Molecular Vision 1998; 4:32 <http://www.molvis.org/molvis/v4/p32> Received 9 October 1998 | Accepted 16 December 1998 | Published 31 December 1998

Download Reprint

Spatial and Temporal Expression of AP-1 Responsive Rod Photoreceptor Genes and bZIP Transcription Factors During Development of the Rat Retina

Lihua He,¹ Martin L. Campbell,² Devesh Srivastava,² Yvonne S. Blocker,² J. Robin Harris,² Anand Swaroop,^3,4 Donald A. Fox^1,2
 
 

¹Department of Biology and Biochemistry and ²College of Optometry, University of Houston, Houston, TX and ³Ophthalmology and ⁴Human Genetics, W. K. Kellogg Eye Center, University of Michigan, Ann Arbor, MI

Correspondence to: Donald A. Fox, Ph.D., College of Optometry, University of Houston, 4901 Calhoun, Houston, TX 77204-6052; Phone: (713) 743-1964; FAX: (713) 743-2053; email:dafox@uh.edu

Abstract

Purpose: The promoter region of the rod-specific ß subunit of cGMP PDE (ß-PDE) and opsin genes contains highly conserved cis-acting elements, which include an AP-1 and/or Nrl response element (NRE: An extended AP-1 like sequence). Transactivation of AP-1 or NRE appears necessary to drive expression of these rod-specific genes during adulthood, however, their role during development is relatively unknown. Therefore, we determined the spatial and temporal relationships between rod morphological and functional development, rod-specific gene expression, and expression of the bZIP transcription factors c-fos, junD and Nrl.

Methods: Retinas from 0-45 day old (PN0-45) dark- and light-adapted Long-Evans rats were used. Morphological development was monitored by light and electron microscopy. Whole retinal trypsin-activated cGMP-PDE activity and rhodopsin content were measured biochemically. The expression of opsin, ß-PDE, c-fos, junD and Nrl mRNAs were determined by Northern blot analysis. The cellular localization of Nrl was examined with in situ hybridization.

Results: The mRNAs for opsin, ß-PDE and c-fos were observed at PN0-2, while cGMP-PDE activity and rhodopsin were detected first at PN5: coincident with rod outer segment development. The developmental pattern of cGMP-PDE activity and rhodopsin accumulation paralleled the expression of ß-PDE and opsin mRNA and all reached their maximal levels by PN45. Nrl expression, for all three transcripts found in the rat retina, was low on PN2 and reached its maximal level at PN14. The c-fos and Nrl expression preceded ß-PDE and opsin mRNA expression by 1-2 days. Nrl expression was detected first in the distal post-mitotic retina at PN5 and then in all nuclear layers during retinal development. Maximal expression shifted from the ganglion cells to the outer nuclear layer as the neural retina matured. In contrast, junD expression was highest at PN0 and declined to a stable level by PN10.

Conclusions: Colocalization of Nrl and c-Fos suggests that expression of rod-specific genes, which utilize AP-1 or NRE sites in their promoter, could be regulated through the formation of Nrl-Fos dimers. We hypothesize that Nrl and c-Fos play a fundamental role in the initiation and regulation of the rod-specific gene expression in developing and adult rod photoreceptors.

Introduction

Biochemical and molecular analyses of mammalian retinal development have revealed that both gene-specific and cell type-specific regulatory factors are involved in rod cell differentiation, proliferation, and growth [1-6]. The identification of extrinsic factors responsible for rod cell fate is progressing rapidly [3,5-7]. However, the molecular mechanisms underlying rod photoreceptor, and especially rod outer segment (ROS), differentiation and growth are less well understood. During rod development, there is a coordinate expression of several rod-specific genes that encode proteins of the phototransduction cascade: rhodopsin, cGMP-PDE, and transducin [1,8,9]. The coordinated temporal expression of rhodopsin, the rod [alpha]-subunit of cGMP-PDE, and the rod [alpha]-subunit of transducin is regulated at the transcriptional level [1,2,10].

Several conserved cis-regulatory elements are located in the promoter regions of the genes encoding rhodopsin, rod beta-subunit of cGMP-PDE (ß-PDE), and rod [gamma] subunit of transducin ([gamma]-T) [4,11-17]. Several transcription regulatory proteins have been identified to bind to these elements. The first transcription factor shown to bind to a conserved sequence element (an extended AP-1 like sequence called Nrl-response element: NRE) in the rhodopsin promoter was Nrl [14], which earlier was isolated by subtraction cloning [15]. Nrl also was able to transactivate rhodopsin promoter activity in cultured cells [14,18]. Recently, Crx (cone-rod homeobox), which is expressed predominantly in photoreceptors and pineal glands in adult mouse, was demonstrated to bind to several sequences in promoter regions of various rod-specific genes and activate promoter activity [16,17]. Importantly, Nrl and Crx synergistically transactivate rhodopsin promoter activity [16].

