CHANGES IN THE EXPRESSION OF ANTI-OXIDANT PROTECTIVE PROTEINS IN THE RAT RETINA DURING PRE- AND POST-NATAL DEVELOPMENT

Weiheng Chen*, D. Margaret. Hunt and Richard C. Hunt

Department of Microbiology and Immunology, University of South Carolina School of Medicine

Columbia, SC 29208

Dr Chen's present address is: Dean A. McGee Eye Institute, 608 Stanton L. Young Blvd. Oklahoma City, OK 73104

Figures are at the end of this paper

Key words: Retinopathy of prematurity; superoxide dismutase; heme oxygenase; catalase; metallothionein

Supported by National Eye Institute Grant EY 10615 to RH and a grant from the South Carolina Consortium for Cardiovascular Diseases and Stroke to RH

Please address correspondence to:

Dr Richard C. Hunt

Department of Microbiology

University of South Carolina School of Medicine

Columbia SC 29208

 

ABSTRACT

Purpose: In retinopathy of prematurity, capillary growth in the retina is attenuated. Subsequent cyclic elevation of oxygen levels leads to renewed capillary growth that may eventually result in retinal detachment. It is hypothesized that the sensitivity of the premature retina to oxidative shock results from its lack of anti-oxidant protective proteins.

Methods: The expression of heme oxygenase-1, metallothionein, superoxide dismutase and catalase mRNAs was measured in retinae of rats from 6 days before birth to 4 days after birth using in situ hybridization and semi-quantitative reverse transcriptase polymerase chain reaction with Southern blotting.

Results: Superoxide dismutase mRNA was expressed to a similar extent at all time points. Metallothionein mRNA expression, which was high at embryonic day 16 (E16) and E18, fell to low levels by the time of birth and remained low at least to 4 days after birth. Catalase mRNA expression was low until birth and rose until at least post-natal day 4. Heme oxygenase-1 mRNA showed low expression at E16 and E18, rose before birth and then diminished.

Conclusions: Four anti-oxidant protein mRNAs expressed by the rat retina show very different patterns of expression. Two of these proteins, heme oxygenase-1 and catalase are expressed at relatively low levels until around the time of birth. The former is important in protection against heme-mediated generation of reactive oxygen species while the latter protects against hydrogen peroxide-generated damage. As a result of the low expression of these mRNAs, and presumably the proteins encoded by them, the premature rat (and probably the premature human) is likely to be born without a full complement of anti-oxidant defenses.

 

 

A newly born infant is exposed to oxidative shock on leaving the hypoxic conditions in utero. In the case of a premature human infant, this shock is exacerbated by the use of oxygen incubators to maintain cerebral health. The retina of the premature infant is incompletely vascularized and at birth, the rise in oxygen levels results in attenuation of capillary growth.1 The resulting retinal hypoxia and the cycling between hyperoxia and normoxia that comes from placing in and removal from the incubator are likely candidates for the causes of the renewed disorganized capillary growth in retinopathy of prematurity (ROP) that may result in retinal detachment. In contrast, the retina of an infant born after a normal gestation period, although still not completely vascularized, does not show symptoms associated with ROP. Thus, the retina of the premature infant, especially those with birth weights below 2 kilograms, appears to be especially sensitive to hyperoxic shock.

One possibility for the sensitivity of the premature retina to oxidative shock is that anti-oxidant proteins develop late in gestation since they are not required in utero, leaving the premature infant without a full protective arsenal. To determine whether this may be the case, the expression of the mRNAs encoding four anti-oxidant proteins, heme oxygenase-1 (HO-1), metallothionein (MT), Cu/Zn superoxide dismutase (SOD) and catalase, was investigated in fetal and neonatal rats from embryonic day 16 (6 days before birth) (E16) until neonatal day 4 (P4). Using in situ hybridization and semi-quantitative PCR with Southern blotting, it was found that the four mRNAs exhibited different kinetics of expression. HO-1 was only expressed at maximal levels shortly before birth while catalase expression rose shortly before birth and continued to rise at least until P4.

 

MATERIALS AND METHODS

In situ hybridization of rat eye sections with probes to detect heme oxygenase-1, metallothionein, catalase, superoxide dismutase and b-actin

Tissue preparation.

