Aly⁄ REF, a factor for mRNA transport, activates RH gene promoter function Hiroshi Suganuma 1 , Maki Kumada 1 , Toshinori Omi 1 , Takaya Gotoh 1 , Munkhtulga Lkhagvasuren 1 , Hiroshi Okuda 1,2 , Toyomi Kamesaki 1,2 , Eiji Kajii 1,2 and Sadahiko Iwamoto 1 1 Division of Human Genetics, Center for Community Medicine, Jichi Medical School, Japan 2 Division of Community Medicine, Center for Community Medicine, Jichi Medical School, Japan The rhesus (Rh) blood group antigens are of consider- able importance in transfusion medicine as well as in newborn or autoimmune hemolytic diseases due to their high antigenicity [1]. The Rh antigens are carried by two distinct but homologous integral membrane proteins of 30–32 kDa, which have been isolated by immunopreci- pitation using anti-Rh antibodies [2]. The corresponding cDNAs, RHCE [3,4] and RHD [5–7] have been cloned. They differ in only 31–35 of 417 amino acid residues and have been mapped in tandem on chromosome 1q34.3–36.1. Genomic organization of the RH locus has revealed that RHD and RHCE face each other at their 3¢ tails, and that the gene SMP1 is interspersed between them [8]. Wagner et al. identified two 9 kb transposon- like DNA segments, called ‘rhesus boxes’ upstream and downstream of the RHD gene. They speculated that these boxes are involved in the regulation of SMP1 because they encode a GC-rich region at the 3¢ ends [9]. However, whether rhesus boxes are involved in regula- tion of the RH locus remains unknown. Rh transcripts are distributed among cells of hema- topoietic lineage with erythroid features [3]. The expression of RhD and CE antigens increases synergis- tically with the maturation of erythroblasts [10]. The promoter sequences of the RH genes support erythroid specific expression of the Rh antigens [11]. The human b-globin locus is composed of five tandem arrayed genes and it is controlled by a locus control region (LCR) localized between positions )21.5 and )6.1 kb from the e globin cap site [12,13]. The LCR can open Keywords Rh blood group antigen; promoter; transcription cofactor; Aly ⁄ REF Correspondence S Iwamoto, Division of Human Genetics, Center for Community Medicine, Jichi Medical School, 3311-1 Minamikawachi- machi, Kawachi-gun, Tochigi 329-0498, Japan Fax: +81 285 44 49 Tel: +81 285 58 7342 E-mail: siwamoto@jichi.ac.jp Note Hiroshi Suganuma and Sadahiko Iwamoto contributed equally to this work (Received 26 November 2004, revised 17 March 2005, accepted 23 March 2005) doi:10.1111/j.1742-4658.2005.04681.x The rhesus (Rh) blood group antigens are of considerable importance in transfusion medicine as well as in newborn or autoimmune hemolytic diseases due to their high antigenicity. We identified a major DNaseI hypersensitive site at the 5¢ flanking regions of both RHD and RHCE exon 1. A 34 bp fragment located at )191 to )158 from a translation start posi- tion, and containing the TCCCCTCCC sequence, was involved in enhan- cing promoter activity, which was assessed by luciferase reporter gene assay. A biotin-labelled 34 bp probe isolated an mRNA transporter pro- tein, Aly ⁄ REF. The specific binding of Aly ⁄ REF to RH promoter in eryth- roid was confirmed by chromatin immunoprecipitation assay. The silencing of Aly ⁄ REF by siRNA reduced not only the RH promoter activity of the reporter gene but also transcription from the native genome. These facts provide second proof of Aly ⁄ REF as a transcription coactivator, initially identified as a coactivator for the TCRa enhancer function. Aly ⁄ REF might be a novel transcription cofactor for erythroid-specific genes. Abbreviations ChIP, chromatin immunoprecipitation; EMSA, electrophoretic mobility shift assay; HEL, human erythroleukemic cell line; HS, hypersensitive sites; LCR, locus control region. 