Promoters of several photoreceptor-specific genes contain an AP-1 binding site, which is similar to NRE [11,13,14]. Results from several laboratories suggest that the AP-1 and/or NRE sites are functionally relevant and necessary for transcription of the rod-specific ß-PDE gene [19] and opsin gene [14,18]. The AP-1 complex consist of dimers of the jun and fos family members but not homodimers of fos. Therefore, AP-1 complexes in the retina may contain Nrl in addition to Jun and Fos proteins. These basic motif-leucine zipper (bZIP) proteins are expressed in a large number of different tissues and regulate the expression of a wide variety of genes [6,20-22]. It should be noted that bZIP proteins like Nrl bind to DNA as homodimers or heterodimers and that heterodimerization enhances the sequence site specificity [20-22]. The identity of Nrl's partner in rod-specific gene regulation is unknown, although it can form dimers with several bZIP proteins (e.g., Fos and Jun) in vitro [21]. In vivo, however, the heterodimerization of Nrl will depend on the availability and consequently expression of specific bZIP proteins. Nrl, c-fos, and junD are expressed in vertebrate photoreceptors as well as other retinal cells [23-25]. For example, in dark-adapted neonatal and adult rat retinas, c-fos expression is localized almost exclusively to the outer nuclear layer (ONL), whereas junD expression is observed throughout all retinal nuclear layers [23,25]. The distribution of Nrl expression in dark-adapted retinas is not known. In the light-adapted neonatal and adult rat retinas, c-fos and junD expression are observed in all retinal nuclear layers [23,25]. Similarly, in light-adapted adult mouse retina Nrl expression is detected in all retinal nuclear layers [24].

Transcription of rod-specific genes is determined by specific cis-regulatory elements in the promoter region and the availability of their cognate binding proteins [10,12,14,16,17,18]. Although Nrl has been suggested to regulate expression of rhodopsin and other genes [14,18], it has not been determined whether AP-1 transcriptional complexes in different rod-specific promoters contain Nrl as a homodimer or a heterodimer. To clarify the role of Nrl and other bZIP transcription factors in regulating expression of AP-1 responsive rod-specific genes, we undertook a comprehensive expression analysis to determine their spatial and temporal relationship during the development of rat retina. Specifically, our aims were to determine: (i) the temporal relationships between the differentiation and development of ROS and the onset and increase in retinal cGMP-PDE activity, rhodopsin concentration, and rod-specific ß-PDE and opsin gene expression, (ii) the temporal and spatial patterns of expression of c-fos, junD and Nrl genes in developing dark-adapted rat retinas, and (iii) the temporal relationships between the developmental expression of ß-PDE, opsin and c-fos, junD and Nrl.

Methods

Materials

All chemicals were of analytical or molecular biological grade and were purchased from Sigma Chemical Co. (St. Louis, MO) or Fisher Scientific (Pittsburgh, PA) unless otherwise noted. The ß-PDE cDNA (provided by Dr. Debora Farber at Jules Stein Eye Institute at UCLA, Los Angeles, CA) was isolated from murine retina and cloned into pBLUESCRIPT KS (-) at the EcoR I cloning site. The mouse opsin cDNA was provided by Dr. Wolfgang Baehr (University of Utah Health Science Center, Salt Lake City, UT). The c-fos probe was derived from a 40 base single stranded synthetic oligonucleotide, based on the sequence from exon 1 of rat c-fos (Calbiochem-Novabiochem Corp., San Diego, CA). The Nrl probe for Northern blot analysis was a 1.9 kb fragment of the mouse Nrl cDNA (MR5) in pBLUESCRIPT KS (-) [26], comprising 131 bp of 5' untranslated region, the coding region and 1 kb of 3' untranslated sequences. The mouse junD cDNA (provided by Dr. Daniel Nathans at The Johns Hopkins University School of Medicine, Baltimore, MD) was cloned into pGEM3 at the EcoR I cloning site. The 18S ribosomal RNA (rRNA) oligo probe was provided by Dr. Cheryl Craft (USC School of Medicine, Los Angeles, CA). All cDNAs were labeled with [[alpha]-³²P] dCTP using random primed DNA labeling kit (Boehringer Mannheim Corp., Indianapolis, IN), and the probes were purified using Sephadex G-50 columns (Boehringer Mannheim Corp.). The c-fos and 18S oligonucleotide probes were labeled with [[gamma]-³²P] ATP using T4 kinase (Promega) and purified using Sephadex G-25 columns (Boehringer Mannheim). For in situ hybridization, the 5'-biotin-labeled Nrl oligonucleotide (5'-AAGATGAGACAGAACAGGATG-3') was derived from the 3' UTR region in exon 3 [26], and was synthesized by Midland Certified Reagent Co. (Midland, TX).

Experimental animals

All experimental and animal care procedures were in compliance with the principles of the American Physiological Society, the NIH Guide for the Care and Use of Laboratory Animals and Maintenance (NIH publication No. 85-123, 1985) and were approved by the Institutional Animal Care Committee of the University of Houston. Rats were housed in a room maintained at 22±1 °C with a 12:12 hr light dark cycle and cage illumination of 5-10 lux as described [27]. Pregnant female Long-Evans hooded rats (Harlan Sprague Dawley, Indianapolis, IN) were monitored daily for birth. Upon giving birth, postnatal day 0 (PN0), litters were culled to eight pups. Only female rats were used in experiments.