Female Sprague-Dawley rats were sacrificed and embryos removed and decapitated. Neonatal rats were also sacrificed by decapitation. Heads (E16 and E18 embryos) or dissected eyes were fixed overnight at 4° in freshly made 4% paraformaldehyde in Dulbecco=s phosphate-buffered saline (PBS). After dehydration in a series of ethanol concentrations, the material was paraffin-embedded. Sections (7mm) were placed on gelatin-coated slides.

Preparation of in situ hybridization probes

Reverse transcriptase-Polymerase chain reaction (RT-PCR)

RT-PCR was performed as previously described.2 For preparation of the probes, RNA was extracted with RNAStat (Tel-Test, Inc., Friendswood, TX) according to the manufacturer=s protocol and cDNAs were amplified from this RNA using the Ahot start@ PCR technique.3 The primer sequences for rat HO-1 were: 5'CTAAGACCGCCTTCCTGCTCA3' and 5' GATTTGGGGCTGCTGGTTTC3' which give rise to a 494 bp PCR fragment; for MT-1, the primers were: 5'TCGGAATGGACCCCAACT3' and 5'GTGGAGGTGTACGGCAAGACT3' which give rise to a 266 bp PCR fragment; for catalase, the primers were: 5'GTGAGAACATTGCCAACCAC3' and 5'CTCGGGAAATGTCATCAAAAG3' which give rise to a 395 bp PCR fragment; for Cu/Zn SOD, the primers were: 5'GCCGTGTGCGTGCTGAA3' and 5'TTTCCACCTTTGCCCAAGTCA3' which give rise to a 383 bp PCR fragment; and for b-actin the primers were: 5'CCTCTATGCCAACACAGTGC3' and 5'AAGCCATGCCAAATGTCTC3' which give rise to a 342 bp PCR fragment.

 

Cloning of PCR fragments into pCR-Script.

PCR fragments were cloned into pCR-Script using the pCR-ScriptJ SK(+) cloning kit (Stratagene, La Jolla, CA) according to the manufacturer=s protocol. Bacterial colonies were tested for the presence of the plasmid using colony PCR.4,5 The identity and orientation of the plasmid were determined by colony PCR using a primer internal to the insert together with vector-specific orientation primers. Internal primers were: for HO-1, 5'AGGAAGGGGGCGAGGAAC3'; for MT-1, 5'CAGCAGCACTGTTCGTCACT3' ; for catalase, 5'ACTGACGTCCACCCTGACTA3'; for SOD, 5'GCAAGCGGTGAACCAGTTGTG3'; and for b-actin, 5'TAACAGTCCGCCTAGAAGCA3'. The insert for each gene was cloned in both orientations so that all riboprobes (sense and anti-sense) could be made using the same promoter.

Transcription of digoxigenin-labeled probe. The template for riboprobe synthesis was made by PCR using a primer corresponding to a region upstream of the pCR-ScriptJ T7 promoter and a primer corresponding to the downstream end of the insert. The PCR product was purified using a MicroconJ 100 filter (Amicon, Beverly, MA) prior to use as a template in transcription. Digoxigenin-labeled sense and anti-sense riboprobes were made using T7 polymerase and MaxiscriptJ components (Ambion, Austin, TX) according to the manufacturer=s protocol except that final concentrations of 1mM ATP, 1mM CTP, 1mM GTP, 0.65mM UTP and 0.35mM digoxigenin-UTP (Boehringer Mannheim, Indianapolis, IN) were used.

In situ hybridization6,7

a) Incubation with the probe: Sections of paraffin-embedded human donor eyes were dewaxed with two changes of xylene (4 min each). They were then rehydrated in a series of ethanol concentrations (100%, 95%, 70%, 50%, 2 min each), after which the sections were rinsed twice for 5 min in phosphate-buffered saline containing 0.1% Tween 20 (PBST). The sections were then digested with proteinase K (3mg/ml in PBST) at room temperature for 12 min. They were rinsed briefly in PBST and washed for 5 min in PBST, after which they were fixed in 4% paraformaldehyde in PBST and washed again for 5 min in PBST. After two further washes in PBST, the sections were incubated in 0.2N HCl for 8 min at room temperature. The sections were then washed in diethyl pyrocarbonate-treated water and incubated for 30 minutes in acetic anhydride (250 ml in 100 ml 0.1M triethanolamine ( pH 7.0)). The sections were washed twice with PBST for 5 min each.