2696 FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS the b-globin locus, enhance transcription and control the timing and choice of gene for transcription within the locus [13]. The LCR is characterized by the presence of five DNaseI hypersensitive sites (HS) that encode multiple DNA motifs for transcription factors. The present study identifies an enhancer element and a protein that promotes its activity in the major HS of both RH genes in an erythroleukemic cell line. Mapping the DNaseI HS of RH loci We investigated the major regulatory region of RH gene expression using in vivo DNaseI hypersensitivity analysis in a human erythroleukemic cell line (HEL), which expresses Rh antigens on the cell surface [14]. We prepared three probes with which to explore two RH loci. Probe 1 that encoded exon 1 and the 5¢ end of intron 1 of the RHD and CE genes revealed 3.7 kb bands (solid arrow head in Fig. 1) along with weakened 20 kb (CE gene) and 18 kb (D gene) bands. Probes 8 and 10 revealed weak bands (grey arrowheads), indica- ting control elements in intron 9. Probes encoding the unique sequence in the rhesus box did not reveal any apparent DNaseI hypersensitivity (data not shown). Thus, the rhesus boxes are not the major controlling region of the RH loci, but the 5¢ flanking sequences of RHCE and RHD exon 1 were the most important regulatory region studied. Functional mapping of regulatory regions in RHD upstream sequence The )12 159 bp sequence from the translation start site of the RHD gene was inserted into the pGL3 Fig. 1. DNaseI hypersensitivity mapping of the RH gene in HEL cells. Probes are indicated in the locus map. Restriction sites of EcoRV and exons of each gene are shown by vertical lines on the horizontal line and gray lines across it, respectively. DNaseI concentrations increase from 0 to 15 lgÆmL )1 (triangles). Extracted DNA was digested by EcoRV. Southern blots show positions of major (arrowheads) and minor (grey arrowheads) HS. H. Suganuma et al. Aly ⁄ REF activates RH gene promoter function FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS 2697 A B Fig. 2. Identification of the functional element by a functional assay of the 5¢ flanking region of the human RH gene. (A) Top, repetitive sequence map. The full-length construct, pGL-RH-12159, is the parental vector to induce deletion derivatives. Constructs were transfected into HEL cells and relative light units were measured. (B) The pGL-RH-238 construct induced further deletions (dashed line) or mutations (grey letters) according to PCR. Relative light units are shown as percentage values of those of parental vector, pGL-RH-238. Rectangles, GATA motifs. Oligo probes used for electrophoretic mobility shift assay are indicated by horizontal bars with numerals. Fig. 3. Purification of DNA binding protein to probe 2. (A) EMSA study of the promoter sequence of the human RH gene. Four double- stranded oligonucleotide probes prepared as shown in Fig. 2B were incubated with nuclear extracts of HEL cells. Competition proceeded with cold competitors indicated above. Numbers with M indicate mutant competitors. The supershift assay included the anti-(GATA-1) mAb. (B) SDS ⁄ PAGE of purified by EMSA probe 2. Samples from each purification cycle were analysed by 5–15% gradient SDS ⁄ PAGE and silver staining. Lane 1, nuclear extract; lanes 2 and 3, 1 ⁄ 10 aliquot of first and second cycle products with mutant probe; lanes 4 and 5, 1 ⁄ 10 aliquot of first and second cycle products with wild-type probe. Left, marker proteins. Bold letters indicate three internal amino sequences of trypsin-digested fragments of protein band in sequence nominated by MASCOT analysis. (C) ChIP assay in HEL and HeLa cells. The chroma- tin enriched by negative control antibody (NegCon), anti-TFIIB or anti-Aly Ig and whole chromatin (imput) were amplified by PCR primers indicated on the left. Aly ⁄ REF activates RH gene promoter function H. Suganuma et al. 2698 FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS A B C H. Suganuma et al. Aly ⁄ REF activates RH gene promoter function FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS 2699 vector. Serial deletion constructs from this parental vector revealed that a DNA