EMX2 Is Epigenetically Silenced and Suppresses Growth in Human Lung Cancer

Oncogene (2010) 29, 5969–5975 & 2010 Macmillan Publishers Limited All rights reserved 0950-9232/10 www.nature.com/onc SHORT COMMUNICATION EMX2 is epigenetically silenced and suppresses growth in human lung cancer

J Okamoto1,2,7, T Hirata1,2,7, Z Chen1,3, H-M Zhou3, I Mikami1,2,HLi1, A Beltran1, M Johansson4,5, LM Coussens4,5, G Clement1, Y Shi1,6, F Zhang1, K Koizumi2, K Shimizu2, D Jablons1,5 and B He1,5

1Thoracic Oncology Program, Department of Surgery, University of California, San Francisco, CA, USA; 2Department of Surgery, Division of Thoracic Surgery, Nippon Medical School, Tokyo, Japan; 3Department of Biological Sciences and Biotechnology, Tsinghua University, Beijing, China; 4Department of Pathology, University of California, San Francisco, CA, USA; 5Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, CA, USA and 6Biosciences Division, SRI International, Menlo Park, CA, USA

Lung cancer is a common cancer and the leading cause of than 80% of lung cancers are non-small cell lung cancer, cancer-related death worldwide. Aberrant activation of which includes distinct histological subtypes: adenocarci- WNT signaling is implicated in lung carcinogenesis. noma, squamous cell carcinoma and large cell carcinoma EMX2, a human homologue of the Drosophila empty (Travis, 2002). The mainstays of conventional treatments spiracles gene is a homeodomain-containing transcription have offered patients only a limited and generally short- factor. The function of EMX2 has been linked to the WNT term beneﬁt. Overall, 5-year survival has remained at a signaling pathway during embryonic patterning in mice. dismal B15% for over two decades (Travis, 2002; Jemal However, little is known about the role of EMX2 in human et al., 2008). The molecular carcinogenesis of lung cancer is tumorigenesis. In this study, we found that EMX2 was characterized by multiple alterations of gene expression dramatically downregulated in lung cancer tissue samples and function. These arise from a series of molecular and and this downregulation was associated with methylation of morphological events affecting oncogenes such as K-ras theEMX2promoter.RestorationofEMX2expressionin and EGFR (Fong et al., 2003) and tumor suppressor genes lung cancer cells lacking endogenous EMX2 expression such as p53 and p16 (Zochbauer-Muller et al., 2000), all suppressed cell proliferation and invasive phenotypes, leading inexorably to abnormal changes in cell signaling inhibited canonical WNT signaling, and sensitized lung transduction pathways. cancer cells to the treatment of the chemo cytotoxic drug The homeobox gene family encodes transcription cisplatin. On the other hand, knockdown of EMX2 factors that regulate morphogenesis and cell differentia- expression in lung cancer cells expressing endogenous tion during embryogenesis by activating or repressing EMX2 promoted cell proliferation, invasive phenotypes the expression of target genes (Boersma et al., 1999). and canonical WNT signaling. Taken together, our study In addition, several homeobox genes have recently been suggests that EMX2 may have important roles as a novel shown to be associated with cancers (Raman et al., 2000; suppressor in human lung cancer. Abate-Shen, 2002; Samuel and Naora, 2005; Yoshida Oncogene (2010) 29, 5969–5975; doi:10.1038/onc.2010.330; et al., 2006). EMX2 is a human homologue of the published online 9 August 2010 Drosophila empty spiracles gene (ems), a homeodomain-containing transcription factor with important Keywords: EMX2; methylation; lung cancer; WNT functions during early development. For example, mice signaling; tumor suppression harboring homozygous mutation of EMX2 (EMX2À/À) exhibit small cerebral hemispheres and olfactory bulbs (Dalton et al., 1989). EMX2 affects the proliferation of adult neural stem cells by regulating the frequency of The American Cancer Society lists lung cancer as the symmetric divisions that generate two stem cells within leading cause of cancer death in the United States, the adult neural stem cell population, and overexpres- exceeding all combined deaths from the next four most sion of EMX2 decreases the frequency of symmetric deadly cancers, that is breast, colon, prostate and divisions (Galli et al., 2002). EMX2 controls mamma- pancreatic cancers (American Cancer Society, 2008). More lian reproduction by adjusting endometrial cell proliferation without affecting differentiation (Taylor and Correspondence: Professor B He, Thoracic Oncology Program, Fei, 2005). Moreover, it has been reported that loss of Department of Surgery, University of California, 2340 Sutter Street, EMX2 function leads to ectopic WNT1 expression in Room N222, San Francisco, CA 94115, USA. the developing mammalian telecephalon, resulting in E-mail: [email protected] 7These authors contributed equally to this work. cortical dysplasia (Ligon et al., 2003). The WNT Received 3 February 2010; revised 11 June 2010; accepted 1 July 2010; pathway is well known to have important roles in published online 9 August 2010 embryogenesis, stem cell renewal and oncogenesis EMX2 in human lung cancer J Okamoto et al 5970