The probe was heated at 70°C at a concentration of 2 mg/ml for 5 min in HYB+ (50% formamide, 5x SSC, 0.1% Tween 20 containing yeast RNA (20 mg/ml) and heparin (10 mg/ml)) and then added to the sections which were incubated overnight with probe in a wet box at 55°C on a shaking platform. They were washed with 2x SSC at room temperature for 5 min and with STE (500 mM NaCl, 20mM Tris -HCl (pH 7.5), 1 mM EDTA) for 1 min. One hundred ml RNAse A (40 mg/ml in STE) were added to each slide which was incubated for 30 min at 37°. The slides were then washed with 2x SSC, 50% formamide at 50° for 5 min, 1x SSC and then 0.5x SSC for 5 min each at room temperature.

b) Detection of the bound probe: Slides were washed for 1 min in 100mM Tris-HCl (pH 7.5) , 150mM NaCl (buffer 1) and then in this buffer containing 2% normal horse serum (NHS) for 30 min. Sheep anti-digoxigenin antibody-conjugated to alkaline phosphatase (Boehringer Mannheim), diluted 1:500 in buffer 1 containing 1% NHS, was added to the slides for 2 hours at room temperature. The slides were then washed twice in buffer 1 followed by a 10 min incubation in 100mM Tris-HCl (pH 9.5), 1mM NaCl, 50mM MgCl2 (buffer 2). The sections were then incubated with NBT/BCIP (Boehringer Mannheim GeniusTM 3 kit) and levamisole (Boehringer Mannheim GeniusTM 3 kit) diluted in buffer 2. The development of the color reaction was stopped by washing in 10mM Tris-HCl (pH 8.0) containing 1mM EDTA. The sections were viewed using Hoffman Contrast Modulation optics.

 

Semi-quantitative Polymerase Chain Reaction

RNA extraction. Pregnant Sprague-Dawley rats were anesthetized with ether to obtain embryos at various stages of fetal development. Embryonic and neonatal rats of various ages were decapitated and the eyes enucleated. The eyes were rinsed with cold phosphate-buffered saline (PBS) (pH 7.4) and extraocular tissues trimmed off using a dissecting microscope. The eyes of six animals at each age were homogenized and RNA was extracted as described above.

Semi-Quantitative Polymerase Chain Reaction

32P-labeled cDNA probe preparation. PCR fragments corresponding to actin, heme oxygenase-1, metallothionein-1, catalase and Cu/Zn superoxide dismutase were cloned into pCRScriptTM Labeled DNA probes were prepared using PCR-amplified cloned fragments and the Prime-a-Gene labeling system (Promega, Madison, WI), according to the manufacturer=s recommendations.

PCR amplification of mRNAs. cDNA was made using 1 mg total RNA, oligo(dT15) and reverse transcriptase and PCR was carried out with the appropriate primers for various numbers of cycles.8 The PCR products were separated on Trevigel (Trevigen, Gaithersburg, MD) and stained with ethidium bromide. The number of cycles that produced sufficient reaction product to be just seen on the gels was determined and two fewer cycles were used for semi-quantitative PCR. In the case of actin this was 15 cycles and, in the case of the other probes, 25 cycles. The number of cycles was chosen to ensure that the PCR was still in the amplification phase and not the plateau phase. Control experiments showed that these conditions resulted in linear amplification.

Southern Hybridization

PCR fragments were subjected to gel electrophoresis in 2% Trevigel and transferred to Hybond N+ membrane (Amersham, Piscataway, NJ) in 0.4M NaOH using capillary action. The membrane was baked for 30 minutes and pre-hybridized for 1 hour in pre-hybridization buffer which contains 6X SSC, 5X Denhardt=s solution, 0.5% SDS and 100mg herring DNA per ml. After hybridization, the membrane was washed in 2X SSC, 0.5%SDS and then in 2X SSC, 0.5%SDS and finally in 0.1X SSC, 0.1% SDS. The blot was exposed to autoradiography film and radioactivity was estimated using an Alpha Imager Digital imaging system (Alpha Innotech, San Leandro, CA). The intensity of the signals were corrected for variation in the amount of RNA in the RT reactions by expressing as a fraction of the signal from the actin probe.

Animals

Rat were managed in accordance with the ARVO Statement on the Use of Animals in Ophthalmic and Vision Research.

