Identification and Characterization of Retinoic Acid Receptor B2 Target Genes in F9 Teratocarcinoma Cells

Retinoids, a group of natural and synthetic analogues of vitamin A (retinol), modulate the differentiation of many cell types. Retinoids are also used for the prevention and treatment of cancer. The actions of retinoids are generally mediated by the retinoic acid receptors (RARs A, B, and ;) and the retinoid X receptors (RXRs A, B, and ;). One of the RARs, RARB, is expressed at reduced levels in many human carcinomas, and F9 RARB2 / cells do not growth arrest in response to RA. To determine if RARB2 regulates the expression of a unique set of genes, through the use of subtractive hybridization and DNA array analysis we have identified and characterized genes that are differentially expressed in F9 RARB2 / teratocarcinoma cells. These genes, which encode transcription factors, cell surface signal transduction molecules, and metabolic enzymes, include c-myc, FOG1, GATA6 , glutamate dehydrogenase, glutathione S-transferase homologue (p28), Foxq1 , Hic5 , Meis1a , Dab2, midkine , and the PDGF-a receptor. These genes are regulated specifically by RARB2 in F9 wild-type (Wt) cells as indicated by their expression profiles in F9 RARB2 / cells as compared to F9 Wt, RARA / , or RAR; / cells, and their responsiveness to specific retinoid receptor agonists. The basal expression levels of some of these genes, such as c-myc , are higher in the F9 RARB2 / cells than in F9 Wt in the absence of exogenous retinoids, suggesting that RARB2 can inhibit gene expression in the absence of a ligand. The RARB2 target genes are transcriptionally activated by retinol, as well as RA, in F9 Wt cells. Because the lack of RARB2 alters both the control of proliferation and differentiation in F9 cells, the genes that we have characterized may mediate key effects of RA, via RARB2, on these processes. Introduction Retinoids, a group of natural and synthetic analogues of vitamin A, exert profound effects on many biological processes, such as vertebrate embryonic development (1) and cell growth and differentiation (2). Vitamin A insufficiency during pregnancy results in the death of the fetus, as well as congenital malformations affecting the eyes and ocular tissues, myocardium, respiratory, urogenital, and circulatory systems (1). Vitamin A deficiency in experimental animals is also associated with a higher incidence of some types of cancer, and with increased susceptibility to chemical carcinogens (3). Experimental models of carcinogenesis have demonstrated the efficacy of pharmacological administration of retinoids in preventing the development of cancers of the skin, oral cavity, lung, mammary gland, prostate, bladder, liver, and pancreas in animals exposed to carcinogenic agents (3). Clinical trials have indicated that retinoids may be useful for prevention of cancers of the upper aerodigestive tract, skin, breast, and ovaries (4). In addition, retinoids have been successfully used in the treatment of acute promyelocytic leukemia (APL) (5, 6), and other cancers (for review, see Refs. 7 and 8). The biological effects of retinoids are primarily mediated by two classes of nuclear retinoid receptors: retinoic acid receptors (RARs) and retinoid X receptors (RXRs) (9, 10). RARs and RXRs are members of the nuclear receptor superfamily that also include estrogen, androgen, thyroid hormone, vitamin D, PPAR, and orphan receptors. The retinoid receptors are encoded by six distinct genes: RARa , RARb , RARc , RXRa , RXRb , and RXRc . Each of these receptors includes several isoforms formed by different splicing and usage of alternative promoters (9, 10). All-trans retinoic acid (RA) binds and activates RARs, and 9-cisretinoic acid (9-cis RA) binds and activates both RARs and RXRs (11). Retinoid receptors activate transcription in a ligand-dependent manner by binding to retinoic acid response elements (RAREs) located in the promoter regions or enhancers of target genes. RAREs generally consist of two directly repeated half-sites of the consensus sequence AGGTCA spaced by 2 or 5 bp (DR2 and DR5 elements) (12). Recently, various coactivators for the nuclear receptors have been identified, including CBP/p300, SRC-1, TIF-2/ Grip1, and ACTR/RAC3/p/CIP. They are recruited by ligandactivated receptors and enhance transcription (13, 14). Many of these coactivators possess acetyltransferase activity that can modify histones and may increase promoter access to DNAbinding proteins and the transcriptional machinery (15–18). The strong conservation of each RAR across vertebrates suggests that each receptor performs unique functions (19). In Received 2/18/03; revised 5/9/03; accepted 5/9/03. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Grant support: NIH grant R01 CA43796 to L.J.G.; a research grant from the American Institute for Cancer Research; and in part by a Cancer Pharmacology Training Grant (CA62948-08) (Y.Z.). Requests for reprints: Lorraine J. Gudas, Room E409, Pharmacology Department, Weill Medical College of Cornell University, 1300 York Avenue, New York, NY 10021. Phone: (212) 746-6250; Fax: (212) 746-8858. E-mail: [email protected] *Current address: Bristol-Myers Squibb Company, New Brunswick, NJ 08903. Copyright D 2003 American Association for Cancer Research. Vol. 1, 619–630, June 2003 Molecular Cancer Research 619 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from our laboratory, we generated F9 teratocarcinoma cells with mutations in both alleles of RARa, RARb2, or RARc by homologous recombination and showed that different RARs have different functions (20–23). Mice homozygous for RARa, RARh, or RARg have also been generated to understand the function of each receptor. Mutant mice lacking RARa, RARh2, RARh, or RARg are viable but exhibit abnormalities in several tissues (24–27). For example, RARh / mice exhibit greatly reduced hippocampal CA1 long-term potentiation (LTP) and long-term depression (LTD) (28), activities which are likely to play a role in learning and memory. The RARb gene has four isoforms: h1, h2, h3, and h4. The h2 isoform is the most abundant RARh isoform and the major RA inducible form (29). An RARE that mediates RA-induced RARh2 gene expression in many different cell types was identified in the promoter region (30, 31). The RARh2 RARE consists of two direct repeats of the core motif sequence AGGTCA separated by five nucleotides (31, 32). Activation of the RARh2 promoter is mediated by RAR/RXR heterodimers (33). A thyroid hormone receptor/RXR heterodimer can also bind strongly to the hRARE and activates the RARh2 promoter in response to RXR ligands (34). RARh exhibits a pattern of expression during development and in the mature organism (35) which is different from those of the other RARs. This suggests that RARh performs specific functions. RARh is not expressed, or is expressed at low levels, in a number of malignant tumors, including lung carcinoma, squamous cell carcinoma of the head and neck, breast cancer cell lines, and esophageal carcinoma (36–40). RARh is encoded by a gene located on the short arm of chromosome 3 (p24) (41). In breast cancer, loss of heterozygosity (LOH) has frequently been detected at chromosome 3p22–25 (42). All of these findings support the concept that the specific loss of RARh expression may be an important event in tumorigenesis. Recent studies have indicated that a decrease in RARh expression results in resistance to the growth inhibitory actions of retinoids. Indeed, transfection of RARh into RARh-negative cervical, breast, and lung cancer cells increased responsiveness to growth inhibition and induction of apoptosis by retinoids (43–47), and the restoration of RARh expression in