Expression of β-Carotene 15,15′ Monooxygenase in Retina and RPE-Choroid

Abstract

purpose. β-Carotene 15,15′ monooxygenase (β-CM) catalyzes the central cleavage of β-carotene to all-trans-retinal, the first step in vitamin A synthesis. This study was conducted to determine the expression of β-CM in the mammalian retina and RPE, to assess its relevance in carotenoid-retinoid metabolism in the retina and RPE.

methods. RT-PCR was used to detect expression of β-CM mRNA in the retina and RPE-choroid of the mouse, cow, human, and monkey and in RPE cells and other cell lines. Immunofluorescence microscopy was used to localize β-CM in mouse and monkey retina with an anti-peptide antibody specific for β-CM.

results. By RT-PCR, β-CM mRNA was detected at a low level in mouse and monkey retina and in the RPE-choroid of the monkey but not of the mouse. Conversely, β-CM mRNA was expressed at a low level in both human and bovine RPE-choroid, but not in the retina of either. RPE primary cultured cells of the monkey also showed β-CM mRNA expression, although the three human lines did not. In addition, of nine other cell lines tested, only COS-7 was positive for β-CM. Immunofluorescence microscopy showed weak immunoreactivity in the inner retina in both the mouse and monkey. β-CM immunoreactivity was not detectable in RPE of the mouse. Use of a long-wavelength exciting and emitting secondary probe to mitigate lipofuscin autofluorescence, facilitated the detection of a low level of β-CM immunoreactivity in monkey RPE.

conclusions. β-CM mRNA and protein are expressed at low levels in the mammalian retina and RPE-choroid. Given the low and variable expression of β-CM in the retina and RPE, it can be concluded that β-CM is not necessary for a conserved retina or RPE-specific function, but may be necessary for a species-specific function.

Vitamin A and its derivatives play a central role in the vertebrate and invertebrate visual cycle. ¹ Retinoids function as the chromophores of the various visual pigments in animals, and photoisomerization of 11-cis-retinal to all-trans-retinal is the initiating event in vision. ² In addition, their importance in development and gene regulation through all-trans and 9-cis isomers of retinoic acid (RA) cannot be ignored. Although RA may be formed by successive oxidation of systemically supplied all-trans-retinol (through serum retinol-binding protein [RBP]), it is possible that oxidation of all-trans-retinal made in situ by cleavage of β-carotene is a local source of RA in tissues where β-carotene is stored. Because animals are unable to synthesize vitamin A de novo, they acquire their supply of vitamin A either as provitamin A carotenoids (β-carotene and some other carotenoids) from plants or as preformed retinoids in the form of retinyl esters from animal sources. ³ β-Carotene 15,15′ monooxygenase (β-CM) cleaves β-carotene into two molecules of all-trans-retinal, which are subsequently metabolized into all-trans-retinol and thence to all-trans-retinyl esters or to all-trans-RA. β-Carotene cleavage activity has been reported highest in the intestinal mucosa and is at high levels in liver, kidney, lung, and fat tissues. ^{4 5} This activity was originally described as a dioxygenase ⁶ but now has been shown mechanistically to be a monooxygenase. ⁷ Although the enzymatic activity has been known for many decades, ⁶ the molecular identity of the protein involved has only recently been determined. β-CM is a member of a family that includes RPE65, a protein required for the regeneration of the 11-cis-retinal chromophore of rhodopsin, ⁸ maize VP14, a 9-cis-epoxycarotenoid cleavage enzyme ⁹ necessary in the synthesis of the plant hormone abscisic acid and bacterial lignostilbene dioxygenase, ¹⁰ which cleaves lignostilbene, a model lignin compound, into two molecules of vanillin. β-CM was originally identified by the use of VP14 ⁹ or RPE65 ^{11 12} as a basis for expressed sequence tag (EST) and genomic searches ^{13 14 15 16} and by sequencing of partially purified protein with carotene cleavage activity from chicken small intestine. ⁵ β-CMs were first described in Drosophila melanogaster ^{13 17} and chicken ⁵ and later in mammals. ^{14 15 16} In addition, von Lintig et al. ¹⁷ have determined that the Drosophila visual mutant ninaB, which exhibits reduced visual sensitivity and an abnormal ERG, is due to a mutation in the β-CM gene that prevents the cleavage of β-carotene into all-trans-retinal, the obligate source of chromophore for opsin in the insectan compound eye. 

