RE-1–silencing transcription factor shows tumor-suppressor functions and negatively regulates the oncogenic TAC1 in breast cancer cells

Bobby Y. Reddya, Steven J. Grecoa, Prem S. Patelb, Katarzyna A. Trzaskaa, and Pranela Rameshwara,1

aDepartment of Medicine–Hematology/Oncology, New Jersey Medical School, University of Medicine and Dentistry of New Jersey, Newark, NJ 07103; and bDepartment of Surgery, Brookdale University Hospital, Brooklyn, NY 11212

Edited by Susan E. Leeman, Boston University School of Medicine, Boston, MA, and approved January 15, 2009 (received for review September 12, 2008) Breast cancer remains the most prevalent cancer among women in (19, 20). An oncogenic property has been indicated by its the United States. , a peptide derived from the TAC1 increase in combination with myc in medulloblastoma (21). A , mediates oncogenic properties in breast and other cancers. tumor-suppressor role has been reported for colon and small-cell TAC1 expression facilitates the entry of breast cancer cells into lung cancers (SCLC) (22–24). In breast, colon, lung, and prostate bone marrow. The transcriptional repressor element 1–silencing cells, REST suppresses the expressions of neuroendocrine transcription factor (REST) has been implicated in both oncogenic and activation of the phosphatidylinositol 3-kinase [PI (3)K] and tumor-suppressor functions. REST binds to the 5؅ untranslated pathway, suggesting negative regulation on cell transformation region of the TAC1 promoter and suppresses its expression. This (19, 23). REST mutation results in a loss of its repressor function, study investigated a role for REST in TAC1 induction in breast with stimulation of the PI (3)K pathway and cell transformation cancer. Western blots and real-time PCR indicated that REST ex- (19, 22, 23). REST and NF␬B synergistically repress TAC1 pression in breast cancer cells was inversely proportional to the expression in mesenchymal stem cells and bone marrow stroma cells’ aggressiveness, for both cell lines and primary breast cancer (15). Here we report on decreased REST expression in TAC1 cells. REST knockdown in low-metastatic T47D cells and nontu- induction in BCCs. The results are verified with primary BC morigenic MCF12A cells resulted in increases in TAC1 induction, tissues from different stages of the disease. proliferation, and migration. These parameters were negatively affected by ectopic expression of REST in highly aggressive MDA- Results MB-231 cells. Together, these findings show a central role for REST REST Expression in BCCs. To explore the function of REST in BC, in the oncogenic function of TAC1 and suggest a tumor-suppressor we first determined REST expression in the cell lines MDA- role for REST in breast cancer. MB-231 and T47D and then compared with the cell line MCF12A. The lowest level in MDA-MB-231 was assigned a REST ͉ substance P ͉ tachykinin value of 1 for relative expression of MCF12A and T47D. There was a significant (P Ͻ 0.05) increase for T47D but lower than for espite advances in diagnostic technologies and treatments, MCF12A (P Ͻ 0.05) (Fig. 1A). Western blots showed consistency Dbreast cancer (BC) continues to be among the leading for protein (Fig. 1 B and C). In summary, REST expression causes of cancer-related deaths among women (1). The TAC1 was reduced with tumorigenicity: MCF12A Ͼ T47D Ͼ MDA- gene has been implicated in the development of breast and other MB-231. cancers (2–4). TAC1 (preprotachykinin-A) is a single-copy gene with 7 exons (5). Through and posttransla- TAC1 Expression in REST Knockdown BCCs. The 5Ј untranslated tional modification, TAC1 produces peptides belonging to the region of TAC1 has a functional REST binding site (15). Because tachykinin family (5–7). Substance P (SP) is the predominant BCCs show high expression of TAC1 (3, 4), we asked whether its peptide produced from TAC1 (8, 9). increase can be explained by decrease in REST expression. We BC cells (BCCs) produce high levels of SP (6). SP binds with stably knocked down REST (wild type) in MDA-MB-231, T47D, varying affinities to 3 G protein–coupled, 7-transmembrane and MCF12A and then studied TAC1 reporter activity by receptors: neurokinin (NK)1, NK2, and NK3 (6, 10). NK1 and transfection with pGL3-TAC1-1.2 (25). Controls included non- NK2 have been reported on BCCs. SP mediates BCC prolifer- transfectants and mutant siRNA (mt) transfectants. RT-PCR ation, imparts radiation resistance, protects from apoptosis, and Western blot verified REST knockdown (Fig. 2 A and B). induces growth- and angiogenic-promoting factors, and facili- REST knockdown resulted in significantly (P Ͻ 0.05) greater tates BC metastasis to bone marrow (2, 4, 11, 12). The truncated luciferase activities as compared with controls (Fig. 2C). Real- NK1 has also been linked to oncogenesis, partly through the time PCR for mRNA and ELISA for SP showed correlations induction of TAC1 (2). between reporter gene activities and endogenous TAC1 expres- Stromal-derived factor-1␣, which is implicated in BC biology, sion (Fig. 2D and Table 1). regulates TAC1 expression (4, 13). Other cytokines also induce TAC1 expression in BCCs via autocrine stimulation (2, 14). RE-1 Effects of REST on the Growth of BCCs. REST knockdown causes an silencer of transcription (REST) represses TAC1 expression in increase in TAC1 expression (Fig. 2). We determined whether mesenchymal stem cells (15). The link between REST and TAC1 REST knockdown could lead to increases in cell proliferation could be relevant to disorders such as tumorigenesis and hip- pocampal seizure (16). REST, also known as neural restrictive silencing factor, is a Author contributions: B.Y.R. and P.R. designed research; B.Y.R., S.J.G., K.A.T., and P.R. zinc finger transcriptional repressor of neuronal genes in non- performed research; S.J.G. and P.S.P. contributed new reagents/analytic tools; S.J.G., P.S.P., neuronal and immature neuronal cells (17). REST facilitates K.A.T., and P.R. analyzed data; and B.Y.R. and P.R. wrote the paper. chromatin remodeling and inhibits by assem- The authors declare no conflict of interest. bling a repressor complex (18), Sin3A/histone deacetylase, and This article is a PNAS Direct Submission. co-REST at the N and C termini, respectively (18). REST has Freely available online through the PNAS open access option. been implicated in both oncogenic and tumor-suppressor roles 1To whom correspondence should be addressed. E-mail: [email protected].