Light and electron microscopy procedures

The histological procedures were conducted essentially as described [28]. Briefly, the animals were sacrificed by decapitation 2 h after light onset. The eyes were removed and fixed by immersion fixation using 3% glutaraldehyde, 2% paraformaldehyde and 0.1% CaCl₂ in 0.1 M cacodylate buffer (pH 7.4). The tissue was embedded in Spurr's epoxy medium. Thin sections were stained with Toluidine Blue and examined using a BH2 Olympus microscope (Leeds Instrument, Inc., Irving, TX) while ultra-thin sections were stained with 3.5% uranyl acetate and Reynold's lead citrate and examined using a JEOL 100-C transmission electron microscope (Tokyo, Japan).

Whole retinal cGMP PDE activity assay

Trypsin-activated cGMP PDE activity was measured according to the procedure described by Srivastava et al [29]. Briefly, aliquots of the retinal homogenate were treated with trypsin (0.1 mg/ml) for 5 min at 4 °C to remove the two inhibitory [gamma] subunits of the PDE [30]. The reaction was terminated with soybean trypsin inhibitor. Aliquots of homogenates containing 2 µg total protein were assayed in the presence of 500 µM cGMP at 30 °C. All rates of hydrolysis were linear with respect to time. Assays were terminated by boiling and then treated with Crotalus atrox venom for 15 minutes at 30 °C to convert the 5'-nucleotide to the nucleoside. Samples were chromatographed, non-adsorbed nucleoside was collected, scintillation cocktail was added, and samples were counted. Protein concentration was determined using the Bradford assay. The values for trypsin-activated PDE, assayed in triplicate, represent the mean±SEM (standard error of the mean) for 4-7 pairs of retinas per age and are expressed as nmoles cGMP hydrolyzed per minute per mg protein.

Rhodopsin measurements

All rats were dark-adapted overnight and sacrificed by decapitation. The neural retinas were removed rapidly and assayed for rhodopsin content. The entire procedure was carried out under dim red light (wavelength >650 nm) according to the procedures described by Fox and Rubinstein [31]. Briefly, rhodopsin was extracted from each of 4-8 neuroretinas per age with 2% Emulphogene BC-720 (Gaf Corp., Wayne, NJ). The pre-bleach and post-bleach spectra were taken from 350 to 700 nm. The absorbance of rhodopsin at its peak wavelength (497-500 nm) was obtained. A separate group of rats were utilized to measure retinal dry weight. The rhodopsin values represent the mean±SEM for 5-7 retinas per age and are expressed as nmole rhodopsin per mg dry weight.

RNA preparation and Northern blot analysis

All rats were dark-adapted overnight and sacrificed by decapitation just at light onset for the following reason. In dark-adapted, compared to light-adapted, retinas: (i) the c-fos expression is almost exclusively in photoreceptors, (ii) junD is shown to be expressed in photoreceptors, (iii) opsin expression is maximal, and (iv) minimal variations are observed due to diurnal expression of opsin and c-fos mRNA [8,23,25,32]. Retinas (n = 2-8 retinas per age per blot) were excised rapidly, frozen in liquid nitrogen and stored at -80 °C until used. Total RNA was prepared according to the published procedure [33]. Samples of total retinal RNA (10 µg) were denatured and separated on a 1% agarose gel containing 10% formaldehyde. After electrophoresis, RNA was transferred to Hybond-N nylon membranes (Amersham Life Science Inc., Arlington Heights, IL) by electroblotting and cross-linked to the membrane with UV light. Blots were prehybridized and hybridized according to standard methods [34] and exposed to Fuji RX film with an intensifying screen at -80 °C. After probing for different cDNAs, the blots were stripped and rehybridized with the 18S rRNA oligonucleotide probe.

The optical densities of the ß-PDE, 1.7 kb opsin, c-fos, junD and the three Nrl mRNA transcripts and of 18S rRNA on the blot were determined with the BioImage Scanner (Millipore Corp., Bedford, MA). The mRNA expression was normalized for loading variance by comparing the respective mRNA transcript optical density values with that of 18S rRNA. Five to seven blots, from independent samples, were analyzed at each age. The mean±SEM optical densities were determined for each age, normalized to the maximal level of gene expression and then plotted as percent of maximal expression per gene±SEM.

In situ Hybridization of Nrl

For in situ hybridization, whole eyes from dark- and light-adapted rats were removed and fixed in 4% paraformaldehyde (in 0.1 M sodium cacodylate buffer, pH 7.2) for 1 h at 4 °C. The eyecups were fixed for an additional hour after removing the lens and vitreous. The cornea and surrounding peripheral sclera were detached and the eyecups were cryoprotected in 30% sucrose overnight at 4 °C before quick-freezing in Tissue Freezing Medium (Electron Microscopy Sciences, Washington, PA) using liquid nitrogen. Cryostat sections of 10 µm thick at each age were mounted on slides and stored at -80 °C until they were used.