120 20 p < 0.001

100

10 80

60 Relative EMX2 Expression 0

10 40 Relative EMX2 Expression 5 20 Relative Methylation 0 0 Normal Tumor 1236910 45 78 (n=64)

1.2 10 1.0 8 0.8 6 0.6 4 0.4 2

0.2 Relative Methylation 0 0 Relative EMX2 Expression H460 H441 A549 A427 H522 H838 H322 H1703 H1299 H2170 H1666 H1975 H460 A549 H838 A427 H522 H322 H441 H1703 H1975 H2170 H1666 H1299 Normal

EMX2 GAPDH DAC -+ -+ -+ -+ -+ -+ -+ -+ -+ -+ -+ -+

H460 A549H1703 H838 H1975 A427 H2170 H1666 H1299 H522 H322 H441 Figure 1 EMX2 expression was downregulated by methylation in lung cancer tissues and cell lines. Fresh samples (lung cancer tissue and its adjacent normal tissue) were collected from patients undergoing surgical resection with approval by the Committee on Human Research at the University of California, San Francisco (UCSF). Samples were promptly snap frozen in liquid nitrogen and stored at À170 1Cbefore use. Total RNA was extracted using TRIzol LS (Invitrogen, Carlsbad, CA, USA). Human lung cancer cell lines were all purchased from American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in RPMI 1640 with 10% fetal bovine serum, penicillin (100 IU/ml)/streptomycin (100 mg/ml)at371C in a humidified 5% CO2 incubator. cDNA synthesis and Taqman PCR were performed as previously described (Raz et al., 2008). Hybridization probes and primers (Supplementary Information Table S1) were purchased from Applied Biosystems (ABI, Foster City, CA, USA). EMX2 expression of samples was calculated by using the 2ÀddCt method (normalizing to their housekeeping gene GAPDH and then comparing to total RNA of adult normallungtissue(BioChain,Hayward,CA,USA)). Quantitative methylation-specific PCR (qMSP) was performed as previously described (Fackler et al., 2004; Grote et al., 2005, 2006). Genomic DNA was extracted with Qiagen DNeasy kits (Qiagen, Valencia, CA, USA) and bisulfite modification of genomic DNA was performed using EZ DNA Methylation-Gold Kits (Zymo Research, Orange, CA, USA). Primers and probes (Supplementary Table S1) were designed using Primer Express and Methyl Primer Express Software v1.0 (ABI) and purchased from Operon (Huntsville, AL, USA). Relative EMX2 methylation levels were determined by using the 2ÀdCt method (normalizing to the housekeeping gene ACTB (Raz et al., 2008)) and then calculating the ratio (tumor/matched normal for tissues; cell line/an adult normal lung tissue (BioChain) for cell lines). Both quantitative RT–PCR and qMSP were done in triplicate using an ABI 7300 Real-time PCR System. (a) Quantitative RT–PCR of 64 tumors and their matched adjacent normal lung tissues. The y axis represents normalized relative EMX2 mRNA expression (arbitrary unit). (b) Quantitative RT–PCR (upper panel) and quantitative MSP (lower panel) of 10 representative tumors (black bars) compared with their matched adjacent normal lung tissues (gray bars). (c) Quantitative RT–PCR analysis and (d) Quantitative MSP analysis in lung cancer cell lines. An adult normal lung tissue was used as a control. Results are means±s.d. (error bars). (e) DAC treatment of lung cancer cell lines. Treatment of cells lines with 5 mM DAC (Sigma, St Louis, MO, USA) was performed as previously described (Mazieres et al., 2004). Total RNA was isolated using Qiagen RNeasy kit 72 h after treatment, and EMX2 expression was examined by semiquantitative RT–PCR (primers (Supplementary Table S1) were purchased from Operon). GAPDH served as control for RNA quality and loading.