4408–4413 ͉ PNAS ͉ March 17, 2009 ͉ vol. 106 ͉ no. 11 www.pnas.org͞cgi͞doi͞10.1073͞pnas.0809130106 Downloaded by guest on September 27, 2021 Fig. 1. REST expression in BCCs. (A) Real-time PCR was performed for REST mRNA. MDA-MB-231 was arbitrarily assigned a value of 1 and the results of T47D and MCF12A presented as the mean Ϯ SD (n ϭ 5) fold change of MDA-MB-231. (B) Western blots were performed for REST with whole-cell extracts. (C) Band densities were normalized with those for ␤-actin. *, P Ͻ 0.05 vs. MDA-MB-231 and MCF12A; **, P Ͻ 0.05 vs. T47D.

and migration. This would be consistent with the transforming ability of TAC1 (3). REST knockdown did not show a significant (P Ͼ 0.05) change in the proliferation of MDA-MB-231 but showed significant (P Ͻ 0.05) increases for T47D and MCF12A (Fig. 3 A–C). REST knockdown showed increases in activated AKT [phospho (P)-AKT] in T47D and MCF12A, which corre- Fig. 2. TAC1 expression in REST knockdown BCCs. REST knockdown BCCs lated with increases in proliferation (Fig. 3D). In the case of (wt) were verified by Western blot (A) and real-time PCR (B). Controls included MDA-MB-231 there were similar band intensities for P-AKT, mutant (mt) siRNA and nontransfectants. mRNA levels for REST (wt) were normalized to 1 for mean fold change Ϯ SD (n ϭ 5). (C) Transfectants in A were regardless of REST expression (Fig. 3D). Similarly, there was no ␤ Ͼ cotransfected with pGL3-TAC1-1.2 and p -gal. Luciferase activities were nor- significant (P 0.05) change in migration for MDA-MB-231 malized with ␤-gal activities and then presented as the mean Ϯ SD (n ϭ 5). (D)

