1196.Full-Text.Pdf

[CANCER RESEARCH 62, 1196–1204, February 15, 2002] retSDR1, a Short-Chain Retinol Dehydrogenase/Reductase, Is Retinoic Acid- inducible and Frequently Deleted in Human Neuroblastoma Cell Lines1

Fabio Cerignoli, Xiaojia Guo, Beatrice Cardinali, Christian Rinaldi, Jessica Casaletto, Luigi Frati, Isabella Screpanti, Lorraine J. Gudas, Alberto Gulino, Carol J. Thiele,3 and Giuseppe Giannini2,3 Department of Experimental Medicine and Pathology, University La Sapienza, 00161 Rome, Italy3 [L. F., I. S., A. G., G. G.]; Department of Experimental Medicine, University of L’Aquila, 67100 L’Aquila, Italy [F. C.]; Neuromed Institute, 86077 Pozzilli, Italy [C. R., L. F.]; Pasteur Institute Cenci-Bolognetti Foundation, 00161 Rome, Italy [I. S.]; Institute of Cell Biology, Consiglio Nazionale delle Ricerche (CNR), 00016 Monterotondo, Italy [B. C.]; Department of Pharmacology, Weill Medical College of Cornell University, New York, New York, 10021 [X. G., L. J. G.]; and Cellular and Molecular Biology Section, National Cancer Institute, NIH, Bethesda, Maryland3 20892 [J. C., C. J. T.]

ABSTRACT enzymes and binding proteins (reviewed in Refs. 2, 3). With the clear exception of the visual cycle, RA appears to be the active metabolite Vitamin A is required for a number of developmental processes and for of retinol for most functions. Free RA does not accumulate in the the homeostatic maintenance of several adult differentiated tissues and cells, and its activity is regulated by at least two different binding organs. In human neuroblastoma (NB) cells as well as some other tumor types, pharmacological doses of retinoids are able to control growth and proteins, CRABPI and -II and by an active metabolism that rapidly induce differentiation in vitro and in vivo. In a search for new genes that converts it into more soluble forms (4–8). RA significantly contrib- are regulated by retinoids and that contribute to the biological effects utes to its own synthesis and degradation through the transcriptional retinoids have on NB cells, we have isolated five differentially expressed regulation of cellular retinol binding proteins, CRABPs, and many transcripts. Here we report on the characterization of one of them enzymes involved in several steps of retinol and RA synthesis and (DD83.1) in NB cell lines. DD83.1 is identical to the human retSDR1, a catabolism (6, 9–16). short chain dehydrogenase/reductase that is thought to regenerate retinol Biologically active vitA metabolites regulate transcription of target from retinal in the visual cycle. Its expression is strongly, but differently, genes through the activation of nuclear receptors of two classes, RARs regulated by retinoids in NB cell lines, and it is widely expressed in human and RXRs, each of which is coded by three different genes (␣, ␤, and tissues, which suggests that it is involved in a more general retinol meta- ␥). A transcriptionally active unit consists of a RXR/RAR heterodimer bolic pathway. Both the retinoic acid-dependent and the exogenous expression of retSDR1 in SK-N-AS cells induce the accumulation of retinyl with required ligand binding on RAR, and occasionally on RXR for a esters, which indicates that it is involved in generating storage forms of more potent genomic response (17–20). A variety of studies have retinol in tissues exposed to physiological retinol concentrations. We also indicated that specific receptor subtypes can serve particular func- show that the human retSDR1 gene, which maps on chromosome 1p36.1, tions. Although there is some redundancy, defective phenotypes, is contained in the DNA fragment deleted in many NB cell lines bearing generated either in cell culture or in animal settings, cannot always be MYCN amplification but is conserved in a cell line with a small 1p deletion fully compensated by the overexpression of other retinoid receptors and normal MYCN. Our observations suggest that retSDR1 is a novel (21, 22). The involvement of RAR␣ in the pathogenesis of promy- regulator of vitamin A metabolism and that its frequent deletion in NB elocytic leukemia has also indicated that spontaneous mutations in the cells bearing MYCN amplification could compromise the sensitivity of retinoid receptors can be associated with cancer development (23). those cells to retinol, thereby contributing to cancer development and progression. More recently, the absence or the reduced expression of specific retinoid receptor isoforms, particularly RAR␤, was found to be relevant for the development and progression of several types of cancer INTRODUCTION (see Refs. 22, 24 and references therein). Natural and synthetic retinoids are being used for the chemoprevention and treatment of many vitA4 or retinol is essential for proper vertebrate embryonic devel- human neoplastic diseases (25). The recent finding that retinoids used opment and growth as well as for the maintenance of several adult in the settings of minimal residual disease could efficiently increase differentiated functions, including vision, fertility, and correct tro- the event-free survival of patients with advanced stage NB (26) phism of the epithelia (1). Dietary-assumed retinol is delivered to the indicates a need for a greater understanding of the sensitivity of these liver, in which more than 90% of the total content of vitA is stored in tumor cells to retinoids. the form of retinyl esters, mostly retinyl palmitate. Retinol is appar- NB is a neural crest-derived tumor. The correct homeostatic control ently the major form of transported vitA and the homeostatic control of vitA metabolism is crucial for proper development and mainte- of circulating and storage retinol is strictly regulated by a plethora of nance of the integrity of neural crest cell structures. Cranial and trunk neural crest derivatives are almost invariably damaged by the terato- Received 2/12/01; accepted 12/28/01. The costs of publication of this article were defrayed in part by the payment of page genic effects of exogenously added retinoids during development (1). charges. This article must therefore be hereby marked advertisement in accordance with Incorrect development of these regions occurs in most double-RA 18 U.S.C. Section 1734 solely to indicate this fact. receptor knockout mice (21, 27), which suggests that either excessive 1 Supported in part by grants from the Associazione Italiana per la Ricerca sul Cancro (AIRC), the National Research Council (CNR), Biotechnology and Oncology Project, the or reduced retinoid signaling are deleterious for proper development Ministry of University, Research and Technology (MURST), the Associazione per la lotta of these regions. More subtle effects of retinoids are also important for al Neuroblastoma (ANB), the MURST-CNR “Biomolecole per la Salute Umana” Pro- gram, the Pasteur Institute Cenci-Bolognetti Foundation, the Italian Ministry of Health, the correct development of the neural crest-derived peripheral nervous and NIH Grant ROI CA77509 (to L. J. G.). system, and the adrenergic cells and enteric neurons are highly re- 2 To whom requests for reprints should be addressed, at Department of Experimental sponsive to retinoids (28, 29). NB cells are among the most sensitive Medicine and Pathology, University La Sapienza, Policlinico Umberto I, Viale Regina Elena, 324, 00161 Rome, Italy. Phone: 39-06-4958637; Fax: 39-06-4461974; E-mail: and frequently used neural crest cells for testing retinoid effects in [email protected]. culture. In fact, many NB cell lines respond to RA with a reduction of 3 These laboratories contributed equally to this work. their proliferation rate and a morphological and biochemical differ- 4 The abbreviations used are: vitA, vitamin A; RA, retinoic acid, ATRA, all-trans-RA; 9CRA, 9-cis-RA; RAR, RA receptor; RXR, retinoid X receptor; CRABP, cellular retinoic acid entiation toward more neuronal phenotypes (30–32). This process is binding protein; NB, neuroblastoma; SRO, shortest region(s) of overlap; IF, immunofluores- partially attributable to the repression of mycN (in MYCN amplified cence; tPA, tissue plasminogen activator; HPLC, high-performance liquid chromatography; hretSDR1, human retSDR1; CHX, cycloheximide; ORF, open reading frame; MoAb, mono- cell lines; Ref. 33), and to the increase in Trk receptors (34, 35) and clonal antibody; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. in the cyclin-dependent kinase inhibitor p27 (36). However, some NB 1196