 

RESULTS

Four anti-oxidant protein mRNAs in the developing rat retina show different kinetics of expression

PCR products amplified from the eyes of rats at various stages of development were subjected to gel electrophoresis, transferred to a nitrocellulose membrane and hybridized to 32P-labeled cDNA probes for HO-1, MT-1, SOD or catalase. Fig 1 shows the resulting Southern blots and fig 2 the densitometry of these blots after the data had been normalized using similar semi-quantitative PCR and Southern blotting for actin cDNA from the same RNA samples. It is assumed that actin mRNA exhibits unchanged expression during the period under investigation. Expression of HO-1 mRNA was low, though clearly present, at days E16 and E18, rose at E20, peaked on the day of birth and then fell during the post-natal period (Fig 1A and 2A). MT-1 mRNA expression exhibited very different kinetics. This mRNA was expressed highly at E16 and then fell so that it was undetectable by E20 (Fig 1B and 2B). The primers were designed using the sequence of MT-1 mRNA and although the sequences of MT-2 and MT-3 are very similar to MT-1, the OligoTM program predicts that they would only hybridize to MT-1 cDNA in the PCR reactions. Other investigations (our data, not shown) have demonstrated the mRNAs for MT-1a, MT-2a and MT-3 in RPE, Müller and ganglion cells.

Cu/Zn SOD was present at all times (Fig 1C and 2C) while catalase was expressed at a very low level in embryonic eyes until the day of birth (Fig 1D and 2D).

 

In situ hybridization of the rat retina to detect anti-oxidant protein mRNAs

To confirm these data and to determine where in the eye the various mRNAs were being expressed, in situ hybridization was carried out to detect mRNAs in sections of rat eyes. The characteristic layers of the adult rat retina do not develop until after birth.9 At E15/E16, there is a single nuclear or neuroblastic layer (NBL) with a marginal zone at the vitreous surface which contains nerve fibers parallel to the inner limiting membrane. By E17, this nerve fiber layer has thickened and there are layers of nuclei near the inner surface of the NBL. These will become ganglion cells. At E18, the inner plexiform layer can be distinguished separating the NBL from the ganglion cells which by E20/21 have completely separated from the NBL. During post-natal days P1 to 4, the retina thickens but it is not until P5 that the outer plexiform layer appears. It is only on P9 that all of the layers of the adult retina are apparent.9

HO-1 mRNA detected by in situ hybridization showed a similar pattern of gross expression to the whole eye HO-1 mRNAs investigated with semi-quantitative PCR. No signal from the anti-sense probe was seen in any part of the eye at E16 or E18 (Fig 3A-C). On E21, staining was observed in the retina particularly in the ganglion cell layer close to the vitreous surface and in the NBL (Fig 3D). Staining was more uniform at birth, with most stain in the NBL (Fig. 3E), and declined to low levels on days P1 and P4 (Fig 3F and G). Minimal staining was observed using the sense probe (Fig 3 H-K).

In situ hybridization for MT mRNA also reflected the data obtained using semi-quantitative PCR. Intense staining of the retina and lens was seen at E16, particularly in the NBL (Fig 4A and B), less intense and more uniform staining of the retina was observed at E18 and E21 but the lens remained intensely stained (Fig 4C-E). Little staining was seen at birth (Fig. 4F) or on days P1 (Fig 4G) and P4 (Fig 4H). Again the sense probe showed little staining on any day (Fig. 4I-L). Due to the similarity between MT-1, MT-2 and MT-3, these in situ hybridizations may detect any or all of these mRNAs.

Catalase mRNA, which was shown by RT-PCR to be absent from the eye before birth, also showed no signal by in situ hybridization at E18 (Fig 5A and B). By P1, staining was seen throughout the retina but most prominently in the ganglion cells (Fig 5D). Again the signal from the sense probe was negative (Fig 5C and E).

In the case of Cu/Zn SOD, staining was observed at all time points investigated. At E18 (Fig 5F), the lens and the retina were intensely stained, particularly towards the outer regions of the NBL (Fig 5G). There was also stain in the RPE cell layer. By P1, the ganglion cell layer and scattered cells in the NBL were most intensely stained (Fig. 5I). No stain was seen using the sense probe for Cu/Zn SOD mRNA (Figs. 5H and J).

