Activation of Sonic Hedgehog Signaling Pathway in ...

Clinical Translational Research

Oncology

Received: December 10, 2008 Accepted after revision: April 13, 2009 Published online: September 7, 2009

Oncology 2009;77:231–243 DOI: 10.1159/000236047

Activation of Sonic Hedgehog Signaling Pathway in Olfactory Neuroblastoma Ling Mao a Yuan-peng Xia a Yu-nan Zhou a Ruo-lian Dai a Xue Yang a Yan-jun Wang b Shu-jie Duan a Xian Qiao a Yuan-wu Mei a Bo Hu a Departments of a Neurology and b Otolaryngology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China

Key Words Apoptosis ⴢ Cell cycle ⴢ Neuroblastoma ⴢ Proliferation ⴢ Sonic hedgehog

Abstract Objectives: Sonic hedgehog (Shh) signaling pathway is associated with tumor development; however, the role of Shh signaling in the development of olfactory neuroblastoma (ONB) is unknown. This study aimed to investigate the relationship between the regulation of Shh signaling and the pathogenesis of ONB. Methods: The expression of Shh signaling components was characterized by immunohistochemistry in human non-tumor olfactory epithelium and ONB specimens, and by RT-PCR and immunoblotting in human ONB cell lines. The impact of the treatment with cyclopamine (a selective inhibitor of the Shh pathway) and/or exogenous Shh on ONB cell proliferation, cycle and apoptosis was examined by MTT, soft agar colony formation and flow cytometry assays, respectively. The influence of Shh signaling on the expression of Shh signaling components and cell cycle-related regulators was determined by immunoblotting and quantitative RT-PCR, respectively. Results: The expression of Pacthed1, Gli1 and Gli2 was detected in 70, 70, and 65% of human ONB specimens, respectively, and in proportion of ONB cell lines, but not in non-tumor olfactory epithelium. Treatment with cyclopamine inhibited the proliferation and colony formation of ONB cells, induced ONB cell cycle arrest and apoptosis, and down-regulated the expres-

© 2009 S. Karger AG, Basel 0030–2414/09/0774–0231$26.00/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/ocl

sion of Pacthed1, Gli1 and cyclin D1, but up-regulated p21 expression in vitro. These regulatory effects of cyclopamine were partially or completely erased by exogenous Shh. Conclusion: These data suggest that the Shh signaling pathway is crucial for the growth of ONB. Copyright © 2009 S. Karger AG, Basel

Introduction

Olfactory neuroblastoma (ONB) is a rare malignant tumor and is thought to arise from the olfactory epithelium in the vault of the nasal cavity [1]. ONB represents approximately 5% of all nasal malignant tumors in people of different ages. Since ONB was described by Berger and Luc in 1924, only about 1,000 cases have been reported in the literature [2–4]. Although current treatments can improve patients’ quality of life and prolong their survival, ONB often recurs soon or several years after treatment [5, 6]. Therefore, the discovery of the pathogenesis and molecular mechanism underlying the regulation of ONB cell growth will be of great significance and may help in developing a new therapy for ONB in humans. The sonic hedgehog (Shh) signaling pathway has been demonstrated to regulate embryo development and tumor cell proliferation in several human neoplasms, such

Ling Mao and Yuan-peng Xia contributed equally to this study.

Dr. Bo Hu Department of Neurology, Union Hospital Tongji Medical College, Huazhong University of Science and Technology Wuhan 430022 (China) Tel./Fax +86 27 8575 5457, E-Mail [email protected]

as the prostate, pancreas and skin carcinomas [7–10]. Shh is a member of the hedgehog family that was originally identified by its homology to the Drosophila melanogaster segment polarity gene, hedgehog. Shh can activate the Shh signaling pathway by binding to its membrane receptor, Patched, consequently inactivating Patched, which is associated with transmembrane protein Smoothened (SMO), and leading to the downstream activation of target genes, such as the transcription factors of the Gli family. The Gli family members can regulate the expression of many target genes involved in the growth and differentiation of cells in a wide range of tissues. Aberrant activation of the Shh signaling pathway has been implicated in the pathogenesis of various cancer types [11–13]. However, there is no available information on the role of the Shh signaling pathway in the pathogenesis of human ONB. In this study, we characterized the expression of Patched, Gli1 and Gli2 in human non-tumor olfactory epithelial tissues, ONB specimens and cell lines by immunohistochemistry, RT-PCR and immunoblotting assays, respectively. Furthermore, we evaluated the impact of Shh and/or cyclopamine, a selective inhibitor of the Shh signaling pathway, on the growth, cell cycle and apoptosis of human ONB cells, and on the expression of Patched, Gli1, Gli2 and cell cycle-related regulators in ONB cell lines in vitro by immunoblotting and RT-PCR assays, respectively. We found that the expression of Patched1, Gli1 and Gli2 was detected in some human ONB samples and cell lines. Treatment with cyclopamine inhibited the proliferation and colony formation of ONB cells, induced ONB cell-cycle arrest and apoptosis, which were associated with the down-regulation of the expression of Patched1, Gli1 and Cyclin D1 and the up-regulation of p21 expression in ONB cell lines. The regulatory effects of cyclopamine were partially or completely erased by the addition of exogenous Shh. Our data suggest that the Shh signaling pathway may be involved in the pathogenesis of human ONB.