The sections from all examined ages were processed together in batches such that all slides were treated identically. Prehybridization and hybridization procedures were carried out according to Protocol 2 from Amersham Life Science, Inc (In situ hybridization manual). The sections were hydrolyzed in 0.02 M HCl, treated with 0.5 µg/µl proteinase K solution at 37 °C and prehybridized for 1 h at room temperature in 2X SSC, 0.05% Tween-20, 2% non-fat milk, 4% normal goat serum (Jackson ImmunoResearch Laboratories, Inc. West Grove, PA) and 1% BSA. Each series of sections was dehydrated in graded alcohol and then hybridized overnight at 55 °C in the solution containing 2X SSC, 5% dextran sulfate, 0.2% non-fat milk and 50% formamide and 25-200 ng/ml of the 5'-biotinylated Nrl probe. After hybridization, slides were washed sequentially and incubated at 4 °C overnight with 4 nm (LM grade) colloidal gold-streptavidin (Jackson ImmunoResearch Labs, Inc.) diluted in PBS-Tween to a final absorbance at 520 nm of 0.01-0.10. Sections were stained with silver enhancement reagent (Amersham Life Science, Inc.) according to manufacture's instructions, washed in water, dehydrated in graded alcohol and mounted. Since a serial dilution of the Nrl probe was utilized, a semi-quantitative comparison of retinal expression of Nrl at each age was obtained for each batch of identically treated slides.

Statistical analysis

All data are presented as means±SEM (standard error of the mean). Data were analyzed using the appropriate analysis of variance and Fisher's Protected Least Significant Difference posthoc comparisons, according to procedures provided by StatView statistical package (Abacus Concepts, Inc., Berkeley, CA). All statistical analyses were performed on untransformed data and the difference between groups was regarded as significant if p < 0.05.

Results

Light and electron microscopy of the developing rat retina

The morphological development of Long-Evans hooded rat retinas, especially the ROS, was examined using light and electron microscopy. On PN4, no ROS were observed in the rat retina. On PN5, the beginning of rudimentary ROS discs were detected with the electron microscope (data not shown) while on PN6 numerous ROS discs were clearly visible (Figure 1). By PN7, the ROS were visible at the light microscopic level. At PN14 and PN21, ROS were significantly increased in length such that they were approximately 50% and 80% of their final adult length obtained by PN45, respectively (Table 1). A similar pattern of ROS development was observed in the peripheral retina: albeit delayed by two days. This pattern of retinal development we observed is similar to that reported for the albino rat [35,36].

Development of cGMP PDE enzyme activity and rhodopsin concentration

Since cGMP-PDE and rhodopsin are good markers for rod photoreceptor differentiation, we examined the trypsin-activated cGMP PDE activity and rhodopsin concentration during retinal development. Both trypsin-activated cGMP PDE activity and rhodopsin were first detected on PN5, increased relatively linearly until PN30, and reached their maximal levels by PN45 (Figure 2, Table 1). The onset of basal cGMP PDE activity also occurred at PN5 and followed a similar pattern of development, although its activity was markedly lower at all ages relative to the trypsin-activated enzyme (Srivastava and Fox, unpublished results). The rhodopsin concentration reached one-half of its maximal level of 1.25±0.09 nmoles rhodopsin per mg dry weight on PN14 (Figure 2). Trypsin-activated cGMP PDE activity reached one-half of its maximal level of 960±53 nmoles cGMP hydrolyzed per min per mg protein between PN15-16 (Figure 2).

Developmental expression patterns of ß-PDE, opsin, junD, c-fos and Nrl

To demonstrate the temporal relationships between rod-specific genes and their cognate regulatory transcription factors, we determined the expression patterns of two rod-specific genes, ß-PDE and opsin, and three AP-1 family genes that are expressed in the ONL: c-fos, junD, and Nrl [23-25,32]. Figure 3 presents representative Northern blots of RNA isolated from PN0-45 dark-adapted rat retinas for these genes as well as 18S rRNA (used as control for RNA loading). A 3.4 kb transcript was the only ß-PDE species of mRNA detected (Figure 3A). Four transcripts of opsin, varying in size from 1.7 to 5.1 kb, were identified in PN7-45 rats after 3 h of exposure with intensifying screens (Figure 3B). After longer exposure times, signals above background level were detected at PN1 and PN2 for opsin and ß-PDE, respectively, though the levels were low compared to the adult levels (Figure 3C and Figure 3D). Similar to the mouse [37], the relative intensity of all the opsin transcripts was relatively constant throughout development. Therefore, the most abundant 1.7 kb transcript was used for all quantitative measurements. The expression of 1.6 kb junD transcript that was detected at all ages examined decreased during development (Figure 3E). A 2.2 kb transcript of c-fos was detected first at PN1 and its expression increased during retinal development (Figure 3F).