In the mammalian eye, the macular region of the primate retina is highly enriched in hydroxycarotenoids including lutein and zeaxanthin, whereas β-carotene is absent. ^{18 19} Lutein and zeaxanthin are thought to perform an antioxidant function, ²⁰ perhaps protecting against the risk of development of age-related macular degeneration. ²¹ Consequently, it is thought that there are mechanisms to enrich lutein and zeaxanthin selectively in the macula, while excluding β-carotene. ^{22 23} It is possible that β-CM may be involved in such a process by scavenging β-carotene. Because a recent report ¹⁶ suggested that β-CM is highly expressed in bovine RPE, we examined its expression in the retina and RPE of the mouse, cow, human, and monkey by reverse transcription-polymerase chain reaction (RT-PCR) and by immunohistochemistry in mouse and monkey retina and RPE, to assess its importance in retina-RPE carotenoid metabolism. In contrast to an earlier report, ¹⁶ our data show that β-CM is variably expressed at low levels or not at all in the inner retina and RPE and are not suggestive of a retina- or RPE-specific role for this enzyme. 

Methods

Animals and Tissues

Animal studies were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Mice were housed under fluorescent lights operating on a 12-hour light-dark lighting cycle and were killed by CO₂ asphyxiation in a closed chamber. For the RT-PCR experiments, fresh eye tissue was obtained from wild-type CD1 mice. The anterior segment and vitreous were removed, the retina was isolated from the RPE-choroid-sclera (eyecup), and the tissues were immediately put on dry ice until needed for RNA isolation. For immunohistochemistry, eyes were obtained from CD1 mice. RPE65-knockout and congenic wild-type mice (C57BL/6/129Sv) were used for some experiments. Monkey eyes were obtained from the vaccine surveillance program of the Center for Biological Evaluation and Research, U.S. Food and Drug Administration (US Department of Health and Human Services, Washington, DC). 

Cell Lines

The human D407 RPE cell line was obtained from Richard C. Hunt ²⁴ (University of South Carolina, Columbia, SC) and the primary culture of monkey RPE cells from Bruce A. Pfeffer (Bausch and Lomb, Valley Cottage, NY). The RPE cell line hTERT-RPE1 was obtained from Clontech (Palo Alto, CA) and ARPE-19, COS-7, HeLa, HS27, HepG2, TCMK-1, GC-2, GC-1, Caco-2, and AML-12 cell lines were obtained from American Type Culture Collection (ATCC; Manassas, VA) and grown according to the respective protocols provided by ATCC. 

RNA Extraction

Total RNA was isolated from the mouse tissues (retina, RPE-choroid-sclera) using a commercial kit (RNAqueous-4PCR; Ambion, Austin, TX). After initial extraction, the RNA was concentrated by precipitation as recommended by the kit protocol and the concentration determined spectrophotometrically. RNA was isolated from human, bovine, and monkey retina and RPE samples by the method of Chomczynski and Sacchi. ²⁵ For RNA isolation, cell lines were grown to confluence, trypsinized for 5 minutes at 37°C, and washed in 1× PBS. The cell pellet was resuspended in a denaturing solution containing 4 M guanidine thiocyanate (Fluka, Buchs, Switzerland), 25 mM sodium acetate (pH 7.0), 0.5% N-lauryl sarcosine, and 0.1 M 2-mercaptoethanol and the RNA extracted according to the single-step method of Chomczynski and Sacchi. ²⁵  