Materials and Methods Clinical Specimens A total of 20 ONB samples were collected from Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China (n = 13), and Huaxi Hospital, Sichuan University, Chengdu, China (n = 7) from 1974 to 2007. The olfactory epithelial samples (n = 12) were collected from patients who were under surgical procedures for benign diseases in the department of otolaryngology, Union Hospital. The diagnosis of ONB

232

Oncology 2009;77:231–243

was reviewed by neuropathologists and confirmed by histological examination. Their pathological characteristics were classified according to the criteria of the Hyams grading system and the Kadish staging system. Written informed consent was obtained from individual patients and experimental protocols were reviewed and approved by the office for scientific research of each hospital. Immunohistochemistry The expression of Patched1, Gli1 and Gli2 in 20 tumors and 12 non-tumor olfactory epithelial samples was characterized simultaneously by routine immunohistochemistry. The paraffin-embedded ONB and olfactory epithelial samples were sectioned at 4 ␮m. After paraffin was removed, the tissue sections were rehydrated and boiled in 10 mM citrate buffer (pH 6.0) for 20 min for retrieving antigens. Following quenching of endogenous peroxidase activity with H2O2 and blocking with normal 10% horse serum for 30 min, the tissue sections were probed with goat antiPatched1 (1: 100; Santa Cruz Biotechnolgy, Santa Cruz, Calif., USA), rabbit anti-Gli1 or Gli2 (1: 300 or 1: 200; Abcam, Cambridge, Mass., USA), respectively, overnight at 4 ° C. After being washed, the sections were incubated with either biotinylated rabbit anti-goat IgG or goat anti-rabbit IgG, and the ABC reagent (Vector Laboratories, Burlingame, Calif., USA). The immunostained sections were visualized with 3, 3ⴕ-diaminobenzidine (Sigma, St. Louis, Mo., USA) and then counterstained with hematoxylin. The sections without the primary antibody were used as negative controls. The relative expression levels of Patched1, Gli1 and Gli2 were determined semi-quantitatively by assessing both the percentage of decorated tumor cells and the staining intensity, as described previously [14]. The relative expression levels of individual proteins for each section were scored as follows: 1 = ^10% positive cells; 2 = 11–50% positive cells; 3 = 51–80% positive cells, and 4 = 180% positive cells. The intensity of the staining was rated as follows: 1 = weak for the majority of positive cells; 2 = moderate, and 3 = strong. A score of the protein expression for the individual cases was equal to the sum of the values of the percentage and intensity scores. Three sections from each case were stained for each target protein and a total of 1,000 cells were counted from at least 4 fields (!400) randomly selected from each section in order to determine the percentage of positive cells. Cytoplasmic and/or membrane immunostaining was considered to be positive for Patched1, while nuclear staining was regarded as positive for Gli1 and Gli2 as the biologically active form of these transcription factors are mainly located in the nucleus. Cell Culture Human olfactory neuroblastoma cell lines, TC-268 and JFEN, were the kind gifts of Dr. Han-Fei Ding (Medical College of Georgia, USA), were maintained in RPMI 1640 medium (Gibco; Invitrogen, Carlsbad, Calif., USA) supplemented with antibiotics and 10% FBS (Gibco). All cells were cultured at 37 ° C in a humidified incubator with 5% CO2. RT-PCR Total RNA was isolated from equal numbers of JFEN and TC268 cells (107 cells per tube) using the RNeasy mini kit (Qiagen, Valencia, Calif., USA) and reversely transcripted into cDNA by using the Transcriptor First Strand cDNA Synthesis kit (Roche,

Mao /Xia /Zhou /Dai /Yang /Wang /Duan / Qiao /Mei /Hu

USA) according to the manufacturers’ instructions. The expression of Patched1, Gli1, Gli2 and control ␤-actin was characterized by PCR by using the cDNA as the template, PuReTaq Ready-to-Go PCR Beads (Amersham Biosciences, Amersham, UK), and specific primers (online suppl. table 1; all online material for this article can be found at www.karger.com/doi/10.1159/000236047). PCR reactions were first denatured at 95 ° C for 4 min and then subjected to 35 cycles of 94 ° C for 40 s, 60 ° C for 30 s and 72 ° C for 45 s, followed by extension at 72 ° C for 10 min. The PCR products were characterized by electrophoresis and imaged. Quantitative RT-PCR JFEN and TC-268 cells at 2 ! 106/flask were treated in duplicate with 20 ␮M cyclopamine and/or 3 ␮g/ml Shh or the vehicle in the medium for 48 h. After the cells were harvested, their total RNA was extracted and reversely transcripted into cDNA. The levels of cyclin D1, p21 and control ␤-actin mRNA transcripts were determined by quantitative RT-PCR by using the SYBR Green PCR Master Mix (Applied Biosystems, Foster City, Calif., USA) and the specific primers (online suppl. table 1, [15]) on an ABI Prism 7700 Sequence Detector (Applied Biosystems). The PCR reactions were denatured at 95 ° C for 15 min and then subjected to 35 cycles of 94 ° C for 30 s, 63 ° C for 30 s, and 72 ° C for 1 min, followed by extension at 72 ° C for 10 min. The individual value was first normalized to that of the control ␤-actin and then the ratio of the relative levels over that of the vehicle-treated cells was calculated. Immunoblotting Assay The levels of Patched1, Gli1, Gli2, Shh and control ␣-tubulin expression in ONB cell lines were characterized by the immunoblotting assay. JFEN and TC-268 cells (106 cells/tube) were lysed in a standard SDS protein lysis buffer. Their protein concentrations were determined with the Bio-Rad protein assay kit, using bovine serum albumin as a reference. Equal cell lysate proteins (50 ␮g) were separated on SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. The membranes were probed with goat anti-Shh (1:200, N-19; Santa Cruz Biotechnology), goat anti-Patched1 (1: 200; Santa Cruz Biotechnology), rabbit anti-Gli1 or Gli2 (1:1,000; Abcam), or mouse anti- ␣-tubulin (1:5,000, B-5-1-2; Sigma). After washing, the bound antibodies were detected by using horseradish peroxidase-conjugated secondary antibodies and visualized using a SuperSignaling West Pico chemiluminescence kit (Pierce). The impact of the activation or inhibition of the Shh signaling pathway on the expression of Patched1, Gli1, Gli2 and control ␣tubulin in ONB cells was determined by immunoblotting assays. JFEN and TC-268 cells at 2 ! 106/flask were treated in duplicate with 20 ␮ M cyclopamine and/or 3 ␮g/ml Shh or the vehicle in the medium for 48 h. The cells were harvested and subjected to immunoblotting analysis. The relative levels of each protein expression to control ␣-tubulin were quantified with a Kodak Image Station 440CF. ONB Cell Proliferation Assay ONB cell proliferation was assessed using the reagent 3-[4, 5dimethylthiazol-2-yl]-2, 5-diphenyltetrazolium bromide (MTT; Sigma), according to the manufacturer’s protocol. Briefly, JFEN and TC-268 cells at 4 ! 103/well were treated in triplicate with different concentrations (5, 10 or 20 ␮M [15]) of cyclopamine or