fragment from )238 to )140 was closely involved in transcriptional activity and that a further upstream sequence acts as a suppressor in HEL cells (Fig. 2A). Deletion and site- directed mutants of the )238 construct (pGL3-RH- 238) showed that the absence of a 34 bp fragment from )191 to )158 (pGL3-RH-238del) decreased the activity to less than 20% of the original (Fig. 2B). Dis- ruption of the GATA motifs, especially the proximal one, decreased the activity to 20% (pGL3-RH- 238Mut3) and a mutation in the C-rich region within the 34 bp sequence decreased the activity by 50% (pGL3-RH-238Mut1). These data suggested that the DNA fragment from )191 to )158, especially the C-rich sequence, is involved in the enhancer activity. Identification of the protein that binds to the putative regulatory region Three overlapping double-stranded oligo probes exam- ined DNA binding proteins at the 34 bp sequence. Among the probes, the shift bands in the electrophore- tic mobility shift assay (EMSA) determined by probe 2 were the most significant (Fig. 3A). The shift bands were partially competed out by a 50-fold excess of wild-type competitor but not by a mutant probe (2 m) or by either of the neighbouring probes 1 and 3, sug- gesting that TCCCCTCCC sequence unique for probe 2 was the motif for the transcription factor. Anti- EKLF and anti-Sp1 antibody did not influence the shift band mobility, while the binding motif of EKLF (CCCACCC) and Sp1 (CCCGCCCC) resembled the C-rich sequence (data not shown). Probe 4 encoding the proximal GATA motif showed an intense band that was super-shifted by the anti-GATA1 antibody, indica- ting the GATA1 is closely involved in the function of this promoter. We used biotin-labelled double-stranded probes to isolate the protein that bound to probe 2. While the mutant probes did not retain any protein after the sec- ond wash, the wild-type probe bound a 35-kDa protein (Fig. 3B). A mass spectrogram of trypsin-digested pep- tides of the band and the internal peptide sequences revealed that the protein recovered by the wild-type probe was Aly ⁄ REF. The binding of Aly ⁄ REF to the RH promoter was verified by chromatin immunoprecipitation (ChIP) assay. Enrichment of a DNA fragment encoding the RH promoter was observed by anti-ALY antibody in two independent experiments using HEL cells, while the RH promoter was not condensed from chromatin of nonerythroid HeLa cells (Fig. 3C). Silencing of Aly ⁄ REF decreased RH promoter activity To determine whether Aly ⁄ REF actually activates the RH promoter, the reporter plasmids pGL3-RH-238 and )158 were induced into HEL cells with a plasmid expressing Aly ⁄ REF siRNAs. The Aly ⁄ REF siRNA encoding nt 283–303 (pSilencerALY283-303) signifi- cantly (P<0.05) decreased the RH promoter activity of the pGL3-RH-238 construct but the decrease was not significant when pGL3-RH-158 was the reporter (Fig. 4A). In contrast to the RH promoter, SV40 pro- moter activity was enhanced by the siRNAs. These results indicated that the C-rich region in the 34 bp sequence is specific for Aly ⁄ REF or for DNA binding proteins associated with Aly ⁄ REF. Quantitative RT-PCR showed that Aly ⁄ REF siRNA decreased the amount of Aly ⁄ REF mRNA (Fig. 4B). The mRNA reduction by pSilencerALY285-303 was greater than that induced by pSilencerALY45-65. The amount of Rh transcripts were significantly decreased when pSilencerALY285-303 was induced in HEL cells, although the decrease induced by pSilencerALY45-65 was not significant, which might reflect its lower efficiency of reduction for Aly ⁄ REF mRNA. The Ct value of GAPDH mRNA was not influenced by the induction of Aly ⁄ REF siRNA (data not shown). Thus, the decreased Rh mRNA and luciferase activity did not result from a dysfunction in mRNA transportation but from a downregulation of RH promoter function caused by a decrease of Aly ⁄ REF protein. Discussion We identified a DNaseI HS region in the 5¢ flanking sequence of exon 1 of both RHCE and RHD, which appeared to be the