-like process involving E-cadherin
and β-catenin, as their cellular localization and expression are altered
in metastatic and late-stage human melanomas (34,35). In fact, AhR
down-modulation in mouse primary keratinocytes and NMuMG cells
and in human HaCaT cells induces an EMT-like process involving
changes in the expression and localization of E-cadherin and
β-catenin (24). Additionally, the tumor suppressor-like role of AhR
in melanoma correlated with the upregulation of caveolin1, which, as
AhR, has cell type-dependent tumor suppressor activity (36–38) and
AhR expression suppresses melanoma progression
Fig. 4. AhR down-modulation increases undifferentiation markers of B16F10 cells. (A) sh-AhR, WT-AhR and CA-AhR cultures were sorted out to collect the
side population cells expressing the adenosine triphosphate-binding cassette transporter ABCB5. These cell fractions were then analyzed by flow cytometry for
CD133+ (left) or CD44+/CD29+ (right). (B) The number of melanospheres formed by each B16F10 cell line under non-adherent culturing conditions (left) and the
average number of cells in each melanosphere (right) were determined. (C) Melanospheres were dissociated and the resulting cell suspensions were analyzed for
CD133+ expression by flow cytometry. The experiments were performed by duplicate in three independent cultures. Data are shown as mean ± SE. Bar 50 µm.
is E-cadherin regulated (27). Moreover, AhR and caveolin1 expression positively correlated with a reduced activation of β1-integrin, a
known promoter of melanoma invasion and dissemination (28,29).
EMT has been associated to stem cell-ness (39) and whereas AhR
knockdown enriched, AhR over-activation depleted the subpopulation
of B16F10 stem-like cells. Because stem cells are generally considered
drivers of tumorigenesis and metastasis (31,32), the antitumoral activity of AhR in melanoma could act by maintaining a reduced pool of
undifferentiated stem-like cells. Indeed, AhR activation by xenobiotics inhibited invasiveness (7) and repressed mammosphere formation
of breast tumor cells (40), whereas AhR inhibition by StemRegenin
increased the number of human hematopoietic stem cells and their
2689
M.Contador-Troca et al.
Fig. 5. Human melanoma cell lines differ in their spreading, migration and invasion potentials. (A) The clonogenic potential of C8161, HBL, A375 and DOR
human melanoma cells was determined in 2D cultures. (B) Cell migration was quantified in the indicated cell lines plus Hmel-1 using wound healing assays 24 h
after plating. (C) Cell invasion was also analyzed in the five cell lines under study by confocal microscopy using matrigel-coated transwells. Duplicate experiments
were done in three cultures of each line. Data are shown as mean ± SE. Bar 50 µm. Note different magnification in A375 and DOR cell lines in panel A.
engraftment in mice (41). We, therefore, propose that AhR suppresses
melanoma by a mechanism reducing the number of stem-like cells
while inhibiting migration and invasion.
The stroma is an additional compartment that controls tumor progression. Contrary to the tumor suppressor-like activity of AhR in melanoma cells, stromal AhR is needed for malignancy since AhR−/− mice
could not sustain tumorigenesis and metastasis of AhR knocked-down
cells. Tissue-/stroma-specific factors must be involved in the process
2690
because even though AhR-expressing lungs allowed increased prometastatic activity at any AhR expression level, primary tumorigenesis was enhanced only for AhR-knockdown melanoma cells. These
AhR-dependent stromal effects likely involve yet unidentified diffusible factors because AhR−/− fibroblasts could not sustain sh-AhR tumor
growth in WT recipient mice and the AhR-null conditioned medium
failed to induce invasion of sh-AhR melanoma cells. Possible candidates are cytokines such as transforming growth factor β because of
AhR expression suppresses melanoma progression
Fig. 6. AHR expression correlates with the migration and invasion of human melanoma cell lines and its levels are reduced in human melanoma. (A) C8161,
A375, HBL, Hmel-1 and DOR cells were analyzed for AHR protein expression by immunoblotting. The expression of β-actin was used as loading control.
(B) AHR expression was rescued in low-expressing C8161 cells (C8161 + AHR) and the amount of cellular receptor was quantified by immunoblotting. (C) The
migration ability of the C8161 + AHR cell line was determined in wound healing assays and quantified with respect to control C8161cells. (D) Low-expressing
C8161 and high-expressing A375 human melanoma cell lines were cultured on a monolayer of γ-irradiated FF-C26 human fibroblasts to analyze their clone
formation potential under a homogeneous fibroblasts background. Duplicate experiments were done in three cultures of each line. Data are shown as mean ± SE.