Shh Signaling in Olfactory Neuroblastoma

the vehicle in 0.5% FBS medium in a 96-well plate for 24, 48 or 72 h, respectively. During the last four 4 h of incubation, the cells were treated with 10 ␮l MTT (5 mg/ml) per well. After the supernatants were discarded, the MTT crystals were dissolved in DMSO, and measured by the absorbance at a wavelength of 570 nm on a microplate reader. ONB cell proliferation was determined as the absorbance of treated versus vehicle-treated control cells and expressed as the proliferation rate (%). The proliferation of vehicletreated ONB cells was designed as 100%. Additionally, JFEN and TC-268 cells at 4 ! 103/well were treated in triplicate with 20 ␮M cyclopamine and/or 3 ␮g/ml Shh or the vehicle for 48 h. Their effects on ONB cell proliferation were determined by MTT assay. ONB Cell Colony Formation Assay The effect of the activation or inhibition of the Shh signaling pathway on anchorage-independent proliferation of ONB cells was examined by the soft agar colony formation assay, as described previously [16]. Briefly, JFEN and TC-268 cells at 1,500 cells/well were mixed in 0.3% Noble agar (in 10% FBS medium) containing 20 ␮M cyclopamine and/or 3 ␮g/ml Shh, and loaded in duplicate onto a solidified bottom layer (0.6% Noble agar in 10% FBS medium containing the same compound and/or reagents) in 6-well plates. After incubation at 37 ° C for 14–21 days, the colonies formed were directly counted or stained with 5 mg/ml MTT and photographed. The total number of visible colonies (usually containing near to or more than 50 cells per colony) was counted. Additionally, the cell number in each colony was counted by randomly selecting 10 colonies and re-suspending them in PBS. Cell Cycling Assay The impact of the activation or inhibition of the Shh signaling pathway on ONB cell cycling was examined by FACS analysis. Briefly, JFEN and TC-268 cells at 106 cells/well were treated in duplicate with 20 ␮ M cyclopamine and/or 3 ␮g/ml Shh or the vehicle in the medium for 48 h. The cells were harvested, washed with PBS twice, and fixed in 75% ethanol overnight at –20 ° C. Subsequently, the cells were digested with 0.2 mg/ml RNase (Sigma) at 70 ° C for 20 min, and stained with 0.5 mg/ml propidium iodide (PI, Sigma) at room temperature for 30 min. The DNA contents were assayed by flow cytometer analysis. The different phases of cells were recorded in histograms and their frequencies were analyzed using the Cellquest software (Becton Dickinson, San Jose, Calif., USA). Apoptosis Assay The effects of the activation or inhibition of the Shh signaling pathway on ONB cell apoptosis were determined by FACS analysis using the annexin V-FITC apoptosis detection kit (Biovision, Mountain View, Calif., USA), according to the manufacturer’s instructions. Briefly, JFEN and TC-268 cells at 106 cells/well were treated in duplicate with 20 ␮ M cyclopamine and/or 3 ␮g/ml Shh or the vehicle in the medium for 48 h. After being washed with binding buffer, the cells were stained with 5 ␮l of FITC-conjugated annexin V and 10 ␮l PI for 15 min at room temperature, and then analyzed by FACS on a flow cytometer (Becton Dickinson). The annexin V+/PI– cells were considered to be at early-stage apoptosis, the annexin V+/PI+ cells were late-stage apoptosis, annexin V–/PI+ cells were necrotic while annexin V–/PI– cells were healthy.

Oncology 2009;77:231–243

233

Statistical Analysis Quantitative data are expressed as mean 8 SD and analyzed by 1-way ANOVA, followed by Dunnett’s post hoc test. The difference in each measure between 2 groups was determined by 2tailed Student’s t test. The correlation between the 2 measures was analyzed by regression analysis. Statistical analysis was performed using the SPSS 13.0 software (SPSS, Chicago, Ill., USA). A p value !0.05 was considered statistically significant.

Results

Expression of Patched1, Gli1 and Gli2 in Human Non-Tumor Olfactory Epithelium, ONB Specimens and ONB Cell Lines The activation of the Shh signaling pathway is associated with the growth of tumors. To determine the potential role of Shh signaling in the development of ONB, the expression of Patched1, Gli1 and Gli2 in human ONB specimens and non-tumor olfactory epithelial samples was characterized by routine immunohistochemistry. A total of 20 ONB patients and 12 non-tumor subjects were recruited in this study and their demographic characteristics, pathological classifications and states of Patched1, Gli1 and Gli2 expression are presented in online supplementary table 2. While the expression of Patched1, Gli1 and Gli2 was not detected in any of the 12 human olfactory epithelial samples from non-ONB subjects (fig. 1a), different levels of Patched1, Gli1 and Gli2 expression were found in 70, 70 and 65% of the 20 ONB specimens, respectively. The expression of Patched1 was mainly located on the cell membrane while that of Gli1 and Gli2 was located predominately in the nucleus (fig. 1b). Quantitative analysis of the expression of these proteins revealed lower levels of membrane-associated Patched1 and higher levels of nuclear Gli1 expression in those with higher degrees of ONB (Kadish stage C or Hyams grade III/IV; 1.9 8 2.42 vs. 5.6 8 2.22 for Patched1 while 6.50 8 0.76 vs. 3.71 8 1.89 for Gli1), when compared with those with lower degrees of ONB (Kadish stage A/B or Hyams grade I/II), but no significant difference in the expression of nuclear Gli2 was found among varying degrees of ONB specimens (table 1). The levels of Patched1 were negatively correlated with degree of ONB, while those of Gli1 were positively correlated (r = –0.975, p ! 0.01 for Patched1, and r = 0.986, p ! 0.01 for Gli1), but they were not associated with the patients’ age, gender or tumor location(s). Furthermore, the Patched1, Gli1, Gli2 and Shh mRNA transcripts were detected in both human ONB cell lines, JFEN and TC-268 (fig. 1c). Patched1, Gli1, Gli2 and Shh 234