RARh-negative lung cancer cell lines inhibited tumorigenicity in nude mice (48, 49). Administration of 13-cis RA to patients with premalignant oral lesions can restore the expression of RARh, as well as reverse the lesions (50). Conversely, the loss of both alleles of the RARb2 in an F9 teratocarcinoma line, generated in this laboratory by homologous recombination, resulted in the loss of growth inhibition by RA (21). We hypothesized that each RAR regulates a specific subset of target genes. Thus, to understand the functions of the transcription factor RARh2, the target genes specifically regulated by RARh2 must be identified and analyzed. Very few RARh target genes have been identified to date. By comparing RA-treated F9 wild-type (Wt) teratocarcinoma cells and F9 RARh2 / cells, we identified some of the RARh2 target genes through the use of subtractive hybridization and DNA expression microarray techniques. The target genes identified include transcription factors, protein tyrosine kinases, homeobox proteins, and oncoproteins. We have initiated the further characterization of eleven target genes in this study. Results Identification of RARb2 Target Genes in F9 Cells We hypothesized that RARh2, which is induced by RA in many cell types, mediated the growth inhibitory actions of RA in F9 cells. To test this hypothesis, an F9 teratocarcinoma cell line that contained targeted disruptions in both alleles of the RARh2 gene was generated by homologous recombination. We showed that the F9 Wt and RARh2 +/ heterozygous lines could undergo RA-induced growth arrest, but that RA did not cause growth arrest in the F9 RARh2 / cells (21). While these data confirmed our hypothesis, more information about the target genes regulated by RARh2 was then required to understand the downstream effects of RARh2. We anticipated that RARh2 positively regulated the transcription of some of its target genes, and negatively regulated other target genes. To isolate RARh2 target genes, F9 Wt cells and F9 RARh2 / cells were treated with 1 AM all-trans RA for 24 h. Subtractive hybridization and microarray analysis were then employed. By comparing the gene expression patterns between F9 Wt cells and the F9 RARh2 / cells, we identified approximately 80 genes from subtractive hybridization and 300 target genes from the microarray analyses (see ‘‘Materials and Methods’’). Approximately one-half of the putative target genes were expressed at higher levels in RA-treated F9 Wt cells as compared to F9 RARh2 / cells, while the remaining genes were expressed at higher levels in RA-treated RARh2 / cells than in F9 Wt. Target Genes Identified by Subtractive Hybridization We examined 20 of the genes isolated by subtractive hybridization by Northern analysis of F9 Wt versus F9 RARh2 / cells and found that 7 out of 20 target genes were reproducibly altered in the Northern blot analysis. We sequenced these seven target genes and identified them using NCBI Blast Programs. Among them, six are known genes (Table 1) and one is an uncharacterized gene. The six known genes are Dab2 (p96 form), glutamate dehydrogenase , Foxq1 , Meis1a , midkine , and glutathione S-transferase homologue (p28). To determine whether or not these genes were specific RARh2 targets, the regulation of these genes by RA was also examined by Northern blot analysis in F9 RARa / and RARg / lines, generated by homologous recombination in our laboratory (22, 23). We also examined RARh2 mRNA expression. As expected, RARh2 mRNA was not expressed in F9 RARh2 / cells and its expression was stimulated by RA in F9 Wt, RARa / , and RARg / cell lines (Fig. 1). Dab2 (disabled-2), one of the two mammalian orthologues of the Drosophila Disabled gene, is thought to be a tumor suppressor in ovarian cancer since its expression is lost or greatly reduced in 85% of breast and ovarian cancers (51). We show here that Dab2 mRNA expression was up-regulated by RA in F9 Wt cells, while Dab2 mRNA levels were not affected by RA in F9 RARh2 / cells (Table 1 and Fig. 1, C and D). RARh2 Target Genes 620 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from Furthermore, Dab2 basal mRNA levels were about 10-fold lower in the F9 RARh2 / cells as compared to Wt. This indicates that in the absence of exogenous RA, RARh2 can positively regulate the Dab2 gene. Dab2 mRNA expression in F9 RARg / cells was also low, both before and after RA treatment, as compared to Wt cells (Fig. 1, C and D), indicating that RARg may also modulate its expression. Foxq1 (formerly Hfh-1L) belongs to the family of the FOX (Forkhead box) transcription factors (previously called HNF-3/ forkhead transcription factors) (52). RA strongly induced Foxq1 expression in F9 Wt cells, but not in F9 RARh2 / cells. For instance, Foxq1 mRNA was induced by 5.6and 7.8-fold after 48 and 72 h RA treatment in F9 Wt cells. By contrast, in F9 RARh2 / cells, after 48 and 72 h, Foxq1 mRNA was only slightly induced (Fig. 1, A and B; Table 1). However, Foxq1 appeared not to be a specific RARh2 target. Foxq1 mRNA expression in F9 RARg / cells was also low, both before and after RA treatment, as compared to Wt cells (Fig. 1, A and B), indicating that RARg may also modulate its expression. Meis1a , a member of a family of genes encoding the TALE (three-amino-acid loop extension) subset of homeodomaincontaining proteins (53), acts as a cofactor of certain Hox proteins and may play a major role in normal hematopoiesis (54). Meis1a mRNA was expressed in F9 Wt, RARa / , and RARg / cell lines in the absence of RA. In contrast, Meis1a mRNA is barely detectable in the untreated F9 RARh2 / cells, indicating that the basal expression level ofMeis1a mRNA in F9 Wt, RARa / , and RARg / cells may be regulated by RARh2 in the absence of exogenously added ligand. After 24 h of RA treatment, the Meis1a mRNA level was increased by 3-, 3-, and 2.5-fold in the F9 Wt, RARa / , and RARg / cell lines, respectively, but it was not changed in the F9 RARh2 / cells (Fig. 1, A and B).Meis1a mRNAwas only slightly induced in the F9 RARh2 / cells, even after 48 or 72 h RA treatment, indicating that Meis1a is a specific RARh2 target gene in these cells. Midkine is a member of a family of heparin-binding growth factors and was originally isolated as a retinoic acid-responsive gene (55). midkine mRNA was increased by RA in F9 Wt, RARa / , and RARg / cells, and after 48 h, RA increased midkine mRNA levels by 5.5-, 4-, and 5-fold, respectively. In contrast, RA led to a slight reduction in the midkine mRNA level in F9 RARh2 / cells (Fig. 1, C and D). Thus, midkine displays the features of a RARh2 specific target gene. Glutamate dehydrogenase (GDH) is a mitochondrial enzyme that catalyzes the oxidative deamination of glutamate to aketoglutarate using NAD or NADP as a cofactor (56). In the brain, besides its function in metabolism, GDH influences glutamate neurotransmitter availability (57). The basal GDH mRNA level was lower in F9 RARh2 / cells than in F9 Wt cells (data not shown). RA treatment caused an induction of glutamate dehydrogenase mRNA expression in F9 Wt cells, but not in F9 RARh2 / cells (Table 1). Finally, glutathione S-transferase homologue (p28) is a small stress response protein which belongs to a family of GST-like (glutathione S-transferase) proteins (58). p28 expression was not affected byRA at either 24 or 48 h inWt, RARa / , or RARg / cells, and its expression was 4-fold lower in F9 RARh2 / cells compared to the other three cell lines (Fig. 1, C and D). Thus, p28 exhibits the properties of an RARh2 specific target gene. Target Genes Identified by Microarray Analyses We used gene microarray analyses to identify RARh2 target genes, and obtained EST clones for the genes showing the largest fold changes