Oncogene EMX2 in human lung cancer J Okamoto et al 5971 (Klaus and Birchmeier, 2008), including that in lung We first examined the mRNA levels of EMX2 in cancer (Mazieres et al., 2004; You et al., 2004; He et al., human lung cancer tissue samples and their matched 2005; Huang et al., 2008; Akiri et al., 2009). There have adjacent normal tissues obtained from 64 patients with been only a limited number of recent studies suggesting lung cancer. Upon comparison, 71.8% (46–64) lung possible involvement of EMX2 in human cancer. For cancer samples analyzed were found to have less EMX2 example, EMX2 may be anti-proliferative in the expression than their matched adjacent normal tissues endometrium, and its expression is decreased in (Figure 1a), and this downregulation was statistically endometrial tumors (Noonan et al., 2001, 2003). significant (mean values of EMX2 expression measured EMX2 also displays methylation but rarely in non- by quantitative RT–PCR were 3.78 and 18.01 in cancer seminomas (Lind et al., 2006). The role of EMX2 in tissues and their matched adjacent normal tissues, tumorigenesis, however, is still largely unknown. In this respectively; Po0.001). Diminished/absent EMX2 study, we seek to investigate the role of EMX2 in human expression was consistently associated with hypermethy- lung cancer. lation of the EMX2 promoter in these cases evaluated by qMSP (Figure 1b showed an example of 10 paired tissue samples). We also analyzed EMX2 expression and the EMX2 promoter methylation status in 12 lung cancer cell lines to verify these results. In all, 10 of the 12 H1299 A427 lines examined were found to lack EMX2 expression Ctrl EMX2 Ctrl shRNA EMX2 shRNA (Figure 1c). Using qMSP, we found that the EMX2 promoter in the same 10 cells lines was also methylated (Figure 1d). Next, using a de-methylating reagent, 5-aza-2’-deoxycytidine (DAC), we restored EMX2 expression in cell lines with initially silenced EMX2 (Figure 1e). These data indicate that epigenetic modification may be one of the important mechanisms to 40 H1299 40 A427 silence EMX2 gene in lung cancer. Interestingly, methylation silencing was recently reported for several 30 p < 0.0001 30 p < 0.001 other homeobox gene family members in cancer. For example, HOXA5 was identified as a direct transcription 20 20 activator of tumor suppressor p53 and HOXA5 silencing by hypermethylation consequently limited p53 10 10 expression in breast cancer (Raman et al., 2000). Cell proliferation (Abs. Day X/Day1) 0 0 1356 1356 Days Days Figure 2 EMX2 suppressed lung cancer cell proliferation and sensitized lung cancer cells to cisplatin. H1299 cells were transfected with pcDNA 3.1/EMX2 mammalian expression vector (subcloned from pCMV6-XL5/EMX2 vector (Origene, Rockville, p < 0.001 p = 0.04 100 MD, USA)). A427 cells were transfected with EMX2 shRNAs (5’-TCAAGCCATTTACCAGGCTTCGGAGGAAG-3’ and 5’-CGG 100 TGGAGAATCGCCACCAAGCAGGCGAG-3’) and non-silencing shRNA (all in pRFP-C-RS vector, Origene). Transfection 50 was done using Lipofectamine2000 (Invitrogen). Transfected cells 50 were re-plated from six-well plates to 10 cm dishes for selection with G418 (500 mg/ml; Invitrogen). Stable transfectants were

Colony number (%) 0 0 maintained in regular medium with G418 (300 mg/ml) before Ctrl EMX2 Ctrl shRNA EMX2 shRNA analyses. (a) Morphology under light microscope ( Â 40). (b) MTS assay of H1299 cells stably transfected with EMX2 (solid H1299 A427 diamonds) and empty pCDNA3.1 vector control (solid squares); and A427 cells stably transfected with EMX2 shRNA (solid squares) and non-silencing shRNA control (solid diamonds). Controls were set as 100%. Proliferation assay was performed by 12 p < 0.0001 plating the stably transfected cells in 96-well plates at a density of 10 500–1000 cells/well in 100 ml of G418 culture medium. Medium was changed every day. Cell viability was evaluated in triplicate by 8 CellTiter 96 AQueous (Promega, Madison, WI, USA). (c) Colony formation assay. In all, 500 individual stably transfected cells were 6 seeded in 10 cm dishes and cultured for 10 days. Colonies were then ﬁxed by 10% formalin, stained with 0.5% crystal violet and 4 counted. (d) Synergistic effect between EMX2 and cisplatin in Cell proliferation