Reverse Transcription-Polymerase Chain Reaction

Total RNA (2 μg) from the retina and RPE-choroid (eyecup) and cell cultures was reverse transcribed with Moloney murine leukemia virus reverse transcriptase with a kit (Retroscript; Ambion), using the random decamer oligonucleotide primers supplied in the kit. The first-strand cDNA product of each reverse transcription reaction was amplified with Taq polymerase (SuperTaq; Ambion), using oligonucleotide primers from highly conserved sequences of β-CM cDNA. The set of primers for β-CM, giving an amplimer of 389 bp, were: forward 5′CTACTTGTCTCACACCATCCCC-3′ and reverse 5′CACCAAAGCTGTGGTAGTAGCT-3′. The Rpe65 primers used were forward 5′-ATGATCGAGAAGAGGATTGTC-3′ and reverse 5′CTGCTTTCAGTGGAGGGATC-3′ and gave a 366-bp amplimer. ²⁶ The GAPDH (Clontech, Palo Alto, CA) primers were: forward 5′-ACCACAGTCCATGCCATCAC-3′ and reverse 5′-TCCACCACCCTGTTGCTGTA-3′ and gave a 452-bp amplimer. PCR conditions were: hold at 94°C for 30 seconds, followed by 40 cycles of amplification (94°C 30 seconds, 55°C 30 seconds, 72°C 1 minute) and 72°C final extension for 5 minutes with a PCR system (GeneAmp 2400; Applied Biosystems, Foster City, CA). Premade first-strand cDNAs derived from mouse tissues (Multiple Choice; OriGene Technologies, Rockville, MD) were also used as the positive control for these PCR amplifications. The presence of the corresponding PCR products was determined by electrophoresis on an ethidium bromide-strained 3% gel (NuSieve; FMC Corp., Philadelphia, PA) in 1× Tris-borate-EDTA (TBE) buffer. 

Immunohistochemistry

Mouse and monkey eyes were dissected and fixed by immersion for 2 hours in freshly prepared 4% formaldehyde in isotonic PBS (pH 7.3). Tissues were washed in chilled isotonic PBS (3 × 20 minutes). To cryoprotect tissues before freezing, tissues were incubated in 10% sucrose in PBS for 60 minutes and then transferred to 20% sucrose in PBS for 60 minutes. Tissues were embedded in optimal cutting temperature embedding medium (Tissue-Tek; Sakura Finetek Inc., Torrance, CA) and frozen in liquid nitrogen. Cryosections (12 μm) were cut and collected on slides (Superfrost/Plus; Fisher Scientific, Fair Lawn, NJ) and air dried overnight. Agarose gel sections of monkey retina were also cut (Vibratome; Ted Pella, Irvine, CA). For immunofluorescence labeling, cryosections were blocked with 5% normal goat serum diluted in ICC buffer (PBS, containing 0.5% BSA, 0.2% Tween-20, 0.1% sodium azide [pH 7.3]) then incubated overnight at 4°C in primary antibodies to β-CM ¹⁴ (1:200) or RPE65 ²⁶ (1:400) diluted in ICC buffer. Diluted preimmune sera from these rabbits and the no-primary control were used to assess nonspecific fluorescence. Sections were washed repeatedly and incubated in the dark for 4 hours with the nuclear dye 4′,6′-diamino-2-phenylindole (DAPI; 1 μg/mL) and a goat anti-rabbit fluorochrome-conjugated secondary antibody (Alexa Fluor 568; Molecular Probes, Eugene, OR) diluted in ICC buffer. For experiments involving monkey tissues, a secondary antibody conjugated with a long-wavelength fluorochrome (Alexa Fluor 680; Molecular Probes) was also used. After repeated washing, sections were covered in mounting medium (Gel Mount; Bio-Meda, Foster City, CA) and secured with a coverslip. Specimens were analyzed on a laser scanning confocal microscope (model SP2; Leica, Deerfield, IL) equipped with Nomarski optics. Immunolabeled and negative control sections were imaged under identical scanning conditions. Files were imported into image-analysis software (Photoshop 5.5; Adobe, San Jose, CA) and converted to a digital format for analysis. 