between F9 Wt cells and F9 RARh2 / cells. Seven out of the seven target genes from microarray analysis were confirmed by semi-quantitative RT-PCR or Northern analysis, indicating that the microarray analysis was generally reliable. These genes are listed in Table 1. Dab2 and Foxq1 were also identified by subtractive hybridization and were discussed above. c-myc is a proto-oncogene that is activated in various animal and human tumors. It is a transcription factor that plays a pivotal role in many biological functions including cell growth and differentiation (59, 60). c-myc mRNA in untreated RARh2 / cells was expressed at a 3.4-fold higher level than in untreated F9 Wt, RARa / , or RARg / cells. Furthermore, c-myc mRNA expression in F9 RARh2 / cells is 4.5-fold higher than in F9 Wt cells after 6 h of RA treatment (Fig. 1, A and B). After 24 h of RA treatment, the expression of c-myc mRNA is reduced by 3.7-, 4-, and 4.9-fold in F9 Wt cells, RARa / , and RARg / cells, respectively, while in the RARh2 / cells, the c-myc mRNA level is still higher than in Wt cells. These data suggest that RARh2 suppresses c-myc mRNA expression in F9 Wt cells even in the absence of exogenously added RA. Table 1. Summary of Genes Differentially Expressed in RA-Treated F9 Wt and F9 RARB2 / Cells GenBank Identity GenBank Accession No. Method of Identification Regulation by RA in Wild-Type F9 Cells RARh2 Specific Myelocytomatosis oncogene (c-myc ) L00039 Microarray # Y Friend of GATA-1 (FOG ) AF006492 Microarray " Y GATA-binding protein 6 U51335 Microarray " Y Glutamate dehydrogenase (GLUD ) X57024 Subtractive hybridization " Y Glutathione S -transferase homologue (p28) MMU80819 Subtractive hybridization NC Y Foxq1 (Hfh-1L) AF010405 Microarray and subtractive hybridization " N Hic-5 L22485 Microarray NC Y Meis1a U33629 Subtractive hybridization " Y Mitogen-responsive phosphoprotein p96 (Dab2 ) U18869 Microarray and subtractive hybridization " N midkine M35833 Subtractive hybridization " Y PDGF-a receptor M57683 Microarray " N Note: NC, no change; Y, yes; N, no. Molecular Cancer Research 621 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from RARh2 Target Genes 622 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from FOG1 is a member of the FOG (for Friend of GATA) family of multitype zinc finger proteins that interact with GATA and play essential roles in development (61). FOG1 is expressed in erythroid cells and megakaryocytes, liver, and testis (61). GATA6 is a member of the GATA family of transcription factors which regulate gene expression during development (62). GATA6, together with GATA4 and GATA5, is involved in the formation of the extra-embryonic and embryonic endoderm, as well as the cardiogenic mesoderm (63). FOG1 and GATA6 mRNAs were increased by RA in the F9 Wt, RARa / , and RARg / cell lines, but were not increased by RA or increased to a lesser extent in the F9 RARh2 / cells (Fig. 1A). Thus, FOG1 and GATA6 are RARh2 specific target genes. Hic5 is a focal adhesion protein that is very similar to paxillin (64). Studies have suggested that Hic5 is involved in the negative regulation of cell growth, since overexpression of Hic5 results in cell growth inhibition (65). Like p28 , the expression of Hic5 is not affected by RA treatment, but basal Hic5 mRNA expression is much higher in F9 Wt cells than in F9 RARh2 / cells (Table 1). PDGF-a receptor (PDGFaR) is one of the two receptors for platelet-derived growth factor (PDGF); it has been suggested that the PDGFaR plays a role in embryonic development (66). It was previously shown that PDGFaR expression is upregulated during RA-induced F9 cell differentiation (67). Our results show that the PDGFaR mRNA level is increased by RA in F9 Wt cells, but not in F9 RARh2 / cells (Table 1). PDGFaR mRNA levels are also lower in F9 RARa / and F9 RARg / cells, indicating that multiple RARs can regulate the expression of this gene. Target Genes are Transcriptionally Regulated by RARb2 To determine if the target genes were transcriptionally regulated by RARh2, we inhibited transcription by using actinomycin D, an inhibitor of RNA polymerase II (68). F9 Wt cells and F9 RARh2 / cells were first treated for 12 h with or without 1 AM RA. At 12 h, RNAwas isolated from some dishes of cells, while actinomycin D (2 Ag/ml) (69) was added to other dishes of cells to inhibit RNA transcription. RNA was isolated after an additional 6 h. The RNA samples were analyzed by Northern blot analysis (Fig. 2). Consistent with results shown in Fig. 1, RA induced target gene expression in F9 Wt cells, but not in the F9 RARh2 / cells. It is also worth noting that the stimulatory effect of RA can be seen as early at 12 h. Treatment of Wt cells with actinomycin D alone resulted in a decrease in target gene mRNA. Furthermore, RA induction of target gene mRNA was completely blocked by treatment with actinomycin D, indicating that ongoing transcription is required for RAinduced expression of target genes. The c-myc mRNA level seen after 18 h of RA treatment in F9 RARh2 / cells was greatly reduced after 6 h of actinomycin D treatment, which is consistent with the previous finding that the half-life of c-myc transcripts in F9 cells is about 40 min (70). RARb Agonists, in the Presence of a RXR Agonist, Stimulate the Expression of Target Genes in F9 Wt but not in F9 RARb2 / Cells To study further the function of RARh2 in the regulation of its target genes, F9 Wt and RARh2 / cells were treated for 48 h with 1 AM RA or RARh selective agonists (100 nM BMS185411 or 1 AM CD2314), and/or a pan-RXR agonist (1 AM BMS188649 [BMS649]) (Fig. 3). Neither BMS185411 nor CD2314, the RARh selective agonists, alone could induce the expression of any target genes in F9 Wt cells. Similarly, BMS649 (pan-RXR) alone had no effect on the expression of target genes. However, BMS649 (pan-RXR) and the RARh selective agonists (BMS185411 or CD2314) synergistically induced the expression of all RARh2 target genes in F9 Wt cells, but not in RARh2 / F9 cells. Thus, RXR-RARh2 heterodimers appear to play a major role in FIGURE 2. Northern blot analysis of the effects of actinomycin D on RARh2 target gene expression. F9 Wt cells and F9 RARh2 / cells were grown for 12 h in the absence ( ) or presence (+) of 1 AM RA. At 12 h, RNA was isolated or 2 Ag/ml transcriptional inhibitor actinomycin D (ACT D ) was added to the medium and RNA was isolated after 6 h. After size fractionation on the gel, the RNA was transferred to nylon and hybridization to different cDNAs for the target genes was performed. This experiment was performed twice with similar results; one experiment is shown. FIGURE 1. Northern blot analysis of the clones obtained from microarray analysis and subtractive hybridization analysis. A. F9 Wt cells, RARh2 / , a / , and g / F9 cells were treated with 1 AM RA for different times and total RNA was extracted. After size fractionation on the gel, the RNA was transferred to nylon and hybridized to different cDNAs for the target genes. h-Actin was used as an RNA loading control. This experiment was repeated three to five times with different RNA preparations and very similar results were obtained. A representative experiment is shown. B. Quantitative analysis of the effect of 1 AM RA on the mRNA levels of some target genes. C. Northern blot analysis of three clones obtained from subtractive hybridization analysis. This experiment was repeated three to five times with different RNA preparations and very similar results were obtained. One representative experiment is shown. D. Quantitative analysis of the effect of 1 AM RA on the mRNA levels of the target genes. Quantitative analyses in panels B and D were performed using NIH Image 1.62 or Image Quant 