(E) Immunohistochemistry for AHR and caveolin1 in human nevi and melanomas. Nevi and melanoma were arranged in a TMA as indicated in Materials and
methods. AHR and caveolin1 protein were detected by immunohistochemistry using specific antibodies. Micrographs from representative biopsies included in
Supplementary Table SI, available at Carcinogenesis Online, are shown. Bar 50 µm.
its functional interaction with AhR (2,42) or hepatocyte growth factor
since it confers resistance to melanoma chemotherapy when produced
in the stroma (43). The fact that the AhR-expressing stroma reduced
blood vessel formation in WT-AhR and CA-AhR but not in sh-AhR
melanoma suggests that melanoma cells with reduced AhR expression may not respond to stromal inhibitory signals. Actually, stromal
AhR seems to contribute to tumor angiogenesis because AhR−/− mice
cannot develop a proper angiogenic response (14). Stromal AhR may
also modulate melanoma growth and dissemination by promoting the
generation of cancer stem-like cells. Diffusible factors secreted by
the AhR-expressing microenvironment could synergistically act with
melanoma cell-autonomous AhR depletion to sustain an undifferentiated phenotype. The cytokines mentioned previously could in fact act
in a paracrine manner on the melanoma cell to sustain stem cell-ness.
Such possibility is currently under investigation.
The AhR is involved in the Treg/Th17 differentiation program that
regulates immune cell activation through the production of Treg cells
(44,45). An interesting possibility is that since Tregs suppress the activation of reactive lymphocytes, their depletion in the AhR-lacking
stroma could induce an augmented immunological response leading to reduced tumor growth and metastasis. The Treg/Th17 balance
is regulated by kynurenine, a high-affinity AhR ligand and a product
of the indoleamine 2,3-dioxygenase and the tryptophan 2,3-dioxygenase enzymes (46). Tryptophan 2,3-dioxygenase-dependent synthesis
of kynurenine from tryptophan was suggested to suppress antitumor
immunity and to support glioblastoma growth, likely by increasing Tregs (47). In this context, kynurenine could impair the immune
response against melanoma cells by promoting AhR-dependent Treg
differentiation, an effect that would increase tumor growth in AhRexpressing but not in AhR-lacking stroma. An additional aspect is
how 6-formylindolo[3,2-b]carbazole, an AhR ligand produced from
tryptophan by ultraviolet exposure (48), can affect melanoma progression. Exposure of AhR-null mice to ultraviolet could help answer this
question.
AHR protein levels were lower in human melanomas with respect
to non-malignant nevi, suggesting that the malignant phenotype correlates with reduced AHR expression and that AHR could have a tumor
suppressor-like role in melanoma. Importantly, AHR shares with
caveolin1 cancer-related properties such as its down-modulation in
advanced human melanomas (29) and its implication in cell adhesion
and migration (26). Consequently, AHR and caveolin1 could act in
concert during melanoma progression and dissemination. Melanoma
2691
M.Contador-Troca et al.
thus increases the group of AhR-inhibited cancers that also includes
acute lymphoblastic leukemia (11), hepatocarcinoma (9), prostate (8)
and mammary cancers (49). In summary, AhR activation has tumor
suppressor-like activity in melanoma cells because its down-modulation promotes tumorigenesis and metastasis. Oppositely, stromal AhR
favors, rather than inhibits, primary melanoma growth and metastasis
to the lungs. Therefore, AhR expression has to be accounted for in
both, tumor mass and surrounding stroma. One mechanism for the
antitumoral activity of AhR could involve the maintenance of a nonmigratory/non-invasive phenotype that is converted to a pro-migratory/pro-invasive mesenchymal phenotype by AhR knockdown. AhR
may also negatively control the pool of stem-like cancer cells eventually driving melanoma progression. Finally, AHR expression appears
to be reduced in invasive human melanoma cells and in melanomas
from human patients, which supports its tumor suppressor-like role.
Altogether, AhR and caveolin1 could be potential molecular markers
in melanoma.
Supplementary material
Supplementary Table S1 and Figures S1–S5 can be found at http://
carcin.oxfordjournals.org/
Funding
Spanish Ministry of Economy and Competitiveness (MINECO;
SAF2008-00462, BFU2011-22678, RD06/0020/1016 and RD12/­
0036/0032) and Junta de Extremadura (GR10008) to P.M.F.-S.;
Spanish MINECO (SAF2009-07172, RD06/0020/0001 and
RD12/0036/0002) and Spanish Association Against Cancer (AECC)
to X.R.B. M.C.-T. was a FPI fellow from the Spanish Ministry of
Science and Innovation. All Spanish funding is cosponsored by the
European Union FEDER program.
Acknowledgements
The support of the Servicio de Técnicas Aplicadas a las Biociencias (STAB) of
the Universidad de Extremadura is greatly acknowledged. We are very grateful to Dr F.Real (CNIO, Madrid, Spain) for providing the γ-irradiated FF-C26
human fibroblasts. The support of the technicians at the Pathology Department
of the Hospital Universitario Infanta Cristina is appreciated.