cells, which can still transduce RA-mediated signaling, are resistant to sucrose, 10 mM Tris-HCl (pH 7.4) and Protease Inhibitor Cocktail (Sigma RA-dependent growth inhibition and differentiation, possibly because Chemical Co.). Nuclei and mitocondria were then removed by a 10-min of a different pattern of expression of HMGI proteins (37, 38). centrifugation at 8,000 ϫ g. The supernatant was further centrifuged at Furthermore, it appears that even in sensitive cells the pathways 105,000 ϫ g for 1.5 h to obtain the microsomal and the cytosolic fraction. For leading to differentiation and growth inhibition can be partially sep- each fraction, an equal amount of proteins normalized on cell number was separated on 13% SDS-PAGE and was analyzed by Western blot using the arated (20, 39). 9E10 anti-myc MoAb and an anti-␣-tubulin MoAb (Oncogene Research Prod- To identify retinoid-responsive genes, whose regulation occurs in ucts). For IF, cells were fixed in 4% formaldehyde, permeabilized with 0.25% association with growth inhibition or differentiation induced by reti- Triton X-100, incubated with either the biotinylated 9E10 antibody or the noids, we applied the RNA fingerprinting methodology to KCNR anti-retSDR1 mouse MoAb followed by an incubation with an antimouse cells treated with different combinations of receptor-selective reti- biotinylated antibody (DAKO Corporation, Carpinteria, CA). Both were re- noids. This analysis resulted in five differentially displayed PCR vealed with streptavidin-FITC (DAKO Corporation). amplicons. In this article, we report on one of them that is identical to retSDR1 overexpressing stable clones were obtained by transfecting SK- the hretSDR1, a short chain dehydrogenase/reductase, that can catal- N-AS cells with either the retSDR1-Myc or the control constructs followed by ize the reduction of retinal in the visual cycle (40). The expression of G418 selection. After 10–20 days, individual clones (including CTR#2 and this gene is strongly, but differently, induced by retinoids in all of the SDR#62) were picked and stably grown under selection. In addition, to minimize the possibility of misinterpretation caused by clonal variability we NB cells tested and is widely expressed among human tissues, which have also used transfected SK-N-AS cells, which, after the proper selection suggests its involvement in a more general retinol metabolic pathway. time were not individually cloned, but established as heterogeneous mixed Consistently, retSDR1 overexpression in SK-N-AS cells strongly nonclonal continuous cell lines (CTRpool and SDRpool). induced the formation retinyl esters. We also demonstrated that the Retinoid Metabolism. Retinol metabolism was measured using [3H]retinol human retSDR1 gene, which maps on chromosome 1p36.1 (40), is by reverse-phase HPLC according to previously described methods (45, 46). contained in the DNA fragment deleted in many NB cell lines bearing The metabolism of nonradiolabeled retinol and retinaldehyde was measured in MYCN amplification, but is conserved in a cell line with a small CTR#3 and SDR#62 clones by reverse phase HPLC followed by photodiode deletion and single copy MYCN. These observations suggest that array detection (45, 46). All of the studies were performed at least twice with retSDR1 might be involved in a general retinol metabolic pathway multiple time points. Appropriate standards were run to assist in the identifi- and that its partial absence in 1p36-deleted NBs might reduce their cation of the endogenous retinoids. Both the cells and the medium were analyzed for retinol content at various times. sensitivity to retinol in vivo.

MATERIALS AND METHODS RESULTS

Cell Culture and Stimulation Experiments. NB cell lines were cultured RNA Fingerprinting of KCNR NB Cells Stimulated with Re- as described previously (20). Stably transfected NB cells (see below) were ceptor-selective Retinoids. We have previously shown that only cultured in the presence of G418 (500 ␮g/ml, Sigma Chemical Co., St. Louis, SR11383, among a group of RAR-subtype selective retinoids, induces MO). ATRA, retinol (Sigma) and the synthetic retinoids were dissolved in growth inhibition of KCNR NB cells to an extent similar to ATRA DMSO and were added to the cells 1 day after seeding at 5 ␮M, unless (20). However, different RAR-subtype-selective retinoids with an specified. RXR-selective retinoid are capable of activating a full-growth inhib- Nucleic Acid Isolation, Northern Blot, and Southern Blot Analysis. itory response in the same cells (20). The RNA fingerprinting per- After the appropriate treatment, cells were washed twice with cold PBS, and formed on KCNR cells treated with different combinations of the total RNA was extracted using the RNeasy system (Qiagen, Hilden, Germany), same retinoids (Fig. 1A) further supported these observations and according to the manufacturer’s instructions. Northern analysis of total RNA (20 ␮g) and Southern analysis of genomic DNA (20 ␮g) were performed as enabled the identification of genes whose expression is modulated by described previously (20). Multiple tissue Northern blots (Clontech, Palo Alto, retinoids in KCNR cells. In fact, through this methodology, we have CA) contained 2 ␮g of polyadenylated RNA from various adult and fetal examined ϳ240 genes expressed in KCNR cells and identified 5 tissues. The DD83.1 fragment or the PCR-cloned (see below) complete coding of them whose expression is regulated by retinoids. Oligo(83), sequence of hretSDR1, were labeled with [␣-32P]dCTP using a random primer oligo(122), and oligo(133) generated amplicons differentially ex- labeling kit (Amersham Pharmacia Biotech, Little Chalfont, United Kingdom) pressed in our assay (Fig. 1A). Band DD83.1, which was clearly according to the manufacturer’s directions and used as cDNA probes for absent in the reaction performed on control cells and on cells treated Northern hybridization. Blots were hybridized [42°C in 50% (v/v) formamide, with the RXR-selective SR11246 and the RAR␥-selective SR11254, ␮ 1 M NaCl, 10% dextran sulfate, 1% SDS, 100 g/ml salmon sperm DNA] and appeared as an additional displayed band in the reaction obtained by washed (0.2ϫ SSC, 0.1% SDS at 60°C–65°C) at high stringency. cells treated with other retinoids and/or retinoid combinations RNA Fingerprinting. For the RNA fingerprinting we followed the proto- col published by Malgaretti et al. (41) and readapted as described previously (Fig. 1A). Basal levels of band DD122.2 were detected in controls, (42–44). Primers used were as follows: 83, 5Ј-CGTGGGCAACCT-3Ј; 122, and it was induced to comparable levels by ATRA, 9CRA, 5Ј-CATGGCTGCCAG-3Ј; 133, 5Ј-CAGTCCTGGCCA-3Ј. Cloned DNA in- SR11383ϩSR11246, Am580ϩSR11246, and SR11254ϩSR11246, serts were sequenced by using an ABI PRISM 377 DNA Sequencer and the and to a lesser extent by the single retinoids. Band DD133.1, which DNA Sequencing kit-Big Dye terminator (PE-Applied Biosystems, War- was absent in the control, SR11246-, and SR11254ϩSR11246-treated rington, Great Britain). The differential regulation of the isolated genes by the cells, was highly represented in 9CRA-treated cells, but was also retinoids was confirmed by semiquantitative reverse transcription (RT)-PCR induced by SR11383ϩSR11246 and Am580ϩSR11246 (Fig. 1A). analysis and/or by Northern blots. The intensity of surrounding bands was used to assess the relative hretSDR1 ORF Cloning, Transfection, Cell Fractionation, and IF. A loading of the lanes and allowed us to consider the strong signal PCR-amplified fragment containing hretSDR1 ORF was cloned in frame with observed in SR11254-treated cells as an artifact caused by overload- a Myc-tag (retSDR1-Myc) in the pcDNA3.1/Myc-His B (Invitrogen, San Diego, CA). retSDR1-Myc was transfected in SK-N-SH neuroblastoma cells, ing of the lane. Band 122.4 was also positively regulated by ATRA, by liposomal transfer using the TransFast reagent (Promega Corporation, 9CRA, and all effective retinoid combinations (Fig. 1A). Interestingly Madison WI), according to the manufacturer’s instructions. The following day, DD122.3, which was not influenced by ATRA, 9CRA, or single cells were processed for cell fractionation or IF experiments. For subcellular retinoid application, was reduced only by the combination of fractionation, cells were lysed in 1 ml of iso-osmotic buffer containing 0.25 M SR11383 and SR11246 (Fig. 1A), which is the retinoid combination 1197