(Fig. 4A) but significant (P Ͻ 0.05) increases for T47D and Levels of TAC1 mRNA in cells from A were determined by real-time PCR. The MEDICAL SCIENCES MCF12A (Fig. 4A). levels in nontransfectants were arbitrarily assigned values of 1 and the exper- To address a role for TAC1, we knocked down REST in imental levels presented as mean fold change Ϯ SD (n ϭ 5). *, P Ͻ 0.05 vs. siRNA MCF12A, which causes an increase in SP (Table 1), and then (mt) and untransfected cells. studied the cells’ proliferation and migration in the presence or absence of specific NK1 and NK2 receptor antagonists (10 nM mental points (Fig. 4B). Similarly, these antagonists showed each). The results show significantly (P Ͻ 0.05) reduced prolif- significant reduction in cell migration (Fig. 4C). In summary, eration in the presence of these antagonists with REST knock- TAC1 has a role in the responses of MCF12A after REST down (wild-type siRNA) as compared with the other experi- knockdown.

Reddy et al. PNAS ͉ March 17, 2009 ͉ vol. 106 ͉ no. 11 ͉ 4409 Downloaded by guest on September 27, 2021 Table 1. SP production in REST knockdown BCCs with varying stages of BC (Table 3). Real-time PCR and SP (pg/mL) Western blots for REST mRNA and protein, respectively, show decreased REST expression as the tumor stage increases REST siRNA (Fig. 7 A and B). Despite the limited cohort of patients, the data correlated with the studies performed with cell lines. This Cells Untransfected wt mt indicates that reduced REST expression correlates with the MDA-MB-231 864 ϩ 17 1375 Ϯ 10* 855 Ϯ 28 aggressiveness of the disease. T47D 315 ϩ 24 780 Ϯ 25* 322 Ϯ 20 MCF12A 0.002 ϩ .001 125 Ϯ 10* 0.002 Ϯ .001 Discussion TAC1 and its major peptide, SP, have been reported to exert an BCCs were stably knocked down for REST. Controls were untransfected (wt) oncogenic role (2, 3, 11). Cytokines have been implicated in the or transfected with mutant (mt) REST. At 80% confluence, culture media were replaced with media containing 2% FCS. After 24 h, media were quantitated induction of TAC1 in BCCs. However, the mechanisms of for SP and the results presented as mean SP Ϯ SD (pg/mL; n ϭ 5). *, P Ͻ 0.05 overexpressed TAC1 in BCCs remain unclear. We report a role vs. REST siRNA (mt) and nontransfectants. for the REST gene in the induction of TAC1 in BC. REST has been shown to exert both oncogenic and tumor-suppressor functions (19, 26). Our results demonstrate a tumor-suppressor Ectopic REST in MDA-MB-231. REST knockdown did not affect the function of REST and that its loss or reduced expression result proliferation and migration of MDA-MB-231 (Figs. 3 and 4). We in TAC1 induction. In addition, REST expression is indirectly asked whether these properties, as well as TAC1 expression, proportional to the tumorigenicity of BC (Figs. 4 and 7). could be reversed by ectopic REST. Because MCF12A and The findings are significant to translational medicine because T47D express significant levels of REST as compared with REST could be a prognostic and diagnostic biomarker for MDA-MB-231 (Fig. 1), they were not included in the studies. carcinogenesis. However, the clinical utility of REST requires Western blot and real-time PCR verified ectopic REST expres- in-depth studies to understand the mechanisms by which REST sion (Fig. 5 A and B). As compared with nontransfectants and mediates tumor suppression. The 5Ј flanking region of TAC1 has vector transfectants, ectopic REST showed a significant (P Ͻ a functional binding site for REST (15). We showed decreased 0.05) decrease in endogenous TAC1 expression for mRNA (Fig. REST expression in advanced stages of BC. In contrast, TAC1 5C) and protein for SP (Table 2). In addition, there was expression is increased in advanced BC (3, 4). This study verified significant (P Ͻ 0.05) decreases in proliferation and migration a cause–effect relationship between REST and TAC1 expres- for ectopic REST expression as compared with vector transfec- sions using both knockdown and expression studies (Figs. 2 and tant and nontransfectant (Fig. 6 A and B). Western blots with 5). Although REST knockdown increased TAC1 expression in all Day-5 cell extracts showed undetectable caspase 3 (not shown) BCCs, its loss did not change the proliferation and migration in but detectable Bcl-2, indicating that reduced cell proliferation the highly aggressive MDA-MB-231 (Figs. 3 and 4). This is was not due to cell death (Fig. 6A, Inset). In addition, daily expected because endogenous REST expression was significantly trypan blue exclusion showed no evidence of cell death. In lower in MDA-MB-231 as compared with T47D (Fig. 1). Thus, summary, ectopic REST expression in MDA-MB-231 resulted in TAC1 expression might have been at its optimum. Hence, a decreased TAC1 induction, proliferation, and migration. threshold level for REST could be required for oncogenic transformation, as well as aggressiveness. REST Expression in Primary Breast Tumors. Because the studies Although knockdown of low levels of endogenous REST presented in Figs. 1–6 support tumor-suppressor functions of could not reverse proliferation and migration in MDA-MB-231, REST, we studied its expression in primary BCCs from patients ectopic expression resulted in decreased TAC1 expression, pro-