Conflict of Interest Statement: None declared.
References
1.Furness,S.G. et al. (2007) The dioxin (aryl hydrocarbon) receptor as a
model for adaptive responses of bHLH/PAS transcription factors. FEBS
Lett., 581, 3616–3625.
2.Roman,A.C. et al. (2012) The aryl hydrocarbon receptor regulates
cell adhesion and migration by interacting with oncogene and growth
factor-dependent signaling. In Pohjanvirta,R. (ed) The AH Receptor
in Biology and Toxicology. John Wiley & Sons, New York, NY, pp.
485–497.
3.Esser,C. et al. (2009) The aryl hydrocarbon receptor in immunity. Trends
Immunol., 30, 447–454.
4.Barouki,R. et al. (2007) The aryl hydrocarbon receptor, more than a xenobiotic-interacting protein. FEBS Lett., 581, 3608–3615.
5.Moennikes,O. et al. (2004) A constitutively active dioxin/aryl hydrocarbon receptor promotes hepatocarcinogenesis in mice. Cancer Res., 64,
4707–4710.
6.Andersson,P. et al. (2002) A constitutively active dioxin/aryl hydrocarbon receptor induces stomach tumors. Proc. Natl Acad. Sci. USA, 99,
9990–9995.
7.Hall,J.M. et al. (2010) Activation of the aryl-hydrocarbon receptor inhibits
invasive and metastatic features of human breast cancer cells and promotes
breast cancer cell differentiation. Mol. Endocrinol., 24, 359–369.
8.Gluschnaider,U. et al. (2010) beta-TrCP inhibition reduces prostate cancer
cell growth via upregulation of the aryl hydrocarbon receptor. PLoS One,
5, e9060.
2692
9.Fan,Y. et al. (2010) The aryl hydrocarbon receptor functions as a tumor
suppressor of liver carcinogenesis. Cancer Res., 70, 212–220.
10. Kawajiri,K. et al. (2009) Aryl hydrocarbon receptor suppresses intestinal
carcinogenesis in ApcMin/+ mice with natural ligands. Proc. Natl Acad.
Sci. USA, 106, 13481–13486.
11. Mulero-Navarro,S. et al. (2006) The dioxin receptor is silenced by promoter hypermethylation in human acute lymphoblastic leukemia through
inhibition of Sp1 binding. Carcinogenesis, 27, 1099–1104.
12. Carvajal-Gonzalez,J.M. et al. (2009) The dioxin receptor regulates the constitutive expression of the vav3 proto-oncogene and modulates cell shape
and adhesion. Mol. Biol. Cell, 20, 1715–1727.
13. 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.
14. Roman,A.C. et al. (2009) Dioxin receptor deficiency impairs angiogenesis by a mechanism involving VEGF-A depletion in the endothelium
and transforming growth factor-beta overexpression in the stroma. J. Biol.
Chem., 284, 25135–25148.
15. Carvajal-Gonzalez,J.M. et al. (2009) Loss of dioxin-receptor expression
accelerates wound healing in vivo by a mechanism involving TGFbeta.
J. Cell Sci., 122(Pt 11), 1823–1833.
16. Leiter,U. et al. (2004) The natural course of cutaneous melanoma. J. Surg.
Oncol., 86, 172–178.
17. Balch,C.M. et al. (2004) An evidence-based staging system for cutaneous
melanoma. CA. Cancer J. Clin., 54, 131–49; quiz 182.
18. Mellman,I. et al. (2011) Cancer immunotherapy comes of age. Nature,
480, 480–489.
19. Gkalpakiotis,S. et al. (2010) Quantitative detection of melanoma-associated antigens by multimarker real-time RT-PCR for molecular staging:
results of a 5 years study. Exp. Dermatol., 19, 994–999.
20. Ribas,A. et al. (2011) Intra-lymph node prime-boost vaccination against
Melan A and tyrosinase for the treatment of metastatic melanoma: results
of a phase 1 clinical trial. Clin. Cancer Res., 17, 2987–2996.
21. Davies,H. et al. (2002) Mutations of the BRAF gene in human cancer.
Nature, 417, 949–954.
22. Braeuer,R.R. et al. (2011) Transcriptional control of melanoma metastasis:
the importance of the tumor microenvironment. Semin. Cancer Biol., 21,
83–88.
23. Fernandez-Salguero,P. et al. (1995) Immune system impairment and
hepatic fibrosis in mice lacking the dioxin-binding Ah receptor. Science,
268, 722–726.
24. Rico-Leo,E.M. et al. (2013) Dioxin receptor expression inhibits basal and
transforming growth factor β-induced epithelial-to-mesenchymal transition. J. Biol. Chem., 288, 7841–7856.