Fig. 1. RNA fingerprinting of retinoid-stimulated KCNR cells and identification of differentially expressed transcripts. A, several differentially displayed bands were identified by the RNA fingerprinting technology from KCNR cells treated with single retinoid and retinoid combinations. B, example of the screening procedure. The hybridization of 10 different DNA preparations coming from colonies containing band DD83.1 with either the control fingerprinting reaction (Probe CTR) or an ATRA-stimulated fingerprinting reaction (Probe ATRA) enabled the identification of colonies 6, 7, and 8 as containing the correct fragment; all of the other cloned fragments are false positives (contaminants). In C, the differential expression of DD83.1, DD122.2, and DD122.3 was confirmed by a reverse transcription-PCR procedure. Total RNA extracted from control- or ATRA-treated cells (or, for DD122.3, cells treated with SR11383ϩSR11246) was retrotranscribed and subjected to semiquantitative PCR amplification. Ribosomal S12 protein transcript was coamplified as an internal control. DD83.1 amplicon is clearly generated after 30 PCR cycles in treated but not in control samples; DD122.2 amplicon is clearly generated after 30 PCR cycles in treated but not in control samples, and its amplification reached a plateau after 35 cycles; DD122.3 amplicon is clearly generated after 35 PCR cycles in control but not in treated samples. In D, the differential expression of DD83.1 in KCNR cells treated with different retinoids and retinoid combinations was further confirmed by Northern blot. GAPDH hybridization to the same blot was used as a loading control. Retinoid selectivity was as follows: Am580, RAR␣; SR11383, RAR␤; SR11254, RAR␥; SR11246 and SR11234, RXR (20). All of the retinoids were used at 0.1 ␮M concentration. that most effectively controls growth and induces differentiation in ibility by retinoids in KCNR cells by Northern blot. As shown in Fig. KCNR cells (20). 1D, this analysis confirmed that retSDR1 is undetectable in untreated According to the method of Consalez et al. (44) to successfully KCNR cells and is induced by ATRA, 9CRA, and all of the retinoid clone the differentially displayed bands, after excision out of the gel, combinations effective in the RNA fingerprinting after 4 days of PCR reamplification and ligation into the pCRII-vector, 10 DNA treatment. Analysis of the retSDR1 mRNA expression in a panel of preparations from each cloned amplicon were blotted on two identical NB cell lines revealed the existence of several transcripts, the expres- membranes and hybridized to the control- and retinoid-induced RNA sion of which was differentially regulated by ATRA (Fig. 2). In fingerprinting reactions previously labeled through random priming particular, after 3 h, we detected a very strong RA-induced increase in (Fig. 1B). Through this method we could clearly distinguish the three different transcripts (1.4-, 1.8-, and 3.5-kb) in SK-N-SH cells differentially displayed cloned inserts (that hybridized only to retin- that persisted even after 3 days of ATRA treatment. A similar situa- oid-induced RNA fingerprinting reactions; Fig. 1B, Lanes 6, 7, and 8) tion occurred in SK-N-AS cells, which lacked the 1.8-kb transcript. In from cloning artifacts (that hybridized to both control- and retinoid- contrast, ATRA induced the 1.8-kb transcript in KCNR cells (Fig. 2A) induced RNA fingerprinting reaction; Fig. 1B, Lanes 1, 2, 3, 4, 5, 9, and in LAN-5 (not shown) at the latest time point and, compared with and 10). The differential expression of DD83.1, DD122.2, and SK-N-SH and SK-N-AS, transcript accumulation reached consistently DD122.3 was further confirmed by a semiquantitative reverse tran- lower levels. Finally, in SK-N-BE cells, we could detect only a slight scription-PCR (Fig. 1C). Sequencing of the cloned amplicons and increase of the 1.4-kb transcript after 3 days of ATRA treatment (Fig. their comparison to nonredundant sequence databanks revealed that 2A). A more detailed time course analysis verified that retSDR1 DD122.2 was a partial clone for the human SH3-binding protein 2, induction by ATRA required a longer time in KCNR cells, in which which maps to chromosome 4p16.3 close to the Huntington’s disease it first appeared after 12 h and reached a plateau between 72 and 96 h; region, whereas DD122.3 was a partial sequence for the human and in both KCNR and SY5Y cells, its expression persisted through- mitochondrial dicarboxylate carrier protein. DD83.1, which demon- out 8 days of ATRA treatment (not shown). Interestingly, protein ␤ strated the typical signature of a dehydrogenase, was identical to the kinase A activation, vitamin D3, dexamethasone, and TGF could not recently reported hretSDR1 mRNA. The other two sequences did not induce retSDR1 mRNA in several NB cell lines (not shown), which have recognizable homologues in nonredundant databases. suggested that the retSDR1 increase is a specific effect of RA. DD83.1/retSDR1 Expression in NB Cells. DD83.1 is a partial In SK-N-AS cells, in which ATRA induced the time-dependent clone spanning from base 289 to 842 of the deposited hretSDR1 accumulation of the 1.4-kb and 3.5-kb transcripts, CHX, a known sequence, a region that includes a large part of its ORF. This fragment inhibitor of protein synthesis, did not induce retSDR1 (Fig. 2B). In and a full coding sequence fragment were used to confirm its induc- addition, it did not affect ATRA induction of retSDR1 in the first 3 h 1198

ported. We detected high levels of expression in fetal kidney, liver, and lung and in adult heart, placenta, lung, liver, kidney, pancreas, thyroid, testis, stomach, trachea, and spinal cord. We also found lower levels in skeletal muscle, intestine, and lymph node. retSDR1 was barely detectable in adrenals, brain, thymus, and hematopoietic tissues. Two different retSDR1 cDNA probes recognized at least four different transcripts of 3.5-, 2.5-, 1.8-, 1.4-kb, and most tissues expressed more than one specie of retSDR1 mRNA (Fig. 4). In heart and fetal kidney, all forms were present, whereas in adult stomach, thyroid, and spinal cord, only the 2.5-, 1.8-, and 1.4-kb transcripts were detected. Interestingly, testis expressed only one form, which was smaller than 1.4 kb.