(Abs. Day X/Day1) H1299. Diamonds, squares, triangles and crosses are treatments of 2 control vector alone, control vector þ cisplatin (0.3 ng/ml), EMX2 0 cDNA alone and EMX2 cDNA þ cisplatin (0.3 ng/ml), respec- 123456 tively. All results are means±s.d. (error bars). Differences between Days groups were compared with a two-sided Student’s t-test. A P-value of p0.05 was considered to be signiﬁcant.

Oncogene EMX2 in human lung cancer J Okamoto et al 5972 The HOXA9 promoter was found to be frequently and support the data observed in H1299 cells. In methylated in non-seminomatous TGCT (Lind et al., addition, by using H1299 stably transfected with 2006). Our data also revealed that in a few cases, EMX2, we examined the chemo-synergistic effect promoter hypermethylation of EMX2 did not correlate between EMX2 and cisplatin, a widely used chemo with decreased EMX2 expression, suggesting that drug in the clinic to treat lung cancer (Figure 2d). We alternative mechanisms may account for EMX2 down- observed that the suppressive potential of moderate regulation. Indeed, a report showed that EMX2 doses of cisplatin was significantly enhanced by re- transcripts were reduced in a subset of endometrial expressing EMX2 in H1299 cells (Po0.0001), indicating cancers investigated with a 35% incidence of LOH for potential future therapeutic role for EMX2 in combina- the 10q25.3–q26.1 region that included the EMX2 gene. tion with current cytotoxic agents in lung cancer. Sequencing analysis uncovered multiple EMX2 variants, To further assess the function of EMX2 in lung cancer including somatic mutations and polymorphisms (Noo- progression, we performed a cell invasion assay and nan et al., 2001). found that the percent invasion of EMX2-stably We next investigated the roles of EMX2 in the growth transfected H1299 cells was significantly reduced of lung cancer cells in which EMX2 was methylation (Po0.05) and that the percent invasion of EMX2 silenced. One week after transfection and subsequent shRNA-stably transfected A427 cells was significantly G418 selection, we found that EMX2 restoration in increased (Po0.05) (Figure 3a). Consistently, using a H1299 led to significant proliferative suppression (MTS: 3D-spheroid model to mimic an in vivo microenviron- Po0.001; colony formation assays: Po0.001) without ment, we observed that EMX2-stably transfected H1299 dramatic morphological changes (Figures 2a–c). In formed smaller spheroids (Po0.05) and EMX2 shRNA- contrast, when endogenous EMX2 expression in A427 stably transfected A427 formed larger spheroids was silenced by anti-EMX2 shRNAs, cell proliferation (Po0.05) (Figure 3b). More importantly, control was stimulated (MTS: Po0.001; colony formation: (non-EMX2 expressing) H1299 cells formed 3D-spher- P ¼ 0.04) with dramatic morphological changes (cells oids with multiple invasive cellular branch-like struc- were larger with more branches after EMX2 shRNA tures indicating broken basal membrane. However, treatment) (Figures 2a–c). These results complement EMX2-stably transfected H1299 formed a less invasive

A427 H1299 A427 H1299 100 40000 * 30000 * * 50 20000 * unit) 10000 Invasion (%)