Results

RT-PCR of Retina and RPE-Choroid Tissue

RT-PCR of total RNA from mouse retina using β-CM primers yielded an amplimer of the expected size of 389 bp that was variable but weak in intensity (Fig. 1 , top, lane 1). However, no amplimer was produced from the mouse RPE-choroid RNA (Fig. 1 , top, lane 2), even with 40 cycles of amplification. Enterohepatic tissues (small intestine and liver), kidney, and testis provided positive first-strand cDNA controls for expression of β-CM. Amplification of the GAPDH control was observed in all tissues (Fig. 1 , bottom). Furthermore, amplification of the RPE65-positive control product was strong in RPE-choroid total RNA (data not shown). All findings taken together, the RT-PCR demonstrates the low expression of β-CM in the mouse neural retina, but not in the mouse RPE-choroid. 

In the monkey, RT-PCR detected β-CM amplimers in both retina and RPE-choroid total RNA (Fig. 2) . The amplimer band was stronger in the retina than in the RPE-choroid (Fig. 2 , left), whereas the GAPDH amplimer observed was equivalent in both (Fig. 2 , right). However, RT-PCR did not detect β-CM amplimers in either human or bovine retina (Fig. 3 , left, lanes 1 and 4), whereas low levels were expressed in both human and bovine RPE-choroid (Fig. 3 , left, lanes 2, 3, and 5) with 40 cycles of amplification. Amplification of the GAPDH control was strong in all these tissues (Fig. 3 , left). 

RT-PCR of Monkey and Human RPE and Other Cell Lines

Because RPE primary cell cultures and cell lines are often used to study aspects of RPE biology, a variety of RPE cell lines were examined for β-CM expression by RT-PCR and compared with some non-RPE cell lines (Fig. 4 , top). Three human derived RPE cell lines (ARPE-19, D407, and hTERT-RPE1), one monkey RPE primary culture, and nine various other cell lines derived from tissues known to express β-CM mRNA (COS-7, HeLa, HS27, HepG2, TCMK-1, GC-2, GC-1, Caco-2, and AML-12) were used. Of the four RPE cell lines, the only cell line in which a β-CM amplimer was detected was the monkey RPE primary culture (Fig. 4 , lane 2). None of the three human RPE cell lines showed evidence of β-CM transcription. In addition, of the nine other non-RPE cell lines only one, COS-7 (Fig. 4 , lane 5), derived from monkey kidney, showed the β-CM amplimer at the appropriate 389-bp size. It is of interest to note that the mouse testis-derived (GC-1 and GC-2), kidney-derived (TCMK-1), and hepatocyte-derived lines (AML-12), whose tissues of origin show strong expression of β-CM (Fig. 1) , did not express this message. The human intestine-derived cell line Caco-2 was also negative for β-CM mRNA. All the cell lines tested showed equivalent transcription of the GAPDH control (Fig. 4 , bottom). 

Immunolocalization of β-CM in Mouse and Monkey Retina

In the mouse retina, immunolabeling was restricted to the inner retina (Fig. 5) . The strongest immunoreactivity was present in the inner plexiform (IPL) and ganglion cell (GCL) layers. No immunoreactivity was detected in the RPE or photoreceptors of the mouse. Preimmune serum and the no-primary serum control showed no staining (data not shown). As a positive control, immunolabeling with an antibody to RPE65 ²⁶ showed strong specific labeling of the RPE (data not shown). 

Immunofluorescence was also performed on monkey retina cryosections using the same probe (Alexa Fluor 568; Molecular Probes; Fig. 6 ). Again, β-CM immunolabeling was present in the inner retina (Fig. 6B) . β-CM immunolabeling was visible in the nerve fiber layer (NFL), the GCL, and the IPL, and, at lower intensity, also in the inner nuclear layer (INL) and outer plexiform layer (OPL). No β-CM-specific immunolabeling was observed in the photoreceptor outer segments (ROS) or their cell bodies in the outer nuclear layer (ONL). Negative controls using preimmune serum (Fig. 6C) or omission of primary antibody (Fig. 6D) showed no staining in the inner retina. However, any possible RPE labeling was obscured by lipofuscin autofluorescence. To minimize background autofluorescence arising from lipofuscin in the monkey RPE, an alternative fluorochrome-conjugated secondary antibody (excitation λ_max 684 nm/emission λ_max 707 nm; Alexa Fluor 680, Molecular Probes), was used in immunolocalization studies (Fig. 7) . A strong positive control was provided by anti-RPE65 antibody using this secondary antibody (Fig. 7A) . Comparison of sections treated with the β-CM primary immune serum (Fig. 7B) and the relevant preimmune serum (Fig. 7C) and the no-primary control (Fig. 7D) indicate a low level of β-CM staining in the RPE. 