Fig. 2. The expression of DD83.1/retSDR1 is regulated by ATRA in different NB cell lines and does not require protein synthesis. Northern blot analysis of total RNAs (20 ␮g) extracted from various NB cell lines treated with RA for different times shows that DD83.1 probe recognizes at least three different transcripts of the indicated lengths (in kb, numbers on the sides). In A, their induction by ATRA is delayed in KCNR and SK-N-BE cells compared with SK-N-SH and SK-N-AS cells. In B, the treatment with the protein synthesis inhibitor CHX (10 ␮g/ml) did not prevent early induction of retSDR1 expression in SK-N-AS cells. However, CHX favors the accumulation of the 1.4-kb isoform and blocks the accumulation of the 3.5-mRNA species at later time points. GAPDH hybridization to the same blots was used as a loading control. of treatment, which suggested a direct regulation by ATRA. However, although it was still completely ineffective by itself, CHX enhanced ATRA-induced accumulation of the 1.4-kb transcript as early as after Fig. 3. Deletion of retSDR1 in NB cell lines. Approximately 20 ␮g of genomic DNA 6 h, thus anticipating the plateau levels otherwise obtained after 24 h extracted from six NB cell lines were EcoRI digested, separated on an agarose gel, transferred onto nylon membrane, and hybridized with the 32P-labeled DD83.1 fragment. of ATRA stimulation (Fig. 2B). Conversely, CHX inhibited the The DNA content of each lane was normalized to the human HMGI-C gene (which is ATRA-induced accumulation of the 3.5-kb specie at 6, 12, and 24 h diploid in these cell lines), and ploidy was calculated after normalization. Notice the diploid content in the retSDR1 gene in SK-N-SH and the derived subclone SY-5Y cells (Fig. 2B). Furthermore, in SK-N-SH cells, CHX did not affect the (which bear no 1p deletion) and SK-N-AS cells (which bears a small 1p interstitial ability of ATRA to induce any retSDR1 transcript at either 6 or 12 h, deletion) and the haploid content in SK-N-BE, KCNR, and LAN-5, which are MYCN whereas doxorubicin, an RNA polymerase inhibitor, abolished the amplified and bear monoallelic large 1p deletions. ATRA-induced expression of both transcripts (not shown), which suggested that at least the 1.4-kb transcript is directly regulated by RA in both SK-N-AS and SK-N-SH cells. retSDR1 Gene Maps within the Region Frequently Deleted in Human NBs. The retSDR1 gene was mapped on human chromosome 1p36.1 (40), an area frequently lost in aggressive NB tumors, which suggests that retSDR1 might cosegregate with the still undiscovered NB tumor suppressor gene(s). In particular, at least two different SRO have been identified in the 1p region. The largest SRO, which would include the 1p36.1, is most frequently deleted in MYCN-amplified tumors (47–49). To test the hypothesis that retSDR1 is deleted in NB tumors, we have analyzed the status of the retSDR1 gene in several NB cell lines, by Southern blot (Fig. 3). We have found that three MYCN-amplified cell lines (LAN-5, KCNR, and SK-N-BE cells) that have a monoallelic deletion in the 1p region (47) also have a haploid content of the retSDR1 gene compared with undeleted cells (SK- N-SH and SY5Y) or compared with a cell line that is known to have a small interstitial deletion (SK-N-AS) not extending to the 1p36.1 band (Fig. 3). Therefore, these results confirm that retSDR1 cosegre- gates with a NB tumor suppressor/deletion region. They also indicate that the delayed and reduced expression that we observed in KCNR, SK-N-BE (Fig. 2), and LAN-5 (not shown) is associated with a retSDR1 haploid DNA content (Fig. 3). DD83.1/retSDR1 Expression in Human Tissues. We assayed the Fig. 4. retSDR1 expression in human tissues. Polyadenylated RNA (2 ␮g) from various expression of retSDR1 in different fetal and adult human tissues (Fig. adult (A, B, C) and fetal (D) human tissues were hybridized with a 32P-probe. Arrows, molecular size markers. At least four different mRNA species can be identified (arrow- 4). Although the retSDR1 cDNA was initially isolated from retina heads, 1.4-, 1.8-, 2.3-, and 3.5-kb) all of which are expressed in heart and fetal kidney. In (40), its expression appears to be less restricted than originally re- A, B, C, and D, GAPDH hybridization to the same blot was used as a loading control. 1199

compartment and was assembled in patched structures (Fig. 5B, a). To further address the issue of retSDR1 localization and to confirm that its expression is induced by ATRA in NB cells, we performed IF studies with a MoAb developed against the human protein and able to recognize overexpressed hretSDR1 (Ref. 40; Fig. 5B, b). The endogenous retSDR1 could be detected by a three-step IF both in control- and in ATRA-treated SK-N-SH cells (Fig. 5B, c and d). The endogenous protein showed an intracellular distribution largely overlapping that of the exogenous one, with localization in patched structures with a strict perinuclear distribution in untreated cells. Under ATRA stimulation, a remarkable enlargement of the immunoreactive regions were observed that were extended through the entire cytoplasm, confirming that ATRA induces a higher content of protein per cell. Retinol Metabolism and Genomic Responses in Cells Express- ing retSDR1. Insect cell extracts expressing retSDR1 could efficiently reduce retinal into retinol in vitro in the presence of [3H]NADPH (40). To assess retinol metabolism in NB cells physio- logically expressing retSDR1, we exposed SK-N-AS cells, pretreated with ATRA for 3 days to induce endogenous retSDR1, and control- ␮ 3 Fig. 5. Expression pattern of exogenously expressed and endogenous retSDR1 protein. treated SK-N-AS cells to 1 M [ H]retinol and harvested cells after In A, the cellular localization of the exogenously expressed and Myc-tagged retSDR1 5–10 h. We found that there was an almost 2-fold increase in the protein was revealed by cellular fractionation followed by Western blotting of the protein accumulation of retinyl esters in the ATRA-pretreated SK-N-AS cells, extracts normalized on cell number and probed with an anti-myc MoAb. Whereas the retSDR1 protein is largely enriched in the microsomal fraction (micro), the ␣-tubulin which turn on the expression of the endogenous retSDR1, compared content is much higher in the cytosolic fraction (Cyto). In B, IF studies on SK-N-AS cells with the control (not shown). We could not detect any other [3H]reti- expressing the Myc-tagged retSDR1 protein showed a perinuclear distribution of the 3 signal using either an anti-myc MoAb (a) or an anti-hretSDR1 MoAb (b). Using the nol metabolites, such as [ H]RA, either in the medium or in the cell anti-hretSDR1 MoAb, we also investigated the endogenous content in retSDR1 protein in extracts (not shown). SK-N-SH cells either before (c) or after (d) treatment with ATRA, and it is consistent with To test whether expression of retSDR1 could change the biological the localization in a membranous cytoplasmic compartment with a typical perinuclear ring distribution in control cells that is far more extended within the cytoplasm in ATRA- responses of NB cells to retinoids, we isolated individual clones or treated cells. heterogeneous mixed nonclonal continuous cell lines (defined pools) of SK-N-AS cells stably expressing retSDR1. Both retSDR1- and Intracellular Localization of retSDR1 Gene Product. To ascer- mock-transfected cells could activate the expression of several genes tain whether the retSDR1 protein is cytosolic or membrane bound in (including the endogenous retSDR1, tPA, RAR␤, and HMGI(Y))in NB cells, we first performed cell fractionation experiments to deter- response to RA treatment (Fig. 6A), thus suggesting that retSDR1 does mine the localization pattern of the exogenously expressed retSDR1/ not significantly modify responsiveness to exogenous RA. Therefore, Myc-tagged protein. By Western blot analysis, we observed that we tested the effect of retSDR1 overexpression on retinol activities. overexpressed protein was strongly enriched in the microsomal frac- Exogenous retinol could efficiently induce the expression of specific tion, and a lower amount was also detected in the cytosolic fraction genes between 0.1- and 1-␮M concentrations in parental SK-N-AS (Fig. 5A). These data are consistent with the in situ pattern of local- cells (not shown). A mock-transfected pool of SK-N-AS (CTRpool) ization of the exogenous retSDR1/Myc-tagged protein detected by IF and one of the selected control clones (CTR2) readily showed an experiments, which mainly localized to a perinuclear membranous increased expression of the endogenous 3.5-kb retSDR1 transcript,

Fig. 6. Altered genomic response to retinol in retSDR1- transfected SK-N-AS cells. In A, Northern analysis of 20 ␮g of total RNA for the indicated samples shows that the expression of the endogenous 3.5-kb transcript for retSDR1 and tPA, RAR␤, and HMGI(Y) mRNAs is normally regulated by RA in retSDR1 (SDR#62) and mock transfected (CTR#2) cells. In B, treatment with 1 or 0.1 ␮M retinol is capable of inducing the expression of the endogenous 3.5-kb retSDR1 transcript, and tPA and RAR␤ transcripts in control cells (CTRpool, CTR#2). In ectopically retSDR1-expressing cells (SDRpool, SDR#62), the same genes are less efficiently regulated by 1 ␮M retinol (ROL) and almost unaffected by 0.1 ␮M retinol. GAPDH hybridization to the same blots was used as a loading control. ExoSDR1 indicates the hybridization of the exogenous retSDR1 transgene in the retSDR1- transfected cell lines. The transgene comigrates with the endogenous 1.4-kb-retSDR1 mRNA, only distinguishable in control-transfected cell lines. In C, the relative expression of retSDR1, tPA, and RAR␤ under either 0.1 or 1 ␮M retinol was normalized for GAPDH expression after densitometric analysis of the films. Values on the Y-axis are reported as fold induction compared with the control levels for each cell type. f, CTRpool; F, CTR#2; Ⅺ, SDRpool; E, SDR#62.