Oncology 2009;77:231–243

proteins were found in JFEN cells, but only Patched1 and Shh were found in TC-268 cells (fig. 1d). Therefore, different levels of Shh signaling pathway activation are displayed in different degrees of human ONB. Shh Signaling Pathway Regulates ONB Cell Proliferation To determine the potential role of the Shh signaling pathway, we examined the effect of treatment with an inhibitor for the Shh signaling pathway on ONB cell proliferation. JFEN and TC-268 cells were treated with different doses of cyclopamine, a selective inhibitor for the Shh signaling pathway, or the vehicle for 24–72 h and their proliferation was determined by MTT assay. Treatment with 5 ␮M cyclopamine for 24 h inhibited the proliferation of JFEN cells by 20%, with increased concentrations greatly enhancing its inhibition of JFEN cell proliferation (fig. 2a). Furthermore, treatment with 5 ␮M cyclopamine for a longer period increased the inhibition of JFEN cell proliferation. A similar pattern of weaker inhibitory effects of cyclopamine was observed on TC-268 cells (fig. 2b). Therefore, the blockage of the Shh signaling pathway inhibited the proliferation of ONB cells in a time- and dose-dependent manner in vitro. Next, we tested the impact of exogenous Shh on ONB cell proliferation and the inhibitory effect of cyclopamine. JFEN cells were treated with 20 ␮M cyclopamine and/or 3 ␮g/ml Shh or the vehicle for 48 h and the states of cell growth were examined under a microscope (fig. 2c) and quantified in figure 2d. Obviously, treatment with exogenous Shh alone did not significantly enhanced JFEN cell proliferation. However, treatment with Shh together

Fig. 1. The characterization of Shh signaling component expression in human non-tumor olfactory epithelial tissues, ONB specimens and cell lines. The expression of Patched1, Gli1 and Gli2 in 12 human non-tumor olfactory epithelial tissues and 20 human ONB specimens was characterized simultaneously by immunohistochemistry. The mRNA transcription of Shh, Patched1, Gli1 and Gli2 genes and their proteins in JFEN and TC-268 cells were characterized by RT-PCR or immunoblotting, respectively. The expressions of ␤-actin or ␣-tubulin were used as controls. Data are representative images from groups of tissues or 3 independent RT-PCR or immunoblotting assays. a Immunohistochemical analysis of Patched1, Gli1 and Gli2 expression in human non-tumor olfactory epithelial tissues. b Immunohistochemical analysis of Patched1, Gli1 and Gli2 expression in human ONB samples. Original magnifications !400. RT-PCR (c) and immunoblotting analysis (d) of Shh, Patched1, Gli1 and Gli2 expression in human ONB cell lines. Shh-N is an active form of 19-kDa Shh.


Normal human olfactory epithelium samples Gli1

Gli2

Gli1

Gli2

Color version available online

Patched1

a Human primary ONB samples Patched1

TC-268

JFEN

TC-268

JFEN

b

Shh Shh-N Shh Patched1 Patched1 Gli1 Gli1 Gli2 Gli2 ␣-Tubulin

Actin c


d

Oncology 2009;77:231–243

235

Table 1. Patched1, Gli1 and Gli2 expression in different degrees of ONB

Positive tumors Kadish stageA Positive tumor Expression level Kadish stage B Positive tumor Expression level Kadish stage C Positive tumor Expression level Low Hyams grade (I/II) Positive tumor Expression level High Hyams grade (III/IV) Positive tumor Expression level

Patched1

Gli1

Gli2

14/20 (70%)

14/20 (70%)

13/20 (65%)

6/8 (75%) 4.2582.87a

4/8 (50%) 4.6082.70

2/8 (25%) 1.3280.54

5/5 (100%) 6.680.89

4/5 (80%) 4.5081.73

5/5 (100%) 1.4581.23

3/7 (42.86%) 1.1481.57a

6/7 (85.71%) 6.1781.17

6/7 (85.71%) 2.0981.45

9/10 (90%) 5.682.22b

6/10 (60%) 3.7181.89c

5/10 (50%) 1.0280.59

5/10 (50%) 1.982.42b

8/10 (80%) 6.5080.76c

8/10 (80%) 2.6781.08

Results with the same superscript letter are significantly different from one another (p < 0.05).

with cyclopamine significantly erased cyclopamine-mediated inhibition of JFEN cell proliferation in vitro. In addition, the effect of exogenous Shh or cyclopamine on the anchorage-independent proliferation of ONB cells was tested in a soft agar colony formation assay. The numbers of total colonies in different groups of JFEN and TC-268 cells were significantly different (p ! 0.05, 1-way ANOVA). While an average of 12 or 30 colonies were observed in JFEN or TC-268 cells treated with cyclopamine, near 50 or 70 colonies were detected after treatment with both cyclopamine and Shh (fig. 3a–d). Moreover, the mean numbers of cells per colony displayed a significant difference. Treatment with cyclopamine alone significantly reduced the number of each colony in both cell lines, while treatment with both cyclopamine and Shh partially or nearly completely erased the inhibition of cyclopamine on the growth of ONB cells in vitro (fig. 3e–h). Together, these data indicated that the blockage of Shh signaling inhibited the anchorage-independent proliferation of ONB cells in vitro. Inhibition of Shh Signaling Induces Cell Cycle Arrest in ONB Cell Lines Inhibition of ONB cell proliferation may be associated with modulating cell cycling. Next, we examined the impact of treatment with cyclopamine and/or Shh on ONB cell cycling. JFEN and TC-268 cells were treated with 20 ␮M cyclopamine and/or 3 ␮g/ml Shh or vehicle for 48 h 236