Fig. 3. Growth curves for REST knockdown BCCs. MDA-MB-231 (A), T47D (B), and MCF12A (C), untransfected or stably transfected with REST siRNA (wt or mt), were analyzed for growth properties. Cells were seeded at 50/mL and viable cells counted daily up to Day 5. The total number of cells are presented as mean Ϯ SD (n ϭ 5). Extracts from transfectants were analyzed for P-AKT in Western blots (D). *, P Ͻ 0.05 vs. siRNA (mt) and untransfected cells.

4410 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0809130106 Reddy et al. Downloaded by guest on September 27, 2021 Fig. 4. Migration of REST knockdown and effects of TAC1.(A) Fluorescence-labeled cells (104) were placed in the inner chamber, and migrated cells were determined from a standard curve, established with BCCs vs. fluorescence intensities. Results are presented as mean Ϯ SD (n ϭ 5). (B) MCF12A was induced to express TAC1 by knockdown of REST. The role of TAC1 peptides were studied in cell proliferation studies by culturing untransfected and mutant (mt) and wild-type (wt) REST siRNA transfectants in the presence or absence of antagonists (10 nM): NK1 (CP-96,345) and NK2 (SR 48968). (C) The cells from B were studied in migration assays. *, P Ͻ 0.05 vs. untransfected and REST siRNA (mt); **, P Ͻ 0.05 vs. REST siRNA (wt) with antagonists.

liferation, and migration (Figs. 5 and 6 and Table 2), suggesting in primary BCCs (Fig. 7). Interestingly, the expression of REST a link between the aggressiveness of MDA-MB-231 and REST was indirectly proportional to the stages of BC. However, it expression. These findings are consistent with the observations should be noted that because stage is not always synonymous with aggressiveness in BC, further studies are needed to validate these highly suggestive preliminary findings. On the basis of the combined results of cell lines and primary cells, it seems that an initial decrease in REST expression might initiate cell transfor- mation via increased TAC1 expression, and continued decrease could cause aggression. Low expression of REST in MDA-MB- 231 might activate pathways, independent of those signaled by TAC1 peptide, as identified by continued activation of AKT, despite REST knockdown (Fig. 3D). Decrease in REST expres- sion has been shown to activate PI (3)K and predispose the cells to transformation (22, 23). Cytokine induction has been implicated in the expression of TAC1 in BCCs (2, 3, 14). Thus, it would be interesting to determine whether cytokines are upstream of the REST–TAC1 axis. IL-1␣ caused rapid decrease in REST expression (15). It is predicted that enhanced cytokine induction may activate NF-␬B, which in turn could lead to the generation of truncated NK1 in BCCs (2). On the basis of these segmented reports, it is possible that there is a complex network involving TAC1, REST, and cytokines in BC biology. The unraveling of such networks could identify potential avenues for therapeutic intervention. In addi- tion to cytokines, neurotrophic factors could also be involved, given that downregulated REST expression in nonneuronal cells causes dysfunction in neurotrophins such as BDNF and their receptors, such as TrkC, which can promote cellular transfor- mation and malignancy in nonneuronal cells (27). REST has also been reported to exert tumor suppression in SCLC (24). Expression of the neuroendocrine peptide arginine vasopressin (AVP) distinguishes SCLC from normal lung and other neoplastic lung cell types, and this anomaly is believed to be prevented by the normal repressor activity of REST at the

