Functional Characterization of the Human RPGR Proximal Promoter

Biochemistry and Molecular Biology Functional Characterization of the Human RPGR Proximal Promoter

Xinhua Shu,*,1,2 Julie R. Simpson,2 Alan W. Hart,2 Zhihong Zeng,3 Sarita Rani Patnaik,1 Philippe Gautier,2 Emma Murdoch,2 Brian Tulloch,2 and Alan F. Wright*,2

PURPOSE. Mutations in the retinitis pigmentosa (RP) GTPase promoter will facilitate understanding of the functional role regulator (RPGR) gene account for more than 70% of X-linked of RPGR in the retina and gene therapy of X-linked RP. (Invest RP cases. This study aims to characterize the proximal Ophthalmol Vis Sci. 2012;53:3951–3958) DOI:10.1167/ promoter region of the human RPGR gene. iovs.11-8811 0 METHODS. The 5 -flanking region (5 kb) of human RPGR was cloned and sequenced. A potential transcription start site and etinitis pigmentosa (RP) is a genetically heterogeneous transcription factor binding motifs were identified by bio- Rgroup of retinal degenerations that affect 1 in 4000 in the informatic analysis. Constructs containing the putative human general population.1,2 Most cases are inherited in an autosomal RPGR promoter region upstream of a luciferase reporter gene dominant, autosomal recessive, X-linked, or mitochondrial were generated and analyzed by transient transfection and manner, but oligogenic inheritance has been established in a luciferase assays. Transgenic mouse lines carrying a 3-kb small proportion of families.3 X-linked RP (XLRP) is one of the human RPGR promoter sequence fused to lacZ were generated most consistently severe forms of RP, with a reported average and RPGR proximal promoter activity was analyzed by X-gal age at onset of 7.2 6 1.7 years.4 XLRP affects 10% to 20% of all staining. RP patients5 and has been genetically mapped to six loci: RP2, 0 RESULTS. Bioinformatic analyses of the human RPGR 5 -flanking RP3, RP6, RP23, RP24, RP34 (http://www.sph.uth.tmc.edu/ region uncovered potential transcription factor binding sites retnet/). The RPGR gene is mutated in the RP3 form of XLRP, and a CpG island. Transient transfection assays with RPGR which accounts for 70% to 80% of affected families.5–8 promoter/luciferase reporter constructs revealed a 980-bp The human RPGR gene is located in chromosomal region fragment (952 to þ28) that produced higher levels of Xp21.1 and spans 172 kilobases.8 There are multiple alterna- luciferase activity. Mutagenesis identified a putative Sp1 tively spliced transcripts, all of which encode an amino (N)- binding site that was critical for regulating transcriptional terminal RCC1-like (RCCL) domain, which is structurally activity. We generated transgenic mice in which a lacZ reporter similar to the RCC1 protein, a guanine nucleotide exchange gene was controlled by the 3-kb upstream region of RPGR. b- factor for the small guanosine triphosphate–binding protein, 9 galactosidase expression was predominantly found in mouse Ran. The RPGR gene that was initially identified contained 19 ex1–19 6,7 retina, brain, and kidney. In the retina, the photoreceptor cell exons (RPGR ), encoding a predicted 90-kDa protein. layer showed the strongest b-galactosidase staining. Exons 2 to 11 encode the RCCL domain, whereas exons 12 to 19 encode a carboxyl (C)-terminal domain rich in acidic CONCLUSIONS. Our study defined the human RPGR proximal residues and ending in an isoprenylation anchorage signal.6,10 promoter region in which a 3-kb fragment contained sufficient Mutations found in RPGRex1–19 account for only 15% to 20% of regulatory elements to control RPGR expression in mouse XLRP patients and subsequent studies revealed many more retina and other tissues. Characterization of the RPGR disease-causing mutations within one or more transcripts containing an alternatively spliced C-terminal exon called ORF15 (RPGRORF15).8 Exon ORF15 encodes a repetitive From the 1Department of Life Sciences, Glasgow Caledonian glycine and glutamic acid–rich domain of unknown function University, Glasgow, United Kingdom; 2MRC Human Genetics Unit, and contains a conserved basic C-terminal domain. The ORF15 Institute of Genetics and Molecular Medicine, Edinburgh, United exon harbors a high frequency of microdeletions, frameshift, Kingdom; and 3Genome Damage and Stability Centre, University of and premature stop mutations.8 In total, 296 RPGR mutations Sussex, Brighton, United Kingdom. have been identified to date, which can give rise to both Supported by RP Fighting Blindness and the Medical Research central and peripheral retinal dystrophies, including X-linked Council(UK)(AFW),theRoyalSocietyofLondon,TENOVUS Scotland,NationalEyeResearchCentre,VisualResearchTrust,the forms of RP (95% of subjects); human cone-rod, cone, and W.H. Ross Foundation, the Rosetrees Trust, the Carnegie Trust for macular dystrophies (3% of subjects); or syndromal forms of the Universities Scotland, and the Nuffield Foundation (XS). XLRP with hearing loss and primary ciliary dyskinesia (2% of Submitted for publication October 17, 2011; revised April 25, subjects).11,12 2012; accepted April 25, 2012. RPGR interacts with a number of photoreceptor and ciliary Disclosure: X. Shu,None;J.R. Simpson,None;A.W. Hart, proteins. The RCCL domain was shown to interact with the None; Z. Zeng,None;S.R. Patnaik,None;P. Gautier,None;E. delta subunit of the rod cyclic GMP phosphodiesterase Murdoch,None;B. Tulloch,None;A.F. Wright,None (PDE6D), a highly conserved protein capable of binding *Each of the following is a corresponding author: Alan F. Wright, several prenylated proteins, including Rab13, Ras, Rap, and MRC Human Genetics Unit, Institute of Genetics and Molecular 13 Medicine, Edinburgh EH4 2XU, United Kingdom; Rho6. The RCCL domain also interacts with the RPGR- [email protected]. interacting protein 1 (RPGRIP1), which, like RPGR, is localized 14–16 Xinhua Shu, Department of Life Sciences, Glasgow Caledonian to the photoreceptor connecting cilium. Mutations in University, 70 Cowcaddens Road, Glasgow G4 0BA, United RPGRIP1 cause a severe early-onset form of retinal degener- Kingdom; [email protected]. ation, Leber’s congenital amaurosis.17,18 The exact function of