1200

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2002 American Association for Cancer Research. retSDR1 AND RETINOL METABOLISM IN NB CELLS tPA, and RAR␤ transcripts when treated with either 0.1 or 1 ␮M retinol (Fig. 6, B and C). In contrast, the retSDR1-transfected pool of SK-N-AS (SDRpool) showed a lower induction of endogenous retSDR1, tPA, and RAR␤ compared with controls at 1 ␮M retinol. In the stably transfected and highly expressing line SDR#62, RAR␤ mRNA was detected at low levels both in basal and in the retinol- induced conditions, and very little induction of the endogenous retSDR1 mRNA was seen even in response to 1 ␮M retinol. Moreover, all of the retSDR1-transfected cells were almost insensitive to 0.1 ␮M retinol (Fig. 6C), which suggests that overexpression of retSDR1 might reduce the amount of retinol available for direct or indirect transcriptional activation, possibly as a consequence of retinol metabolism toward retinyl esters, as we previously observed in ATRA- pretreated SK-N-AS cells. Because it is known that retinoids induce a number of gene changes in addition to retSDR1 in NB cells, we decided to evaluate how retSDR1 expression specifically affects retinoid metabolism in retSDR1-transfected SK-N-AS NB cells. Mock-transfected cells (CTR#3; Fig. 7, A, B, and C) and SDR#62 cells (Fig. 7, D, E, and F) were treated with control solvent, (Fig. 7, A and D)1␮M retynaldehyde (Fig. 7, B and E), or 1 ␮M retinol (Fig. 7, C and F) for 24 h, harvested and analyzed for retinoid metabolism by reverse-phase HPLC and a photodiode array detector. Under basal conditions (control solvent; Fig. 7, A and D), we detected no retinoids peak in the array profiles of either the medium (not shown) or the extracts of the CTR#3 and SDR#62 lines. Incubation with 1 ␮M retynaldehyde for 24 h resulted in no detectable increase in retinoic acid (Fig. 7, B and E, at 21 min) and in a similar increase in retinol in either CTR#3 or SDR#62 (Fig. 7, B and E, at 30.5 min), whereas we detected a 5-fold increase in retinyl esters only in SDR#62 (Fig. 7E,at50–58 min). After incubation with 1 ␮M retinol for 24 h, we could not detect any RA production in the medium (not shown) or extracted from the cells from either CTR#3 or SDR#62 lines; in contrast, we observed a 20-fold increase in the amount of retinyl esters only in clones SDR#62 (Fig. 7F,at50–58 min) and SDR#44 (not shown), but not in the CTR#3 line (Fig. 7C) and CTRpool (not shown). Overall, our data indicate that either endogenous or exogenously expressed retSDR1 favors the metabolism of retinol and retinal to storage forms of vitA. Fig. 7. Retinol metabolism in retSDR1-transfected SK-N-AS cells. CTR#3 (A, B, C) and SDR#62 (D, E, F) cell lines were cultured in the presence of control solvent (A, D), in 1 ␮M retinaldehyde for 24 h (B, E)orin1␮M retinol for 24 h (C, F). Approximately DISCUSSION 60% of the retinol (ϳ600 nM) remained in the medium for both CTR#3 and SDR#62 clones after 24 h. Photodiode array profiles of the various retinoids are depicted. Only the retSDR1 Is a Broadly Expressed and RA-inducible Regulator intracellular retinoid profiles are shown. Standards (not shown) eluted: at 21 min, of Retinol Metabolism. Apart from its special role in the production all-trans-retinoic acid; at 30.5 min, all-trans retinol; at 34 min, all-trans-retinaldehyde; at 50–58 min, retinyl esters. of visual pigments, retinol is required for many developmental processes and for the homeostatic maintenance of several adult differentiated tissues and organs like epidermis and testis (1). Because of the hypothesis. Our data also indicate that whether retSDR1 is physio- major role of vitA during mammalian development and adult life, a logically expressed (induced by RA) or selectively overexpressed in plethora of different proteins have developed for the tight control of the (retSDR1-transfected) SK-N-AS NB cells, retinyl esters accumu- intake, transport, storage, and metabolism of vitA and its derivatives late after treatment with retinaldehyde or retinol. This clearly differs (2, 3). In the search for new dehydrogenases, Haeseleer et al. (40) from the observation that retSDR1 expressed in insect cells causes the have recently identified retSDR1 as a short chain dehydrogenase/ reduction of retinal to retinol in the presence of NADPH in vitro (40), reductase expressed in the cone outer segment in the retina, and have and more work will be required to better clarify this issue. Our future shown that it can catalyze the reduction of all-trans-retinal to all- studies will also include a detailed examination of retinoid metabo- trans-retinol, using NADPH as a cofactor in vitro (40). In our screen lism in NB cells, because none has been performed to date. We have for novel genes regulated by retinoids that contribute to the biological reported that a number of human cancer cell lines, including squa- effects of retinoids on NB cells, we isolated a cDNA corresponding to mous cell carcinoma and breast cancer cell lines cultured in the the hretSDR1. RetSDR1 mRNA and protein are not exclusively presence of [3H]retinol, are unable to accumulate retinyl esters in expressed in the retina. On the contrary, we have detected its expres- response to exogenous retinol (45, 46). Interestingly, we detected sion in a variety of human tissues and the existence of EST sequences constitutive retSDR1 expression in breast cancer cell lines5 that were from a large variety of tissue sources, which suggests a more general capable of retinol esterification (e.g., ZR-75–1 and MDA-MB-468), role for this enzyme in vitA metabolism. Its relatively high expression but not in breast cancer cell lines reported to be incapable of forming in liver and other tissues that actively metabolize retinol (pancreas, lung, kidney, testis), both in the adult and in the fetus, supports this 5 F. Cerignoli and G. Giannini, unpublished observations. 1201