Table 2. SP production in REST-expressing MDA-MB-231 Cells SP (pg/mL) MEDICAL SCIENCES

Fig. 5. Ectopic expression of REST in MDA-MB-231. REST was stably ex- Untransfected 890 Ϯ 15 pressed in MDA-MB-231. Controls were untransfected or stable transfectants REST expression 545 Ϯ 22* with vector. Efficient expression was determined by Western blots for REST Vector transfectant 904 Ϯ 5 with whole-cell extracts and normalized for ␤-actin (A) and real-time PCR for REST (B). Parallel studies were done for TAC1 mRNA by real-time PCR (C). The REST was expressed in MDA-MB-231. Controls were untransfected or trans- untransfected cells were assigned a value of 1 for fold change (mean Ϯ SD; n ϭ fected with vector alone. Culture media were quantitated for SP as for Table 5) of the other experimental points. *, P Ͻ 0.05 vs. the other experimental 1 and the results presented as mean Ϯ SD (n ϭ 5). points. *P Ͻ 0.05 vs. untransfected and vector transfectants.

Reddy et al. PNAS ͉ March 17, 2009 ͉ vol. 106 ͉ no. 11 ͉ 4411 Downloaded by guest on September 27, 2021 Fig. 7. REST expression in primary breast biopsies. Real-time PCR and Western blots for REST mRNA (A) and protein (B), respectively, in breast tissues Fig. 6. REST expression on the growth and migration of MDA-MB-231. REST at varying stages of BC. mRNA levels for nonmalignant cells were assigned was expressed as for Fig. 5 and then analyzed for proliferation (A) and values of 1 and the results presented as mean Ϯ SD (n ϭ 5). Western blots were migration (B). Inset in A shows Western blots for Bcl2 and normalization for normalized for ␤-actin, and the results are shown as normalized band densi- ␤-actin. The results represent mean Ϯ SD (n ϭ 5). *, P Ͻ 0.05 vs. untransfected ties. *, P Ͻ 0.05 vs. nonmalignant tissues. and vector transfectants.