Downloaded from iovs.arvojournals.org on 09/28/2021 3952 Shu et al. IOVS, June 2012, Vol. 53, No. 7

RPGRIP1 is unknown, but it is necessary for photoreceptor disc formation and morphogenesis.19 The disease course in the Rpgrip1 knockout (KO) mouse was more severe than that seen in mice lacking RPGR,19,20 which is consistent with the differential severity of these two disorders in humans. RPGR was absent from the connecting cilia in Rpgrip1 KO mice, suggesting that Rpgrip1 is necessary for the proper localization of RPGR.19 The C-terminal domain of the RPGRORF15 isoform was found to interact with nucleophosmin,21 a chaperone involved in many cellular processes, including centrosome duplication, folding of denatured proteins and histone chaper- oning. Co-immunoprecipitation studies showed that RPGR interacts with several basal body/axonemal proteins (CEP290/ NPHP6, NPHP5, IFT88, gamma-tubulin, 14-3-3 epsilon, RPGRIP1L) and microtubule transport proteins (kinesin II– related proteins KIF3A and KAP3, dynein heavy and interme- diate chains, dynactin subunits p150-Glued, and p50-dynami- tin), supporting a role for RPGR in microtubular organization and transport between photoreceptor inner and outer segments.22–24 RPGR is widely expressed and shows a complex expression pattern. At the mRNA level, RPGR transcripts were detected in different tissues, including brain, eye, kidney, lung, and testis in several different species.25–29 At the protein level, RPGR has 0 been detected in retina, trachea, brain, and testis. In human, FIGURE 1. (A) Schematic representation of the 5 region of the human mouse, and bovine retina, RPGR mainly localizes to photore- RPGR gene investigated using reporter constructs generated in the ceptor connecting cilia,29 but expression has also been pGL-3 vector. (B) The corresponding activities of the luciferase reported in outer segments in some species.30 RPGR is reporter gene in RPE1 and HEK 293T cell lines. The promoter-less expressed in the transitional zone of motile cilia and within plasmid pGL3-Basic was used as a negative control and the pGL3- 29,31 Control plasmid containing an SV40 promoter and enhancer was used human and monkey cochlea. Overexpression of mouse as a positive control. The Renilla luciferase plasmid was used as an RPGR results in male infertility because of defects in flagellar internal control for the normalization of transfection efficiency. The formation.32 The severity of the flagellar defect is correlated activities of the reporter gene are expressed as -fold change relative to with increased RPGR copy number, suggesting that RPGR the activity of pGL3-Basic (an activity value of 1.0). expression is tightly controlled. Our study aimed to increase our understanding of the regulation of human RPGR expres- primer, as shown in Supplementary Table S1 (see Supplementary sion and to provide appropriate expression in therapeutic Material, http://www.iovs.org/lookup/suppl/doi:10.1167/iovs. strategies for treating patients with RPGR mutations. 11-8811/-/DCSupplemental). The five PCR fragments: 5098 bp (5070 to þ28 bp), 3093 bp (3065 to þ28 bp), 2005 bp (1977 to þ28 bp), 1508 bp (1480 to þ28 bp), and 980 bp (952 to þ28 bp) MATERIALS AND METHODS were ligated into the pGEM-T Easy vector (Promega, Southampton, Bioinformatic Analysis of Human RPGR Promoter UK), sequenced, and then subcloned into the pGL3 basic vector (Fig. 2A). An additional 222-bp (268 to 47 bp) fragment containing two A 5098 nucleotide bp sequence upstream from human RPGR exon 1 putative Sp1 binding sites was cloned into the pGEM-T Easy vector was analyzed using Gene2promoter software (Genomatix, Munich, using RPGR222bp For and RPGR222bp Rev primers (see Supplemen- Germany), which provides access to promoter sequences of all genes tary Table S1, http://www.iovs.org/lookup/suppl/doi:10.1167/iovs. annotated in the available genomes and predicts the genomic context 11-8811/-/DCSupplemental), then sequenced and subcloned into the of eukaryotic polymerase II promoter sequences (see Supplementary bpGL3 basic vector (222bp-Luc). Material and Supplementary Fig. S1, http://www.iovs.org/lookup/ suppl/doi:10.1167/iovs.11-8811/-/DCSupplemental). If a transcription- Mutagenesis of the Sp1 Binding Sites of Human al start site (TSS) is predicted to be within 200 bp downstream of the RPGR Promoter predicted promoter region, the identified region was marked as a true promoter region. CpG islands in the 50 flanking region were predicted Mutations in two putative Sp1 binding sites in the 222bp-Luc promoter using CpGplot/CpGreport software (http://www.ebi.ac.uk/Tools/ fragment were introduced using a QuickChange site-directed muta- sequence.html). The criteria used to determine a potential island were genesis kit (Stratagene, Stockport, UK). Three mutants were generated: an observed/expected ratio of CpG greater than 0.6; a %Cþ%G greater Sp1A Mut, Sp1B Mut, and Sp1AB Mut (Fig. 3B). The introduction of the than 50.0%; and a window length longer than 200 bp. Vertebrate mutations was verified by DNA sequencing. transcription factor binding sites were identified using MatInspector software, which is a platform for the identification of transcription Cell Culture and Transfections factor binding sites in multiple genomes (Genomatix). Conservation of the putative RPGR promoter region during evolution was analyzed The human telomerase-transformed RPE1 cell line was grown in using Clustalw software (http://www.ebi.ac.uk/). Dulbecco’s modified Eagle’s medium (DMEM):F-12 medium supplemented with 10% fetal calf serum (FCS) and penicillin/streptomycin. Reporter Constructs Human embryonic kidney (HEK) 293T cells were grown in DMEM supplemented with 10% FCS and penicillin/streptomycin. The cells Five fragments that were upstream of the human RPGR exon 1, with were grown in 24-well plates until confluent and transiently progressive 50 to 30 deletions, were amplified using human genomic transfected with the RPGR promoter luciferase reporter constructs, DNA as template (Fig. 2), with five Forward primers and one Reverse together with Renilla luciferase control plasmid, using Lipofectamine