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2002 American Association for Cancer Research. retSDR1 AND RETINOL METABOLISM IN NB CELLS retinyl esters (45, 46). This suggests that there is a correlation between ing the 1p-deleted tumors with another growth advantage. It is tempt- retinol esterification and retSDR1 expression in cancer cell lines. ing to speculate that the reduced content of several gene products The strong induction of retSDR1 by RA is common to all NB cell involved in growth control and survival (including p73, RIZ, TNFR2, lines that we have tested and also to a variety of other nonneuroblastic ID3, DAN, E2F2) caused by large 1p chromosomal deletions might human cell lines.6 This induction, therefore, can be interpreted as the add further growth or survival advantages to neoplastic cells. The attempt of cells to accumulate local retinol storage in the case of monoallelic deletion of retSDR1 that we observed in several NB cell retinol availability. RetSDR1 induction by vitA derivatives appears lines is associated with the reduced and delayed accumulation of the highly specific because it was not evoked by either protein kinase A corresponding mRNAs in response to RA. Thus, it could provide a ␤ 6 activation or treatment with vitamin D3, dexamethasone, or TGF . selective growth and survival advantage by altering the metabolic Our findings complement several observations indicating that many pathways that control the local production of biologically active vitA cells possess a RA-inducible retinol metabolic enzyme system that metabolites. In support of this hypothesis, we have shown that can generate either transcriptionally active retinol metabolites or stor- retSDR1 is an enzyme involved in the generation of storage forms of age forms of vitA (46, 50–52). Consistent with this, RA transcrip- retinol that may be important cell growth and differentiation regula- tionally or posttranscriptionally regulates the expression of a number tors. The possibility that retSDR1 might be involved in a growth/ of proteins relevant to its own metabolism and activity, including tumor suppressive pathway, perhaps through the further inactivation cellular retinol-binding proteins, CRABPs, LRAT, RALDHs, RARs, of the single allele remaining in many 1p-deleted human cancers, and now retSDR1 (6, 9–16). RA also induces its own metabolism in should also be considered. embryonal carcinoma and in some neoplastic cells in culture (5, 6, 46, In conclusion, we have provided evidence that retSDR1 is a novel 53, 54). The biochemical reactions through which retSDR1 promotes regulator of vitA metabolism involved in the production of a local retinyl ester accumulation are not well characterized and will be the storage form of retinol, retinyl esters in NB cells. RetSDR1 is induced subject of future studies. by RA in a wide array of cell lines derived from different human Haploinsufficiency at retSDR1 Locus Might Contribute to Can- tissues, and it is frequently deleted in MYCN-amplified NB cell lines. cer. Quantitative or qualitative alterations in proteins involved in vitA It is possible that its deletion in NB cells and in a number of other metabolism and signal transduction are involved in cancer develop- human tumors might compromise their capability to form local retinyl ment. Spontaneous mutations in the retinoid receptors can be associ- esters for retinol storage. In the absence of local stores, and particu- ated with cancer or even involved in its molecular pathogenesis, as in larly under the low concentrations of circulating retinol that can be the case of RAR␣ and promyelocytic leukemia (23). An absence or found in NB and other cancer patients (65), the production of vitA active metabolites would be blunted. This might impair an important reduced expression of RAR␤ was observed in several types of cancer growth-inhibitory pathway and thus contribute to cancer development (see Refs. 22, 24 and references therein), and its presence seems to be and progression. required for growth inhibition and differentiation (22). vitA and some of its derivatives are being actively and successfully used in cancer chemoprevention and therapy (25). ACKNOWLEDGMENTS The retSDR1 gene, which encodes an enzyme widely expressed among human tissues and is involved in retinol metabolism, is located We thank Dr. Jhon C. Saari (University of Washington School of Medicine, on chromosome 1p36.1 (40), a region very frequently rearranged in Seattle, WA) for the anti-hretSDR1 MoAb and Dr. Oreste Segatto (Istituto human cancer. Heterogeneous deletions and translocations of chro- Regina Elena, Rome, Italy) for the biotinylated anti-myc MoAb. We also thank P. Harley (National Cancer Institute), and M. Zani (University La Sapienza) mosome 1p were described in different neoplasias, but they appear for excellent technical assistance. with maximum frequency in NB, melanoma, and MEN2A-associated neoplasia (55). Extensive deletions of 1p are associated with advanced stage and/or a more aggressive neoplasia and poor prognosis, sug- REFERENCES gesting the presence of gene(s) the deletion of which may contribute 1. Morriss-Kay, G. M., and Ward, S. J. Retinoids and mammalian development. Int. to tumor progression (56–58). In NBs, the extent of 1p deletions is Rev. Cytol., 188: 73–131, 1999. extremely heterogeneous, and at least two potential NB suppressor 2. Napoli, J. L. Interactions of retinoid binding proteins and enzymes in retinoid metabolism. Biochim. Biophys Acta, 1440: 139–162, 1999. loci are present in two different areas (59–61). Large chromosomal 3. Duester, G. Function of alcohol dehydrogenase and aldehyde dehydrogenase gene deletions from 1p35–pter are observed in most MYCN-amplified families in retinoid signaling. Adv. Exp. Med. Biol., 463: 311–319, 1999. 4. Donovan, M., Olofsson, B., Gustafson, A. L., Dencker, L., and Eriksson, U. The tumors, associated with more aggressive behavior and advanced stage, cellular retinoic acid binding proteins. J. Steroid. Biochem. Mol. Biol., 53: 459–465, whereas smaller interstitial deletions involving 1p36.1–1p36.3 are 1995. mainly detected in single copy-MYCN tumors (47–49, 57, 62). 5. Williams, J. B., and Napoli, J. L. Metabolism of retinoic acid and retinol during differentiation of F9 embryonal carcinoma cells. Proc. Natl. Acad. Sci. USA, 82: Rather than the biallelic deletion of one gene, haploinsufficiency at 4658–4662, 1985. several loci leading to a dosage defect involving multiple gene prod- 6. Chen, A. C., and Gudas, L. J. An analysis of retinoic acid-induced gene expression and metabolism in AB1 embryonic stem cells. J. Biol. Chem., 271: 14971–14980, ucts is the most common outcome of the microsatellite instability 1996. observed in gastrointestinal cancer (63). Although several candidates 7. Roberts, A. B., Frolik, C. A., Nichols, M. D., and Sporn, M. B. Retinoid-dependent have been cloned out of the 1p SRO, there is no conclusive evidence induction of the in vivo and in vitro metabolism of retinoic acid in tissues of the vitamin A-deficient hamster. J. Biol. Chem., 254: 6303–6309, 1979. that any of them is the bona fide NB suppressor gene. It is possible 8. van der Leede, B. M., van den Brink, C. E., Pijnappel, W. W., Sonneveld, E., van der that the sum of mono- or biallelic deletions at many loci within the Saag, P. T., and van der Burg, B. Autoinduction of retinoic acid metabolism to polar derivatives with decreased biological activity in retinoic acid-sensitive, but not in large 1p deletion might provide a growth or survival advantage to retinoic acid-resistant human breast cancer cells. J. Biol. Chem., 272: 17921–17928, preneoplastic cells and, therefore, contribute to tumor development 1997. and/or progression. For example, the deletion of the MEMO1 gene, 9. Haq, R., and Chytil, F. Retinoic acid rapidly induces lung cellular retinol-binding protein mRNA levels in retinol deficient rats. Biochem. Biophys. Res. Commun., which maps at 1p35–36.1, is associated with and might be involved in 156: 712–716, 1988. controlling MYCN amplification and/or expression (64), thus provid- 10. Randolph, R. K., and Ross, A. C. Vitamin A status regulates hepatic lecithin: retinol acyltransferase activity in rats. J. Biol. Chem., 266: 16453–16457, 1991. 11. Matsuura, T., Gad, M. Z., Harrison, E. H., and Ross, A. C. Lecithin:retinol acyltrans- 6 G. Giannini and C. J. Thiele, unpublished observations. ferase and retinyl ester hydrolase activities are differentially regulated by retinoids 1202