Oncology 2009;77:231–243

and the status of ONB cell cycling was determined by FACS analysis (fig. 4). Clearly, a significantly higher frequency of JFEN cells that had been treated with cyclopamine were at subG1 and G0/G1 phases while a lower frequency were at S phase, as compared with that of vehicletreated cells. After treatment with both cyclopamine and Shh, the frequency of JFEN cells at subG1 and G0/G1 phases was significantly reduced, as compared with the frequency of the cells treated with cyclopamine alone. Similar, but less effective, blockage patterns of Shh signaling on cell cycling were observed on TC-268 cells. These data indicated that the blockage of Shh signaling induced ONB cell cycle arrest, which was partially erased by the addition of Shh in vitro. Inhibition of Shh Signaling Induces ONB Cell Apoptosis in vitro We further determined the effect of treatment with the blockage of the Shh signaling pathway and/or exogenous Shh on ONB cell apoptosis by FACS analysis. Treatment with 3 ␮g/ml Shh did not significantly modulate the frequency of apoptotic JFEN and TC-268 cells, which was similar to that of vehicle-treated controls (fig. 5). Treatment with 20 ␮M cyclopamine significantly elevated the frequency of apoptotic JFEN and TC-268 cells, as compared with that of vehicle-treated controls. However, treatment with both cyclopamine and Shh significantly reduced the frequency of apoptotic JFEN and TC-268 Mao /Xia /Zhou /Dai /Yang /Wang /Duan / Qiao /Mei /Hu

120

Vehicle

C5

C10

120

C20

80 60

100

*

* *

*

*

*

40

*

20

a

0

TC-268 proliferation (%)

JFEN proliferation (%)

100

24 h

V

* 48 h

* *

*

* *

*

*

*

60

*

40 20

* 72 h

80

b

0

24 h

48 h

72 h

C20 120

S3

C20 + S3

JFEN proliferation (%)

100

**

80 60

*

40 20

d

0

V

C20

S3

C20 + S3

c

Fig. 2. The effects of the Shh signaling pathway on ONB cell proliferation. JFEN and TC-268 cells were treated with 5, 10 or 20 ␮ M cyclopamine or vehicle-treated for 24, 48 or 72 h, respectively. Their proliferation was determined by MTT assay. Additionally, JFEN cells were treated with 20 ␮ M and/or 3 ␮g/ml Shh or the vehicle for 48 h and their growth was examined under a microscope and quantitatively analyzed by MTT assay. The proliferation of control cells at each time point was designated as 100%. Data shown are representative images or expressed as mean percent 8 SD of proliferation by individual groups of cells at each

time point relative to the controls of 3 separated experiments. Percent proliferation of JFEN cells (a). Percent proliferation of TC268 cells (b). Representative images of JFEN cells (c). Quantitative analysis of JFEN cell proliferation (d). V = Vehicle treatment; C5 = cells treated with 5 ␮ M cyclopamine; C10 = cells treated with 10 ␮ M cyclopamine; C20 = cells treated with 20 ␮ M cyclopamine; S3 = cells treated with 3 ␮g/ml Shh; C20 + S3 = cells treated with both 20 ␮ M cyclopamine and 3 ␮g/ml Shh. * p ! 0.05 vs. vehicle; ** p ! 0.05 vs. cyclopamine.


Oncology 2009;77:231–243

237

JFEN V

C20

V

C20

S3

C20 + S3

S3

Number of JFEN cell colonies

70

C20 + S3

60

40 30 20

a

10

c

a

b

50

0

V

C20

S3

C20 + S3

b JFEN

Number of TC-268 cell colonies

90

d

TC-268

V

C20

V

C20

S3

C20 + S3

S3

C20 + S3

b

80 70 60 50 a

40 30 20 10 0

V

C20

S3

C20 + S3

e

g

300 250

b

200 150 100 a 50 0

V

C20

S3

C20 + S3

Number of TC-268 cells per colony

Number of JFEN cells per colony

350

f

h

600 500 b 400 a 300 200 100 0

V

Fig. 3. The effect of the Shh signaling pathway on colony formation of ONB cells in vitro. JFEN and TC-268 cells were treated with 20 ␮ M cyclopamine and/or 3 ␮g/ml Shh or the vehicle in a soft agar culture system for 2–3 weeks. The colonies formed were stained with MTT and imaged for counting or directly counted for average number of ONB cells per colony by randomly selecting 10 colonies and then re-suspending the cells. Images are representative of each group of colonies. Original magnifications !100. Histograms show mean 8 SD of each group from 3 independent experiments. MTT staining of JFEN cell colonies (a). MTT stain-

238

Oncology 2009;77:231–243

C20

S3

C20 + S3

ing of TC-268 cell colonies (b). Number of JFEN cell colonies (c). Number of TC-268 cell colonies (d). Unstained JFEN cell colony (e). Unstained TC-268 cell colony (f). Number of JFEN cells per colony (g). Number of TC-268 cells per colony (h). V = Vehicle treatment; C5 = cells treated with 5 ␮ M cyclopamine; C10 = cells treated with 10 ␮ M cyclopamine; C20 = cells treated with 20 ␮ M cyclopamine; S3 = cells treated with 3 ␮g/ml Shh; C20 + S3 = cells treated with both 20 ␮ M cyclopamine and 3 ␮g/ml Shh. a p ! 0.05 vs. vehicle; b p ! 0.05 vs. cyclopamine.


Color version available online

TC-268

V

80

C20

S3

C20 + S3

80 a b

a

60

JFEN cells (%)

b

40

b

a 20

b

b

TC-268 cells (%)

60

40

b 20

a

b

a

a

0

subG1

G0/G1

a

b

a S

G2/M

b

0

subG1

a

G0/G1

S

G2/M

Fig. 4. The effect of the Shh signaling pathway on ONB cell cycling. JFEN and TC-268 cells were treated with 20 ␮ M cyclopamine (C20) and/or 3 ␮g/ml Shh (S3) or the vehicle alone for 48 h and the status of cell cycling was analyzed by FACS analysis. Cell cycle analysis of JFEN cells (a). Cell cycle analysis of TC-268 cells (b). Data are expressed as mean percent 8 SD of each phase of cells from 3 independent experiments. a p ! 0.05 vs. vehicle; b p ! 0.05 vs. cyclopamine.