0 Size (arbitrary 0 Ctrl EMX2 Ctrl EMX2 Ctrl EMX2 Ctrl EMX2 shRNA shRNA shRNA shRNA

H1299 A427

100 p < 0.001 12.3% 100 11.7% p < 0.0001 32.1% 40.5% 22.7% 33.3%

50 50 37.2% 51.4% 65.6%

Cells (%) 54.4% Cells (%) 22.3% Fraction of Total Fraction of Total 16.5% 0 0 Control EMX2 Ctrl shRNA EMX2 shRNA Figure 3 EMX2 suppressed invasive phenotypes of lung cancer cells. (a) Invasion assay using trans-well chamber with and without matrigel (BD BioCoat Matrigel Invasion Chamber, BD Biosciences, Lexington, KY, USA) was performed in triplicate for each stable transfectant according to the manufacturer’s protocol. Cells from five different fields of each insert membrane were counted under a light microscope ( Â 40) and percent invasion was determined as follows: % invasion ¼ (mean # of cells invading through matrigel insert membrane/mean # of cells migrating through control insert membrane without matrigel) Â 100. (b, c) Analyses of 3D cultures of stably transfected cells. Eight-chambered culture slides (BD) were coated with 35 ml growth-factor reduced Cultrex Basement Membrane Extract (Trevigen, Gaithersburg, MD, USA) per well and left to solidify for 15 min. H1299 or A427 cells were treated with trypsin and resuspended in regular culture medium with serum. Cultrex was added to a total concentration of 2%, and 500 ml of the cell suspension was added to each chamber of the matrigel-coated slide. Medium was replaced every 2 days. After 1 week, 100 þ acini were measured for size and graded for disruption. (b) Quantification of size of spheroids. (c) and (d) Phenotypes of spheroids (H1299 and A427) were categorized into three types (round (solid filling), asymmetric (dense dotted) and disrupted (sparse dotted)) and quantified. Representative phenotypes in each treatment were also shown. Symbol (*) in each graph represents statistical significance (Po0.05) by two-sided Student’s t-test.

Oncogene EMX2 in human lung cancer J Okamoto et al 5973 a b 120 30 p < 0.001 100 20 80 60 10

40 TOP/FOP

(arbitrary unit) 0 20 Relative EMX2 Expression (%) Ctrl EMX2 0 CtrlEMX2 Ctrl EMX2 H1703 c 300 shRNA shRNA p = 0.004 H1299 A427 200

d β-Catenin 100

Cyclin D1 TOP/FOP (%) 0 β-Actin Ctrl EMX2 Ctrl EMX2 shRNA shRNA Ctrl EMX2 shRNA shRNA

H1299 A427 f Control EMX2

e EMX2 PTGS2 S100A4 GAPDH SGPP2 TMEM92 β-Catenin A_32_P171793 AQP3 Cyclin D1 S100P CCL20 β-actin CXCL3 IFI27 DAC - + - + 2

H1299 A427 -2 0.4 1.2 1.5 1.7 -0.4 -1.7 -1.5 -1.2 0.67 0.93 0.13 -0.93 -0.67 -0.13 Figure 4 EMX2 suppressed canonical WNT signaling in lung cancer cells. (a) Quantitative RT–PCR of EMX2 expression in cell lines stably transfected with control or EMX2 expression vector (in H1299) and with non-silencing control or EMX2-specific shRNA (in A427). H1703 served as an EMX2 expression level control. (b, c) TOP/FOP luciferase assays (performed 24 h after transfection as previously described in Clement et al., 2008) in H1299 cells stably transfected with EMX2 cDNA and in A427 cells stably transfected with EMX2 shRNA, respectively. (d) Western blotting of key canonical WNT downstream effector (cytosolic b-catenin; antibody from BD Biosciences) and target protein (Cyclin D1; antibody from Cell Signaling Technology, Danvers, MA, USA). b-Actin (antibody from Sigma) was used as protein control. Cytosolic proteins were prepared by using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce Biotechnology, Rockford, IL, USA) according to the manufacturer’s protocol. (e) The effect of DAC on key canonical WNT downstream effectors. Semiquantitative RT–PCR was used to confirm reactivation of the EMX2 expression after treatment of cells lines with 10 mM DAC. DAC treatment and semiquantitative RT–PCR were performed as described in Figure 1e. GAPDH served as control for RNA quality and loading. Western blotting of cytosolic b-catenin and Cyclin D1 was performed as described in Figure 4d. b-Actin was used as protein control. (f) Microarray profiling of lung cancer cell line H1299 stably transfected with EMX2 and empty vector control. Partial heatmap plot of hierarchical clustering including S100A4 and S100P is shown. Total RNA quality was assessed using a Pico Chip on an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA). RNA was amplified and labeled with Cy3-CTP or Cy5-CTP using the Agilent low RNA input fluorescent linear amplification kits following the manufacturer’s protocol. Labeled cRNA was assessed using the Nandrop ND-100, and equal amounts of Cy3- and Cy5-labeled target were hybridized to Agilent whole-human genome 44 K ink-jet arrays. Hybridization samples were randomized on the 3 Â 44 K format to correct any batch bias. Hybridizations were performed for 14 h according to the manufacturer’s protocol. Arrays were scanned using the Agilent microarray scanner and raw signal intensities were extracted with Feature Extraction v9.5 software (Agilent). rounder phenotype (Po0.001) (Figure 3c). In contrast, established, we used real-time PCR to examine EMX2 A427 cells treated with control shRNA formed more expression. H1703 cell line expressing endogenous round/less invasive spheroids than those treated with EMX2 at similar levels to that of adult normal lung EMX2 shRNA (Po0.001) (Figure 3d). Together, our tissue (Figure 1c) was used as an EMX2 expression level results suggest that EMX2 may have a role as a novel control. We confirmed that EMX2 expression in EMX2- suppressor of malignant lung cell progression or stably transfected H1299 was compatible with the metastasis. To test this hypothesis and explore mechan- physiological expression in normal lung, and determined isms involved in EMX2 function in lung cancer, we that EMX2 levels in shRNA-stably transfected A427 examined the canonical WNT signaling pathway known were significantly downregulated (Figure 4a). Interest- to be aberrantly activated in lung cancer and important ingly, we observed that EMX2 restoration in H1299 for proliferation, survival and metastasis of lung cancer cells led to decreased WNT signaling-dependent tran- cells (Mazieres et al., 2004; You et al., 2004; He et al., scription activity (Figure 4b). In contrast, shRNA 2005; Huang et al., 2008; Akiri et al., 2009; Nguyen silencing of EMX2 in A427 cells resulted in increased et al., 2009). EMX2 cDNA and anti-EMX2 shRNA transcription activity of the canonical WNT signaling were stably transfected into both H1299 cells lacking pathway (Figure 4c). In these stable lines, analysis of the EMX2 expression and into EMX2-expressing A427 canonical WNT downstream effector cytosolic b-catenin cells, respectively. After stable transfectants were and its target Cyclin D1, corroborated the association