Three Nrl transcripts of 2.7 kb, 2.0 kb, and 1.3 kb were found in rat retinas (Figure 3G). All three transcripts were expressed at low but detectable levels on PN0-2 (data for PN0 not shown) and their expression increased rapidly during development (Figure 4). As illustrated in Figure 4A, the 2.0 kb transcript was approximately twice as abundant as either the 2.7 or 1.3 kb transcripts. Relative to their maximal level of expression (90-100%) that occurred initially on PN14, the transcripts from dark-adapted rats were at ~20-30% on PN3 and 40-60% on PN5-7 (Figure 4B; Table 1). The expression of the three Nrl transcripts in light-adapted developing and adult rats (data not shown) were similar to those in the dark-adapted rats (Figure 3G and Figure 4).

ß-PDE and opsin expression were low at PN2 (<1% of adult maximum) and increased over 2.0 log units during development (Figure 5). Relative to their maximal level of expression at PN45, expression for both genes was at ~10% on PN7, at 50% between PN12-14 and at 80-90% on PN21 (Figure 4; Table 1). The mRNA expression for both genes preceded the detection of their respective proteins by 2-3 days (compare Figure 2 and Figure 5). The expression of junD was highest at PN0-2, decreased rapidly during the next ten days and reached a steady-state level of ~30% of its maximum on PN14. In marked contrast, c-fos expression was low during early postnatal development and increased to a maximal level of expression at PN45. Relative to its maximal level of expression, c-fos expression was at ~6% on PN0-2, at 50% on PN14 and at 60-70% on PN17 (Table 1). Thus, the expression of c-fos and Nrl preceded the expression of both rod-specific genes by 1-2 days (compare Figure 3, Figure 4, and Figure 5; only data for the 2.0 kb Nrl transcript is presented). Interestingly, using in situ hybridization, Ohki et al. [25] did not detect c-fos expression in rat retina until after PN10. Compared to the expression in dark-adapted retinas (Figure 3 and Figure 5), the expression of ß-PDE and junD in light-adapted retinas was not different while opsin and c-fos expression were significantly decreased (data not shown).

Spatiotemporal Expression Pattern of Nrl During Retinal Development

Nrl has previously been implicated in regulation of rhodopsin expression [14,18]. Therefore, we studied the spatiotemporal expression pattern of the Nrl gene in developing and adult dark-adapted and light-adapted rat retinas using in situ hybridization. Figure 6 presents representative micrographs from light-adapted rat retinas from a series of slides whose sections were all incubated in the same concentration of Nrl probe. At PN5, Nrl expression was just above the background level of detection and it localized to the inner nuclear layer (INL) and ganglion cell layer (GCL) (Figure 6A). Expression in all nuclear layers increased during development such that, at PN10 (Figure 6B) and PN15 (Figure 6C) expression appeared equal in all nuclear layers. The maximal level of Nrl expression was reached at PN15 (Figure 6C; Table 1). During retinal development, the highest level of Nrl expression shifted from the GCL and INL to the ONL (Compare Figures 6A, B with Figures 6C, D and E). Figure 6F is a PN45 retina hybridized with sense probe, which showed no signal at all. Similarly, no signal was observed on slides incubated in the absence of Nrl, colloidal gold-streptavidin or silver enhancement reagent (data not shown). There were no differences in retinal Nrl expression between light-adapted and dark-adapted retinas at any age (data not shown).

Discussion

The objective of this study was to determine the temporal relationships between rod morphological and functional development, rod-specific gene expression, and expression of the bZIP transcription factor genes, c-fos, junD and Nrl, in pigmented rat retinas. There are three major conclusions. First, the developmental pattern of cGMP-PDE activity and rhodopsin accumulation paralleled the expression of ß-PDE and opsin mRNA, and the onset of gene expression preceded the appearance of the proteins by 2-3 days. Second, the c-fos and Nrl expression preceded ß-PDE and opsin mRNA expression by 1-2 days. Thereafter, the transcripts for these four genes were detected continuously and in all the developmental stages examined. In contrast, the expression of junD was highest at birth and decreased to a steady-state level during neonatal development. Third, Nrl expression was detected first in the distal post-mitotic retina and then in all nuclear layers during retinal development. Maximal expression shifted from the GCL to ONL as the neural retina matured.

This is the first study to determine the developmental pattern of trypsin-activated retinal cGMP-PDE activity in any species and of rod-specific ß-PDE mRNA in the rat. Rat retinal basal and trypsin-activated cGMP PDE activity were detected first on PN5. This is coincident with the appearance of rod outer segments (ROS) but 2-3 days after the initial expression of ß-PDE mRNA. The expression of ß-PDE mRNA, cGMP-PDE activity, and ROS length reached their adult and maximal values by PN45. The rod cGMP-PDE immunoreactivity was first detected at PN5, with maximum staining reached at PN10 [9]. This apparent discrepancy in the timing of maximal activity and expression may be due to the use of a single high concentration of antibody in the immunohistochemistry study. The ability of trypsin to activate basal cGMP-PDE activity at PN5 suggested that both [gamma] subunits of this heterotetrameric protein were bound to the catalytic [alpha] and ß subunits when its initial activity was detected [30]. A similar increasing pattern of basal retinal cGMP-PDE activity was observed in developing mice [38,39]. The onset and developmental profile of ROS elongation, rhodopsin accumulation and increase in opsin mRNA in our pigmented rat retinas are similar to those found for albino rat and mouse retinas [37,40,41].