1.2. The mRNA levels are normalized to h-actin mRNA and are expressed relative to the value derived from Wt cells not treated with RA. Columns, means from three to five experiments; bars, SE. Y-axes , arbitrary density units. Molecular Cancer Research 623 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from mediating the induction of the expression of RARh2 target genes. This is consistent with the conclusion that RXR-RAR heterodimers are the functional units transducing the retinoid signal (9). Expression of RARb2 Target Genes Is Induced in F9 Wt Cells Treated With RACT, but not in F9 RARb2 / Cells RACT (retinoic acid, dibutyryl cAMP, and theophylline) treatment causes F9 Wt cells to differentiate into parietal endoderm, while RA alone results in primitive endoderm (71). We examined the regulation of the expression of some of the RARh2 target genes in F9 cells treated with RACT. F9 Wt or F9 RARh2 / cells were treated with RACT, or different combinations of RARh agonist and/or RXR agonist in the presence of CT (dibutyryl cAMP and theophylline) for 48 h. The RNA samples were analyzed by Northern blot analysis (Fig. 4). As expected, CT alone did not have an effect on expression of target genes (72). Like RA, RACT induced the expression of target genes in F9 Wt cells. Addition of the panRXR agonist BMS649 decreased the Foxq1 , FOG1 , and Meis1a expression induced by RACT in F9 Wt cells, but increased the GATA6 expression induced by RACT. In the presence of CT, BMS649 (pan-RXR) and the RARh selective agonists (BMS185411 or CD2314) induced the expression of FIGURE 3. Expression of RARh2 target genes is increased by a RARh-specific agonist and pan-RXR agonist. A. F9 Wt cells and RARh2 / cells were treated for 48 h with 1 AM RA, 1 AM or 100 nM RARh-specific agonist (BMS185411 or CD2314 ), 1 AM pan-RXR agonist (BMS188649), or different combinations of drugs. Total RNA was then extracted and size fractionated on the gel. The RNA was transferred onto nylon and hybridized to different cDNAs for the target genes. The experiments were repeated with different RNA preparations. h-Actin is used as an RNA loading control. One representative experiment is shown here. B. Quantitation of the effects of different agonists on the mRNA levels of the target genes in F9 Wt cells. The mRNA levels are normalized to h-actin mRNA and are expressed relative to the value derived from Wt cells not treated with RA. Columns, means from three experiments; bars, SE. Y-axes , arbitrary units. FIGURE 4. Expression of RARh2 target genes in F9 Wt cells and RARh2 / cells treated with RACT. F9 Wt cells and F9 RARh2 / cells were treated for 48 h with 1 AM RA, 100 nM (CD2314 ) or 1 AM (BMS185411 ) RARh-specific agonist, 1 AM pan-RXR agonist, or different combinations in the presence of CT [dibutyryl cAMP (250 AM) and theophylline (250 AM)]. Total RNA (10 Ag) was then extracted and size fractionated on the gel. The RNA was transferred onto nylon and hybridized to different cDNAs for the target genes. The experiments were performed two times with different RNA preparations. h-Actin is used as an RNA loading control. Quantitation of some of the mRNAs is shown below the figure. RARh2 Target Genes 624 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from Molecular Cancer Research 625 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from all RARh2 target genes in F9 Wt cells. In addition, retinol (Rol; vitamin A) induced the expression of all of the RARh2 target genes in F9 Wt cells. In contrast, in F9 RARh2 / cells, expression of FOG1 , Foxq1 , GATA6 , and Meis1a mRNA was very low with various treatments (Fig. 4). We also studied the regulation of expression of the Hoxa1 gene, a gene we have studied previously (23, 69, 73, 74). Unlike the RARh2 target genes, a similar pattern of Hoxa1 mRNA expression was observed in F9 Wt and F9 RARh2 / cells (Fig. 4). Hoxa1 mRNA expression was induced by RACT in both F9 Wt and the RARh2 / cells. Furthermore, Hoxa1 mRNA expression was not significantly increased by the RXR agonist (BMS649) and RARh selective agonists (BMS185411 or CD2314) in the presence of CT (Fig. 4). Rol only slightly induced Hoxa1 mRNA expression in F9 Wt cells; thus, Rol treatment was more effective in inducing several of the RARh2 target genes, especially GATA6 and Foxq1 , than in inducing Hoxa1 mRNA, consistent with our prior genetic data that the Hoxa1 gene is primarily transcriptionally regulated by RARg and not RARh2 in F9 cells (23). Discussion In this study, using subtractive hybridization and microarray analyses, we identified some of the RARh2 targets in F9 cells. To our knowledge, this is the first time that specific RARh2 target genes have been identified and confirmed by a comparison of Wt, RARh2 / , RARa / , and RARg / cell lines. Most of the target genes are regulated by RA, though our data suggest that some genes, such as Dab2 , Hic5 , and p28 , are also regulated by RARh2 in the absence of exogenously added RA. The RARh2 target genes are also induced by Rol (Fig. 4) to a greater extent than a primary RA target gene such as Hoxa1 , which is not an RARh target (23). As we have shown previously that Rol is not metabolized to RA in F9 cells (75), a Rol metabolite, 4-oxoRol, is likely to be involved in the regulation of these target genes in F9 cells. We have not yet determined the number of primary versus secondary RARh2 target genes, as this requires much more extensive characterization of the promoters of the genes. The target genes identified in this study encode a wide range of proteins that regulate a variety of biological processes. This suggests that, in addition to the regulation of cell proliferation, the transcription factor RARh2 participates in the regulation of many biological processes via the transcription of multiple target genes. We found that some target genes are RARh2 specific (Table 1) in that the F9 RARh2 / cells exhibited reduced or no expression of these target genes, but the F9 RARa / and F9 RARg / cell lines exhibited expression levels similar to those in F9 Wt cells. These data provide evidence that each RAR isoform has distinct functions. Target Genes and Cell Growth and Cancer Consistent with our data showing that the induction of RARh2 is necessary for the growth arrest of F9 cells, two of the target genes we identified, c-myc and Dab2 , from this study have been shown to play major roles in cell growth and proliferation. The c-myc proto-oncogene, the cellular homologue of the retroviral v-myc oncogene, is activated in various animal and human tumors (76). c-Myc is a transcription factor which regulates many biological processes, including the cell cycle and cell differentiation (77). The c-Myc protein can regulate some cyclins, cyclin-dependent kinases (CDKs), and CDK inhibitors, which are cell cycle regulators (78). Deregulated c-myc expression often results in increased cyclin E and cyclin A expression (79, 80). In addition, expression of c-myc can decrease p27 levels and interfere with its function (81). Our data indicate that expression of c-myc mRNA is repressed by RARh2 in F9Wt cells, even in the absence of exogenous RA, and that c-myc mRNA is expressed at a much higher level in untreated and 24 h RA treated F9 RARh2 / than F9 Wt cells (Fig. 1). We conclude that in F9 Wt cells, a low level of RARh2 expression (without exogenous RA addition but in the presence of low levels of Rol from serum) inhibits c-myc mRNA expression. The reduction of c-myc mRNA by RA in F9 Wt cells has been observed in several studies, and down-regulation of c-myc mRNA by RA is associated with RA-induced growth arrest of F9 cells (82). The increased level of c-myc mRNA in RARh2 / cells may play a role in preventing these cells