Oncogene EMX2 in human lung cancer J Okamoto et al 5974 between EMX2 and the canonical WNT pathway in S1) supported a possible association between reduced lung cancer (Figure 4d). After treatment of H1299 cells EMX2 expression and canonical WNT activation in with DAC, a de-methylation agent, we found that both lung cancer. It is possible that EMX2 functions as a cytosolic b-catenin and Cyclin D1 were downregulated, transcriptional repressor in a similar manner in lung confirming our hypothesis (Figure 4e). After DAC cancer as it does during embryonic development. treatment in A427 cells, we also observed some down- However, other co-factors may have important roles in regulation of these WNT downstream effectors EMX2-related transcription regulation. The identifica- (Figure 4e). Our interpretation is that DAC is not a tion of direct target(s) of EMX2 and its interacting specific de-methylation agent for a single gene such as proteins is needed to further elucidate the function of EMX2. Thus, expression of other tumor suppressors EMX2 and its relationship with canonical WNT that are known to be methylation silenced in lung cancer signaling in lung cancer. such as WNT pathway antagonists WIF-1 and SFRP In conclusion, this is the first demonstration of the (Mazieres et al., 2004; Fukui et al., 2005) should also be importance of EMX2 as a suppressor in lung carcino- re-activated after the DAC treatment. This would genesis. Epigenetic silencing of the EMX2 expression contribute to the downregulation of those WNT down- may be important for aberrant activation of the stream effectors. Moreover, our microarray analysis canonical WNT signaling in lung cancer and consequent revealed that restoration of EMX2 expression in lung proliferation and metastasis of lung cancer cells. cancer cells downregulated metastasis genes such as S100P and S100A4, members of the EF-hand calcium- binding protein family (Figure 4f). These proteins have Conflict of interest been reported to induce metastasis in rodent mammary model systems for breast cancer and are apparently The authors declare no conflict of interest. associated with poor patient outcome in breast, colon, lung and esophageal carcinomas (Donato, 2001; Heiz- mann et al., 2002; Zimmer et al., 2003; Marenholz et al., Acknowledgements 2004). It is possible that EMX2 regulates transcription of these metastasis genes in lung cancer by directly or This study was supported by grant from Joan’s Legacy: uniting indirectly downregulating canonical WNT signaling. against Lung Cancer Research Grant, NIH/NCI grant This is supported by recent evidence that S100P and R01CA125030 and the Eileen D Ludwig endowed for Thoracic S100A4 may be direct downstream targets of canonical Oncology Research (to BH); the Bonnie J Addario Lung WNT signaling (Stein et al., 2006; Ganesan et al., 2008). Cancer Foundation, the Kazan, McClain, Abrams, Fernan- Taken together, our results suggest that EMX2 may dez, Lyons, Greenwood, Harley and Oberman Foundation and the Barbara Isackson Lung Cancer Research Fund (DJ); suppress lung cancer, either directly or indirectly, Swedish Cancer Institute (MJ); NIH/NCI grants through regulation of the canonical WNT signaling R01CA130980, R01CA13256 and DOD BCRP Era of Hope activity. In addition, a preliminary analysis of co- Scholar Award (W81XWH-06-1-0416) (to LMC). Microarray expression of several WNT family genes along with profiling was done by the UCSF Shared Microarray Core EMX2 in lung cancer cell lines (Supplementary Figure Facilities. We thank M Roshni Ray for editing this paper.