The tissue- and cell type-specific expression of genes requires the interaction of various DNA-binding regulatory proteins to their cognate cis-sequence elements in the promoter region [42,43]. The presence of conserved AP-1 and/or NRE as well as other DNA-sequence elements in the promoter regions of ß-PDE, opsin and [gamma]-transducin suggests that multiple transcription factors are involved in the regulation of these rod-specific genes during rod development and maintenance of rod function. Significant regulatory activity associated with AP-1 and/or NRE sites is observed for ß-PDE [18] and rhodopsin [12].

The activity of the rhodopsin promoter is modulated by Nrl [14,18] and Crx in a synergistic manner [16]. Crx appears to bind to three different sites, Ret 1, Ret 4 and BAT-1, in the rhodopsin promoter [12,16]. These sites are close to NRE in rhodopsin promoter [12,16]. bZIP proteins like Nrl bind to DNA as homodimers or heterodimers. Moreover, heterodimerization enhances the sequence site specificity [20-22]. The identity of Nrl's partner in rod-specific gene regulation is unknown, although it forms dimers with several bZIP proteins in vitro [21]. The heterodimerization of Nrl in vivo will depend on the availability and expression of specific bZIP proteins. In addition to Nrl, c-fos and junD are expressed in developing and adult rat photoreceptors [23-25], but c-jun is not expressed in ONL [23]. The expression pattern of junD was opposite to the expression of rod-specific genes and of c-fos and Nrl, indicating that it may not be involved in the activation of rod specific genes at AP-1 and/or NRE sites. Our results show that the onset and developmental pattern of c-fos and Nrl expression parallels that of rod-specific genes, ß-PDE and opsin. Using dark-adapted rats, c-fos expression was localized almost exclusively to the photoreceptors [25,32]. Moreover, and importantly, the early expression of c-fos should be independent of light since the first electroretinographic (ERG) response is not detectable until PN12-13, the age at eye opening [35, 40]. Our studies in the developing and adult retina are the first to demonstrate that Nrl and junD retinal expression are not regulated by light. These results also confirm that c-fos expression in developing and adult retinas is light-regulated, while junD expression in the adult retina is not influenced by light [23,25,32,44].

Interestingly, adult c-fos deficient mice (c-fos-/-), compared to either c-fos+/+ littermates or control mice, have a 22% loss of rod photoreceptor cells in the central and peripheral retina while the number of cones is unchanged (He, Campbell and Fox; unpublished data). A search of the literature and Genbank revealed that no AP-1 like sites are present in the promoter region of cone-specific opsins or PDEs [45-49]. This may explain a normal number of cones in the c-fos -/- mice. Our morphometric data on c-fos-/- mice is consistent with the results presented by Hafezi et al. [50] and with the observations of Reme and co-workers (personal communications) that adult c-fos-/- mice have a 20% decrease in rhodopsin content as well as decreased and delayed ERG a-waves and b-waves.

In summary, our findings suggest that Nrl and c-Fos play an important role in the regulation of the rod-specific gene expression in developing and adult rod photoreceptors. Since c-Fos homodimers can not bind DNA, we hypothesize that AP-1 like sequence elements in promoters of rod-specific genes bind to heterodimers of c-Fos and Nrl and/or to Nrl homodimers. Our observation that not all the rod photoreceptor cells are lost in c-fos-/- mice and that c-fos expression is decreased in ONL during light onset suggests that other bZIP proteins may also form heterodimers with Nrl to transactivate gene expression at AP-1 sites. Further investigations are in progress to identify retinal Nrl-interacting proteins that specify expression of genes in rod photoreceptors.

Acknowledgements

We thank Drs. Richard L. Hurwitz, Deborah Kimbrell, David M. Sherry, George M. Stancel, and Ms. Sandra E. Tirrell for discussions and technical assistance. We also thank The University of Texas Health Science Center: Houston for the generous use of the BioImage scanner. This investigation was supported in part by grants from the National Institutes of Health [ES 03183 (DAF), EY11115 (AS), T35 EY070888 (UHCO for MLC)], a University of Houston PEER Grant [DAF], and the Foundation Fighting Blindness [AS]. AS also is supported by a Research to Prevent Blindness Lew R. Wasserman Merit Award.

References

1. Treisman JE, Morabito MA, Barnstable CJ. Opsin expression in the rat retina is developmentally regulated by transcriptional activation. Mol Cell Biol 1988; 8:1570-9.

2. Timmers AM, Newton BR, Hauswirth WW. Synthesis and stability of retinal photoreceptor mRNAs are coordinately regulated during bovine fetal development. Exp Eye Res 1993; 56:257-65.