from undergoing growth arrest on RA treatment. Since changes in the c-myc gene or its expression are associated with approximately one-seventh of U.S. cancer deaths (76), experiments will be carried out to determine if the loss of RARh2 expression in tumors causes the overexpression of c-myc transcripts and concomitant abnormal regulation of tumor cell growth. RA has been shown to regulate Dab2 expression during embryonic development (83). Moreover, Dab2 has been suggested to be a major mediator of RA cell growth inhibition. Transient transfection of Dab2 was sufficient to uncouple MAPK activation and c-fos expression, and was associated with growth inhibition in F9 cells (84). Since we found Dab2 to be a RARh2 target in this study, RARh2 could mediate the growth inhibitory effects of RA by up-regulating Dab2 expression. Studies of GATA6-deficient mouse embryos suggest that Dab2 expression is regulated by GATA6 (85). Since we have shown that GATA6 is also a RARh2 specific target, Dab2 may be an indirect target of RARh2 and may be regulated by RARh2 through GATA6. The lack of a known RARE in the Dab2 promoter (86) supports this suggestion. The observation that the expression of Dab2 mRNA is eliminated in 85–95% of breast and ovarian tumors suggests that it is a tumor suppressor gene (51). Indeed, overexpression of Dab2 reduced the tumorigenicity of carcinoma cells (87). Conversely, the reduction of RARh2 levels in human breast tumors (88) could result in the down-regulation of Dab2 , which in turn could alter the control of cell proliferation. Target Genes and Development Several targets identified in this study provide insight into the roles of RARh2 in development. Meis1a is a member of the homeobox gene family of transcription factors and has been suggested to play a role in peripheral neural system development. Correct expression of Meis1a is essential for correct patterning in the embryonic peripheral nervous system (89). Our finding that Meis1a is a target of RARh2 is consistent with the finding that RARh is expressed and plays a role in the developing nervous system (90). RARh2 Target Genes 626 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from FOG1 is the founding member of the FOG family of proteins that interact with GATA transcription factors and modulate their activity (61). FOG1 knockout mice die at embryonic day 11.5 of gestation because of arrested erythroid cell maturation and complete failure of megakaryopoiesis (91). As FOG1 was shown to be a target of RARh2 in these F9 cells, RARh2 may play a role in megakaryopoiesis. The GATA6 transcription factor is involved in the formation of the extra-embryonic and embryonic endoderm, as well as in cardiogenic mesoderm (92). GATA6 is essential for lung epithelial differentiation in the mouse embryo (63). Our result that GATA6 is a RARh2 specific target suggests that RARh has a role in the lung epithelial cell differentiation. Consistent with this, RARh is expressed in the epithelia of the lungs (93). In the midgestation period, midkine is expressed in the endoderm-derived lung epithelium and the adjacent mesenchyme, which suggests a key role for midkine in the regulation of lung development (94). Interestingly, RARh is also expressed in the endoderm-derived lung epithelium and the adjacent mesenchyme at the same stage of embryogenesis. In the embryo, PDGF-A is expressed primarily by epithelial cells, whereas the PDGF-a receptor is expressed primarily in mesenchymal cells. This ligand-receptor pair is thought to be important to epithelial-mesenchymal signaling in the lung during development (95). These findings, together with our results that midkine and the PDGF-a receptor are RARh2 target genes, suggest that RARh plays a role in epithelialmesenchymal signaling during lung development. Foxq1 is expressed throughout embryonic development of the mouse (52). Although its function remains unclear, it is likely to play a crucial role during embryogenesis, given its similarity to other family members. Another family member, Foxp2 , is involved in the development of speech (96). Further characterization of the RARh2 target genes identified in this study will be helpful in increasing our understanding of the functions of RARh2 in a variety of biological processes. Future studies should also provide important insights into the molecular mechanisms by which the silencing of the RARh2 gene leads to cancer. Materials and Methods Materials The RARh agonist BMS185411 and the pan-RXR agonist BMS188649 are from Bristol-Myers Squibb (Buffalo, NY). The RARh agonist CD2314 is from CIRD Galderma (Sophia Antipolis, Valbonne, France). Actinomycin D was purchased from Sigma Chemical Co. (St. Louis, MO) and used at a concentration (2 Ag/ml) shown to inhibit RNA synthesis (69). Cell Culture, RNA Extraction, and Northern Blot Analysis F9 Wt, RARh2 / , RARa / , and RARg / cells were maintained in DME supplemented with 10% bovine calf serum and 2 mM glutamine. Total cellular RNA was extracted using RNA-STAT 60 (Tel-Test, Friendswood, TX) following the manufacturer’s protocol. Northern blot analysis was performed as described previously (97). Briefly, total RNA (10 Ag) was resolved on 1% agarose gels containing 5% formaldehyde and RNA was transferred to nylon filters. Nylon filters were prehybridized, and hybridized at 42jC in 50% (v/v) formamide, 5 SSC, 50 mM NaH2PO4-Na2HPO4 (pH 7.4), 5 mM EDTA, and 0.1 mg/ml salmon sperm DNA. cDNA probes of target genes were labeled with [P]dCTP using a random primer labeling kit (Roche, Indianapolis, IN). cDNA inserts for target genes were either from the F9 cDNA library described below or were commercially available EST clones. EST clones for GATA6 (AI894084), PDGF-a receptor (BE630572), and FOG1 (AW763708) were obtained from Incyte (Palo Alto, CA). EST clones for c-myc (BG243402) and Hic5 (BF536993) were from ATCC. Northern blots were quantitated by phosphorimager analysis, and the software NIH Image 1.62 or Image Quant 1.2. Subtractive Hybridization A PCR-Select cDNA Subtraction Kit (Clontech, Palo Alto, CA) was used to carry out subtractive hybridization analysis. F9 Wt teratocarcinoma cells and the F9 RARh2 / line were treated with 1 AM all-trans RA for 24 h and total RNA was extracted. After reverse transcription, cDNAs from the two cell lines were subjected to subtractive hybridization according to the manufacturer’s protocol. Products from the secondary PCR reactions were radiolabeled with [a-P]dCTP and hybridized to duplicate filters containing phage clones from a cDNA library in a Lambda ZAP vector; this cDNA library was synthesized from F9Wt cells treated with RA for 72 h. After autoradiography, the filter patterns of F9 Wt cells and of F9 RARh2 / cells were compared for each spot. Differentially hybridized clones were picked and the inserts were subjected to Northern blot analysis. DNA Microarray Analysis Total cellular RNA was extracted from F9 Wt teratocarcinoma cells and the F9 RARh2 / cells, both of which were treated with 1 AM all-trans RA for 24 h. Microarray analysis was carried out according to the Affymetrix Genechip expression analysis technical manual. Briefly, total RNA from both cell lines was reverse transcribed into cDNA. After second-strand synthesis, cDNAs were then in vitro transcribed into cRNAwith biotinylated ribonucleotides (Enzo Diagnostics, Farmingdale, NY). cRNA (20 Ag) was fragmented by heating at 94jC for 35 min. A cocktail containing fragmented cRNA, control oligonucleotide B2, control cRNA (Biotin B, Biotin C, Biotin D, and Cre), and herring sperm DNA was hybridized to microarray chips (MG-U74Av2) for 16 h at 45jC. After washing and staining, the distribution of fluorescent material on the array was measured using a laser scanner. The resultant image was processed with MAS (Microarray Suite) software (Affymetrix, Santa Clara, CA). The microarray experiments were repeated three times with different RNA preparations. Only genes which exhibited different expression patterns between the F9 Wt and F9 RARh2 / cells in three independent array experiments were further analyzed. Microarray Data Analysis The Microarray Core of the Weill Medical College of Cornell performed data preparation and data analysis. Thompson and Gudas, unpublished data. Molecular Cancer Research 627 Research. on November 6, 2017. © 2003 American Association for Cancer mcr.aacrjournals.org Downloaded from The expression level of each gene and the fold change between genes from F9 Wt cells and F9 RARh2 / cells were calculated using a computer program (GeneSpring, Redwood, CA). This computer program was also used for functional gene grouping analyses. AcknowledgmentsWe thank members of the Gudas laboratory for scientific discussions. References1. Ross, S. A., McCaffery, P. J., Drager, U. C., and De Luca, L. M. Retinoids inembryonal development. Physiol. Rev., 80: 1021–1054, 2000. 