References

Abate-Shen C. (2002). Deregulated homeobox gene expression in Fackler MJ, McVeigh M, Mehrotra J, Blum MA, Lange J, Lapides A cancer: cause or consequence? Nat Rev Cancer 10: 777–785. et al. (2004). Quantitative multiplex methylation-specific PCR assay Akiri G, Cherian MM, Vijayakumar S, Liu G, Bafico A, Aaronson for the detection of promoter hypermethylation in multiple genes in SA. (2009). Wnt pathway aberrations including autocrine Wnt breast cancer. Cancer Res 64: 4442–4452. activation occur at high frequency in human non-small-cell lung Fong KM, Sekido Y, Gazdar AF, Minna JD. (2003). Lung cancer. carcinoma. Oncogene 28: 2163–2172. 9: molecular biology of lung cancer: clinical implications. Thorax American Cancer Society. (2008), Cancer facts and figures. 58: 892–900. Boersma CJ, Bloemen M, Hendriks JM, van Berkel EA, Olijve W, van Fukui T, Kondo M, Ito G, Maeda O, Sato N, Yoshioka H et al. Zoelen EJ. (1999). Homeobox proteins as signal transduction (2005). Transcriptional silencing of secreted frizzled related protein intermediates in regulation of NCAM expression by recombinant 1 (SFRP 1) by promoter hypermethylation in non-small-cell lung human bone morphogenetic protein-2 in osteoblast-like cells. Mol cancer. Oncogene 24: 6323–6327. Cell Biol Res Commun 2: 117–124. Galli R, Fiocco R, De Filippis L, Muzio L, Gritti A, Mercurio S et al. Clement G, Guilleret I, He B, Yagui-BeltrańA,LinYC,YouLet al. (2008). (2002). Emx2 regulates the proliferation of stem cells of the Epigenetic alteration of the Wnt inhibitory factor-1 promoter occurs early adult mammalian central nervous system. Development 129: in the carcinogenesis of Barrett’s esophagus. Cancer Sci 99: 46–53. 1633–1644. Dalton D, Chadwick R, McGinnis W. (1989). Expression and Ganesan K, Ivanova T, Wu Y, Rajasegaran V, Wu J, Lee MH et al. embryonic function of empty spiracles: a Drosophila homeo box (2008). Inhibition of gastric cancer invasion and metastasis by gene with two patterning functions on the anterior-posterior axis of PLA2G2A, a novel beta-catenin/TCF target gene. Cancer Res 68: the embryo. Genes Dev 3: 1940–1956. 4277–4286. Donato R. (2001). S100: a multigenic family of calcium-modulated Grote HJ, Schmiemann V, Geddert H, Bocking A, Kappes R, Gabbert proteins of the EF-hand type with intracellular and extracellular HE et al. (2006). Methylation of RAS association domain family functional roles. Int J Biochem Cell Biol 33: 637–668. protein 1a as a biomarker of lung cancer. Cancer 108: 129–134.