characterize the molecular mechanisms of REST dysregulation AVP promoter region (24). Interestingly, SCLCs express high during different stages of BC, as well as triple-negative BCCs, as levels of a splice variant of REST, which antagonizes REST- identified by MDA-MB-231. Investigation of such pathways will mediated repression of AVP, thereby promoting tumorigenesis lead to further knowledge of developmental and pathologic of neuronal phenotype (19, 26, 28). Similarly, a truncated form processes in neuroendocrine systems and cancer cell biology. of REST has been shown to downregulate the expression of Eventually, such progress will serve to improve diagnosis and REST in colon cancer (23). Because REST recruits cofactors treatment of BC. through its C and N termini (26), it is possible that the truncated Materials and Methods forms might hinder tumor suppression owing to their inability to Reagents. Rabbit anti-REST, anti-AKT, and anti-P-AKT were purchased from recruit cofactors. In this study, we did not observe any of the Upstate Cell Signaling, HRP–goat antirabbit IgG from BD PharMingen, mouse REST isoforms. anti-␤-actin and HRP-goat antimouse IgG from Sigma. CP-96,345 and SR 48968 REST has been known to modulate the expressions of several are nonpeptide antagonists for NK1 and NK2 receptors, respectively. CP- genes. This study shows a tumor-suppressor role in BC, with a 96,345 was obtained from Pfizer and SR 48968 from Sanofi Recherche. link to TAC1 expression (Fig. 4), thereby initiating studies to Cell Lines and Primary Breast Cells. The cell lines MCF12A, T47D, and MDA- MB-231 were purchased from the American Type Culture Collection and Table 3. Profile of breast cancer patients 1–19 propagated according to their instructions. MCF12A is a nontumorigenic breast epithelial cell line. T47D, a low-metastatic cell line, was derived from Patients Age (y) status (ER/PR/HER2) Stage pleural effusion of ductal carcinoma. MDA-MB-231, a highly metastatic and invasive cell line, was derived from pleural effusion of an adenocarcinoma. 1–3 42–72 Negative biopsies — Breast tissues from surgical specimens were acquired from Brookdale Uni- ϩ ϩ Ϫ 4–7 53–80 / / I versity Hospital (Brooklyn, NY). At the time of surgery, patients were not 848 ϩ/ϩ/ϩ I subjected to any treatment. The studies were approved by the institutional 9, 11 44, 92 ϩ/ϩ/Ϫ II review board of the University of Medicine and Dentistry of New Jersey and 10, 12 47, 76 ϩ/ϩ/ϩ II Brookdale University Hospital. Tissues from 4 patients were provided by the 13, 14 55, 59 Ϫ/Ϫ/Ϫ II Cooperative Human Tissue Network. The demographics of patients are shown 15 67 ϩ/ϩ/ϩ III in Table 3. 16–19 46–83 ϩ/ϩ/Ϫ III Vectors and Stable Transfectants. Construction of the siRNA vectors pPMSKH1- All samples were obtained from female subjects whose staging followed RESTwt and pPMSKH1-RESTmt has been previously described (15). Briefly, for standard guidelines (30). ER, estrogen receptor; PR, progesterone receptor; the REST siRNA wild-type sequence (pPMSKH1-RESTwt) spans ϩ1550/ϩ1568: HER2, c-erbB-2. 5Ј gtg taa tct aca gta tca c 3Ј (GenBank no. NM࿝005612). Mutant siRNA

4412 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0809130106 Reddy et al. Downloaded by guest on September 27, 2021 (pPMSKH1-RESTmt) changed 3 nucleotides (italics): 5Ј gtg taa tttatagtg tca c the UN-SCAN-IT gel software (Silk Scientific) and then normalized with bands 3Ј.p␤-gal and pTK-Hyg were obtained from Clontech. TAC1 reporter vector, for housekeeping genes. pGL3-PPT-I-1.2, which contains Intron 1 and upstream sequences, has been described (25). REST expression vector, pCMV6-XL4-REST, and the empty Substance P ELISA. Media were quantitated for SP levels by competitive ELISA vector were obtained from OriGene. as previously described (29). Briefly, 96-well plates were coated with a complex ␮ Stable knockdown of REST was done as previously described (4). Briefly, of streptavidin-biotinylated SP. Equal volumes (50 L) of the samples and cells were cotransfected with pPMSKH1-RESTwt or pPMSKH1-RESTmt and optimum rabbit anti-SP were added to triplicate wells. Bound anti-SP was pHyg and then selected with 25 ␮g/mL hygromycin. Ectopic REST expression in screened with alkaline phosphatase-conjugated goat antirabbit IgG and Sigma 104 phosphatase substrate. MDA-MB-231 was done with pCMV6-XL4-REST or vector alone and then selected with 0.5 ␮g/mL neomycin. REST expressions were verified by Growth Curve. The method for growth curve analysis was performed as Western blots and RT-PCR for protein and mRNA, respectively, as previously previously described (2). Briefly, BCCs (100/mL) were seeded in 60-mm tissue described (15). culture dishes and allowed to adhere overnight. The total number of viable cells was counted daily up to Day 5. Reporter Gene Assay. Cells were cotransfected with p␤-gal and pGL3-PPT-I/1.2 with Effectene (Qiagen) (13). Cell extracts were analyzed for total protein with Migration Assay. BCC migration was assessed in a Boyden chamber with 8-␮m an assay kit from Bio-Rad. Normalizations were computed with the ratio of inserts as previously described (4). Briefly, cells were labeled with CellTracker luciferase to ␤-gal activity as previously described (25). Green 5-chloromethylfluorescein diacetate according to the manufacturer’s instructions (Invitrogen). Labeling efficiency, as determined by fluorescence Real-Time RT-PCR. Real-time PCR was performed with 200 ng cDNA and TAC1- microscopy, indicated Ͼ95% labeling efficiency. Labeled BCCs (104) were or REST-specific primers as previously described (15). The reactions used the added to the inner chamber in sera-free culture media. After 2 h, migration Platinum SYBR Green qPCR SuperMix UDG Kit (Invitrogen). Amplifications was determined as described using Typhoon (Amersham Pharmacia, Molec- were done on a 7300 Real Time PCR System (Applied Biosystems. The cycling ular Dynamics), and then the mean fluorescence intensity was calculated with profiles were 94 °C for 15 s and 60 °C for 45 s. The reactions were normalized Image Quant software. The percentage of cell migration was calculated on a standard curve of total BCCs vs. fluorescence intensity. by amplifying the same cDNA with ␤-actin primers.