3. Kelley MW, Turner JK, Reh TA. Retinoic acid promotes differentiation of photoreceptors in vitro. Development 1994; 120:2091-102.

4. DesJardin LE, Hauswirth WW. Developmentally important DNA elements within the bovine opsin upstream region. Invest Ophthalmol Vis Sci 1996; 37:154-65.

5. Ezzeddine ZD, Yang X, DeChiara T, Yancopoulos G, Cepko CL. Postmitotic cells fated to become rod photoreceptors can be respecified by CNTF treatment of the retina. Development 1997; 124:1055-67.

6. Levine EM, Roelink H, Turner J, Reh TA. Sonic hedgehog promotes rod photoreceptor differentiation in mammalian retinal cells in vitro. J Neurosci 1997; 17:6277-88.

7. Neophytou C, Vernallis AB, Smith A, Raff MC. Muller-cell-derived leukaemia inhibitory factor arrests rod photoreceptor differentiation at a postmitotic pre-rod stage of development. Development 1997; 124:2345-54.

8. Bowes C, van Veen T, Farber DB. Opsin, G-protein and 48-kDa protein in normal and rd mouse retinas: developmental expression of mRNAs and proteins and light/dark cycling of mRNAs. Exp Eye Res 1988; 47:369-90.

9. Colombaioni L, Strettoi E. Appearance of cGMP-phosphodiesterase immunoreactivity parallels the morphological differentiation of photoreceptor outer segments in the rat retina. Vis Neurosci 1993; 10:395-402.

10. DesJardin LE, Timmers AM, Hauswirth WW. Transcription of photoreceptor genes during fetal retinal development. Evidence for positive and negative regulation. J Biol Chem 1993; 268:6953-60.

11. Tao L, Pandey S, Simon MI, Fong HK. Structure of the bovine transducin gamma subunit gene and analysis of promoter function in transgenic mice. Exp Eye Res 1993; 56:497-507.

12. Chen S, Zack DJ. Ret 4, a positive acting rhodopsin regulatory element identified using a bovine retina in vitro transcription system. J Biol Chem 1996; 271:28549-57.

13. Di Polo A, Rickman CB, Farber DB. Isolation and initial characterization of the 5' flanking region of the human and murine cyclic guanosine monophosphate-phosphodiesterase beta-subunit genes. Invest Ophthalmol Vis Sci 1996; 37:551-60.

14. Rehemtulla A, Warwar R, Kumar R, Ji X, Zack DJ, Swaroop A. The basic motif-leucine zipper transcription factor Nrl can positively regulate rhodopsin gene expression. Proc Natl Acad Sci U S A 1996; 93:191-5.

15. Swaroop A, Xu JZ, Pawar H, Jackson A, Skolnick C, Agarwal N. A conserved retina-specific gene encodes a basic motif/leucine zipper domain. Proc Natl Acad Sci U S A 1992; 89:266-70.

16. Chen S, Wang QL, Nie Z, Sun H, Lennon G, Copeland NG, Gilbert DJ, Jenkins NA, Zack DJ. Crx, a novel Otx-like paired-homeodomain protein, binds to and transactivates photoreceptor cell-specific genes. Neuron 1997; 19:1017-30.

17. Furukawa T, Morrow EM, Cepko CL. Crx, a novel otx-like homeobox gene, shows photoreceptor-specific expression and regulates photoreceptor differentiation. Cell 1997; 91:531-41.

18. Kumar R, Chen S, Scheurer D, Wang QL, Duh E, Sung CH, Rehemtulla A, Swaroop A, Adler R, Zack DJ. The bZIP transcription factor Nrl stimulates rhodopsin promoter activity in primary retinal cell cultures. J Biol Chem 1996; 271:29612-8.

19. Di Polo A, Lerner LE, Farber DB. Transcriptional activation of the human rod cGMP-phosphodiesterase beta-subunit gene is mediated by an upstream AP-1 element. Nucleic Acids Res 1997; 25:3863-7.

20. Angel P, Karin M. The role of Jun, Fos and the AP-1 complex in cell-proliferation and transformation. Biochim Biophys Acta 1991; 1072:129-57.

21. Kerppola TK, Curran T. Maf and Nrl can bind to AP-1 sites and form heterodimers with Fos and Jun. Oncogene 1994; 9:675-84.

22. Karin M, Liu ZG, Zandi E. AP-1 function and regulation. Curr Opin Cell Biol 1997; 9:240-6.

23. Imaki J, Yamashita K, Yamakawa A, Yoshida K. Expression of jun family genes in rat retinal cells: regulation by light/dark cycle. Brain Res Mol Brain Res 1995; 30:48-52.

24. Liu Q, Ji X, Breitman ML, Hitchcock PF, Swaroop A. Expression of the bZIP transcription factor gene Nrl in the developing nervous system. Oncogene 1996; 12:207-11.

25. Ohki K, Yoshida K, Harada T, Takamura M, Matsuda H, Imaki J. c-fos gene expression in postnatal rat retinas with light/dark cycle. Vision Res 1996; 36:1883-6.