Discussion

β-CM enzymatic activity is most highly expressed in tissues involved in primary vitamin A biosynthesis from provitamin A carotenoids, such as gut and liver, and at lower levels in other tissues. ^{4 5} The function of β-CM in tissues outside the enterohepatic axis, such as kidney and testis, although it may be involved in local production of retinoids at a secondary level, remains unclear. This lack of clarity would, a priori, describe the possible role of β-CM in ocular tissues. However, a recent report ¹⁶ describes β-CM as being highly expressed in bovine RPE and as perhaps playing a role as subsidiary source of retinoids in the visual cycle. As acutely and exclusively dependent as the visual cycle has been shown to be on all-trans-retinol supplied from hepatic stores through RBP ^{27 28} it is rather unlikely that local production involving β-CM could be a significant source of retinoids. In this study, we examined the expression of β-CM in mammalian retina and RPE, in comparison with some other tissues, as well as in a variety of RPE and non-RPE cell lines. Our data point to a variable level of expression and the absence of a discernible pattern of expression of β-CM in retina and RPE and are not suggestive of an obligate role for this enzyme in the retina and RPE. 

First, both immunofluorescence microscopy and RT-PCR showed low expression of β-CM in the mouse retina and failed to demonstrate the expression of β-CM in the mouse RPE. Second, weak expression was observed in human, monkey, and bovine RPE, by RT-PCR and by immunofluorescence (monkey). Finally, whereas a monkey RPE primary culture was shown to express β-CM by RT-PCR, three human RPE cell lines did not. (By itself, this last finding is not surprising, because some genes, such as RPE65, normally expressed in vivo in RPE are expressed at much lower level in RPE cell lines in vitro. ²⁹ ) It appears that β-CM is not consistently expressed in either RPE or retina and, when it is, only at low levels. This argues against a direct retina-RPE-specific role for this protein, its crucial indirect role in the enteric biosynthesis of all-trans-retinol notwithstanding. However, Yan et al. ¹⁶ have reported that β-CM is highly expressed in the human and bovine RPE cells but not in retina of either species. Although our results for RT-PCR of β-CM in retina of these species are in agreement, the low level of amplimers in the RPE-choroid (even with 40 cycles of amplification) contradicts their conclusion that β-CM is highly expressed in RPE. The extremely low level in bovine RPE-choroid could be due to primer mismatches, because the sequence of the bovine β-CM cDNA is not known. We do not think this likely, because the primers were chosen from a region of the cDNA conserved between mouse and human, and, overall, the human genome is more divergent from that of the mouse than from that of the cow. In addition, we have found, by RT-PCR and immunofluorescence, that β-CM is expressed in the retina in both mouse and monkey. Furthermore, the human RPE cell lines that we tested with RT-PCR, ARPE-19, hTERT-RPE1, and D407, did not show β-CM in the RPE, although it was present in the monkey RPE primary cell culture. Taken together, our findings and those of Yan et al. ¹⁶ are somewhat at odds, because we do not find it to be highly expressed at all in RPE of these species. Nor is expression of β-CM high in those species with retinal expression. Based on these data, we conclude that β-CM is expressed at a low level in retina and RPE in a species- rather than tissue-specific manner. 