major regulatory region within RH loci. Neither in vivo DNaseI hypersensitivity analysis nor luciferase assays suggested that loci control regions are located in rhesus boxes. In the 5¢ flanking sequence of RH exon 1, the C-rich sequence acts as an enhancer element, whereas the GATA element was the most important motif in the same manner as other eryth- roid-specific genes. We isolated Aly ⁄ REF protein using double-stranded oligo probes encoding the C-rich sequence. The binding of Aly ⁄ REF at the RH promoter in HEL cells was con- firmed by ChIP assay. Its specific binding to the RH pro- moter in erythroid cells was shown by its background level accumulation from chromatin of the cervical cancer cell line, HeLa, which does not express Rh antigen. Aly ⁄ REF has two RNA binding domains and is involved in RNA transportation from the nucleus Aly ⁄ REF activates RH gene promoter function H. Suganuma et al. 2700 FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS [15,16], but it was initially identified as a coactivator of LEF-1 and AML-1 for TCRa enhancer function [17]. Bruhn et al. have shown that Aly⁄ REF itself has no specific DNA or RNA sequence motif and estima- ted that it acts as a context dependent coactivator. However, they also remarked that the ternary com- plex of Aly ⁄ REF with LEF-1 and DNA was depend- ent on the flanking sequence of the LEF binding motif, suggesting that Aly ⁄ REF also weakly interacts with the sequences [17]. Our data suggested that Aly ⁄ REF interacts with a preferential sequence. The mutation in TCCCCTCCC reduced the promoter activity by 50% (Fig. 2B), as it lost the ability to compete out the shift band in EMSA (Fig. 3A) and the mutant biotin labelled probe failed to retain Aly ⁄ REF protein (Fig. 3B). Another explanation for these phenomena is that an unknown factor interacts with the TCCCCTCCC motif and that an overwhelm- ing molar excess of Aly ⁄ REF protein interacts with it as a scaffold. These notions remain to be investi- gated. The knockdown experiment with Aly ⁄ REF siRNA in HEL cells reduced the RH promoter activity of both the reporter plasmid and the native genome. The reduction was indeed subtle, which might result from low efficiency of pSilencer transfection into HEL cells, and from partial magnitude of Aly ⁄ REF activity on the promoter as shown by mutageneis analysis at the motif (Fig. 2B). In contrast to the RH promoter, the reduction in Aly ⁄ REF significantly enhanced SV40 A B Fig. 4. Interference of siRNA in Aly ⁄ REF. (A) HEL cells were transfected with pSilenc- er harbouring a random control sequence, ALY45-65 or ALY285-303 and reporter pGL3 inserted with RH-158, RH-238 or the SV40 promoter indicated below and internal con- trol pRL-CMV vector. Relative luciferase assays were performed 24 h later. Relative luciferase activities normalized by pSilencer- control in each reporter vector are shown with mean (bars) and SD (lines) values of triplicated transfection experiments. (B) HEL cells were transfected solely with pSilencer constructs indicated below and Aly ⁄ REF or Rh mRNA levels were assessed by quantita- tive RT-PCR. Relative mRNA amounts nor- malized by pSilencer-control are shown with SD values of triplicated transfection experi- ments. Asterisks indicate statistical signifi- cance: *P < 0.05 and **P < 0.01. H. Suganuma et al. Aly ⁄ REF activates RH gene promoter function FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS 2701 promoter activity. Bruhn et al. have also shown the promoter specific enhancement by Aly ⁄ REF; ALY antisense oligonucleotide decreased TCRa promoter function, but did not affect Rous sarcoma virus promoter [17]. These data suggested that Aly ⁄ REF protein is actually involved in the expression of Rh protein depending on the promoter sequence. In the same way as glycophorin A and B [18], two RH genes are synergistically expressed along with erythroid maturation and they might be under the con- trol of individual enhancers in each gene. The C-rich