Statistical Analyses. Statistical evaluations of data were performed with anal- Western Blots and Densitometries. Cell extracts were analyzed by Western ysis of variance and the Tukey-Kramer multiple comparisons test. A P value of blots on gradient SDS-PAGE (BioRad), as previously described (2). Proteins Ͻ0.05 was considered significant. were transferred onto polyvinyl membranes and then incubated with primary and secondary antibodies at dilutions of 1:1,000 and 1:2,000, respectively. ACKNOWLEDGMENTS. This work was supported by a grant from the De- Bands were detected by chemiluminescence. HeLa cell nuclear extract served partment of Defense, University Cancer Center, and New Jersey Cancer as a positive control for REST detection. Band densities were determined with Commission.

1. Jemal A, et al. (2003) Cancer Statistics, 2003. CA Cancer J Clin 53:5–26. 16. Spencer EM, et al. (2006) Regulation and role of REST and REST4 variants in modulation 2. Patel HJ, Ramkissoon SH, Patel PS, Rameshwar P (2005) Transformation of breast cells of gene expression in in vivo and in vitro in epilepsy models. Neurobiol Dis 24:41–52. by truncated neurokinin-1 receptor is secondary to activation by preprotachykinin-A 17. Lunyak VV, Rosenfeld MG (2005) No rest for REST: REST/NRSF regulation of neurogen- peptides. Proc Natl Acad Sci USA 102:17436–17441. esis. Cell 121:499. 3. Rao G, et al. (2004) Facilitating role of preprotachykinin-I gene in the integration of 18. Ballas N, Grunseich C, Lu DD, Speh JC, Mandel G (2005) REST and its corepressors breast cancer cells within the stromal compartment of the bone marrow: A model of mediate plasticity of neuronal gene chromatin throughout neurogenesis. Cell early cancer progression. Cancer Res 64:2874–2881. 121:645–657. 4. Corcoran KE, et al. (2008) Mesenchymal stem cells in early entry of breast cancer into 19. Majumder S (2006) REST in good times and bad: Roles in tumor suppressor and bone marrow. PLoS ONE 3:e2563. oncogenic activities. Cell Cycle 5:1929–1935. 5. Rameshwar P, Oh HS, Yook C, Gascon P, Chang VT (2003) Substance p-fibronectin- 20. Weissman AM (2008) How much REST is enough? Cancer Cell 13:381–383. cytokine interactions in myeloproliferative disorders with bone marrow fibrosis. Acta 21. Su X, et al. (2006) Abnormal expression of REST/NRSF and Myc in neural stem/ Haematol 109:1–10. progenitor cells causes cerebellar tumors by blocking neuronal differentiation. Mol 6. Pennefather JN, et al. (2004) Tachykinins and tachykinin receptors: A growing family. Life Sci 74:1445–1463. Cell Biol 26:1666–1678. 7. Leeman SE, Aronin N, Ferris C (1982) Substance P and neurotensin. Recent Prog Horm 22. Zhao JJ, et al. (2003) Human mammary epithelial cell transformation through the Res 38:93–132. activation of phosphatidylinositol 3-kinase. Cancer Cell 3:483–495. 