Immunohistochemistry studies in the monkey, using anti-β-CM antibody, showed localization of β-CM in the inner retina, as in the mouse. In addition, considerable fluorescence was seen in the RPE cell layer. Because fluorescence of similar intensity was seen in the preimmune and no-primary control, it must be concluded that most, if not all, of this signal was due to lipofuscin autofluorescence. To address the question of β-CM expression in the highly autofluorescent monkey RPE, ³⁰ an alternate fluorochrome (Alexa Fluor 680; Molecular Probes) was used. This dye has a fluorescence excitation λ_max at 684 nm and emission λ_max at 707 nm, well away from the 380 to 440 nm excitation λ_max and 590 to 650 nm emission λ_max of lipofuscin fluorophores. Using this dye, we detected a weak signal in the monkey RPE sections incubated with primary antibody, whereas control sections incubated with preimmune or no primary antibody had no discernible signal. This signal could be attributable to β-CM immunoreactivity. 

In general, the role of β-CM in the metabolism of carotenoids and retinoids in the retina and RPE, although unclear, does not appear to be an essential one. It is well known that the macula lutea of the primate retina is highly enriched in carotenoids. Hydroxycarotenoids (e.g., lutein and zeaxanthin) are the main carotenoids in the macula (for review see Landrum and Bone ²² ), whereas β-carotene is present only at trace levels. ¹⁹ β-CM activity in crude homogenates has been shown to be inhibited by lutein, ^{31 32} and therefore these macular hydroxycarotenoids would be resistant to cleavage. It is possible that the function of β-CM in the RPE and retina (when it occurs) is to prevent the accumulation of β-carotene in these tissues. Although its role in the retina-RPE may still be that of a carotenoid scavenger, we do not think that it is involved in any major way in vitamin A biosynthesis per se in these tissues. It is possible that any all-trans-retinal generated by β-CM in primate and bovine RPE is to supply all-trans-retinal for the orphan retinal G-protein-coupled receptor (RGR) opsin ^{33 34} or to be further oxidized to all-trans-RA, though no evidence exists for either outcome. 

In conclusion, we have determined the localization and expression of β-CM in the retina and RPE of several mammalian species. In all cases, expression, where present, is weak and inconsistent in species distribution. These findings do not support the hypothesis that β-CM mutations in humans may lead to visual deficits, as is the case in the ninaB mutation of Drosophila β-CM, ¹⁷ and raise doubts about its overall importance in retinal-RPE retinoid and carotenoid metabolism. It seems safe to conclude, however, that any such role is not a retina-RPE-specific one common to all species, but it may be species specific. The absence of expression of β-CM mRNA in most of the tested cell lines (both RPE and non-RPE) suggests that its expression is under tight transcriptional control, even in cell lines derived from tissues that normally express it. This suggestion is borne out by analysis of the mouse β-CM gene promoter function (Boulanger et al., manuscript submitted, 2002). 

²  These authors contributed equally to the work and therefore should be regarded as equivalent senior authors.

 Supported in part by a grant from the V. Kann Rasmussen Foundation, Denmark (SLB) and a Career Development Award from Research to Prevent Blindness (SLB).

 Submitted for publication February 15, 2002; revised July 29, 2002; accepted August 9, 2002.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Expression of β-CM mRNA in mouse retina and RPE-choroid-sclera in comparison with other tissues. PCR amplification was performed on cDNA reverse transcribed from total RNA samples. Lane 1: retina; lane 2: RPE-choroid-sclera; lane 3: testis; lane 4: small intestine; lane 5: liver; and lane 6: kidney. Top: A 389-bp amplimer for β-CM was detected in retina but not in RPE-choroid-sclera. Expression of β-CM was lower in retina than in other tissues surveyed. Bottom: GAPDH control primers displayed a GAPDH amplimer in all tissues with the expected size of 452 bp. PCR products were loaded on 3% gel, and DNA bands were stained with ethidium bromide. A 100-bp DNA ladder was used to identify the size of the products.

Figure 2.

 

Expression of β-CM mRNA in monkey retina and RPE-choroid. PCR amplification was performed on cDNA reverse transcribed from total RNA samples. Left: A 389-bp amplimer for β-CM was detected in monkey retina and in monkey RPE-choroid. Expression of β-CM was low in retina and lower in RPE-choroid. Right: GAPDH expression in the mouse retina and RPE. RT-PCR performed on monkey RPE-choroid and retina RNA using GAPDH control primers displayed an amplimer in both tissues with the expected band of 452 bp. PCR products were visualized as described in Figure 1 .