Downloaded from iovs.arvojournals.org on 09/28/2021 IOVS, June 2012, Vol. 53, No. 7 RPGR Proximal Promoter Region 3953

FIGURE 2. Effect of mutating predicted Sp1 sites on RPGR promoter activity in RPE1 and 293T cell lines. (A) The sequence conservation of the two most conserved Sp1 sites (SP1A, SP1B) 50 of the human RPGR gene in 5 mammalian species. (B) Schematic representation of SP1A and SP1B mutation constructs. (C) The promoter activities of wild-type and SP1A or SP1B mutant reporter constructs. The promoter-less plasmid pGL3-Basic was used as a negative control, the pGL3-control plasmid containing an SV40 promoter was used as a positive control. The Renilla luciferase plasmid was used as an internal control for the normalization of transfection efﬁciency. The activities of the reporter gene are expressed as -fold change relative to the activity of pGL3-basic (an activity value of 1.0). Compared with wild type, SP1A, SP1B and SP1AB showed signiﬁcantly reduced activities in both RPE1 and 293T cell lines (** P < 0.0001, t-test).

2000 (Invitrogen/Life Technologies, Paisley, UK). The pGL3-Control cytoplasm was removed, and the nuclear pellet was washed and vector containing SV40 promoter and enhancer sequences was used as resuspended in 200 lL PBS. For the depletion of Sp1, 0.1 mg HeLa cell a positive control and the empty pGL3-Basic vector as a negative nuclear extract was precleared with 50 lL protein G-Sepharose beads control. (Sigma) for 2 hours at 48C, then incubated with 20 lL anti-Sp1 antibody (0.2 lg/lL, sc-59; Santa Cruz Biotech, Santa Cruz, CA) on a carousel at Luciferase Assays 48C overnight. This was followed by addition of 30 lL of protein G- Sepharose beads and incubation at 48C for 1 hour with gentle agitation. Cells transiently transfected with the luciferase constructs were The Sp1-binding protein G-Sepharose beads were removed by cultured for a further 48 hours and washed twice with PBS. Cell centrifugation so that the supernatant contained Sp1-depleted nuclear lysates were prepared by adding lysis buffer (Passive Lysis buffer; extract. For preparation of the probe, 30-bp oligonucleotides compris- Promega) and incubated for 30 minutes; 20 lL of cell lysates were ing Sp1 binding sequence (wild type or mutant, as underlined below) transferred to a 96-well plate containing 100 lL luciferase assay reagent were commercially synthesized. The oligonucleotide sequences are as (LAR II, Promega). The activity of firefly luciferase was measured first follows: Sp1A wild type, 50-CATTCCCAAGCTCCGCCCCCGTTGCCCG- and Renilla luciferase activity was measured after addition of 100 lLof TA-30; Sp1A Mutant, 50-CATTCCCAAGCTGAATTCCCGTTGCCCGTA-30); Stop & Glo reagent (Promega). Experiments were performed at least Sp1B wild type, 50-GGCCTCCGTTCCCCTCCCCAACGGCGCCTG-30; three times. Sp1B mutant, 50-GGCCTCCGTTGAATTCCCCAACGGCGCCTG-3. Those oligonucleotides were annealed and labeled with a-32P dCTP using 0 Electrophoretic Mobility Shift Assay (EMSA) Klenow DNA polymerase, the 3 -labeled probes were purified by G25- columns (GE Healthcare, Little Chalfont, Buckinghamshire, UK). The Nuclear extracts were made from HeLa cells, which have Sp1 Sp1 binding reactions were carried out by incubating the indicated expression and are commonly used for Sp1 binding assays. Briefly, 2 labeled-probe (7.5 nM) with 1 lL of HeLa cell nuclear extract (16 lg/lL) · 107 HeLa cells were harvested by trypsinization and resuspended in 1 in binding buffer (20 mM Tris pH 7.5, 100 mM NaCl, 0.2 mg/mL BSA,

mL fractionation buffer (15 mM Tris pH 7.5, 0.3 M sucrose, 15 mM NaCl 10% glycerol, 2 mM MgCl2, and 1 mM DTT) at room temperature for 20 , 5 mM MgCl2, 0.1 mM EGTA, 0.5 mM dithiothreitol [DTT], 0.1 mM minutes. All reactions were terminated by the addition of loading buffer phenylmethylsulfonylﬂuoride). An equal volume of fractionation buffer containing 50% glycerol and 5· loading dye (Qiagen, Crawley, West containing 0.4% IGEPAL (Sigma, Irvine, UK) was added and the Sussex, UK). The DNA-protein complexes were separated by electro- suspension was mixed and incubated on ice for 10 minutes. The lysate phoresis on a 4% native polyacrylamide gel in 0.4· Tris borate-EDTA was then layered onto 5 mL extraction buffer containing 1.2 M sucrose (TBE) at 120 V for 2 hours at room temperature, vacuum-dried, and then and spun at 10,000g for 20 minutes at 48C. The top layer containing the autoradiographed.

Downloaded from iovs.arvojournals.org on 09/28/2021 3954 Shu et al. IOVS, June 2012, Vol. 53, No. 7

samples were washed three times in PBS containing 0.01% Na- deoxycholate, 0.02% Nonidet P40, 5 mM EGTA (pH 8.0), and 2 mM

MgCl2, stained overnight at 378C in a solution of 0.5 mg/mL 5-bromo-4- chloro-3-indolyl b-D-galactopyranoside (X-gal) containing 2 mM MgCl2,5 mM K3Fe(CN)6, 5 mM K4Fe(CN)6, 0.01% Na deoxycholate, and 0.02% NP40. After washing three times, tissues were postﬁxed in 4% PFA. Sections of the stained eye were cut at a thickness of 6 lmanddewaxed. Images were captured using a microscope and analyzed using IPLab software (Digital Imaging Systems, Bourne End, UK). For immunostaining to detect b-galactosidase in the retina of transgenic and nontransgenic mice, the sections were de-waxed, blocked with 2% BSA-PBS, incubated with mouse antirhodopsin antibody (1:200; Abcam, Cambridge, UK ) and rabbit anti-b-galactosidase antibody (1:200; Cortex Biochem, Inc., San Leandro, CA), and subsequently incubated with Texas-Red or FITC-conjugated secondary antibody. Sections were mounted in Vectashield (Vector Laboratories Ltd, Peterborough, UK) containing 4-6-diamidino-2-phenylindole (DA- PI). Images were captured using a confocal microscope (LSM 510; Carl Zeiss Inc., Welwyn Garden City, UK).

RESULTS Bioinformatic Characterization of the RPGR Upstream Region

FIGURE 3. EMSA showing Sp1 binding to its two predicted sites in the To identify a potential promoter in the upstream region of the RPGR promoter. The gel shift assay was performed with the nuclear human RPGR gene, 5094 bp of sequence (5066 to þ28 bp) extract isolated from HeLa cells and with radiolabeled probes derived including RPGR exon 1 was analyzed using Gene2Promoter from the RPGR promoter region containing putative Sp1 binding sites. software (Genomatix). A potential promoter of 641 bp ( 683 The Sp1/DNA complex occurred with wild-type oligonucleotides (WT- to 42 bp) and two TSSs (142 and 168 bp) were identified. probe, Sp1A: lane 2; Sp1B: lane 8), no complex was formed with To investigate the predicted TSSs further, expressed sequence mutant oligonucleotides (Mut-probe, Sp1Amut: lane 5; Sp1Bmut: lane 11). The anti-Sp1 antibody depletion assay showed ablation of the Sp1/ tags (ESTs), including the upstream region of RPGR, were DNA complex (Lanes 3 and 9). In the control (no nuclear extract searched, and 40 ESTs were found and aligned to the RPGR added), no complex was seen (lanes 1, 4, 7, 10). NB, nonspecific upstream sequence. Four potential TSSs (197, 175, 168, binding. and 142 bp) were identified, including the most commonly used 50 transcription start site (142 bp), which was 142 nucleotides upstream of the start codon (see Supplementary Generation and Analysis of Transgenic Mice Fig. S1, http://www.iovs.org/lookup/suppl/doi:10.1167/iovs. All studies were performed in accordance with the ARVO Statement for 11-8811/-/DCSupplemental). the Use of Animals in Ophthalmic and Vision Research and under the To identify the regulatory elements in the upstream region guidance of the Medical Research Council in Responsibility in the Use of of RPGR, 1508 bp (1480 to þ28 bp) of sequence was analyzed Animals for Medical Research (July 1993) and UK Home Office Project using the MatInspector program to search the TRANSFAC License 60/3785. For each micro-injection session, 10 F1 (B6CBAF1/Crl) database; 142 possible transcription-factor binding sites were female mice (approximately 4 weeks of age) were superovulated by identified within the sequence. Some selected consensus injecting 5 units of pregnant mare serum gonadotropin (National sequences for transcription factor binding sites are shown in Hormone and Pituitary Program, Bethesda, MD) on day 1 (2 PM) followed Supplementary Figure S1 (see Supplementary Fig. S1, http:// by a second intraperitoneal injection of human chorionic gonadotropin www.iovs.org/lookup/suppl/doi:10.1167/iovs.11-8811/-/ 0 on day 3 at 12 noon. At 4 PM on day 3, injected females were mated with DCSupplemental). The RPGR 5 -upstream region contained F1 (B6CBAF1/Crl) male studs and checked the following morning (day three E-box sites (1486 to 1456, 253 to 241, and 123 to 4) for a copulation plug. Superovulated females were killed on day 4 and 111 bp); three Octamer sites (1030 to 1014, 992 to 966, embryos were harvested by flushing oviducts with fresh FHM (Millipore) and 588 to 572 bp); three Sp1 sites (683 to 669, 236 to

medium and collected embryos were transferred into CO2-equilibrated 226, and 196 to 177 bp) and one AP4 site (506 to 490 KSOM medium (Millipore) overlaid with mineral oil (Sigma) and bp). In addition, a possible TATA box was identiﬁed at position

incubated at 378Cin5%CO2 until required. Batches of 40 to 50 715 to 699 bp and a CCAAT box at position 598 to 584 embryos were microinjected with linearized double-stranded DNA (3093 bp. Many blocks of sequences are conserved in the RPGR 50- bp RPGR promoter region [3065 toþ28 bp]), fused with a downstream upstream region between human and mouse, including most lacZ gene, into one of the pronuclei of the 0.5-day embryo. Injected of the above mentioned potential transcription factor binding embryos were incubated overnight at 378Cin5%CO2 in KSOM medium. sites (see Supplementary Fig. S2, http://www.iovs.org/lookup/ On day 5, viable 2-cell embryos were then transferred into 0.5-day suppl/doi:10.1167/iovs.11-8811/-/DCSupplemental). pseudopregnant CD1 recipient females. Pregnancies were allowed to go We screened the RPGR gene upstream sequence of 5098 bp to term and at 3 weeks of age pups were screened to determine positive (5070 to þ28 bp) for CpG islands using the CpG island transgenic founders by PCR using ear clip DNA. Primers for the PCR searcher software (http//www.ebi.ac.uk). One strongly pre- screen were a forward primer 50 GTTGCGCAGCCTGAATGGCG 30 and a dicted CpG island of 249 bp from 276 to 27 bp was reverse primer 50GCCGTCACTCCAACGCAGCA 30. Transgenic founders identified with 67.5% CG content. The observed CpG/ were crossed to CD1 mice to generate F1 progeny. Eyes, brains, kidneys, expected CpG ratio in this region was greater than 0.65 (see and lungs were dissected from transgenic mice at the age of 3 months Supplementary Fig. S3, http://www.iovs.org/lookup/suppl/ and fixed in 4% paraformaldehyde (PFA) overnight at 48C. The fixed doi:10.1167/iovs.11-8811/-/DCSupplemental). This suggests

Downloaded from iovs.arvojournals.org on 09/28/2021 IOVS, June 2012, Vol. 53, No. 7 RPGR Proximal Promoter Region 3955

FIGURE 4. Human RPGR gene 3098-bp upstream promoter-driven lacZ activities in different tissues from transgenic mice. No lacZ staining with X- gal was present in nontransgenic mice.

that this CpG island may have an important role in regulating and 293T cells. The mutation introduced into Sp1 binding site RPGR transcription. B (SP1Bmut) reduced luciferase activity by 47.0% in RPE1 cells and 32% in HEK 293T cells, respectively (P < 0.0001). The Analysis of the Transcriptional Activity of the mutation introduced into Sp1 binding site A (SP1Amut) RPGR Upstream Region dramatically reduced luciferase activity, by 70% in RPE1 cells and 80% in HEK 293T cells, respectively (P < 0.0001). The To investigate the transcriptional activity of the upstream combination of these two mutations, SP1A and SP1B, caused sequence of the human RPGR gene, a series of luciferase only a slightly greater reduction in luciferase activity, by 90% in reporter gene constructs were generated in the pGal3 vector RPE1 cells and 84% in HEK 293T cells, compared with the (Fig. 1A). The 293T or RPE1 cells were transiently transfected SP1A single mutant (Fig. 2C). This suggests that the predicted with individual reporter gene constructs or with the pGL3- Sp1 binding site SP1A is likely to be a critical RPGR promoter basic control, together with a Renilla luciferase transfection element. To demonstrate that Sp1 interacts with the two control plasmid. The luciferase activity data from three putative elements, the radiolabeled oligonucleotide probes independent experiments were normalized for transfection spanning each of the two Sp1 sites were prepared (sequences efficiency. In both 293T and RPE1 cell lines, luciferase activities shown in Materials and Methods). Gel-shift experiments increased progressively with deletions in a 50 to 30 direction showed HeLa nuclear extracts produced a band shift when (Fig. 1), The fragment containing 5070 to þ28 bp directed incubated with wild-type probes (SP1A and SP1B), but no band luciferase expression 10.5-fold and 12.2-fold higher than the was observed with the mutant probes (SP1A and SP1B background pGL3-Basic vector in RPE1 and 293T cells, mutants) (Fig. 3). The band shift disappeared when Sp1- respectively. The highest luciferase activities of the five depleted nuclear extract was incubated with wild-type probes, constructs was directed by the most proximal fragments of suggesting Sp1-specific binding (Fig. 3). 222 bp (268 to 47 bp) and 980 bp (952 to þ28 bp), which gave 44-fold higher expression in RPE1 cells and 38.2-fold Analysis of RPGR Promoter Region–Driven lacZ higher expression in HEK293 T cells than the background Expression in Transgenic Mice pGL3-Basic vector (Fig. 1). To analyze the expression pattern of lacZ directed by the Definition of Two Sp1 Sites within the RPGR putative RPGR promoter in vivo, we made a construct Promoter Region containing the upstream sequence (3065 to þ28 bp) of the RPGR gene coupled to the lacZ gene/SV40 poly(A) signal There are three putative SP1 binding sites in the 1508-bp sequence. Microinjection of the construct into fertilized sequence (1480 to þ28 bp) (see Supplementary Fig. S1, oocytes generated several founder lines. We analyzed the http://www.iovs.org/lookup/suppl/doi:10.1167/iovs.11-8811/ expression of the RPGR promoter-lacZ gene by X-gal staining -/DCSupplemental). The first putative Sp1 binding site for b-galactosidase activity in a variety of tissues from the consensus sequence (SP1; 683 to 669 bp) showed less resultant transgenic mice, including brain, lung, kidney, and evolutionary conservation than the other two Sp1 binding site eye. The transgene was highly expressed in brain (olfactory consensus sequences, SP1B (236 to 226 bp) and SP1A (196 lobes, cerebrum, cerebellum, and pineal gland, dorsal view) to 187 bp) across mammalian species (Fig. 2A and data not and retina, and was expressed at a lower level in trachea, renal shown). We therefore decided to identify the function of the pelvis, and renal cortex (Fig. 4). In sections of the X-gal–stained two most conserved Sp1 binding sites (SP1A and SP1B) in the eyes, the blue signal was seen in retina but not in RPE, choroid, RPGR promoter by constructing mutated derivatives of these sclera, lens, or cornea. X-gal staining of retinal sections showed sites in the RPGR promoter fragment (268 to 47 bp) cloned that the transgene was mainly expressed in the outer nuclear into the pGL3-Basic vector (Fig. 2B) and transfected into RPE1 layer (rods and cones) and demonstrated a dorsal to ventral

Downloaded from iovs.arvojournals.org on 09/28/2021 3956 Shu et al. IOVS, June 2012, Vol. 53, No. 7

islands.37 Most promoter-associated CpG islands are unmethy- lated, whereas methylation of the CpG sites in the promoter of a gene may inhibit its expression. The CpG island–searching software strongly predicted a CpG island in the 50 region of the human RPGR gene. The four predicted TSSs are in the CpG island, consistent with the broad distribution of TSSs in CpG islands.34 The multiple TSSs and CpG island in the RPGR promoter region may contribute to the regulation of its expression in different tissues. The transcriptional activity of the 50 flanking region of the human RPGR gene was assayed in the 293T and RPE1 cell lines, which showed that the 980-bp fragment upstream of the ATG start codon had the highest luciferase reporter activity, suggesting that it contains the DNA elements required for RPGR transcription.33 The transcriptional activity was almost abol- ished when two predicted Sp1 sites (SP1A, SP1B) were mutated. The position of the Sp1 binding motif (especially SP1A) appears to be critical for the high basal activity of the RPGR promoter. FIGURE 5. Dorsal-ventral expression of human RPGR promoter-driven This Sp1 site is also highly conserved during evolution (Fig. 2A) lacZ activity detected by X-gal staining in retinal sections from and lies close to the most common TSS (142 bp). Sp1 is a transgenic mice is shown in (A). Stronger lacZ activity was present member of the Sp family of transcription factors, which bind to in the outer nuclear layer in the dorsal retina (B) than in the ventral retina (C). No dorsal-ventral gradient distribution of staining was the consensus GC box element and regulate gene expression lacZ 38 seen in the inner nuclear or ganglion cell layers. during both physiological and pathological processes. Sp1 is ubiquitously expressed in mammalian cell lines and tissues and both Sp1 and Sp4 are expressed in retina.39,40 Sp1 has a role in gradient across the retina. Weaker X-gal staining was also seen regulating the transcription of photoreceptor-specific genes: it in the inner nuclear and ganglion cell layers (Fig. 5). can activate the rod opsin promoter and competitively represses Immunostaining with anti-b-galactosidase antibody showed Sp4-mediated activation of the b-subunit of rod-specific cGMP- that b-galactosidase was expressed in the outer nuclear layer phosphodiesterase (b-PDE) promoter when co-expressed with at a high level and in the inner nuclear layer at a low level (see Sp4. Sp1, Sp3, and Sp4 interact with the photoreceptor-enriched Supplementary Fig. S4, http://www.iovs.org/lookup/suppl/ Crx transcription factor and exhibit functional synergy with Crx doi:10.1167/iovs.11-8811/-/DCSupplemental), consistent with at the rod opsin promoter but not the b-PDE promoter.41 Crx is the expression pattern of the transgene detected by lacZ a central regulator of many photoreceptor genes and plays a staining. critical role in photoreceptor cell differentiation and mainte- nance.42 Mutations in CRX cause cone-rod dystrophy, RP, and Leber’s congenital amaurosis.43 Crx regulates photoreceptor DISCUSSION gene transcription through functional interactions with photoreceptor-specific transcription factors (Nrl, Nr2e3, and Crx) and In an effort to understand the molecular basis of RPGR gene with ubiquitously expressed transcription factors, such as Sp regulation, we have characterized the proximal promoter family members.41-42 Our results demonstrate that Sp1 regulates region of human . Mammalian promoters can be RPGR RPGR transcription, although other photoreceptor-specific separated into two classes: TATA box enriched promoters, in transcription factors, such as Crx and Nrl, may also be involved which a TATA box directs transcription from a defined site; and in its regulation. TATA-less promoters, which contain multiple TSSs.33 The TATA RPGR transcripts have been detected in several tissues from box was the first identified eukaryotic core promoter element human and other species.25–28 The 3093-bp (3065 to þ28 bp) and its consensus sequence binds the TATA-box binding 33 RPGR promoter can direct the expression of the lacZ reporter protein, which is a subunit of the pre-initiation complex. gene in retina, brain, kidney, and trachea (Fig. 4), consistent The TATA box is usually located 25 to 34 bp upstream of the with the known expression pattern of native RPGR. In retina, TSS and directs the transcriptional apparatus to a single defined the 3093-bp promoter was able to drive reporter 33,34 RPGR lacZ nucleotide position. The TATA box is uncommon in expression, most strongly in the photoreceptor cell layer but vertebrates since only a minority of mouse and human genes also weakly in the inner nuclear and ganglion cell layers (Fig. 33 have classical TATA box promoters. Most mammalian genes 5), in agreement with earlier reports.30,44 One explanation as have TATA-less promoters with multiple TSSs spreading over to why RPGR mutations only affect retinal function despite this 35 50 to 100 bp. EST data showed that RPGR has four TSSs and a gene’s wider tissue expression is because RPGR has several predicted TATA box (715 to 699 bp) in its promoter but it is different isoforms, only some of which are involved in ocular located approximately 500 bp upstream from the four TSSs, disease (exon ORF15- containing isoforms). RPGR is predom- suggesting that RPGR has a TATA-less promoter. Genes with inantly expressed in ciliated cells, which are abundant in TATA-less promoters are usually expressed more widely in respiratory tract and renal tubules as well as retina, but the tissues than those with a functional TATA box. The expression ORF15-containing isoform(s), is most abundant in nonmotile pattern of RPGR in adult tissues25–28 further supports the cilia, which are present in a restricted set of tissues, including nature of the RPGR promoter as a TATA-less promoter. photoreceptors and cochlear cells.21,29,30 The proximal pro- CpG islands represent genomic regions of DNA with a high moter that we have identified is likely to regulate basal GþC content and a high frequency of CpG dinucleotides.36 A expression of both ORF15 and non-ORF15–containing iso- CpG island is usually defined as a region of at least 200 bp in forms. length, with a GC percentage greater than 50% and an Another interesting finding is that the RPGR promoter- observed/expected CpG ratio greater than 60%.36 Approxi- driven lacZ reporter shows a gradient of expression from mately 70% of human promoters contain a high CpG content dorsal to ventral in the retina (Fig. 5). In mice, M opsin has a and 50% of human promoters are associated with CpG similar gradient of expression from dorsal to ventral but S opsin