35

16 a

14

25 20 b 15 10

TC-268 cell apoptosis (%)

JFEN cell apoptosis (%)

30

5

a

0

a

12

b

10 8 6 4 2

V

C20

S3

C20 + S3

b

0

V

C20

S3

C20 + S3

Fig. 5. The effect of the Shh signaling pathway on ONB cell apoptosis. JFEN and TC-268 cells were treated with

20 ␮ M cyclopamine and/or 3 ␮g/ml Shh or the vehicle for 48 h and their apoptosis was determined by FACS analysis. The Annexin V+/PI– cells were considered to be at early-stage apoptosis, the Annexin V+/PI+ cells were at a late-stage apoptosis, Annexin V–/PI+ cells were necrotic while Annexin V–/PI– cells were healthy. a Analysis of JFEN cells. b Analysis of TC-268 cells. Data are expressed as mean percent 8 SD of each group of cells from 3 independent experiments. a p ! 0.05 vs. vehicle; b p ! 0.05 vs. cyclopamine.


Oncology 2009;77:231–243

239

1.0

V

C20

S3

C20 + S3

Relative protein levels

0.8

Patched1 Gli1 Gli2

0.6

b

0.4 b 0.2

a

a

Tubulin V

a

b

JFEN

1.4

0

Patched1

2.5

Gli1

a

Gli2

JFEN

1.2 1.0 b

0.8 0.6

a

0.4

Relative p21 mRNA levels

Relative cyclin D1 mRNA levels

C20 S3 C20 + S3

2.0 b 1.5

1.0

0.5 0.2 0

V

C20

C20 + S3

d

TC-268

1.4

Relative cyclin D1 mRNA levels

S3

b

1.0 a

0.8 0.6 0.4

240

0

V

C20

S3

C20 + S3

TC-268 a

1.4

1.2

0.2

e

0

1.6

Relative p21 mRNA levels

c

b

1.2 1.0 0.8 0.6 0.4 0.2

V

C20

S3

Oncology 2009;77:231–243

C20 + S3

f

0

V

C20

S3

C20 + S3


cells, as compared with that of cyclopamine treatment alone. Therefore, the blockage of the Shh signaling pathway induced ONB cell apoptosis in vitro, which was partially rescued by exogenous Shh. Inhibition of Shh Signaling Modulates the Expression of Shh Signaling Components and Cell Cycle Regulators in ONB Cells To understand the mechanism underlying the effect of Shh and/or cyclopamine on ONB cell proliferation and cycling, the expression levels of Shh signaling events (Pathched1, Gli1 and Gli2 and cycling regulators cyclin D1 and p21) were characterized by immunoblotting and quantitative RT-PCR assays, respectively. As shown in figure 6a, b, treatment with 3 ␮g/ml Shh did not significantly modulate the expression of Patched1, Gli1 and Gli2 in JFEN cells, as the relative levels of them to control ␣-tubulin were similar to those of vehicle-treated control cells. Treatment with 20 ␮M cyclopamine alone for 48 h significantly down-regulated the expression of Patched1 and Gli1 but not Gli2, as compared with that of vehicletreated cells. More importantly, treatment with both cyclopamine and Shh partially rescued the cyclopaminemediated down-regulation of Patched1 and Gli1 expression in JFEN cells in vitro. Apparently, the blockage of Shh signaling down-regulates the expression of some Shh signaling events in ONB cells in vitro. Quantitative analysis of cyclin D1 and p21 mRNA transcripts revealed that treatment with cyclopamine alone down-regulated cyclin D1, but up-regulated p21 gene transcription in both JFEN and TC-268 cells in vitro. These regulatory effects of cyclopamine treatment were partially mitigated by the addition of exogenous Shh in both ONB cell lines.

Fig. 6. The effect of the Shh signaling pathway on the expression

of Shh signaling components and cell cycle-related regulators in ONB cells. JFEN cells were treated with 20 ␮ M cyclopamine and/ or 3 ␮g/ml Shh or vehicle for 48 h, and the expression of Patched1, Gli1, Gli2 and control ␣-tubulin was characterized by immunoblotting. Additionally, JFEN and TC-268 cells were treated as described above and the relative levels of cyclin D1 and p27 mRNA transcripts were determined by quantitative RT-PCR. Images are representative of each group. Data in histograms is expressed as mean value 8 SD of each group of cells relative to the value of control. a Immunoblotting of Shh signaling events. b Relative levels of proteins. c Relative levels of cyclin D1 in JFEN cells. d Relative levels of p21 in JFEN cells. e Relative levels of cyclin D1 in TC-268 cells. f Relative levels of p21 in TC-268 cells. a p ! 0.05 vs. vehicle; b p ! 0.05 vs. cyclopamine.


Discussion

The Shh signaling pathway has been associated with the development of tumors. Patched1, associated with the transmembrane protein SMO, is engaged by Shh, which releases SMO, leading to the activation of Shh signaling [11]. Gli family members are transcription activators or repressors, and regulate the expression of Shh signaling targeted genes [13]. To explore the potential role of Shh signaling in the development of ONB, we examined the expression of Patched1, Gli1 and Gli2, major components of the Shh signaling pathway, in 12 human non-tumor olfactory epithelial samples and 20 ONB specimens by immunohistochemistry. We found that no single protein was detected in any of the non-tumor olfactory epithelial samples. Furthermore, we found that Patched1, Gli1 and Gli2 were expressed in 70, 70 and 65% of ONB specimens, respectively. The levels of Patched1 or Gli1 appeared to be either negatively or positively correlated with the pathological degrees of ONB. The lower levels of Patched1 and higher levels of Gli1 expression in higher degrees of ONB suggest that Shh signaling may predominately regulate the progression of ONB. Notably, the Shh signaling pathway has been shown to regulate the growth of several kinds of cells [17]. Surprisingly, we did not detect its component expression in non-tumor olfactory epithelial tissues. The failure may be due to the very low level of Shh in non-tumor olfactory epithelial tissues that may be not detected by the available antibodies. Alternatively, the Shh signaling pathway may not regulate non-tumor olfactory epithelial cells. The detection of Patched1, Gli1 and Gli2 in a higher frequency of ONB suggests that Shh signaling regulates the development of human ONB. It is possible that initial oncogenic factors may activate the oncogen genes, such as transcription factors, which induce the expression of Patched1 and SMO, and in turn regulate the development of ONB. However, whether the Shh signaling components are expressed in normal olfactory epithelial cells and what factors regulate their expression during the development of ONB remain to be further determined. We are interested in examining whether SMO and Gli3 are expressed in human ONB in future studies. Furthermore, we detected the Patched1, Gli1, Gli2 and Shh mRNA transcripts in both JFEN and TC-268 ONB cell lines and the Patched1, Gli1, Gli2 and Shh proteins in JFEN cells, while only Patched1 and Shh were detected in TC-268 cells. In consistence with this observation, although 40% of specimens showed the expression of 3 proteins, some other specimens displayed 1 or 2 proteins Oncology 2009;77:231–243