8. Reddy BY, Trzaska KA, Murthy RG, Navarro P, Rameshwar P (2008) Neurokinin recep- 23. Westbrook TF, et al (2005) A genetic screen for candidate tumor suppressors identifies tors as potential targets in breast cancer treatment. Curr Drug Discov Technol 5:15–19. REST. Cell 121:837–848. 9. Nowicki M, Ostalska-Nowicka D, Miskowiak B (2008) In vitro substance P-dependent 24. Coulson JM, Fiskerstrand CE, Woll PJ, Quinn JP (1999) Arginine vasopressin promoter induction of bone marrow cells in common (CD10) acute lymphoblastic leukaemia. regulation is mediated by a neuron-restrictive silencer element in small cell lung Leuk Res 32:97–102. cancer. Cancer Res 59:5123–5127. 10. Tansky MF, Pothoulakis C, Leeman SE (2007) Functional consequences of alteration of 25. Qian J, et al. (2001) Cloning of human preprotachykinin-I promoter and the role of N-linked glycosylation sites on the neurokinin 1 receptor. Proc Natl Acad Sci USA cyclic adenosine 5Ј-monophosphate response elements in its expression by IL-1 and 104:10691–10696. stem cell factor. J Immunol 166:2553–2561. 11. Hennig IM, Laissue JA, Horisberger U, Reubi JC (1995) Substance-P receptors in human 26. Coulson JM (2005) Transcriptional regulation: Cancer, neurons and the REST. Curr Biol primary neoplasms: Tumoral and vascular localization. Int J Cancer 61:786–792. 15:R665–R668. 12. Taborga M, Corcoran KE. Fernandes N, Ramkissoon SH, Rameshwar P (2007) G-coupled 27. Downward J (2005) RNA interference libraries prove their worth in hunt for tumor protein receptors and breast cancer progression: Potential drug targets. Mini Rev Med suppressor genes. Cell 121:813–815. Chem 7:245–251. 28. Coulson JM, Edgson JL, Woll PJ, Quinn JP (2000) A splice variant of the neuron- 13. Corcoran KE, Malhotra A, Molina CA, Rameshwar P (2008) Stromal-derived factor- restrictive silencer factor repressor is expressed in small cell lung cancer: A potential 1{alpha} induces a non-canonical pathway to activate the endocrine-linked Tac1 gene role in derepression of neuroendocrine genes and useful clinical marker. Cancer Res in non-tumorigenic breast cells. J Mol Endocrinol 40:113–123. 14. Oh HS, et al. (2004) Bone marrow stroma influences transforming growth factor-{beta} 60:1840–1844. production in breast cancer cells to regulate c-myc activation of the preprotachykinin-I 29. Greco SJ, Rameshwar P (2007) Enhancing effect of IL-1{alpha} on neurogenesis from gene in breast cancer cells. Cancer Res 64:6327–6336. adult human mesenchymal stem cells: Implication for inflammatory mediators in

15. Greco SJ, Smirnov SV, Murthy RG, Rameshwar P (2007) Synergy between the RE-1 regenerative medicine. J Immunol 179:3342–3350. MEDICAL SCIENCES silencer of transcription and NF{kappa}B in the repression of the 30. Singletary SE, et al. (2002) Revision of the American Joint Committee on Cancer staging gene TAC1 in human mesenchymal stem cells. J Biol Chem 282:30039–30050. system for breast cancer. J Clin Oncol 20:3628–3636.

Reddy et al. PNAS ͉ March 17, 2009 ͉ vol. 106 ͉ no. 11 ͉ 4413 Downloaded by guest on September 27, 2021