2. Gudas, L. J., Sporn, M. B., and Roberts, A. B. Cellular biology andbiochemistry of the retinoids. In: M. B. Sporn, A. B. Roberts, and D. S. Goodman(eds.), The Retinoids: Biology, Chemistry, and Medicine, Second edition,pp. 443– 520. New York: Raven Press, 1994. 3. Moon, R. C., Mehta, R. G., and Rao, K. V. N. Retinoids and cancer inexperimental animals. In: M. B. Sporn, A. B. Roberts, and D. S. Goodman (eds.),The Retinoids: Biology, Chemistry and Medicine, pp. 573–596. New York:Raven Press, 1994. 4. Lotan, R. Retinoids in cancer chemoprevention. FASEB J., 10: 1031–1039,1996. 5. Warrell, R. P., Jr., de The, H., Wang, Z-Y., and Degos, L. Acute promyelocyticleukemia. N. Engl. J. Med., 329: 177–189, 1993. 6. Mandelli, F., Avvisati, G., and Lo Coco, F. Advances in the understanding andmanagement of acute promyelocytic leukemia. Rev. Clin. Exp. Hematol., 6: 60–71; discussion 86–67, 2002. 7. Miller, W. H. The emerging role of retinoids and retinoic acid metabolismblocking agents in the treatment of cancer. Cancer, 83: 1471–1482, 1998. 8. Hansen, L. A., Sigman, C. C., Andreola, F., Ross, S. A., Kelloff, G. J., andDe Luca, L. M. Retinoids in chemoprevention and differentiation therapy.Carcinogenesis, 21: 1271–1279, 2000. 9. Kastner, P., Mark, M., and Chambon, P. Nonsteroid nuclear receptors: what aregenetic studies telling us about their role in real life? Cell, 83: 859–869, 1995. 10. Mangelsdorf, D. J., Thummel, C., Beato, M., Herrlich, P., Schutz, G.,Umesono, K., Blumberg, B., Kastner, P., Mark, M., Chambon, P., and Evans, R. M.The nuclear receptor superfamily: the second decade. Cell, 83: 835–839, 1995. 11. Allenby, G., Bocquel, M. T., Saunders, M., Kazmer, S., Speck, J.,Rosenberger, M., Lovey, A., Kastner, P., Grippo, J. F., Chambon, P., and Levin,A. A. Retinoic acid receptors and retinoid X receptors: interactions withendogenous retinoic acids. Proc. Natl. Acad. Sci. USA, 90: 30–34, 1993. 12. Chambon, P. A decade of molecular biology of retinoic acid receptors.FASEB J., 10: 940 –954, 1996. 13. Torchia, J., Glass, C., and Rosenfeld, M. G. Co-activators and co-repressorsin the integration of transcriptional responses. Curr. Opin. Cell Biol., 10: 373–383, 1998. 14. Freedman, L. P. Increasing the complexity of coactivation in nuclear receptorsignaling. Cell, 97: 5 –8, 1999. 15. Bannister, A. J. and Kouzarides, T. The CBP co-activator is a histoneacetyltransferase. Nature, 384: 641 –643, 1996. 16. Ogryzko, V. V., Schiltz, R. L., Russanova, V., Howard, B. H., and Nakatani,Y. The transcriptional coactivators p300 and CBP are histone acetyltransferases.Cell, 87: 953–959, 1996. 17. Spencer, T. E., Jenster, G., Burcin, M. M., Allis, C. D., Zhou, J., Mizzen,C. A., McKenna, N. J., Onate, S. A., Tsai, S. Y., Tsai, M-J., and O’Malley, B. W.Steroid receptor coactivator-1 is a histone acetyltransferase. Nature, 389:194– 197, 1997. 18. Glass, C. K. and Rosenfeld, M. G. The coregulator exchange in transcrip-tional functions of nuclear receptors. Genes Dev., 14: 121–141, 2001. 19. Kastner, P., Leid, M., and Chambon, P. The role of nuclear retinoic acidreceptors in the regulation of gene expression. In: Vitamin A in Health andDisease, pp. 189 –238. New York: Marcel Dekker, Inc., 1994. 20. Taneja, R., Bouillet, P., Boylan, J. F., Gaub, M. P., Roy, B., Gudas, L. J., andChambon, P. Reexpression of retinoic acid receptor (RAR) g or overexpression ofRAR a or RAR h in RAR g-null F9 cells reveals a partial functional redundancybetween the three RAR types. Proc. Natl. Acad. Sci. USA, 92: 7854–7858, 1995. 21. Faria, T. N., Mendelsohn, C., Chambon, P., and Gudas, L. J. The targeteddisruption of both alleles of RARh(2) in F9 cells results in the loss of retinoicacid-associated growth arrest. J. Biol. Chem., 274: 26783 –26788, 1999.22. Boylan, J. F., Lufkin, T., Achkar, C. C., Taneja, R., Chambon, P., and Gudas,L. J. Targeted disruption of retinoic acid receptor a (RAR a) and RAR g results inreceptor-specific alterations in retinoic acid-mediated differentiation and retinoicacid metabolism. Mol. Cell. Biol., 15: 843 –851, 1995. 23. Boylan, J. F., Lohnes, D., Taneja, R., Chambon, P., and Gudas, L. J. Loss ofretinoic acid receptor g function in F9 cells by gene disruption results in aberrantHoxa-1 expression and differentiation upon retinoic acid treatment. Proc. Natl.Acad. Sci. USA, 90: 9601– 9605, 1993. 24. Mendelsohn, C., Larkin, S., Mark, M., Le Meur, M., Clifford, J., Zelent, A.,and Chambon, P. RARh isoforms: distinct transcriptional control by retinoic acidand specific spatial patterns of promoter activity during mouse embryonicdevelopment. Mech. Dev., 45: 227– 241, 1994. 25. Lufkin, T., Lohnes, D., Mark, M., Dierich, A., Gorry, P., Gaub, M. P.,Le Meur, M., and Chambon, P. High postnatal lethality and testis degenerationin retinoic acid receptor a mutant mice. Proc. Natl. Acad. Sci. USA, 90:7225– 7229, 1993. 26. Lohnes, D., Kastner, P., Dierich, A., Mark, M., LeMeur, M., and Chambon, P.Function of retinoic acid receptor g in the mouse. Cell, 73: 643 –658, 1993. 27. Luo, J., Pasceri, P., Conlon, R. A., Rossant, J., and Giguere, V. Mice lackingall isoforms of retinoic acid receptor b develop normally and are susceptible to theteratogenic effects of retinoic acid. Mech. Dev., 53: 61–71, 1995. 28. Chiang, M. Y., Misner, D., Kempermann, G., Schikorski, T., Giguere, V.,Sucov, H. M., Gage, F. H., Stevens, C. F., and Evans, R. M. An essential role forretinoid receptors RARh and RXRg in long-term potentiation and depression.Neuron, 21: 1353– 1361, 1998. 29. Mendelsohn, C., Mark, M., Dolle, P., Dierich, A., Gaub, M. P., Krust, A.,Lampron, C., and Chambon, P. Retinoic acid receptor h 2 (RAR h 2) null mutantmice appear normal. Dev. Biol., 166: 246 –258, 1994. 30. Sucov, H. M., Murakami, K. K., and Evans, R. M. Characterization of anautoregulated response element in the mouse retinoic acid receptor type h gene.Proc. Natl. Acad. Sci. USA, 87: 5392– 5396, 1990. 31. de Thé, H., Vivanco-Ruiz, M., Tiollais, P., Stunnenberg, H., and Dejean, A.Identification of a retinoic acid responsive element in the retinoic acid receptor hgene. Nature, 343: 177–180, 1990. 32. Hoffmann, B., Lehmann, J. M., Zhang, X-K., Hermann, T., Graupner, G., andPfahl, M. A retinoic acid receptor specific element controls the retinoic acidreceptor-h promoter. Mol. Endocrinol., 4: 1727–1736, 1990. 33. Zhang, X., Hoffmann, B., Tran, P. B. V., Graupner, G., and Pfahl, M.Retinoid X receptor is an auxiliary protein for thyroid hormone and retinoic acidreceptors. Nature, 355: 441 –446, 1992. 