region with which Aly ⁄ REF interacts is a control element for both RH loci. Various transcription factors that interact with erythroid specific gene promoters or enhancers have been identified. GATA-1, GATA-2, NF-E2, EKLF, FOG, LMO, SCL and Lbd have been characterized as transcription factors of erythroid specific genes [19,20] and Aly ⁄ REF might be a novel addition to this group. Rh antigens, especially RhD, cause harmful hemo- lysis in neonates and during transfusion medicine [1]. Understanding the regulatory mechanisms of Rh anti- gen expression will help to avoid harmful hemolytic reactions in newborns or in auto-immune hemolytic diseases. Experimental procedures DNaseI hypersensitivity analysis HEL cells obtained from the Health Science Research Resources Bank (Osaka, Japan) was maintained in RPMI1640 containing 10% (v ⁄ v) fetal bovine serum. Nuclei from HEL cells were isolated as described [21]. Aliquots (0.25 mL) of nuclei at a density of 2 · 10 7 ÆmL )1 were digested with increasing concentrations of DNaseI (0–15 lgÆmL )1 ) and then DNA was extracted as des- cribed. Purified DNA (10 lg) was digested with EcoRV, resolved by electrophoresis on 0.8% agarose gels and transferred onto Hybond-N+ nylon membranes (Amer- sham Pharmacia Biotech, Little Chalfont, Buckingham- shire, UK). The blots were initially hybridized with a human RH exon 1 probe (Fig. 1, probe 1). After auto- radiography, the membranes were stripped and succes- sively rehybridized with probes 8 and 10, encoding exons 8 and 10, respectively. Plasmid constructs and transduction and reporter enzyme assay The )12 159 bp sequence from the translation start site of the RHD gene (GenBank accession no. AB029152) [22] that encodes almost 85% of the rhesus box was cloned as described and inserted into the pGL3 vector. Serial deletion derivatives were constructed from this par- ental vector using the restriction sites. Site-directed muta- genesis or segmental deletion in the predicted motifs of the minimum promoter proceeded using the PCR-based method of Imai et al. [23] as described. For transfection, 5 · 10 5 HEL cells were cotransfected with the pGL3-RH plasmid and pRL-CMV using LipofectAMINE (Invitro- gen, Carlsbad, CA). Twenty-four hours later, cells were harvested and relative light units (firefly ⁄ Renilla light units) were measured using a Dual-Luciferase reporter assay system (Promega, Madison, WI, USA) and a TD-20 ⁄ 20 luminometer (Turner Designs, Sunnyvale, CA, USA). At least three independent transfection assays were performed. EMSA EMSA reactions were performed using nuclear extracts of HEL cells and [ 32 P]dCTP-labelled double-stranded probes (Fig. 2B). For supershift assays, 1 lL of rat mAb anti- GATA1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was added. Samples were incubated for 30 min at room temperature and resolved on 5% acrylamide gels in 0.5· TBE buffer at room temperature. The gels were dried and exposed to X-ray film. Purification and identification of the protein that binds to the regulatory motif The protein that bound to probe 2 in the EMSA was puri- fied using avidin-biotin as described [24] with some modifi- cation. The biotinylated probe was composed of a chemically synthesized oligonucleotide with a biotin-phos- phoramidite tail (lower strand) and an unlabelled upper strand oligonucleotide as follows: biotin-5¢-GGGACTAT GATGATGGGGAGGGGAGGAAATGT-3¢ and 5¢-ACA TTTCCTCCCCTCCCCATCATAGTCCC-3¢. To exclude the possibility of nonspecific protein binding, we prepared the following mutant probe: biotin- 5¢-GGGACTATGAT GATGGGTTTGTGAGGAAATGT-3¢ and 5¢-ACATTTC CTCACAAACCCATCATAGTCCC-3¢. HEL cell nuclear proteins carried by the probes were separated using strept- avidin-magnetic beads (Qiagen, Valencia, CA, USA). Bead suspensions were washed and then trapped proteins were extracted in high-salt buffer until a 10% aliquot of the extracted protein resolved as a single protein band in SDS ⁄ PAGE as shown by silver staining. The protein band was excised