26. Farjo Q, Jackson AU, Xu J, Gryzenia M, Skolnick C, Agarwal N, Swaroop A. Molecular characterization of the murine neural retina leucine zipper gene, Nrl. Genomics 1993; 18:216-22.

27. Fox DA, Farber DB. Rods are selectively altered by lead: I. Electrophysiology and biochemistry. Exp Eye Res 1988; 46:597-611.

28. Fox DA, Chu LW. Rods are selectively altered by lead: II. Ultrastructure and quantitative histology. Exp Eye Res 1988; 46:613-25.

29. Srivastava D, Fox DA, Hurwitz RL. Effects of magnesium on cyclic GMP hydrolysis by the bovine retinal rod cyclic GMP phosphodiesterase. Biochem J 1995; 308:653-8.

30. Deterre P, Bigay J, Forquet F, Robert M, Chabre M. cGMP phosphodiesterase of retinal rods is regulated by two inhibitory subunits. Proc Natl Acad Sci U S A 1988; 85:2424-8.

31. Fox DA, Rubinstein SD. Age-related changes in retinal sensitivity, rhodopsin content and rod outer segment length in hooded rats following low-level lead exposure during development. Exp Eye Res 1989; 48:237-49.

32. Yoshida K, Kawamura K, Imaki J. Differential expression of c-fos mRNA in rat retinal cells: regulation by light/dark cycle. Neuron 1993; 10:1049-54.

33. Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 1979; 18:5294-9.

34. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 1989.

35. Weidman TA, Kuwabara T. Postnatal development of the rat retina. An electron microscopic study. Arch Ophthalmol 1968; 79:470-84.

36. Braekevelt CR, Hollenberg MJ. The development of the retina of the albino rat. Am J Anat 1970; 127:281-301.

37. al-Ubaidi MR, Pittler SJ, Champagne MS, Triantafyllos JT, McGinnis JF, Baehr W. Mouse opsin. Gene structure and molecular basis of multiple transcripts. J Biol Chem 1990; 265:20563-9.

38. Lolley RN, Farber DB. Cyclic nucleotide phosphodiesterases in dystrophic rat retinas: guanosine 3',5' cyclic monophosphate anomalies during photoreceptor cell degeneration. Exp Eye Res 1975; 20:585-97.

39. Farber DB, Park S, Yamashita C. Cyclic GMP-phosphodiesterase of rd retina: biosynthesis and content. Exp Eye Res 1988; 46:363-74.

40. Bonting SL, Caravaggio LL, Gouras P. The rhodopsin cycle in the developing retina. I. Relation of rhodopsin content, electroretinogram and rod structure in the rat. Exp Eye Res 1961; 1:14-24.

41. Gonzalez-Fernandez F, Van Niel E, Edmonds C, Beaver H, Nickerson JM, Garcia-Fernandez JM, Campochiaro PA, Foster RG. Differential expression of interphotoreceptor retinoid-binding protein, opsin, cellular retinaldehyde-binding protein, and basic fibroblastic growth factor. Exp Eye Res 1993; 56:411-27.

42. Yoon SO, Chikaraishi DM. Tissue-specific transcription of the rat tyrosine hydroxylase gene requires synergy between an AP-1 motif and an overlapping E box-containing dyad. Neuron 1992; 9:55-67.

43. Benbow U, Brinckerhoff CE. The AP-1 site and MMP gene regulation: what is all the fuss about? Matrix Biol 1997; 15:519-26.

44. Nir I, Agarwal N. Diurnal expression of c-fos in the mouse retina. Brain Res Mol Brain Res 1993; 19:47-54.

45. Nathans J, Thomas D, Hogness DS. Molecular genetics of human color vision: the genes encoding blue, green, and red pigments. Science 1986; 232:193-202.

46. Wang Y, Macke JP, Merbs SL, Zack DJ, Klaunberg B, Bennett J, Gearhart J, Nathans J. A locus control region adjacent to the human red and green visual pigment genes. Neuron 1992; 9:429-40.

47. Chiu MI, Nathans J. A sequence upstream of the mouse blue visual pigment gene directs blue cone-specific transgene expression in mouse retinas. Vis Neurosci 1994; 11:773-80.

48. Piriev NI, Viczian AS, Ye J, Kerner B, Korenberg JR, Farber DB. Gene structure and amino acid sequence of the human cone photoreceptor cGMP-phosphodiesterase alpha' subunit (PDEA2) and its chromosomal localization to 10q24. Genomics 1995; 28:429-35.

49. Feshchenko EA, Andreeva SG, Suslova VA, Smirnova EV, Zagranichny VE, Lipkin VM. Human cone-specific cGMP phosphodiesterase alpha' subunit: complete cDNA sequence and gene arrangement. FEBS Lett 1996; 381:149-52.

50. Hafezi F, Steinbach JP, Marti A, Munz K, Wang ZQ, Wagner EF, Aguzzi A, Reme CE. The absence of c-fos prevents light-induced apoptotic cell death of photoreceptors in retinal degeneration in vivo. Nat Med 1997; 3:346-9.