34. Wu, Q., Dawson, M. I., Zheng, Y., Hobbs, P. D., Agadir, A., Jong, L., Li, Y.,Liu, R., Lin, B., and Zhang, X. K. Inhibition of trans -retinoic acid-resistanthuman breast cancer cell growth by retinoid X receptor-selective retinoids. Mol.Cell. Biol., 17: 6598–6608, 1997. 35. Zelent, A., Mendelsohn, C., Kastner, P., Krust, A., Garnier, J. M., Ruffenach,F., Leroy, P., and Chambon, P. Differentially expressed isoforms of the mouseretinoic acid receptor h generated by usage of two promoters and alternativesplicing. EMBO J., 10: 71–81, 1991. 36. Hu, L., Crowe, D. L., Rheinwald, J. G., Chambon, P., and Gudas, L. J.Abnormal expression of retinoic acid receptors and keratin 19 by humanoral and epidermal squamous cell carcinoma cell lines. Cancer Res., 51:3972– 3981, 1991. 37. Qiu, H., Zhang, W., El-Naggar, A. K., Lippman, S. M., Lin, P., Lotan, R., andXu, X. C. Loss of retinoic acid receptor-h expression is an early event duringesophageal carcinogenesis. Am. J. Pathol., 155: 1519–1523, 1999. 38. Swisshelm, K., Ryan, K., Lee, X., Tsou, H. C., Peacocke, M., and Sager, R.Down-regulation of retinoic acid receptor h in mammary carcinoma cell lines andits up-regulation in senescing normal mammary epithelial cells. Cell Growth &Differ., 5: 133– 141, 1994. 39. Xu, X. C., Ro, J. Y., Lee, J. S., Shin, D. M., Hong, W. K., and Lotan, R.Differential expression of nuclear retinoid receptors in normal, premalignant, andmalignant head and neck tissues. Cancer Res., 54: 3580–3587, 1994. 40. Zhang, X. K., Liu, Y., Lee, M. O., and Pfahl, M. A specific defect in theretinoic acid response associated with human lung cancer cell lines. Cancer Res.,54: 5663–5669, 1994. 41. Mattei, M. G., de The, H., Mattei, J. F., Marchio, A., Tiollais, P., and Dejean,A. Assignment of the human hap retinoic acid receptor RAR h gene to the p24band of chromosome 3. Hum. Genet., 80: 189 –190, 1988. 42. Deng, G., Lu, Y., Zlotnikov, G., Thor, A. D., and Smith, H. S. Loss ofheterozygosity in normal tissue adjacent to breast carcinomas. Science, 274:2057– 2059, 1996. 43. Weber, E., Ravi, R. K., Knudsen, E. S., Williams, J. R., Dillehay, L. E.,Nelkin, B. D., Kalemkerian, G. P., Feramisco, J. R., and Mabry, M. Retinoic acid-RARh2 Target Genes628 Research.on November 6, 2017. © 2003 American Association for Cancermcr.aacrjournals.orgDownloaded from mediated growth inhibition of small cell lung cancer cells is associated withreduced myc and increased p27 expression. Int. J. Cancer, 80: 935 –943,1999. 44. Seewaldt, V. L., Johnson, B. S., Parker, M. B., and Collins, S. J. Expressionof retinoic acid receptor h mediates retinoic acid-induced growth arrest andapoptosis in breast cancer cells. Cell Growth & Differ., 6: 1077–1088, 1995. 45. Si, S. P., Lee, X., Tsou, H. C., Buchsbaum, R., Tibaduiza, E., and Peacocke,M. RAR h 2-mediated growth inhibition in HeLa cells. Exp. Cell Res., 223:102– 111, 1996. 46. Liu, Y., Lee, M-I., Wang, H-G., Li, Y., Hashimoto, Y., Klaus, M., Reed, J. C.,and Zhang, X-K. Retinoic acid receptor h mediates the growth-inhibitory effect ofretinoic acid by promoting apoptosis in human breast cancer cells. Mol. Cell.Biol., 16: 1138– 1149, 1996. 47. Li, Y., Dawson, M. I., Agadir, A., Lee, M. O., Jong, L., Hobbs, P. D., andZhang, X. K. Regulation of RAR h expression by RARand RXR-selectiveretinoids in human lung cancer cell lines: effect on growth inhibition andapoptosis induction. Int. J. Cancer, 75: 88–95, 1998. 48. Houle, B., Rochette-Egly, C., and Bradley, W. E. Tumor-suppressive effect ofthe retinoic acid receptor h in human epidermoid lung cancer cells. Proc. Natl.Acad. Sci. USA, 90: 985– 989, 1993. 49. Toulouse, A., Loubeau, M., Morin, J., Pappas, J. J., Wu, J., and Bradley,W. E. RARh involvement in enhancement of lung tumor cell immunogenicityrevealed by array analysis. FASEB J., 14: 1224–1232, 2000. 50. Lotan, R., Xu, X. C., Lippman, S. M., Ro, J. Y., Lee, J. S., Lee, J. J.,and Hong, W. K. Suppression of retinoic acid receptor-h in premalignant orallesions and its up-regulation by isotretinoin. N. Engl. J. Med., 332: 1405–1410,1995. 51. Fazili, Z., Sun, W., Mittelstaedt, S., Cohen, C., and Xu, X. X. Disabled-2 inactivation is an early step in ovarian tumorigenicity. Oncogene, 18: 3104–3113, 1999. 52. Hong, H. K., Noveroske, J. K., Headon, D. J., Liu, T., Sy, M. S., Justice,M. J., and Chakravarti, A. The winged helix/forkhead transcription factor Foxq1regulates differentiation of hair in satin mice. Genesis, 29: 163–171, 2001. 53. Moskow, J. J., Bullrich, F., Huebner, K., Daar, I. O., and Buchberg, A. M.Meis1 , a PBX1-related homeobox gene involved in myeloid leukemia in BXH-2mice. Mol. Cell. Biol., 15: 5434–5443, 1995. 54. Pineault, N., Helgason, C. D., Lawrence, H. J., and Humphries, R. K.Differential expression of Hox , Meis1 , and Pbx1 genes in primitive cellsthroughout murine hematopoietic ontogeny. Exp. Hematol., 30: 49–57, 2002. 55. Tomomura, M., Kadomatsu, K., Matsubara, S., and Muramatsu, T. A retinoicacid-responsive gene, MK , found in the teratocarcinoma system. Heterogeneity ofthe transcript and the nature of the translation product. J. Biol. Chem., 265:10765 –10770, 1990. 56. Tzimagiorgis, G., Adamson, M. C., Kozak, C. A., and Moschonas, N. K.Chromosomal mapping of glutamate dehydrogenase gene sequences to mousechromosomes 7 and 14. Genomics, 10: 83–88, 1991. 57. Zaganas, I., Waagepetersen, H. S., Georgopoulos, P., Sonnewald, U.,Plaitakis, A., and Schousboe, A. Differential expression of glutamate dehydrogen-ase in cultured neurons and astrocytes from mouse cerebellum and cerebralcortex. J. Neurosci. Res., 66: 909–913, 2001. 58. Kodym, R., Calkins, P., and Story, M. The cloning and characterization of anew stress response protein. A mammalian member of a family of theta classglutathione s -transferase-like proteins. J. Biol. Chem., 274: 5131–5137, 1999. 59. Dang, C. V. c-Myc target genes involved in cell growth, apoptosis, andmetabolism. Mol. Cell. Biol., 19: 1–11, 1999. 60. Stanton, L. W., Fahrlander, P. D., Tesser, P. M., and Marcu, K. B. Nucleotidesequence comparison of normal and translocated murine c-myc genes. Nature,310: 423–425, 1984. 61. Tsang, A. P., Visvader, J. E., Turner, C. A., Fujiwara, Y., Yu, C., Weiss, M. J.,Crossley, M., and Orkin, S. H. FOG, a multitype zinc finger protein, acts as acofactor for transcription factor GATA-1 in erythroid and megakaryocyticdifferentiation. Cell, 90: 109–119, 1997. 62. Narita, N., Heikinheimo, M., Bielinska, M., White, R. A., and Wilson, D. B.The gene for transcription factor GATA-6 resides on mouse chromosome 18and is expressed in myocardium and vascular smooth muscle. Genomics, 36:345– 348, 1996. 63. Keijzer, R., van Tuyl, M., Meijers, C., Post, M., Tibboel, D., Grosveld, F.,and Koutsourakis, M. The transcription factor GATA6 is essential for branchingmorphogenesis and epithelial cell differentiation during fetal pulmonary develop-ment. Development, 128: 503–511, 2001. 64. Thomas, S. M., Hagel, M., and Turner, C. E. Characterization of a focaladhesion protein, Hic-5, that shares extensive homology with paxillin. J. Cell.Sci., 112 (Pt 2): 181–190, 1999.65. Nishiya, N., Tachibana, K., Shibanuma, M., Mashimo, J. I., and Nose, K.Hic-5-reduced cell spreading on fibronectin: competitive effects between paxillinand Hic-5 through interaction with focal adhesion kinase. Mol. Cell. Biol., 21:5332– 5345, 2001. 66. Wang, C. and Song, B. Cell-type-specific expression of the platelet-derivedgrowth factor a receptor: a role for GATA-binding protein. Mol. Cell. Biol., 16:712 –723, 1996. 67. Wang, C., Kelly, J., Bowen-Pope, D. F., and Stiles, C. D. Retinoic acidpromotes transcription of the platelet-derived growth factor a-receptor gene. Mol.Cell. Biol., 10: 6781–6784, 1990. 