and have distinct distributions between hepatocyte and nonparenchymal cell fractions Gulino, A. HMGI(Y), and HMGI-C genes are expressed in neuroblastoma cell of rat liver. J. Nutr., 127: 218–224, 1997. lines and tumors and affect retinoic acid responsiveness. Cancer Res., 59: 2484– 12. Kurlandsky, S. B., Duell, E. A., Kang, S., Voorhees, J. J., and Fisher, G. J. Auto- 2492, 1999. regulation of retinoic acid biosynthesis through regulation of retinol esterification in 38. Giannini, G., Kim, C. J., Marcotullio, L. D., Manfioletti, G., Cardinali, B., Cerignoli, human keratinocytes. J. Biol. Chem., 271: 15346–15352, 1996. F., Ristori, E., Zani, M., Frati, L., Screpanti, I., and Gulino, A. Expression of the 13. White, J. A., Guo, Y. D., Baetz, K., Beckett-Jones, B., Bonasoro, J., Hsu, K. E., HMGI(Y) gene products in human neuroblastic tumours correlates with differentiation Dilworth, F. J., Jones, G., and Petkovich, M. Identification of the retinoic acid- status. Br. J. Cancer., 83: 1503–1509, 2000. inducible all-trans-retinoic acid 4-hydroxylase. J. Biol. Chem., 271: 29922–29927, 39. Matsumoto, K., Wada, R. K., Yamashiro, J. M., Kaplan, D. R., and Thiele, C. J. TrkB 1996. Expression of brain-derived neurotrophic factor and p145 affects survival, dif- 14. White, J. A., Beckett-Jones, B., Guo, Y. D., Dilworth, F. J., Bonasoro, J., Jones, G., ferentiation, and invasiveness of human neuroblastoma cells. Cancer Res., 55: 1798– and Petkovich, M. cDNA cloning of human retinoic acid-metabolizing enzyme 1806, 1995. (hP450RAI) identifies a novel family of cytochromes P450. J. Biol. Chem., 272: 40. Haeseleer, F., Huang, J., Lebioda, L., Saari, J. C., and Palczewski, K. Molecular 18538–18541, 1997. characterization of a novel short-chain dehydrogenase/reductase that reduces all- 15. Stoner, C. M., and Gudas, L. J. Mouse cellular retinoic acid binding protein: cloning, trans-retinal. J. Biol. Chem., 273: 21790–21799, 1998. complementary DNA sequence, and messenger RNA expression during the retinoic 41. Malgaretti, N., Pozzoli, O., Bosetti, A., Corradi, A., Ciarmatori, S., Panigada, M., acid-induced differentiation of F9 wild type and RA-3–10 mutant teratocarcinoma Bianchi, M. E., Martinez, S., and Consalez, G. G. Mmot1, a new helix-loop-helix cells. Cancer Res., 49: 1497–1504, 1989. transcription factor gene displaying a sharp expression boundary in the embryonic 16. Durand, B., Saunders, M., Leroy, P., Leid, M., and Chambon, P. All-trans and 9-cis mouse brain. J. Biol. Chem., 272: 17632–17639, 1997. 42. Napolitano, M., Marfia, G. A., Vacca, A., Centonze, D., Bellavia, D., Di Marcotullio, retinoic acid induction of CRABPII transcription is mediated by RAR-RXR het- L., Frati, L., Bernardi, G., Gulino, A., and Calabresi, P. Modulation of gene expres- erodimers bound to DR1 and DR2 repeated motifs. Cell, 71: 73–85, 1992. sion following long-term synaptic depression in the striatum. Brain Res. Mol. Brain 17. Minucci, S., Leid, M., Toyama, R., Jeannet, J-P. S., Peterson, V. J., Horn, V., Res., 72: 89–96, 1999. Ishmael, J. E., Bhattacharyya, N., Dey, A., Dawid, I. B., and Ozato, K. Retinoid 43. Vacca, A., Di Marcotullio, L., Giannini, G., Farina, M., Scarpa, S., Stoppacciaro, A., X Receptor (RXR) within the RXR-retinoic acid receptor heterodimer binds its Calce, A., Maroder, M., Frati, L., Screpanti, I., and Gulino, A. Thrombospondin-1 is ligands, and enhances retinoid-dependent gene expression. Mol. Cell. Biol., 17: a mediator of the neurotypic differentiation induced by EGF in thymic epithelial cells. 644– 655, 1997. Exp. Cell Res., 248: 79–86, 1999. 18. Chen, J. Y., Clifford, J., Zusi, C., Starrett, J., Tortolani, D., Ostrowski, J., Reczek, 44. Consalez, G. G., Corradi, A., Ciarmatori, S., Bossolasco, M., Malgaretti, N., and P. R., Chambon, P., and Gronemeyer, H. Two distinct actions of retinoid-receptor Stayton, C. L. A new method to screen clones from differential display experiments ligands. Nature (Lond.), 382: 819–822, 1996. prior to RNA studies. Trends Genet., 12: 455–456, 1996. 19. Kastner, P., Mark, M., Ghyselinck, N., Krezel, W., Dupe, V., Grondona, J. M., and 45. Guo, X., and Gudas, L. J. Metabolism of all-trans-retinol in normal human cell strains Chambon, P. Genetic evidence that the retinoid signal is transduced by heterodimeric and squamous cell carcinoma (SCC) lines from the oral cavity and skin: reduced RXR/RAR functional units during mouse development. Development (Camb.), 124: esterification of retinol in SCC lines. Cancer Res., 58: 166–176, 1998. 313–326, 1997. 46. Chen, A. C., Guo, X., Derguini, F., and Gudas, L. J. Human breast cancer cells and 20. Giannini, G., Dawson, M. I., Zhang, X., and Thiele, C. J. Activation of three distinct normal mammary epithelial cells: retinol metabolism and growth inhibition by the RXR/RAR heterodimers induces growth arrest and differentiation of neuroblastoma retinol metabolite 4-oxoretinol. Cancer Res., 57: 4642–4651, 1997. cells. J. Biol. Chem., 272: 26693–266701, 1997. 47. Cheng, N. C., Van Roy, N., Chan, A., Beitsma, M., Westerveld, A., Speleman, F., and 21. Mark, M., Ghyselinck, N. B., Wendling, O., Dupe, V., Mascrez, B., Kastner, P., and Versteeg, R. Deletion mapping in neuroblastoma cell lines suggests two distinct Chambon, P. A genetic dissection of the retinoid signalling pathway in the mouse. tumor suppressor genes in the 1p35–36 region, only one of which is associated with Proc. Nutr. Soc., 58: 609–613, 1999. N-myc amplification. Oncogene, 10: 291–297, 1995. 22. Faria, T. N., Mendelsohn, C., Chambon, P., and Gudas, L. J. The targeted disruption 48. Caron, H., Peter, M., van Sluis, P., Speleman, F., de Kraker, J., Laureys, G., Michon, of both alleles of RAR␤(2) in F9 cells results in the loss of retinoic acid-associated J., Brugieres, L., Voute, P. A., Westerveld, A., et al. Evidence for two tumour growth arrest. J. Biol. Chem., 274: 26783–26788, 1999. suppressor loci on chromosomal bands 1p35–36 involved in neuroblastoma: one 23. de The, H. Altered retinoic acid receptors. FASEB J., 10: 955–960, 1996. probably imprinted, another associated with N-myc amplification. Hum. Mol. Genet., 24. Liu, Y., Lee, M. O., Wang, H. G., Li, Y., Hashimoto, Y., Klaus, M., Reed, J. C., and 4: 535–539, 1995. Zhang, X. Retinoic acid receptor ␤ mediates the growth-inhibitory effect of retinoic 49. Fong, C. T., Dracopoli, N. C., White, P. S., Merrill, P. T., Griffith, R. C., Housman, acid by promoting apoptosis in human breast cancer cells. Mol. Cell. Biol., 16: D. E., and Brodeur, G. M. Loss of heterozygosity for the short arm of chromosome 1138–1149, 1996. 1 in human neuroblastomas: correlation with N-myc amplification. Proc. Natl. Acad. 25. Lotan, R. Retinoids in cancer chemoprevention. FASEB J., 10: 1031–1039, 1996. Sci. USA, 86: 3753–3757, 1989. 26. Matthay, K. K., Villablanca, J. G., Seeger, R. C., Stram, D. O., Harris, R. E., Ramsay, 50. Lane, M. A., Chen, A. C., Roman, S. D., Derguini, F., and Gudas, L. J. Removal of N. K., Swift, P., Shimada, H., Black, C. T., Brodeur, G. M., Gerbing, R. B., and LIF (leukemia inhibitory factor) results in increased vitamin A (retinol) metabolism Reynolds, C. P. Treatment of high-risk neuroblastoma with intensive chemotherapy, to 4-oxoretinol in embryonic stem cells. Proc. Natl. Acad. Sci. USA, 96: 13524– radiotherapy, autologous bone marrow transplantation, and 13-cis-retinoic acid. Chil- 13529, 1999. dren’s Cancer Group. N. Engl. J. Med., 341: 1165–1173, 1999. 51. Achkar, C. C., Derguini, F., Blumberg, B., Langston, A., Levin, A. A., Speck, J., 27. Kastner, P., Mark, M., and Chambon, P. Nonsteroid nuclear receptors: what are Evans, R. M., Bolado, J., Jr., Nakanishi, K., Buck, J., et al. 4-Oxoretinol, a new genetic studies telling us about their role in real life? Cell, 83: 859–869, 1995. natural ligand and transactivator of the retinoic acid receptors. Proc. Natl. Acad. Sci. 28. Rockwood, J. M., and Maxwell, G. D. An analysis of the effects of retinoic acid and USA, 93: 4879–4884, 1996. other retinoids on the development of adrenergic cells from the avian neural crest. 52. Blumberg, B., Bolado, J., Jr., Derguini, F., Craig, A. G., Moreno, T. A., Chakravarti, Exp. Cell Res., 223: 250–258, 1996. D., Heyman, R. A., Buck, J., and Evans, R. M. Novel retinoic acid receptor ligands 29. Chandrasekaran, V., Zhai, Y., Wagner, M., Kaplan, P. L., Napoli, J. L., and Higgins, in Xenopus embryos. Proc. Natl. Acad. Sci. USA, 93: 4873–4878, 1996. 53. Sonneveld, E., van den Brink, C. E., van der Leede, B. M., Schulkes, R. K., D. Retinoic acid regulates the morphological development of sympathetic neurons. Petkovich, M., van der Burg, B., and van der Saag, P. T. Human retinoic acid (RA) J. Neurobiol., 42: 383–393, 2000. 4-hydroxylase (CYP26) is highly specific for all-trans-RA and can be induced 30. Abemayor, E., and Sidell, N. Human neuroblastoma cell lines as models for the in through RA receptors in human breast and colon carcinoma cells. Cell Growth vitro study of neoplastic and neuronal cell differentiation. Environ. Health Perspect., Differ., 9: 629–637, 1998. 80: 3–15, 1989. 54. Abu-Abed, S. S., Beckett, B. R., Chiba, H., Chithalen, J. V., Jones, G., Metzger, 31. Reynolds, C. P., Kane, D. J., Einhorn, P. A., Matthay, K. K., Crouse, V. L., Wilbur, D., Chambon, P., and Petkovich, M. Mouse P450RAI (CYP26) expression and J. R., Shurin, S. B., and Seeger, R. C. Response of neuroblastoma to retinoic acid in retinoic acid-inducible retinoic acid metabolism in F9 cells are regulated by vitro and in vivo. Prog. Clin. Biol. Res., 366: 203–211, 1991. retinoic acid receptor ␥ and retinoid X receptor ␣. J. Biol. Chem., 273: 2409– 32. Thiele, C. J. Biology of pediatric peripheral neuroectodermal tumors. Cancer Metas- 2415, 1998. tasis Rev., 10: 311–319, 1991. 55. Schwab, M., Praml, C., and Amler, L. C. Genomic instability in 1p and human 33. Thiele, C. J., Reynolds, C. P., and Israel, M. A. Decreased expression of N-myc malignancies, Genes Chromosomes Cancer, 16: 211–229, 1996. precedes retinoic acid induced morphological differentiation of human neuroblas- 56. Caron, H., van Sluis, P., de Kraker, J., Bokkerink, J., Egeler, M., Laureys, G., Slater, toma. Nature (Lond.), 313: 404–406, 1985. R., Westerveld, A., Voute, P. A., and Versteeg, R. Allelic loss of chromosome 1p as 34. Kaplan, D. R., Matsumoto, K., Lucarelli, E., and Thiele, C. J. Induction of TrkB by a predictor of unfavorable outcome in patients with neuroblastoma. N. Engl. J. Med., retinoic acid mediates biologic responsiveness to BDNF and differentiation of human 334: 225–230, 1996. neuroblastoma cells. Neuron., 11: 321–331, 1993. 57. Iolascon, A., Lo Cunsolo, C., Giordani, L., Cusano, R., Mazzocco, K., Boumgartner, 35. Lucarelli, E., Kaplan, D. R., and Thiele, C. J. Selective regulation of TrkA and TrkB M., Ghisellini, P., Faienza, M. F., Boni, L., De Bernardi, B., Conte, M., Romeo, G., receptors by retinoic acid and interferon-␥ in human neuroblastoma cell lines. J. Biol. and Tonini, G. P. Interstitial and large chromosome 1p deletion occurs in localized Chem., 270: 24725–24731, 1995. and disseminated neuroblastomas and predicts an unfavourable outcome. Cancer 36. Matsuo, T., and Thiele, C. J. p27Kip1: a key mediator of retinoic acid induced growth Lett., 130: 83–92, 1998. arrest in the SMS-KCNR human neuroblastoma cell line. Oncogene, 16: 3337–3343, 58. Stockton, D. W., Ross, H. L., Bacino, C. A., Altman, C. A., Shaffer, L. G., and 1998. Lupski, J. R. Severe clinical phenotype due to an interstitial deletion of the short arm 37. Giannini, G., Di Marcotullio, L., Ristori, E., Zani, M., Crescenzi, M., Scarpa, S., of chromosome 1: a brief review. Am. J. Med. Genet., 71: 189–193, 1997. Piaggio, G., Vacca, A., Peverali, F. A., Diana, F., Screpanti, I., Frati, L., and 59. Martinsson, T., Sjoberg, R. M., Hallstensson, K., Nordling, M., Hedborg, F., and 1203