and digested with trypsin. The peptide fragments were examined by MALDI-TOF-MS and LC-MS ⁄ MS performed at the ProPhoenix Institute (Hiro- shima, Japan). The mass fingerprint data were analysed by mascot and the internal amino acid sequences were com- pared with the NCBInr databases. Aly ⁄ REF activates RH gene promoter function H. Suganuma et al. 2702 FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS ChIP assay HEL and HeLa cells were fixed for 10 min in RPMI1640 con- taining 1% (v ⁄ v) formaldehyde. The nuclei from the fixed cells were sonicated in the presence of protease inhibitor cocktail and the sheared chromatins were incubated with mouse anti-ALY Ig (ImmuQuest, Cleveland, UK), anti- TFIIB or negative control IgG provided by Active Motif (Carlsbad, CA, USA). Enrichment of chromatin fragments by the antibodies was assessed through PCR reactions in the linear stage of amplification using primers for RH (5¢-ACATTTCCTCCCCTCCCCATCATAGTCCCT-3¢ and 5¢-ACACCCGCCAAAGGCCTTATCTCAG-3¢), GAPDH primers or negative control primers. SiRNA interference We prepared the dsRNA constructs pSilencerALY45-64 and pSilencerALY285-303 against the human Aly ⁄ REF gene by cloning inverted repeat sequences into pSilencer 2.0-U6 (Ambion Inc., Austin, TX, USA). The dsRNA template consisted of 19 bp target sequences derived from Aly ⁄ REF mRNA and a 9 bp linker (TTCAAGAGA) for transcription of the short hairpin dsRNA. The construct numerals of pSilencerALY45-65 and pSilencerALY285-303 corresponded to the nucleotide number of Aly ⁄ REF mRNA (GenBank accession no. AF047002). A plasmid with a random sequence was used as a negative control dsRNA. The pSilencer constructs were cotransfected into HEL cells with pGL3-RH-158, pGL3-RH-238 or the pGL3-SV40 promoter and the pRL-CMV vector. Relative luciferase assays were performed 24 h thereafter, which was determined as the optimum by a time course experi- ment. The HEL cells were also transfected solely with the pSilencer constructs. Twenty-four hours later, Aly ⁄ REF and Rh transcripts were quantified by real-time PCR using SYBR Green PCR Master Mix and an ABI PRISM 7900HT (Applied Biosystems, Foster City, CA, USA). The primer sequences to detect Aly ⁄ REF and Rh transcripts were for Aly ⁄ REF: sense, 5¢-CTGGTCGCAGCTTAGG AACAG-3¢ and antisense, 5¢-AATGTTCATGGGGCGGC CATC-3¢, for RH: sense, 5¢-GCAACGATACCCAGTTT GTC-3¢ and antisense, 5¢-AGTTGACACTTGGCCAGA AC-3¢. The relative amounts of Aly ⁄ REF and RH were assessed as differences in the threshold of the amplification curve of the target gene from the internal control, GAPDH (delta Ct), and in the delta Ct from the control siRNA construct (delta-delta Ct). Acknowledgements We are grateful to Mr T. Oyamada and Ms. T. Hatano for valuable technical assistance. This work was supported by Grants-in-Aid for Scientific Research for the Ministry of Education, Science and Culture of Japan (Nos. 15590587 for Dr. H. Okuda, 15591018 for Dr. T. Kamesaki and 15590586 for Dr. S. Iwamoto). References 1 Mollison PL, Engelfreit CP & Contreras M (1993) Blood Transfusion in Clinical Medicine, 9th edn. Black- well, Oxford. 2 Moore S, Woodrow CF & McClelland DBL (1982) Isolation of membrane components associated with human red cell antigens Rh (D), (c), (E) and Fy. Nature 295, 529–531. 3 Cherif-Zahar B, Bloy C, Le Van Kim C, Blanchard D, Bailly P, Hermand P, Salmon C, Cartron J-P & Colin Y (1990) Molecular cloning and protein structure of a human blood group Rh polypeptide. Proc Natl Acad Sci USA 87, 6243–6247. 4 Avent ND, Ridgwell K, Tanner MJ & Anstee DJ (1990) cDNA cloning of a 30 kDa erythrocyte membrane pro- tein associated with Rh (Rhesus) -blood-group-antigen expression. Biochem J 271, 821–825. 5 Le Van Kim C, Mouro I, Cherif-Zahar B, Raynal V, Cherrier C, Cartron J-P & Colin Y (1992) Molecular cloning and primary structure of the human blood group RhD polypeptide. Proc Natl Acad Sci USA 89 , 10925–10929. 