68. Cohen, S. M., Cobb, E. R., Cordeiro-Stone, M., and Kaufman, D. G.Identification of chromosomal bands replicating early in the S phase of normalhuman fibroblasts. Exp. Cell Res., 245: 321 –329, 1998. 69. LaRosa, G. J. and Gudas, L. J. Early retinoic acid-induced F9 teratocarci-noma stem cell gene ERA-1: alternate splicing creates transcripts for a homeobox-containing protein and one lacking the homeobox. Mol. Cell. Biol., 8: 3906–3917, 1988. 70. Sejersen, T., Rahm, M., Szabo, G., Ingvarsson, S., and Sumegi, J. Similaritiesand differences in the regulation of N-myc and c-myc genes in murine embryonalcarcinoma cells. Exp. Cell Res., 172: 304 –317, 1987. 71. Strickland, S., Smith, K. K., and Marotti, K. R. Hormonal induction ofdifferentiation in teratocarcinoma stem cells: generation of parietal endoderm byretinoic acid and dibutyryl cAMP. Cell, 21: 347 –355, 1980. 72. Wang, S-Y., LaRosa, G., and Gudas, L. J. Molecular cloning of genesequences transcriptionally regulated by retinoic acid and dibutyryl cyclic AMP incultured mouse teratocarcinoma cells. Dev. Biol., 107: 75–86, 1985. 73. Langston, A. W. and Gudas, L. J. Identification of a retinoic acid responsiveenhancer 3V of the murine homeobox gene Hox 1.6 . Mech. Dev., 38: 217–228,1992. 74. Langston, A. W., Thompson, J. R., and Gudas, L. J. Retinoic acid-responsiveenhancers located 3Vof the Hox A and Hox B homeobox gene clusters. Functionalanalysis. J. Biol. Chem., 272: 2167– 2175, 1997. 75. Achkar, C. C., Derguini, F., Blumberg, B., Langston, A., Levin, A. A.,Speck, J., Evans, R. M., Bolado, J., Jr., Nakanishi, K., and Buck, J. 4-Oxoretinol,a new natural ligand and transactivator of the retinoic acid receptors. Proc. Natl.Acad. Sci. USA, 93: 4879– 4884, 1996. 76. Dang, C. V., Resar, L. M., Emison, E., Kim, S., Li, Q., Prescott, J. E.,Wonsey, D., and Zeller, K. Function of the c-Myc oncogenic transcription factor.Exp. Cell Res., 253: 63– 77, 1999. 77. Eisenman, R. N. Deconstructing myc . Genes Dev., 15: 2023–2030, 2001. 78. Amati, B., Alevizopoulos, K., and Vlach, J. Myc and the cell cycle. Front.Biosci., 3: D250–D268, 1998. 79. Barrett, J. F., Lewis, B. C., Hoang, A. T., Alvarez, R. J., Jr., and Dang, C. V.Cyclin A links c-Myc to adhesion-independent cell proliferation. J. Biol. Chem.,270: 15923 –15925, 1995. 80. Hanson, K. D., Shichiri, M., Follansbee, M. R., and Sedivy, J. M. Effects ofc-myc expression on cell cycle progression. Mol. Cell. Biol., 14: 5748–5755,1994. 81. Vlach, J., Hennecke, S., Alevizopoulos, K., Conti, D., and Amati, B. Growtharrest by the cyclin-dependent kinase inhibitor p27Kip1 is abrogated by c-Myc.EMBO J., 15: 6595–6604, 1996. 82. Dean, M., Levine, R. A., and Campisi, J. c-myc regulation during retinoicacid-induced differentiation of F9 cells is posttranscriptional and associated withgrowth arrest. Mol. Cell. Biol., 6: 518– 524, 1986. 83. Sheng, Z., Smith, E. R., He, J., Tuppen, J. A., Martin, W. D., Dong, F. B., andXu, X. X. Chromosomal location of murine disabled-2 gene and structuralcomparison with its human ortholog. Gene, 268: 31–39, 2001. 84. Smith, E. R., Capo-chichi, C. D., He, J., Smedberg, J. L., Yang, D. H.,Prowse, A. H., Godwin, A. K., Hamilton, T. C., and Xu, X. X. Disabled-2mediates c-Fos suppression and the cell growth regulatory activity of retinoicacid in embryonic carcinoma cells. J. Biol. Chem., 276: 47303 –47310,2001. 85. Morrisey, E. E., Musco, S., Chen, M. Y., Lu, M. M., Leiden, J. M., andParmacek, M. S. The gene encoding the mitogen-responsive phosphoproteinDab2 is differentially regulated by GATA-6 and GATA-4 in the visceralendoderm. J. Biol. Chem., 275: 19949–19954, 2000. 86. Sheng, Z., He, J., Tuppen, J. A., Sun, W., Fazili, Z., Smith, E. R., Dong, F. B.,and Xu, X. X. Structure, sequence, and promoter analysis of human disabled-2gene (DAB2). Genomics, 70: 381– 386, 2000. 87. Sheng, Z., Sun, W., Smith, E., Cohen, C., and Xu, X. X. Restoration ofpositioning control following Disabled-2 expression in ovarian and breast tumorcells. Oncogene, 19: 4847–4854, 2000.Molecular Cancer Research 629 Research.on November 6, 2017. © 2003 American Association for Cancermcr.aacrjournals.orgDownloaded from 88. Widschwendter, M., Berger, J., Daxenbichler, G., Muller-Holzner, E.,Widschwendter, A., Mayr, A., Marth, C., and Zeimet, A. G. Loss of retinoicacid receptor h expression in breast cancer and morphologically normal adjacenttissue but not in the normal breast tissue distant from the cancer. Cancer Res., 57:4158–4161, 1997. 89. Kurant, E., Pai, C. Y., Sharf, R., Halachmi, N., Sun, Y. H., and Salzberg, A.Dorsotonals/homothorax, the Drosophila homologue of meis1 , interactswith extradenticle in patterning of the embryonic PNS. Development, 125:1037–1048, 1998. 90. Smith, S. M. and Eichele, G. Temporal and regional differences in theexpression pattern of distinct retinoic acid receptor-h transcripts in the chickembryo. Development, 111: 245 –252, 1991. 91. Tsang, A. P., Fujiwara, Y., Hom, D. B., and Orkin, S. H. Failure ofmegakaryopoiesis and arrested erythropoiesis in mice lacking the GATA-1transcriptional cofactor FOG. Genes Dev., 12: 1176– 1188, 1998. 92. Koutsourakis, M., Langeveld, A., Patient, R., Beddington, R., and Grosveld,F. The transcription factor GATA6 is essential for early extraembryonicdevelopment. Development, 126: 723 –732, 1999.93. Dollé, P., Ruberte, E., Leroy, P., Morriss-Kay, G., and Chambon, P. Retinoicacid receptors and cellular retinoid binding proteins. I. A systematic study of theirdifferential pattern of transcription during mouse organogenesis. Development,110: 1133– 1151, 1990. 94. Mitsiadis, T. A., Salmivirta, M., Muramatsu, T., Muramatsu, H., Rauvala, H.,Lehtonen, E., Jalkanen, M., and Thesleff, I. Expression of the heparin-bindingcytokines, midkine (MK) and HB-GAM (pleiotrophin) is associated withepithelial-mesenchymal interactions during fetal development and organogenesis.Development, 121: 37–51, 1995. 95. Betsholtz, C. and Raines, E. W. Platelet-derived growth factor: a keyregulator of connective tissue cells in embryogenesis and pathogenesis. KidneyInt., 51: 1361–1369, 1997. 96. Lai, C. S., Fisher, S. E., Hurst, J. A., Vargha-Khadem, F., and Monaco, A. P.A forkhead-domain gene is mutated in a severe speech and language disorder [seecomments]. Nature, 413: 519 –523, 2001. 97. Boylan, J. F. and Gudas, L. J. Overexpression of the cellular retinoic acidbinding protein-I (CRABP-I) results in a reduction in differentiation-specific geneexpression in F9 teratocarcinoma cells. J. Cell Biol., 112: 965–979, 1991.RARh2 Target Genes630 Research.on November 6, 2017. © 2003 American Association for Cancermcr.aacrjournals.orgDownloaded from 2003;1:619-630.Mol Cancer ResYong Zhuang, Teresa N. Faria, Pierre Chambon, et al.Pharmacology Training Grant (CA62948-08) (Y.Z.).Institute for Cancer Research; and in part by a CancerR01 CA43796 to L.J.G.; a research grant from the AmericanNIH grant11Target Genes in F9 Teratocarcinoma Cells2βIdentification and Characterization of Retinoic Acid Receptor Updated versionhttp://mcr.aacrjournals.org/content/1/8/619Access the most recent version of this article at: Cited articleshttp://mcr.aacrjournals.org/content/1/8/619.full.html#ref-list-1This article cites 92 articles, 46 of which you can access for free at: Citing articles/content/1/8/619.full.html#related-urlsThis article has been cited by 11 HighWire-hosted articles. Access the articles at: E-mail alertsrelated to this article or journal.Sign up to receive free email-alerts SubscriptionsReprints [email protected] atTo order reprints of this article or to subscribe to the journal, contact the AACR Publications [email protected] atTo request permission to re-use all or part of this article, contact the AACR Publications Research.on November 6, 2017. © 2003 American Association for Cancermcr.aacrjournals.orgDownloaded from