Kogner, P. Delimitation of a critical tumour suppressor region at distal 1p in short arm of chromosome 1 in neuroblastoma, Genes Chromosomes Cancer, 10: neuroblastoma tumours. Eur. J. Cancer, 33: 1997–2001, 1997. 275–281, 1994. 60. Takeda, O., Homma, C., Maseki, N., Sakurai, M., Kanda, N., Schwab, M., Nakamura, 63. Schwartz, S., Jr., Yamamoto, H., Navarro, M., Maestro, M., Reventos, J., and Y., and Kaneko, Y. There may be two tumor suppressor genes on chromosome arm Perucho, M. Frameshift mutations at mononucleotide repeats in caspase-5 and other 1p closely associated with biologically distinct subtypes of neuroblastoma. Genes target genes in endometrial and gastrointestinal cancer of the microsatellite mutator Chromosomes Cancer, 10: 30–39, 1994. phenotype. Cancer Res., 59: 2995–3002, 1999. 61. White, P. S., Maris, J. M., Sulman, E. P., Jensen, S. J., Kyemba, S. M., Beltinger, 64. Cheng, N. C., Chan, A. J., Beitsma, M. M., Speleman, F., Westerveld, A., and C. P., Allen, C., Kramer, D. L., Biegel, J. A., and Brodeur, G. M. Molecular analysis Versteeg, R. A human modifier of methylation for class I HLA genes (MEMO-1) of the region of distal 1p commonly deleted in neuroblastoma. Eur. J. Cancer, 33: maps to chromosomal bands 1p35–36.1. Hum. Mol. Genet., 5: 309–317, 1996. 1957–1961, 1997. 65. Fiore, P., Castagnola, E., Marchese, N., Dufour, C., Garaventa, A., Mangraviti, S., 62. Schleiermacher, G., Peter, M., Michon, J., Hugot, J. P., Vielh, P., Zucker, J. M., and Cornaglia-Ferraris, P. Retinol (vitamin A), and retinal-binding protein serum Magdelenat, H., Thomas, G., and Delattre, O. Two distinct deleted regions on the levels in children with cancer at onset. Nutrition, 13: 17–20, 1997.

1204

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2002 American Association for Cancer Research. retSDR1, a Short-Chain Retinol Dehydrogenase/Reductase, Is Retinoic Acid-inducible and Frequently Deleted in Human Neuroblastoma Cell Lines

Fabio Cerignoli, Xiaojia Guo, Beatrice Cardinali, et al.

Cancer Res 2002;62:1196-1204.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/62/4/1196

Cited articles This article cites 63 articles, 29 of which you can access for free at: http://cancerres.aacrjournals.org/content/62/4/1196.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/62/4/1196.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/62/4/1196. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.