241

among the different stages of ONB (online suppl. table 2). Notably, those 2 cell lines were also derived from different patients with varying stages of ONB [18]. These data suggest that the Shh signaling components are differentially expressed in the different stages of ONB and different levels of Shh signaling activation regulate the development and progression of ONB in humans. Shh signaling is crucial for stem cell proliferation [15]. We found that treatment with cyclopamine, a selective inhibitor of the Shh signaling pathway, inhibited the proliferation of JFEN and TC-268 cells in a dose- and timedependent manner in vitro, similar to the results of previous reports in other types of tumor cells [17, 19]. Interestingly, JFEN cells were much more sensitive to the inhibition of cyclopamine than were TC-268 cells. The different susceptibilities of those cells may stem from their differential expressions of Shh signaling components. The higher susceptibility to the inhibition of cyclopamine in JFEN cells suggests that they may be more dependent on the Shh signaling for proliferation. Furthermore, we found that the addition of exogenous Shh ligands neither enhanced the anchorage-independent proliferation of JFEN and TC-268 cells nor affected their cell cycling and spontaneous apoptosis in vitro. These data suggest that Shh produced by those cell lines may be sufficient in supporting their proliferation in our experimental conditions. Given that Shh activates the Shh signaling pathway in an autocrine manner in many types of cells [20], the presence of Shh and activated Shh-N, and the exogenous Shh-independent activation of Shh signaling in JFEN cells suggest that the Shh signaling pathway may also regulate the proliferation of JFEN cells in an autocrine fashion. Importantly, treatment with both cyclopamine and Shh partially or completely erased cyclopamine-mediated inhibition of JFEN and TC-268 cell proliferation in vitro. Cyclopamine functions to regulate the balance of active and inactive forms of SMO, inhibiting Shh signaling [21]. Therefore, exogenous Shh may bind to Patched1, increase the active form of SMO, and reduce the inhibition of cyclopamine on ONB cell proliferation. However, the precise mechanisms underlying the action of exogenous Shh ligands in antagonizing the inhibitory effect of cyclopamine on ONB cell proliferation remain to be further investigated. To understand the mechanisms by which the blockage of the Shh signaling inhibits the proliferation of ONB cells, we examined the impact of treatment with cyclopamine and/or Shh on JFEN and TC-268 cell cycle and apoptosis. We found that treatment with cyclopamine 242

Oncology 2009;77:231–243

induced JFEN and TC-268 cell cycle arrest at the G0/G1 phases and a higher frequency of apoptotic JFEN and TC-268 cells, which may contribute to the inhibitory effect of cyclopamine on ONB cell proliferation. Notably, treatment with cyclopamine down-regulated the expression of Patched1 and Gli1, but not Gli2 expression in JFEN cells while treatment with exogenous Shh reversed the down-regulatory effect of cyclopamine. Consequently, treatment with exogenous Shh antagonized the effect of cyclopamine by reducing the frequency of apoptotic ONB cells and promoting ONB cell cycle progression, consistent with promoting ONB cell proliferation. In addition, we found that treatment with cyclopamine significantly inhibited the transcription of cyclin D1 and promoted higher levels of p21 expression in ONB cells. The D-cyclin’s expression and activity are critical to cell cycle progression from the G1 to the S phase [22], while higher levels of p21/CIP1 expression are usually associated with the arrest of the cell cycle at G1, thus blocking S-phase entry by inactivating CDKs [23]. The combination of reduced cyclin D1 expression and up-regulated p21 expression by cyclopamine induces ONB cell cycle arrest at the G1 phase. It has been shown that Shh signaling can up-regulate the expression of anti-apoptotic BcL-2, supporting the survival of medulloblastoma cells and promoting neuronal precursor cell cycling by upregulating the expression of cyclins [24]. Morton et al. [25] found that active Shh signaling elevated the levels of cyclin D1 and decreased the levels of p21 expression in pancreatic tumor. Conceivably, induction of cell cycle arrest and apoptosis by the blockage of the Shh signaling pathway suggests that the Shh signaling pathway may act in a similar manner, by promoting cell cycling and preventing apoptosis of ONB cells. Therefore, our findings may provide new insights into understanding the regulation of the Shh signaling pathway on the growth of ONB. In summary, our data, for the first time, demonstrated that the Patched1, Gli1 and Gli2 were expressed by a high frequency of ONB tumors and by some ONB cells. The blockage of Shh signaling inhibited the proliferation of ONB cells and induced ONB cell cycle arrest and apoptosis, which was associated with modulating the expression of Shh signaling components and cell cycle-related regulators. Addition of exogenous Shh partially or completely reduced blockage-mediated inhibition of ONB cell proliferation. Therefore, our data suggest that the Shh signaling pathway is a crucial regulator of the growth of ONB and our findings may aid in the design of new therapies for the intervention of human ONB. Mao /Xia /Zhou /Dai /Yang /Wang /Duan / Qiao /Mei /Hu

Acknowledgments This study was supported by grants from the National Natural Science Foundation of China (No. 30600189). We thank Dr. HanFei Ding (Cancer Center, Medical College of Georgia, USA) for

his kind provision of ONB cell lines and technical assistance. We also thank Dr. Ying Tang (Department of Pathology, Huaxi Hospital, Sichuan University, Chengdu, China) for her assistance in pathological analysis.