Oncogene EMX2 in human lung cancer J Okamoto et al 5975 Grote HJ, Schmiemann V, Geddert H, Rohr UP, Kappes R, Gabbert Noonan FC, Mutch DG, Ann Mallon M, Goodfellow PJ. (2001). HE et al. (2005). Aberrant promoter methylation of p16(INK4a), Characterization of the homeodomain gene EMX2: sequence RARB2 and SEMA3B in bronchial aspirates from patients with conservation, expression analysis, and a search for mutations in suspected lung cancer. Int J Cancer 116: 720–725. endometrial cancers. Genomics 76: 37–44. He B, Lee AY, Dadfarmay S, You L, Xu Z, Reguart N et al. (2005). Raman V, Tamori A, Vali M, Zeller K, Korz D, Sukumar S. (2000). SFRP4 is silenced by hypermethylation and induces apoptosis in HOXA5 regulates expression of the progesterone receptor. J Biol b-catenin-deﬁcient human mesothelioma cells. Cancer Res 65: 743–748. Chem 275: 26551–26555. Heizmann CW, Fritz G, Schafer BW. (2002). S100 proteins: structure, Raz D, Ray MR, Kim J, He B, Taron M, Skrzypski M et al. (2008). functions and pathology. Front Biosci 7: d1356–d1368. A multi-gene assay is prognostic of survival in patients with Huang CL, Liu D, Ishikawa S, Nakashima T, Nakashima N, early-stage lung adenocarcinoma. Clin Cancer Res 14: Yokomise H et al. (2008). Wnt1 overexpression promotes 5565–5570. tumour progression in non-small cell lung cancer. Eur J Cancer Samuel S, Naora H. (2005). Homeobox gene expression in cancer: 44: 2680–2688. insights from developmental regulation and deregulation. Eur Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T et al. (2008). J Cancer 16: 2428–2437. Cancer statistics. CA Cancer J Clin 58: 71–96. Stein U, Arlt F, Walther W, Smith J, Waldman T, Harris ED et al. Klaus A, Birchmeier W. (2008). Wnt signaling and its impact on (2006). The metastasis-associated gene S100A4 is a novel target of development and cancer. Nat Rev Cancer 8: 387–398. beta-catenin/T-cell factor signaling in colon cancer. Gastroenterol- Ligon KL, Echelard Y, Assimacopoulos S, Danielian PS, Kaing S, ogy 131: 1486–1500. Grove EA et al. (2003). Loss of Emx2 function leads to ectopic Taylor HS, Fei X. (2005). Emx2 regulates mammalian reproduction expression of Wnt1 in the developing telencephalon and cortical by altering endometrial cell proliferation. Mol Endocrinol 19: dysplasia. Development 130: 2275–2287. 2839–2846. Lind GE, Skotheim RI, Fraga MF, Abeler VM, Esteller M, Lothe RA. Travis WD. (2002). Pathology of lung cancer. Clin Chest Med 23: (2006). Novel epigenetically deregulated genes in testicular cancer include 65–81, viii. homeobox genes and SCGB3A1 (HIN-1). J Pathol 210: 441–449. Yoshida H, Broaddus R, Cheng W, Xie S, Naora H. (2006). Marenholz I, Heizmann CW, Fritz G. (2004). S100 proteins in mouse Deregulation of the HOXA10 homeobox gene in endometrial and man: from evolution to function and pathology (including an carcinoma: role in epithelial-mesenchymal transition. Cancer Res 2: update of the nomenclature). Biochem Biophys Res Commun 322: 889–897. 1111–1122. You L, He B, Xu Z, Uematsu K, Mazieres J, Mikami I et al. (2004). Mazieres J, He B, You L, Xu Z, Lee AY, Mikami I et al. (2004). Wnt A novel anti-human Wnt-2 monoclonal antibody induces pro- inhibitory factor-1 is silenced by promoter hypermethylation in grammed cell death in human non-small-cell lung cancer cells. human lung cancer. Cancer Res 64: 4717–4720. Oncogene 23: 6170–6174. Nguyen DX, Chiang AC, Zhang XH, Kim JY, Kris MG, Ladanyi M Zimmer DB, Wright Sadosky P, Weber DJ. (2003). Molecular et al. (2009). WNT/TCF signaling through LEF1 and HOXB9 mechanisms of S100-target protein interactions. Microsc Res Tech mediates lung adenocarcinoma metastasis. Cell 138: 51–62. 60: 552–559. Noonan FC, Goodfellow PJ, Staloch LJ, Mutch DG, Simon TC. Zochbauer-Muller S, Wistuba II, Minna JD, Gazdar AF. (2000). (2003). Antisense transcripts at the EMX2 locus in human and Fragile histidine triad (FHIT) gene abnormalities in lung cancer. mouse. Genomics 81: 58–66. Clin Lung Cancer 2: 141–145.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene