Endogenous CRX expression and IRBP promoter activity in retinoblastoma cells

Brain Research 916 (2001) 136–142 www.elsevier.com / locate / bres

Research report

Endogenous CRX expression and IRBP promoter activity in retinoblastoma cells a, b a a Jeffrey H. Boatright *, Diane E. Borst , Eva Stodulkova , John M. Nickerson a Emory Eye Center, Atlanta, GA, USA Uniformed Services University of Health Sciences, Bethesda, MD, USA

b

Accepted 23 July 2001

Abstract Purpose: To determine whether antisense oligonucleotides (AODNs) targeted against CRX, a photoreceptor-specific trans-acting factor, suppress CRX expression and interphotoreceptor retinoid binding protein (IRBP) promoter activity. Methods: Cultures of human retinoblastoma cells were transfected with chloramphenicol acetyltransferase (CAT) reporter plasmids containing a mouse IRBP promoter and AODNs directed against CRX. RT–PCR using primers specific to CRX, OTX2, GAPDH, or RNase H was conducted on total RNA isolated from retinoblastoma cells at various times following transfection with AODNs. Results: Transfection of retinoblastoma cells with IRBP promoter CAT constructs alone produced high activity. Co-transfection with AODNs suppressed IRBP promoter activity in a concentration-dependent manner, with half-maximal effect produced at about 2 nM AODN concentration. Transfection with CAT constructs containing an SV40 promoter produced high activity that was unaffected by co-transfection with AODNs. RT–PCR products were obtained for all target sequences. CRX RT–PCR product from cells transfected with AODNs was greatly diminished following transfection with an AODN whereas control transcripts, including that of OTX2, were relatively unaffected. Conclusions: The CRX-specific AODNs specifically and potently suppressed CRX expression and IRBP promoter activity, as measured by RT–PCR and transient transfection assays, respectively. Little or no effect was seen on controls. These data suggest that endogenous CRX is required for IRBP promoter activity in retinoblastoma cells.  2001 Elsevier Science B.V. All rights reserved. Keywords: IRBP; Retinoblastoma; Antisene; CRX; Quantitative PCR

1. Introduction Interphotoreceptor retinoid binding protein (IRBP) is a large glycolipoprotein found abundantly in the interphotoreceptor matrix of many vertebrates [6,7,18,19]. IRBP protein is detected only in photosensitive tissues [23] and its mRNA is detected uniquely in photoreceptor cells and pinealocytes [20,21]. The 59 flanking sequence of the IRBP gene directs promoter activity only in the retina and pineal [3,16]. A short fragment, 2123 to 118 relative to transcription start, of the IRBP gene is sufficient to direct retina-specific expression in transgenic mice [5,10]. Shorter 59 flanking fragments (270 to 1101) retain neuronal specificity but lose the stricter photoreceptor cell-type specificity [1,3]. An element of the promoter, often re*Corresponding author. Department of Ophthalmology, Emory University, Room B5511, 1365B Clifton Road NE, Atlanta, GA 30322, USA. Tel.: 11-404-778-4113; fax: 11-404-778-2231. E-mail address: [email protected] (J.H. Boatright).

ferred to as the CRX binding element [1,10,12], at positions 256 to 250 in the mouse IRBP promoter, is essential for IRBP promoter activity [1,4,5,8,10,12]. Two homeobox transcription factors have been proposed to bind to this site, CRX [1,8,10,12] and OTX2 [4,11]. In yeast one-hybrid screens against a bovine retinal cDNA library, CRX cDNA clones were detected using multiple copies of the opsin Ret4 consensus element as bait [8]. Similar screening experiments with a human retina cDNA library detected OTX2 but no CRX cDNA clones when using the CRX binding element as the bait [11]. Forced expression of either OTX2 or CRX transactivates the IRBP promoter in non-retina cells and the CRX binding element is required for this activity [4,10,11,15]. Mutation of the CRX binding element in a 2123 to 118 IRBP fragment precludes reporter gene expression in transgenic mice [5,10]. These results indicate that the CRX binding element is required for IRBP promoter activity; OTX2 and CRX may play roles in endogenous IRBP transcription by binding these sites.

0006-8993 / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 01 )02884-0

J.H. Boatright et al. / Brain Research 916 (2001) 136 – 142

Fig. 1. Proposed mechanism for AODN effect and considerations in oligonucleotide design.

Cepko and colleagues [13] examined mice lacking the CRX gene. The CRX-deficient mice exhibited changes in mRNA levels in many retina-specific mRNAs, abnormal circadian rhythms, and diminished ERGs, suggesting that the CRX gene is critical to the survival and function of the photoreceptor cell. However, little change was detected in the mRNA levels of IRBP and several other retina genes. Thus, the CRX gene can be compensated for or is not necessary for the normal mRNA levels of these genes. Compensatory mechanisms may include the substitution of the OTX2 gene for the absent CRX, though no evidence concerning OTX2 mRNA levels were reported by either Northern or microarray analyses [13,17]. We wondered if a transient decrease in endogenous Table 1

137

CRX expression might affect the activity of the IRBP promoter in a retinoblastoma cell line. To gain insight into whether IRBP promoter activity is sensitive to endogenous CRX levels, we sought to determine the effects of antisense oligonucleotides (AODNs) designed against human CRX on CRX expression and on the activity of a truncated (270 to 1101, relative to transcription start) IRBP promoter [1,2] transiently transfected into human retinoblastoma cells (WERI-Rb1). AODN strategies are based in part on the mechanism of RNaseH hydrolyzing the RNA strand of RNA–DNA hybrids formed when an AODN binds its complementary, target mRNA [9,14,22] (Fig. 1). In the present experiments, we found that AODNs designed against CRX mRNA reduce the levels of CRX mRNA and IRBP promoter activity while having little effect on OTX2 and GAPDH mRNAs. We also found that RNaseH1a, but not RNAseH2, exist in WERI cells and that properly designed propyne- and phosphorothioate-modified AODNs are efficacious, potent, and selective tools for studying gene regulation in retinoblastoma cells.

2. Methods

2.1. AODN Design Three AODN sequences are given in Table 1A. These oligonucleotides contained phoshorothioates replacing the phosphates and the pyrimidines contained C-5 propynyl groups [22]. The targets were selected near putative ribosome binding sites and the beginning of translation. We used MacVector (Oxford Molecular Group, Madison, WI) to design the oligonucleotides to minimize binding to OTX2 and other mRNAs. The oligonucleotides were synthesized locally using propyne–phosphorothioate bases from Cruachem (Annovis, Aston, PA) and used with and without purification by ion pair high-performance liquid chromatography (HPLC).

2.2. Plasmid construction The CAT plasmid vectors used were based on the promoterless CAT vector pBLCAT3 (GenBank Accession number X64409), which served as a negative control. The positive control contained the SV40 promoter and enhancer regions (pCAT Control; Promega, Madison, WI) and is referred to as pSV40. The IRBP promoter / CAT reporter construct was made by inserting into pBLCAT3 a BamHI fragment from 21783 to 1101 (relative to transcription start) of the murine IRBP sequence (Genbank accession M32734). A nested set of deletions based on this plasmid was prepared by the ExoIII / Mung bean nuclease method. This generated a series of plasmids, one of which had a 59 end corresponding to position 270 of the murine IRBP, and it is referred to as p70CAT.

138

J.H. Boatright et al. / Brain Research 916 (2001) 136 – 142

2.3. Cell culture, transient transfections, and CAT assays Cell lines were obtained from the American Type Culture Collection (ATCC; Manassas, VA) and established as per ATCC instructions. Cells were maintained at 378C in 5% CO 2 . WERI-Rb1 retinoblastoma (ATCC HTB-169) cells were grown in RPMI 1640 medium (Mediatech, Herdon, VA) and 293 human embryonic kidney cells (HEK 293; ATCC CRL-1573) were grown in EMEM medium (ATCC), supplemented with 10% fetal bovine serum (HyClone; Logan, UT), 100 units / ml penicillin G, 100 mg / ml streptomycin (GIBCO-BRL, Grand Island, NY) and 1 mM HEPES (GIBCO-BRL), with renewal every 3 to 4 days and 1:4 splitting weekly. Cells were plated at 5310 5 cells per 35 mm culture well in 1 ml of growth medium and transfected the same day (WERI cells) or the following day (HEK 293 cells) were transfected with Superfect (Qiagen, Valencia, CA) as per the manufacturer instructions: 1 mg of plasmid DNA (in 1 to 2 ml volume) was added to 1 ml Superfect in a total of 100 ml of OptiMem medium (GIBCO BRL). This solution was added dropwise to a well of cells. When AODNs were co-transfected with plasmids, 1 to 2 ml of the AODN at varying concentrations (see Results) was added with the plasmid DNA into the 100 ml of OptiMem medium. Following a 18, 48, or 66 h incubation (see Results) at 378C in 5% CO 2 , cells were harvested and assayed for CAT activity as previously described [2]. Briefly, cell cytoplasm solutes were mixed with tritiated acetyl-Coenzyme A and chloramphenicol and layered under an organic scintillation liquid. CAT activity was measured as 3 the accumulation of H-acetylchloramphenicol over time on an LS6500 scintillation counter (Beckman Instruments, Schaumberg, IL). CAT activity is reported as CAT enzyme units310 2 5 , and represent the mean of 3–9 repeats6S.E.M. per group per experiment (as detailed in Results).

2.4. Total RNA Preparation Total RNA was prepared from cultured cells by adding 1.0 ml RNAzol B (TEL TEST, Friendswood, TX) directly to the plate. The cells were scraped, homogenized, and 200 ml chloroform was added. This was centrifuged at 12,0003g, 48C for 15 min. The aqueous layer was isolated and 1 volume isopropanol was added. The sample was centrifuged again at 7,5003g, 48C for 8 min. The RNA pellet was washed with 2 volumes 70% ethanol, resuspended in 10 ml DEPC-treated water, and stored at 2808C until use.

2.5. Reverse transcription coupled PCR ( RT–PCR) RT–PCR was performed on total RNA from WERI cells using a OneStep RT–PCR kit (QIAGEN, Valencia, CA). The primer pairs are given in Table 1B and C. The RT–PCR reaction mixture (50 ml total) contained 33 ml

RNase free water, 10 ml 5X OneStep RT–PCR buffer, 2 ml dNTP mix, 1 ml (5 mM) each of the forward and reverse primers, 2 ml OneStep Enzyme Mix, and 1 ml RNA template (ranging in concentration from 0.001 to 1 mg / ul). The reaction mixture was heated to 508C for 30 min for first strand synthesis, 958C for 15 min to denature the reverse transcriptase, then cycled at 948C for 30 s, 558C for 45 s, and 728C for 1 min for a total of 30 cycles. After cycling, the tubes were incubated at 728C for 10 min to complete as many strands as possible. The samples were examined on a 5% polyacrylamide gel in TAE buffer.

2.6. Real time reverse transcription PCR ( RT-RT–PCR) RT-RT–PCR for several mRNAs was carried out with the same primers and reverse transcription steps as above. A Bio-Rad iCycler iQ real time PCR detection system (Bio-Rad, Hercules, CA) was utilized at the PCR stage. We employed a SYBR green measurement to detect the accumulation of double-stranded DNA during PCR. PECetus (Foster City, CA) SYBR Master Mix and protocols were followed. Relative mRNA concentrations were estimated based on the difference in the number of cycles required to reach a threshold. The threshold was taken as a value above background corresponding to 10 times rms noise of the fluorescence intensity in a window usually from 2 to 10 cycles. The formula used was: mRNA concentration after antisense treatment, relative to no treatment 5 (2 [T as 2T no ] ), where T as 5 threshold cycle number for antisense treatment and T no 5 threshold cycle number for no antisense treatment. RT-RT–PCR data are percent means of treatment groups6S.E.M.

2.7. Statistics Numerical data are group means6S.E.Ms. Statistical differences were determined by Student’s unpaired t-test or by simple analysis of variance followed by Newman– Keuls post hoc analysis as noted in Results.

3. Results

3.1. AODNs suppress p70 CAT reporter activity Fig. 2 (left panel) shows the results of co-transfecting WERI cell cultures with p70CAT and three different AODNs at 10, 30, and 100 nM final concentration. CAT activity was reduced to background levels by 100 nM concentrations of any of the AODNs, though A1 AODN was less potent than A3 or A4 AODNs. Fig. 2 (right panel) shows an expanded concentration–response relationship for A3 AODN. WERI cells were co-transfected with

J.H. Boatright et al. / Brain Research 916 (2001) 136 – 142

139

Fig. 2. Effect of CRX AODNs on IRBP promoter activity in WERI-Rb1 cells. Left Panel: Human WERI-Rb1 retinoblastoma cells were transfected with a CAT reporter construct containing a 270 to 1101 fragment (relative to transcription start) of the mouse IRBP gene as promoter (‘IRBP’). Other cells were transfected with CAT constructs containing no promoter (‘None’). Additional groups were co-transfected with the IRBP/ CAT construct and with AODNs specific to human CRX (‘1A1’, ‘1A3’, or ‘1A4’) at 10, 30, and 100 nM final concentrations. Co-transfection with any of the three CRX AODNs suppressed IRBP promoter activity (‘IRBP’ was greater than all groups, P,0.001, N53 per group). Right Panel: WERI-Rb1 cells were co-transfected with the IRBP/ CAT construct and with increasing concentrations of A3 AODN. IRBP promoter activity was suppressed with increasing A3 concentrations. Suppression was statistically significant at 1.0 nM of A3, and half-maximal effect was about 2 nM. N53 per group; *P,0.05 vs. ‘0.0’ concentration of A3.

p70CAT and increasing amounts of the A3 AODN. At lower doses than previously tested (1 nM), the A3 AODN treatment significantly suppressed IRBP promoter activity. The concentration of the A3 AODN at half maximal response was about 2 nM.

3.2. Specificity of AODN response Several experiments were conducted to determine whether the effects of the AODNs are specific or whether AODN treatment is generally suppressive in the transient transfection protocol and promoter / reporter assay. Whereas 10 nM A3 AODN suppressed p70CAT activity by about 90%, a mutated version of A3 had no effect, even at a 100-fold greater concentration (Fig. 3, left panel). The mutation of A3 consisted of the same bases as A3, but in a scrambled order (see Table 1). In a similar experiment, co-transfection with A4 AODN suppressed p70CAT activity by 85%, whereas a mutated A4 yielded about a 30% inhibition (Fig. 3, right panel). This inhibition may indicate that the effects of A4 AODN are less specific or it could be due to residual similarity between A4 and A4Mu. The need to select AODNs with maximal amounts of pyrimidines can restrict AODN mutation design options. To determine whether AODN treatment nonspecifically affected gene promoter activity or CAT expression and enzymatic activity, the effects of AODNs on CAT activity driven by an SV40 promoter were tested in HEK 293 cells and WERI cells. The SV40 promoter drove CAT expression in HEK 293 cells (Fig. 4, left panel) and WERI cells (Fig. 4, right panel). Co-transfection with any of the three AODNs produced no significant effect on SV40-driven CAT activity in either cell line. Note that treatment with

A4 produced a discernable but not statistically significant suppression in HEK 293 cells in three experiments.

3.3. RT–PCR analysis of WERI cell mRNAs RT–PCR was carried out to learn whether CRX, OTX2, GAPDH, or RNase H mRNAs were present in WERI retinoblastoma cells cultured under the present conditions. The RT-RT–PCR primers were designed to completely span one intron, so that a product of the expected size reflected the existence and identification of a given mRNA. Table 1 lists the PCR primers and the product size expected based on the sequences of the gene and its mRNA. The mRNAs for CRX, OTX2, GAPDH, and RNaseH1a (GenBank Accession Z97029) were present in WERI cells (Fig. 5). The mRNA for RNaseH2 (GenBank Accession AF039652) was not detected, suggesting that this mRNA is not expressed in WERI cells. Endpoint RT–PCR analysis at 18 h post-transfection showed that antisense treatment (AODN A3, 10 nM) reduced the level of CRX mRNA and simultaneously reduced the amount of IRBP promoter activity (Fig. 6). The antisense treatment had little effect on GAPDH mRNA level. We additionally examined the effect of A3 AODN on mRNAs and IRBP promoter activity in RT-RT–PCR experiments. WERI cells were transfected with A3 (10 nM) and total RNA prepared 48 to 66 h following transfection. Treatment with A3 decreased CRX mRNA to 1364% of untreated levels, OTX2 mRNA to 73616% of untreated levels, and GAPDH mRNA to 84618% of untreated levels (N54 per treatment group). A3 treatment suppressed CRX mRNA levels significantly more than the

140

J.H. Boatright et al. / Brain Research 916 (2001) 136 – 142

Fig. 3. Effect of mutant AODNs on IRBP promoter activity in WERI-Rb1 cells. Human WERI-Rb1 retinoblastoma cells were transfected with a CAT reporter construct containing either a 270 to 1101 fragment (relative to transcription start) of the mouse IRBP gene as promoter (‘IRBP’) or no promoter (‘None’). Additional groups were co-transfected with the IRBP/ CAT construct and with AODNs and mutants of AODNs. Left panel: Co-transfection with A3 AODN (‘1A3’; 10 nM) suppressed IRBP promoter activity, while a mutated version of A3 AODN (‘1A3Mu’; 1000 nM) had no effect. All comparisons significantly different (P,0.001) except ‘None’ vs. ‘1A3’ and ‘IRBP’ vs. ‘1A3Mu’. N56–9 per group. Right panel: Co-transfection with A4 AODN (‘1A4’; 100 nM) suppressed IRBP promoter activity more than 85% compared to ‘IRBP’, while a mutated version of A4 AODN (‘1A4Mu’; 100 nM) suppressed activity about 30%. All comparisons significantly different (P,0.001). N54 per group.

other two mRNAs (P,0.05). In the same experiment, we measured p70CAT activity and found that without A3 co-transfection, CAT activity was 238667310 25 units whereas treatment with A3 caused activity to drop to 1961.4310 25 units, a significant decrease (P,0.0001 by unpaired t-test).

4. Discussion Antisense technology has been used with variable results to reduce mRNA levels selectively, yet specific antisense oligonucleotides (AODNs) are effective and reproducibly reduce the level of target mRNAs and hence protein expression [9]. Here we applied antisense technology to decrease CRX mRNA levels without substantially altering

the mRNA levels of OTX2, another homeodomain transcription factor, or that of a ‘housekeeping’ gene, GAPDH. Additionally, the mRNA for RNaseH1 was present in WERI cells (though RNaseH2 was not), consistent with the proposal that the antisense mechanism of action of our AODNs was mediated by RNaseH activity, as illustrated in Fig. 1. Treatment with CRX AODNs also potently and selectively suppressed IRBP promoter activity. Together, these data are consistent with a model in which reduction of CRX expression results in reduction in IRBP promoter activity, suggesting that endogenously, IRBP expression involves CRX. Wagner and co-workers [14,22] demonstrated that phosphorothioate oligodeoxynucleotides containing C-5 propyne analogs of uridine and cytidine bind to RNA with high affinity and are potent inhibitors of gene expression

Fig. 4. Effect of CRX AODNs on SV40 promoter activity. Human 293 kidney cells (left panel) or human WERI retinoblastoma cells (right panel) were co-transfected with AODNs specific to human CRX (’1A1‘, ‘1A3’, or ‘1A4’) and with a CAT reporter construct containing either the SV40 promoter (‘SV40’) or no promoter (‘None’). The oligonucleotides had no statistically significant effect on SV40 promoter activity. All groups greater than ‘None’ in either experiment, P,0.05. N53–6 per group for both experiments.

J.H. Boatright et al. / Brain Research 916 (2001) 136 – 142

141

Fig. 5. RT–PCR using total RNA isolated from retinoblastoma cells and primers specific for human CRX, OTX2, GAPDH, RNase H1a, and RNase H2. PCR products were electrophoresed on a 1% agarose gel. These data suggest that all of the genes of interest are expressed in the WERI cell cultures with the exception of RNase H2. OTX2 ‘A’ and ‘B’ are products of two separate sets of RT–PCR primers specific for different targets in the OTX2 gene. Likewise, two additional sets of CRX primers (making a total of three primer pairs) produce correctly-sized products.

[22]. Additionally, this group recommends testing three to six AODNs in order to increase the probability of having at least one that is efficacious, potent, and specific. We based our AODN design on their strategies. We found that one of our AODNs (A3) was very effective at 1 to 3 nM concentrations. It exhibited little nonspecific effect even at 100 nM concentrations, having no obvious deleterious effects on CAT activity driven by an SV40 promoter in HEK 293 or WERI cells, and little effect on OTX2, RNaseH1a, and GAPDH mRNAs from the WERI cells. We speculate that because of the efficacy and selectivity of the propyne and phosphorothioate modified AODNs in our cell

culture system, these types of antisense tools may be effective in more complex conditions or even in animal experiments. It should be noted, though, that as found by other laboratories [9,14,22], the potency and selectivity of AODNs are not completely predictable. For instance, A1 AODN was much less potent than the other AODNs (Fig. 2) and A4 AODN may have been less selective (Fig. 3). Results with a CRX-deficient [13] mouse showing little effect on the mRNA level of IRBP suggested that CRX might have little to do with the overall expression of IRBP. Other experiments showed that OTX2 might play a more prominent role [4,11]. However, several studies show that

Fig. 6. CRX AODNs suppress CRX expression. Left Panel: WERI cells were co-transfected with or without AODNs specific to human CRX. Total RNA was extracted with RNAzol B 18 h later and used as template for RT–PCRs performed with primers specific for human CRX or the housekeeping gene GAPDH. PCR products were electrophoresed on a 1% agarose gel. Antisense-treated cells showed much less CRX PCR product (lane 4) compared to untreated (lane 3). There was no difference in GAPDH PCR product between treated (lane 6) and untreated (lane 5). These data suggest that CRX is expressed in WERI cells and that transfection of WERI cells with CRX AODNs specifically suppresses CRX mRNA levels. These data were corroborated using RT-RT–PCR (see text). Lanes 7 and 8 are RT–PCRs in which the reverse transcriptase was inactivated and CRX primers were used. Lane 7 PCR used RNA as template and shows no CRX product, whereas Lane 8 PCR used WERI genomic DNA and shows a CRX product. These data suggest that the CRX RT–PCR products of Lanes 3 and 4 are not due to genomic DNA contaminating the RNA extractions. Lane 1 is 100 bp ladder DNA and Lane 2 is fX174 RF DNA / Hae III fragments. Right Panel: Companion CAT assay for the RT–PCR. Aliquots of cells from the experiment in the left panel were assayed for IRBP promoter activity. Co-transfection with A3 AODN suppressed promoter activity. (*P,0.05 vs. other groups; N53 per group). A model in which endogenous CRX stimulates IRBP promoter activity may explain these data. Suppression of CRX expression by AODNs precludes this activation, suggesting that endogenous CRX is requisite for stimulated IRBP promoter activity in retinoblastoma cells.

142

J.H. Boatright et al. / Brain Research 916 (2001) 136 – 142

the ‘CRX binding element’ itself is essential for IRBP promoter activity, and that forced CRX expression markedly enhances IRBP promoter activity in HEK 293, HeLa, and other systems [10,15]. Taken together, these results suggest that the expression of IRBP is regulated by many factors, but may not necessarily include CRX. Hence, we posed the question, not of the necessity for CRX in IRBP promoter activity, but rather whether CRX is actually used to activate the IRBP promoter endogenously. Thus, we used a simple IRBP promoter fragment of just 70 nucleotides and monitored its activity while transiently perturbing the level of CRX mRNA with a highly specific AODN. We chose the shortest promoter that we knew to be active in order to exclude as many other cis-elements as possible and thus preclude conflicting or compensating transactivation regulatory mechanisms [1,11]. We measured CAT at relatively short periods of time following transfection to avoid effects of other potential regulatory mechanisms such as mRNA or protein stabilization / degradation. We used a plasmid, rather than an integrated transgene or endogenous gene, because these could have additional chromatin structures and context that might mask CRX’s putative role. While this reflects a model circumstance, the results show that the decrease in CRX mRNA paralleled a decrease in IRBP promoter-mediated CAT activity. We conclude that the decrease in CRX mRNA results in a change in the amount of active CRX protein in the WERI retinoblastoma cells and that a transient decrease in CRX protein activity is sufficient to greatly decrease IRBP promoter activity in the p70CAT plasmid.

[6]

[7]

[8]

[9] [10]

[11]

[12]

[13]

[14]

[15]

[16]

Acknowledgements Funded by NIH P30 EY6360, P30 EY1765, R01 EY12514, R01 EY9378, and R01 EY9769, Research to Prevent Blindness, Fight For Sight, the Foundation Fighting Blindness, the Knights Templar Foundation, and a Research Program Grant from the Emory Eye Center. References [1] J.H. Boatright, D.E. Borst, J.W. Peoples, J. Bruno, C.L. Edwards, J.S. Si, J.M. Nickerson, A major cis activator of the IRBP gene contains CRX-binding and Ret-1 / PCE-I elements, Mol. Vis. 3 (1997) 15–24. [2] J.H. Boatright, R. Buono, J. Bruno, R. Lang, J.S. Si, T. Shinohara, J.W. Peoples, J.M. Nickerson, The 59 flanking regions of mouse IRBP and arrestin have promoter activity in primary embryonic chicken retina cell cultures, Exp. Eye Res. 64 (1997) 269–277. [3] J.H. Boatright, B.E. Knox, E. Stodulkova, H. Nguyen, D.E. Borst, J.M. Nickerson, A region of the IRBP 59 flanking sequence is needed to restrict promoter activity to photoreceptor cells, FEBS Lett. (2001), in press. [4] N. Bobola, P. Briata, C. Ilengo, N. Rosatto, C. Craft, G. Corte, R. Ravazzolo, OTX2 homeodomain protein binds a DNA element necessary for interphotoreceptor retinoid binding protein gene expression, Mech. Dev. 82 (1999) 165–219. [5] N. Bobola, E. Hirsch, A. Albini, F. Altruda, D. Noonan, R.

[17]

[18]

[19]

[20]

[21]

[22]

[23]

Ravazzolo, A single cis-acting element in a short promoter segment of the gene encoding the interphotoreceptor retinoid-binding protein confers tissue-specific expression, J. Biol. Chem. 270 (1995) 1289– 1294. A.H. Bunt-Milam, J.C. Saari, Immunocytochemical localization of two retinoid-binding proteins in vertebrate retina, J. Cell Biol. 97 (1983) 703–712. A. Carlson, D. Bok, Promotion of the release of 11-cis-retinal from cultured retinal pigment epithelium by interphotoreceptor retinoidbinding protein, Biochemistry 31 (1992) 9056–9062. S. Chen, Q.L. Wang, Z. Nie, H. Sun, G. Lennon, N.G. Copeland, D.J. Gilbert, N.A. Jenkins, D.J. Zack, Crx, a novel Otx-like pairedhomeodomain protein, binds to and transactivates photoreceptor cell-specific genes, Neuron 19 (1997) 1017–1030. S.T. Crooke, Molecular mechanisms of action of antisense drugs, Biochim. Biophys. Acta 1489 (1999) 31–44. Y. Fei, S. Matragoon, S.B. Smith, P.A. Overbeek, S. Chen, D.J. Zack, G.I. Liou, Functional dissection of the promoter of the interphotoreceptor retinoid-binding protein gene: the cone-rodhomeobox element is essential for photoreceptor-specific expression in vivo, J. Biochem. (Tokyo) 125 (1999) 1189–1199. S.L. Fong, W.B. Fong, Elements regulating the transcription of human interstitial retinoid-binding protein (IRBP) gene in cultured retinoblastoma cells, Curr. Eye Res. 18 (1999) 283–291. T. Furukawa, E.M. Morrow, C.L. Cepko, Crx, a novel otx-like homeobox gene, shows photoreceptor-specific expression and regulates photoreceptor differentiation, Cell 91 (1997) 531–541. T. Furukawa, E.M. Morrow, T. Li, F.C. Davis, C.L. Cepko, Retinopathy and attenuated circadian entrainment in Crx-deficient mice, Nat. Genet. 23 (1999) 466–470. A.J. Gutierrez, M.D. Matteucci, D. Grant, S. Matsumura, R.W. Wagner, B.C. Froehler, Antisense gene inhibition by C-5-substituted deoxyuridine-containing oligodeoxynucleotides, Biochemistry 36 (1997) 743–778. A. Kimura, D. Singh, E.F. Wawrousek, M. Kikuchi, M. Nakamura, T. Shinohara, Both PCE-1 / RX and OTX / CRX interactions are necessary for photoreceptor-specific gene expression, J. Biol. Chem. 275 (2000) 1152–1160. G.I. Liou, S. Matragoon, J. Yang, L. Geng, P.A. Overbeek, D.P. Ma, Retina-specific expression from the IRBP promoter in transgenic mice is conferred by 212 bp of the 59-flanking region, Biochem. Biophys. Res. Commun. 181 (1991) 159–165. F.J. Livesey, T. Furukawa, M.A. Steffen, G.M. Church, C.L. Cepko, Microarray analysis of the transcriptional network controlled by the photoreceptor homeobox gene Crx, Curr. Biol. 10 (2000) 301–310. T.-I.L. Okajima, D.R. Pepperberg, H. Ripps, B. Wiggert, G.J. Chader, Interphotoreceptor retinoid-binding protein: role in the delivery of retinol to the pigment epithelium, Exp. Eye Res. 49 (1989) 629–644. T.I. Okajima, D.R. Pepperberg, H. Ripps, B. Wiggert, G.J. Chader, Interphotoreceptor retinoid-binding protein promotes rhodopsin regeneration in toad photoreceptors, Proc. Natl. Acad. Sci. USA 87 (1990) 6907–6911. K. Porrello, S.P. Bhat, D. Bok, Detection of interphotoreceptor retinoid binding protein (IRBP) mRNA in human and cone-dominant squirrel retinas by in situ hybridization, J. Histochem. Cytochem. 39 (1991) 171–176. T. van Veen, A. Katial, T. Shinohara, D.J. Barrett, B. Wiggert, G.J. Chader, J.M. Nickerson, Retinal photoreceptor neurons and pinealocytes accumulate mRNA for interphotoreceptor retinoidbinding protein (IRBP), FEBS Lett. 208 (1986) 133–137. R.W. Wagner, M.D. Matteucci, J.G. Lewis, A.J. Gutierrez, C. Moulds, B.C. Froehler, Antisense gene inhibition by oligonucleotides containing C-5 propyne pyrimidines, Science 260 (1993) 1510–1513. B. Wiggert, L. Lee, M. Rodrigues, H. Hess, T.M. Redmond, G.J. Chader, Immunochemical distribution of interphotoreceptor retinoidbinding protein in selected species, Invest. Ophthalmol. Vis. Sci. 27 (1986) 1041–1049.