References 1 Finkelstein SD, Hirose T, VandenBerg SR: Olfactory neuroblastoma; in Kleihues P, Cavenee WK (eds): Pathology and Genetics of Tumours of the Nervous System. Lyon, International Agency for Research on Cancer Press, 2000, pp 150–152. 2 Dulguerov P, Allal AS, Calcaterra TC: Esthesioneuroblastoma: a meta-analysis and review. Lancet Oncol 2001;2:683–690. 3 Berger L, Luc R: L’esthésionéroépithéliome olfactif. Bull Assoc Franc Etude Cancer 1924;13:410–421. 4 Broich G, Pagliari A, Ottaviani F: Esthesioneuroblastoma: a general review of the cases published since the discovery of the tumour in 1924. Anticancer Res 1997;17:2683–2706. 5 Lund VJ, Howard D, Wei W, Spittle M: Olfactory neuroblastoma: past, present, and future? Laryngoscope 2003;113:502–507. 6 Morita A, Ebersold MJ, Olsen KD, Foote RL, Lewis JE, Quast LM: Esthesioneuroblastoma: prognosis and management. Neurosurgery 1993;32:706–714; discussion 714–715. 7 Thayer SP, di Magliano MP, Heiser PW, Nielsen CM, Roberts DJ, Lauwers GY, Qi YP, Gysin S, Fernández-del Castillo C, Yajnik V, Antoniu B, McMahon M, Warshaw AL, Hebrok M: Hedgehog is an early and late mediator of pancreatic cancer tumorigenesis. Nature 2003;425:851–856. 8 Karhadkar SS, Bova GS, Abdallah N, Dhara S, Gardner D, Maitra A, Isaacs JT, Berman DM, Beachy PA: Hedgehog signaling in prostate regeneration, neoplasia and metastasis. Nature 2004;431:707–712. 9 Couvé-Privat S, Le Bret M, Traiffort E, Queille S, Coulombe J, Bouadjar B, Avril MF, Ruat M, Sarasin A, Daya-Grosjean L: Functional analysis of novel sonic hedgehog gene mutations identified in basal cell carcinomas from xeroderma pigmentosum patients. Cancer Res 2004;64:3559–3565.


10 Daya-Grosjean L, Couvé-Privat S: Sonic hedgehog signaling in basal cell carcinomas. Cancer Lett 2005;225:181–192. 11 Hooper JE, Scott MP: Communicating with Hedgehogs. Nat Rev Mol Cell Biol 2005; 6: 306–317. 12 Riobo NA, Manning DR: Pathways of signaling transduction employed by vertebrate Hedgehogs. Biochem J 2007; 403:369–379. 13 Rubin LL, de Sauvage FJ: Targeting the Hedgehog pathway in cancer. Nat Rev Drug Discov 2006;5:1026–1033. 14 Diensthuber M, Potinius M, Rodt T, Stan AC, Welkoborsky HJ, Samii M, Schreyögg J, Lenarz T, Stöver T: Expression of bcl-2 is associated with microvessel density in olfactory neuroblastoma. J Neurooncol 2008; 89: 131–139. 15 Peacock CD, Wang Q, Gesell GS, CorcoranSchwartz IM, Jones E, Kim J, Devereux WL, Rhodes JT, Huff CA, Beachy PA, Watkins DN, Matsui W: Hedgehog signaling maintains a tumor stem cell compartment in multiple myeloma. Proc Natl Acad Sci USA 2007; 104:4048–4053. 16 Freedman VH, Shin SI: Cellular tumorigenicity in nude mice: correlation with cell growth in semi-solid medium. Cell 1974; 3: 355–359. 17 Berman DM, Karhadkar SS, Maitra A, Montes De Oca R, Gerstenblith MR, Briggs K, Parker AR, Shimada Y, Eshleman JR, Watkins DN, Beachy PA: Widespread requirement for hedgehog ligand stimulation in growth of digestive tract tumors. Nature 2003;425:846–850.

18 Cavazzana AO, Navarro S, Noguera R, Reynolds PC, Triche TJ: Olfactory neuroblastoma is not a neuroblastoma but is related to primitive neuroectodermal tumor. Adv Neuroblastoma Res 1988;2:463–473. 19 Feldmann G, Dhara S, Fendrich V, Bedja D, Beaty R, Mullendore M, Karikari C, Alvarez H, Iacobuzio-Donahue C, Jimeno A, Gabrielson KL, Matsui W, Maitra A: Blockade of hedgehog signaling inhibits pancreatic cancer invasion and metastases: a new paradigm for combination therapy in solid cancers. Cancer Res 2007;67:2187–2196. 20 Jacob L, Lum L: Deconstructing the hedgehog pathway in development and disease. Science 2007;318:66–68. 21 Incardona JP, Gaffield W, Kapur RP, Roelink H: The teratogenic veratrum alkaloid cyclopamine inhibits sonic hedgehog signaling transduction. Development 1998; 125: 3553– 3562. 22 Kenney AM, Rowitch DH: Sonic hedgehog promotes G (1) cyclin expression and sustained cell cycle progression in mammalian neuronal precursors. Mol Cell Biol 2000; 20: 9055–9067. 23 Gartel AL, Serfas MS, Tyner AL: p21-negative regulator of the cell cycle. Proc Soc Exp Biol Med 1996; 213:138–149. 24 Bar EE, Chaudhry A, Farah MH, Eberhart CG: Hedgehog signaling promotes medulloblastoma survival via Bc/II. Am J Pathol 2007;170:347–355. 25 Morton JP, Mongeau ME, Klimstra DS, Morris JP, Lee YC, Kawaguchi Y, Wright CV, Hebrok M, Lewis BC: Sonic hedgehog acts at multiple stages during pancreatic tumorigenesis. Proc Natl Acad Sci USA 2007; 104: 5103–5108.

Oncology 2009;77:231–243

243