Downloaded from iovs.arvojournals.org on 09/28/2021 IOVS, June 2012, Vol. 53, No. 7 RPGR Proximal Promoter Region 3957

has a gradient of expression from ventral to dorsal.45 RPGR KO allows the identification of mutations underlying Leber mice show mislocalization of both M (green) and S (blue) cone congenital amaurosis. Eur J Hum Genet. 2001;9:561–571. opsins,20 suggesting that the dorsal-to-ventral gradient of RPGR 19. Zhao Y, Hong DH, Pawlyk B, et al. The retinitis pigmentosa expression is unlikely to be functionally related to its proposed GTPase regulator (RPGR)-interacting protein: subserving role in the transport of cone opsins.20 RPGR function and participating in disk morphogenesis. Proc Natl Acad Sci U S A. 2003;100:3965–3970. References 20. Hong DH, Pawlyk BS, Shang J, Sandberg MA, Berson EL, Li T. A retinitis pigmentosa GTPase regulator (RPGR)-deficient mouse 1. Wright AF, Chakarova CF, Abd El-Aziz M, Bhattacharya SS. model for X-linked retinitis pigmentosa (RP3). Proc Natl Acad Photoreceptor degeneration—genetic and mechanistic dissec- Sci U S A. 2000;97:3649–3654. tion of a complex trait. Nat Rev Genet. 2010;11:273–284. 21. Shu X, Fry AM, Tulloch B, et al. RPGR ORF15 isoform co- 2. Bramall A, Wright AF, Jacobson SG, McInnes RR. The genomic, localizes with RPGRIP1 at centrioles and basal bodies and interacts with nucleophosmin. . 2005;14: biochemical and cellular responses of the retina in inherited Hum Mol Genet 1183–1197. photoreceptor degenerations, and prospects for the treatment of these disorders. Annu Rev Neurosci. 2010;33:441–472. 22. Khanna H, Hurd TW, Lillo C, et al. RPGR-ORF15, which is mutated in retinitis pigmentosa, associates with SMC1, SMC3, 3. Kajiwara K, Berson EL, Dryja TP. Digenic retinitis pigmentosa and microtubule transport proteins. J Biol Chem. 2005;280: due to mutations at the unlinked peripherin/RDS and ROM1 33580–33587. loci. Science. 1994;264:1604–1608. 23. Chang B, Khanna H, Hawes N, et al. In-frame deletion in a 4. Hussels-Maumenee I, Pierce ER, Bias WB, Schleutermann DA. novel centrosomal/ciliary protein CEP290/NPHP6 perturbs its Linkage studies of typical retinitis pigmentosa and common interaction with RPGR and results in early-onset retinal markers. Am J Hum Genet. 1975;27:505–508. degeneration in the rd16 mouse. Hum Mol Genet. 2006;15: 5. Breuer DK, Yashar BM, Filippova E, et al. A comprehensive 1847–1857. mutation analysis of RP2 and RPGR in a North American 24. Khanna H, Davis EE, Murga-Zamalloa CA, et al. A common cohort of families with X-linked retinitis pigmentosa. Am J allele in RPGRIP1L is a modifier of retinal degeneration in Hum Genet. 2002;70:1545–1554. ciliopathies. Nat Genet. 2009;41:739–745. 6. Meindl A, Dry K, Herrmann K, et al. A gene (RPGR) with 25. Kirschner R, Rosenberg T, Schultz-Heienbrok R, et al. RPGR homology to the RCC1 guanine nucleotide exchange factor is transcription studies in mouse and human tissues reveal a mutated in X-linked retinitis pigmentosa (RP3). Nat Genet. retina-specific isoform that is disrupted in a patient with X- 1996;13:35–42. linked retinitis pigmentosa. Hum Mol Genet. 1999;8:1571– 7. Roepman R, van Duijnhoven G, Rosenberg T, et al. Positional 1578. cloning of the gene for X-linked retinitis pigmentosa 3: 26. Shu X, Zeng Z, Eckmiller MS, et al. Developmental and tissue homology with the guanine-nucleotide exchange factor expression of Xenopus laevis RPGR. Invest Ophthalmol Vis RCC1. Hum Mol Genet. 1996;5:1035–1041. Sci. 2006;47:348–356. 8. Vervoort R, Lennon A, Bird AC, et al. Mutational hot spot 27. Shu X, Zeng Z, Gautier P, et al. Zebrafish RPGR is required for within a new RPGR exon in X-linked retinitis pigmentosa. Nat normal retinal development and plays a role in dynein-based Genet. 2000;25:462–466. retrograde transport processes. Hum Mol Genet. 2010;19: 9. Renault L, Nassar N, Vetter I, et al. The 1.7 A crystal structure 657–670. of the regulator of chromosome condensation (RCC1) reveals 28. Zhang Q, Acland GM, Wu WX, et al. Different RPGR exon a seven-bladed propeller. Nature. 1998;392:97–101. ORF15 mutations in Canids provide insights into photorecep- 10. Wright AF, Shu X. Focus on molecules: RPGR. Exp Eye Res. tor cell degeneration. Hum Mol Genet. 2002;11:993–1003. 2007;85:1–2. 29. Hong DH, Pawlyk B, Sokolov M, et al. RPGR isoforms in 11. Shu X, Black GC, Rice JM, et al. RPGR mutation analysis and photoreceptor connecting cilia and the transitional zone of motile cilia. . 2003;44:2413–2421. disease: an update. Hum Mutat. 2007;28:322–328. Invest Ophthalmol Vis Sci 30. Mavlyutov TA, Zhao H, Ferreira PA. Species-specific subcellular 12. Shu X, McDowall E, Brown AF, Wright AF. The human retinitis localization of RPGR and RPGRIP isoforms: implications for pigmentosa GTPase regulator gene variant database. Hum the phenotypic variability of congenital retinopathies among Mutat. 2008;29:605–608. species. Hum Mol Genet. 2002;11:1899–1907. 13. Linari M, Ueffing M, Manson F, Wright A, Meitinger T, Becker J. 31. Iannaccone A, Breuer DK, Wang XF, et al. Clinical and The retinitis pigmentosa GTPase regulator, RPGR, interacts immunohistochemical evidence for an X linked retinitis with the delta subunit of rod cyclic GMP phosphodiesterase. pigmentosa syndrome with recurrent infections and hearing Proc Natl Acad Sci U S A. 1999;96:1315–1320. loss in association with an RPGR mutation. J Med Genet. 2003; 14. Boylan JP, Wright AF. Identification of a novel protein 40:e118. interacting with RPGR. Hum Mol Genet. 2000;9:2085–2093. 32. Brunner S, Colman D, Travis AJ, et al. Overexpression of RPGR 15. Roepman R, Bernoud-Hubac N, Schick DE, et al. The retinitis leads to male infertility in mice due to defects in flagellar pigmentosa GTPase regulator (RPGR) interacts with novel assembly. Biol Reprod. 2008;79:608–617. transport-like proteins in the outer segments of rod photore- 33. Sandelin A, Carninci P, Lenhard B, Ponjavic J, Hayashizaki Y, ceptors. Hum Mol Genet. 2000;9:2095–2105. Hume DA. Mammalian RNA polymerase II core promoters: 16. Hong DH, Yue G, Adamian M, Li T. Retinitis pigmentosa insights from genome-wide studies. Nat Rev Genet. 2007;8: GTPase regulator (RPGRr)-interacting protein is stably associ- 424–436. ated with the photoreceptor ciliary axoneme and anchors 34. Wray GA, Hahn MW, Abouheif E, et al. The evolution of RPGR to the connecting cilium. J Biol Chem. 2001;276: transcriptional regulation in eukaryotes. Mol Biol Evol. 2003; 12091–12099. 20:1377–1419. 17. Dryja TP, Adams SM, Grimsby JL, et al. Null RPGRIP1 alleles in 35. Frith MC, Valen E, Krogh A, Hayashizaki Y, Carninci P, Sandelin patients with Leber congenital amaurosis. Am J Hum Genet. A. A code for transcription initiation in mammalian genomes. 2001;68:1295–1298. Genome Res. 2008;18:1–12. 18. Gerber S, Perrault I, Hanein S, et al. Complete exon-intron 36. Gardiner-Garden M, Frommer M. CpG islands in vertebrate structure of the RPGR-interacting protein (RPGRIP1) gene genomes. J Mol Biol. 1987;196:261–282.