6 Arce M, Thompson ES, Wagner S, Coyne KE, Ferdman BA & Lublin DM (1993) Molecular cloning of RhD cDNA derived from a gene present in RhD-positive, but not RhD-negative individuals. Blood 82, 651–655. 7 Kajii E, Umenishi F, Iwamoto S & Ikemoto S (1993) Isolation of a new cDNA clone encoding an Rh poly- peptide associated with the Rh blood group system. Hum Genet 91, 157–162. 8 Wagner FF & Flegel WA (2000) RHD gene deletion occurred in the Rhesus box. Blood 95, 3662–3668. 9 Wagner FF & Flegel WA (2002) RHCE represents the ancestral RH position, while RHD is the duplicated gene. Blood 99, 2272–2273. 10 Southcott MJ, Tanner MJ & Anstee DJ (1999) The expression of human blood group antigens during erythropoiesis in a cell culture system. Blood 93, 4425–4435. 11 Cherif-Zahar B, Le Van Kim C, Rouillac C, Raynal V, Cartron J-P & Colin Y (1994) Organization of the gene (RHCE) encoding the human blood group RhCcEe antigens and characterization of the promoter region. Genomics 19, 68–74. 12 Tuan D & London IM (1984) Mapping of DNase I-hypersensitive sites in the upstream DNA of human embryonic epsilon-globin gene in K562 leukemia cells. Proc Natl Acad Sci USA 81, 2718–2722. H. Suganuma et al. Aly ⁄ REF activates RH gene promoter function FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS 2703 13 Forrester WC, Takegawa S, Papayannopoulou T, Stamatoyannopoulos G & Groudine M (1987) Evidence for a locus activation region: the formation of developmentally stable hypersensitive sites in globin-expressing hybrids. Nucleic Acids Res 15, 10159–10177. 14 Iwamoto S, Yamasaki M, Kawano M, Okuda H, Omi T, Takahashi J, Tani Y, Omine M & Kajii E (1998) Expression analysis of human Rhesus blood group antigens by gene transduction into erythroid and non- erythroid cells. Int J Hematol 68, 257–268. 15 Luo ML, Zhou Z, Magni K, Christoforides C, Rappsil- ber J, Mann M & Reed R (2001) Pre-mRNA splicing and mRNA export linked by direct interactions between UAP56 and Aly. Nature 413, 644–647. 16 Gatfield D & Izaurralde E (2002) REF1 ⁄ Aly and the additional exon junction complex proteins are dispen- sable for nuclear mRNA export. J Cell Biol 159, 579–588. 17 Bruhn L, Munnerlyn A & Grosschedl R (1997) ALY, a context-dependent coactivator of LEF-1 and AML-1, is required for TCRalpha enhancer function. Genes Dev 11, 640–653. 18 Nemoto Y, Terajima M, Shoji W & Obinata M (1996) Regulatory function of delta ⁄ YY-1 on the locus control region-like sequence of mouse glycophorin gene in erythroleukemia cells. J Biol Chem 271, 13542–13548. 19 Cantor AB & Orkin SH (2002) Transcriptional regula- tion of erythropoiesis: an affair involving multiple part- ners. Oncogene 21, 3368–3376. 20 Perry C & Soreq H (2002) Transcriptional regulation of erythropoiesis. Fine tuning of combinatorial multi- domain elements. Eur J Biochem 269, 3607–3618. 21 Iwa.moto S, Suganuma H, Kamesaki T, Omi T, Okuda H & Kajii E (2000) Cloning and characterization of erythroid-specific DNase I-hypersensitive site in human rhesus-associated glycoprotein gene. J Biol Chem 275, 27324–27331. 22 Okuda H, Suganuma H, Tsudo N, Omi T, Iwamoto S & Kajii E (1999) Sequence analysis of the spacer region between the RHD and RHCE genes. Biochem Biophys Res Commun 263, 378–383. 23 Imai Y, Matsushima Y, Sugimura T & Terada M (1991) A simple and rapid method for generating a dele- tion by PCR. Nucleic Acids Res 19, 2785. 24 Otsuka F, Iwamatsu A, Suzuki K, Ohsawa M, Hamer DH & Koizumi S (1994) Purification and characteriza- tion of a protein that binds to metal responsive elements of the human metallothionein IIA gene. J Biol Chem 269, 23700–23707. 2704 FEBS Journal 272 (2005) 2696–2704 ª 2005 FEBS Aly ⁄ REF activates RH gene promoter function H. Suganuma et al. . Aly⁄ REF, a factor for mRNA transport, activates RH gene promoter function Hiroshi Suganuma 1 , Maki Kumada 1 , Toshinori Omi 1 , Takaya Gotoh 1 ,. biotin-phos- phoramidite tail (lower strand) and an unlabelled upper strand oligonucleotide as follows: biotin-5¢-GGGACTAT GATGATGGGGAGGGGAGGAAATGT-3¢ and 5¢-ACA TTTCCTCCCCTCCCCATCATAGTCCC-3¢.