Figure 3.

 

Expression of β-CM mRNA in human and bovine retina and RPE-choroid. PCR amplification was performed on cDNA reverse transcribed from total RNA samples. Lane 1: human retina; lane 2: human RPE-choroid; lane 3: human RPE-choroid; lane 4: bovine retina; lane 5: bovine RPE-choroid; lane 6: no-cDNA control. Left: A 389-bp amplimer for β-CM was detected in human and bovine RPE (very weak) but not in retina (left). Right: GAPDH expression in the mouse retina and RPE. RT-PCR performed using GAPDH control primers displayed an amplimer with the expected size of 452 bp in all tissues. PCR products were visualized as described in Figure 1 .

Figure 4.

 

Expression of β-CM mRNA in RPE and other cell lines. RT-PCR was performed on total RNA from three human-derived RPE lines, a monkey RPE line primary culture, and nine various other tissue cell lines (tissue of origin in parentheses). Lanes l: ARPE-19 (human RPE); lane 2: monkey RPE; lane 3: D407 (human RPE); lane 4: hTERT-RPE1 (human RPE); lane 5: COS-7 (monkey kidney); lane 6: HeLa, (human cervical adenocarcinoma); lane 7: HS27 (human bone marrow); lane 8: HepG2 (human hepatocellular carcinoma); lane 9: TCMK-1 (mouse kidney); lane 10: GC-2 (mouse testis); lane 11: GC-1 (mouse testis); lane 12: Caco-2, (human intestine); lane 13: AML-12 (mouse hepatocyte). Top: RT-PCR performed with β-CM primers. The only RPE cell line in which a β-CM amplimer was seen was the monkey RPE primary culture. In addition, of the nine other tissues, only one, COS-7 derived from monkey kidney, showed the β-CM amplimer at the appropriate 389-bp size. Bottom: RT-PCR performed with GAPDH primers. All the cell lines showed equivalent expression of GAPDH mRNA.

Figure 5.

 

Confocal immunofluorescence localization of β-CM in C57BL/6 mouse retina. (A) Differential interference contrast (DIC) image of mouse retinal cryosection. (B) β-CM immunolabeling (red) was restricted to the inner retina. In this region, the GCL and IPL showed moderate levels of immunofluorescence. Very low-intensity immunolabeling was also present in the INL and OPL. No detectable labeling was observed in the RPE or in the ROS or their cell bodies in the ONL. Nuclei in the neural retina and RPE were labeled with the DNA-binding fluorescent dye DAPI (blue).

Figure 6.

 

Confocal immunofluorescence localization of β-CM in monkey retina. (A) DIC image of monkey retinal cryosection. (B) β-CM immunolabeling is restricted to the inner retina. In this region, the GCL and IPL showed moderate to high levels of immunofluorescence. Lower-intensity immunolabeling was also present in the INL and OPL layer. The RPE showed strong lipofuscin autofluorescence in the absence of antibody labeling. No β-CM-specific immunolabeling was observed in association with the ROS or their cell bodies in the ONL. Nuclei in the neural retina and RPE are labeled with the DNA-binding fluorescent dye DAPI (blue). Negative controls with (C) preimmune serum and (D) omission of primary antibody.

Figure 7.

 

Confocal immunofluorescence localization of β-CM in monkey RPE. Agarose gel sections of monkey retina-RPE were prepared and labeled with a long-wavelength fluorophore to mitigate interference due to lipofuscin autofluorescence. Labeling in the retina is as shown in Figure 5 . In the RPE, a strong positive control was provided by anti-RPE65 antibody (A). Comparison of sections treated with the β-CM primary immune serum (B) and the relevant preimmune serum (C) and no-primary controls (D) indicated a low level of staining in the RPE. Nuclei in the neural retina and RPE were labeled with the DNA-binding fluorescent dye DAPI (blue).

The authors thank Amanda Bundek for excellent technical assistance and Carl Haugen, Transgenics and Gene Manipulation Section, Laboratory of Molecular and Developmental Biology, National Eye Institute for the mice used in the study.