Downloaded from iovs.arvojournals.org on 09/28/2021 3958 Shu et al. IOVS, June 2012, Vol. 53, No. 7

37. Antequera F, Bird A. Number of CpG islands and genes in 42. Hennig AK, Peng GH, Chen S. Regulation of photoreceptor human and mouse. Proc Natl Acad Sci U S A. 1993;90:11995– gene expression by Crx-associated transcription factor net- 11999. work. Brain Res. 2008;1192:114–133. 38. Tan NY, Khachigian LM. Sp1 phosphorylation and its 43. Rivolta C, Berson EL, Dryja TP. Dominant Leber congenital regulation of gene transcription. Mol Cell Biol. 2009;29: 2482–2488. amaurosis, cone-rod degeneration, and retinitis pigmentosa 39. Hagen G, Muller S, Beato M, Suske G. Cloning by recognition caused by mutant versions of the transcription factor CRX. site screening of two novel GT box binding proteins: a family Hum Mutat. 2001;18:488–498. of Sp1 related genes. Nucleic Acids Res. 1992;20:5519–5525. 44. Trifunovic D, Karali M, Camposampiero D, Ponzin D, Banfi S, 40. Lerner LE, Gribanova YE, Ji M, Knox BE, Farber DB. Nrl and Sp Marigo V. A high-resolution RNA expression atlas of retinitis nuclear proteins mediate transcription of rod-specific cGMP- pigmentosa genes in human and mouse retinas. Invest phosphodiesterase b-subunit gene. J Biol Chem. 2001;276: Ophthalmol Vis Sci. 2008;49:2330–2336. 34999–35007. 41. Lerner LE, Peng G, Gribanova YE, Chen S, Farber D. Sp4 is 45. Applebury ML, Antoch MP, Baxter LC, et al. The murine cone expressed in retinal neurons, activates transcription of photoreceptor: a single cone type expresses both S and M photoreceptor-specific genes, and synergizes with Crx. J Biol opsins with retinal spatial patterning. Neuron. 2000;27:513– Chem. 2005;280:20642–20650. 523.

Downloaded from iovs.arvojournals.org on 09/28/2021