25. Rossi,L. et al. (2011) Hematopoietic stem cell characterization and isolation. Methods Mol. Biol., 750, 47–59.
26. Rey-Barroso,J. et al. (2013) The dioxin receptor controls β1 integrin activation in fibroblasts through a Cbp-Csk-Src pathway. Cell Signal., 25,
848–859.
27.Torres,V.A. et al. (2007) E-cadherin is required for caveolin-1-mediated downregulation of the inhibitor of apoptosis protein survivin via reduced betacatenin-Tcf/Lef-dependent transcription. Mol. Cell. Biol., 27, 7703–7717.
28. Pinon,P. et al. (2011) Integrins: versatile receptors controlling melanocyte
adhesion, migration and proliferation. Pigment Cell Melanoma Res., 24,
282–294.
29. Trimmer,C. et al. (2010) CAV1 inhibits metastatic potential in melanomas
through suppression of the integrin/Src/FAK signaling pathway. Cancer
Res., 70, 7489–7499.
30. Fukunaga-Kalabis,M. et al. (2010) Tenascin-C promotes melanoma progression by maintaining the ABCB5-positive side population. Oncogene,
29, 6115–6124.
31. Quintana,E. et al. (2008) Efficient tumour formation by single human melanoma cells. Nature, 456, 593–598.
32. Farnie,G. et al. (2007) Mammary stem cells and breast cancer–role of
Notch signalling. Stem Cell Rev., 3, 169–175.
33. Pohjanvirta,R. (2012) The AH Receptor in Biology and Toxicology. John
Wiley & Sons, New York, NY.
34. Pećina-Slaus,N. et al. (2007) E-cadherin and beta-catenin expression patterns in malignant melanoma assessed by image analysis. J. Cutan. Pathol.,
34, 239–246.
35. Tucci,M.G. et al. (2007) Involvement of E-cadherin, beta-catenin, Cdc42
and CXCR4 in the progression and prognosis of cutaneous melanoma.
Br. J. Dermatol., 157, 1212–1216.
36. Capozza,F. et al. (2012) Genetic ablation of Cav1 differentially affects
melanoma tumor growth and metastasis in mice: role of Cav1 in Shh heterotypic signaling and transendothelial migration. Cancer Res., 72, 2262–2274.
AhR expression suppresses melanoma progression
37. Urra,H. et al. (2012) Caveolin-1-enhanced motility and focal adhesion
turnover require tyrosine-14 but not accumulation to the rear in metastatic
cancer cells. PLoS One, 7, e33085.
38. Arpaia,E. et al. (2012) The interaction between caveolin-1 and RhoGTPases promotes metastasis by controlling the expression of alpha5integrin and the activation of Src, Ras and Erk. Oncogene, 31, 884–896.
39. Thiery,J.P. et al. (2009) Epithelial-mesenchymal transitions in development and disease. Cell, 139, 871–890.
40. Zhao,S. et al. (2012) Activation of the aryl hydrocarbon receptor represses
mammosphere formation in MCF-7 cells. Cancer Lett., 317, 192–198.
41. Boitano,A.E. et al. (2010) Aryl hydrocarbon receptor antagonists promote
the expansion of human hematopoietic stem cells. Science, 329, 1345–1348.
42. Gomez-Duran,A. et al. (2009) Fitting a xenobiotic receptor into cell
homeostasis: how the dioxin receptor interacts with TGFbeta signaling.
Biochem. Pharmacol., 77, 700–712.
43. Straussman,R. et al. (2012) Tumour micro-environment elicits innate
resistance to RAF inhibitors through HGF secretion. Nature, 487, 500–504.
44. Quintana,F.J. et al. (2008) Control of T(reg) and T(H)17 cell differentiation
by the aryl hydrocarbon receptor. Nature, 453, 65–71.
45.Veldhoen,M. et al. (2008) The aryl hydrocarbon receptor links TH17-cellmediated autoimmunity to environmental toxins. Nature, 453, 106–109.
46. Platten,M. et al. (2012) Tryptophan catabolism in cancer: beyond IDO and
tryptophan depletion. Cancer Res., 72, 5435–5440.
47. Adams,S. et al. (2012) The kynurenine pathway in brain tumor pathogenesis. Cancer Res., 72, 5649–5657.
48. Bergander,L. et al. (2003) Characterization of in vitro metabolites of the
aryl hydrocarbon receptor ligand 6-formylindolo[3,2-b]carbazole by liquid
chromatography-mass spectrometry and NMR. Drug Metab. Dispos., 31,
233–241.
49. Wang,T. et al. (2011) Activation of the aryl hydrocarbon receptor by TCDD
inhibits mammary tumor metastasis in a syngeneic mouse model of breast
cancer. Toxicol. Sci., 124, 291–298.
Received May 20, 2013; revised June 27, 2013; accepted July 6, 2013
2693