 

DISCUSSION

ROP has become an increasingly important disease because of the large number of very low birth weight premature babies that survive.1 In the first phase of ROP there is an attenuation of the development of the retinal vasculature as a result of the rise in oxygen levels from the hypoxia in utero to normoxia after birth and further hyperoxia as a consequence of oxygen therapy. In the second phase, vasoproliferation results from retinal hypoxia in the under vascularized region of the retina when the infant is exposed to normoxic conditions. There is evidence that cycling between hyperoxia and normoxia, rather than continuous hyperoxia, is important in the development of ROP.10-12 Cycling between hyperoxia and normoxia has parallels in ischemia/reperfusion in the retina, a situation that is known to lead to reactive oxygen intermediate (ROI) formation 13 and it is probable that ROI production is also important in ROP. ROI may lead to angiogenesis directly or indirectly via the release of angiogenic growth factors. Many hypoxic tissues,14 including the retina 15,16 produce vascular permeability factor/vascular endothelial growth factor (VEGF), the level of which correlates with ocular angiogenesis 17 and ROIs such as superoxide elevate VEGF synthesis in RPE cells.18 Inhibition of VEGF synthesis prevents ischemia-associated neovascularization the primate eye.19

The retina is especially vulnerable to damage as a result of ROI formation. It is the most highly oxygenated tissue in the body and, after birth, is exposed to light which can generate ROI.20 Once the photoreceptors have developed, shed outer segment membranes are degraded by RPE cells, a process that generates H2O2 as a result of long chain fatty acid b oxidation. Moreover, the high level of polyunsaturated fatty acids in the retina 21,22 makes it very susceptible to ROI-mediated damage. Hence, it is not surprising that the eye is rich in anti-oxidant protection proteins which comprise two classes according to the way they deal with ROI. Production of free radicals such as superoxide or hydroxyl radicals is catalyzed by transition metals (notably iron) or by transition metal complexes such as heme or hemoglobin. ROI production may be ameliorated by sequestering these catalysts as complexes with secreted proteins including transferrin, haptoglobin and hemopexin. These proteins are supplied to most tissues by the liver but, in the case of the retina, which is cut off from the circulation by the blood-retinal barrier, they are made locally.2,23-25 In addition, most cells possess an arsenal of enzymes and other antioxidants that break down ROI after their formation. The retina is rich in these proteins 26 which include SOD,27,28 the glutathione redox cycle enzymes 29,30 and catalase.31,32 In addition there are non-enzymic antioxidants in the eye including zinc, which may protect -SH groups of proteins,33 and ascorbic acid.34 Vitamin E, b-carotene and MT are also present.26 MT, the function of which remains obscure, is found at high concentrations in nervous tissue (including the retina and the brain). 35-37 It is a zinc-binding protein which may act as a free radical-scavenging anti-oxidant by virtue of its high sulfhydryl group content.35 There are at least 12 human MT genes and four known classes of MT, each of which may have several isoforms. MT-1 and MT-2 are expressed widely in the body35,37 while MT-3 appears restricted to the brain and the retina.35

Few studies have been carried out on the development of anti-oxidant defenses during embryogenesis and neonatal life. It has been found that MT increases in the rat 38 and human 39 brain after birth but in other tissues, such as the liver,40 MT is found at high levels in the fetus and subsequently declines, as it appears to do in the rat retina from the current studies. High MT in some fetal tissues might reflect its participation in the utilization of certain essential metals during periods of extensive growth and differentiation as in the developing retina.

HO is the rate limiting enzyme in the oxidation of heme to bilirubin and one form, HO-1, is normally only expressed in stress situations that produce ROI including light exposure, hyperoxia and hypoxia as well as in hemorrhage.41 This enzyme is developmentally regulated in fetal rat liver where, with the exception of a narrow peak of high expression at E16, there appears to be a steady rise in HO-1 activity from E15 until birth. There is a further rise during the first post-natal week, after which HO-1 activity falls rapidly.42-44 A similar rise in HO-1 in late embryonic life occurs in the rat retina although here the peak of expression is earlier than in the liver, i.e. on the day of birth. This might reflect the sudden rise in oxygenation that is likely to occur in the highly vascularized retina at birth.

In the rat, phagocytosis of outer segment disc membranes starts during the second post-natal week and it is unlikely that the full extent of fatty acid metabolism and the related H2O2 generation will occur until this time. As a result, it would be expected that catalase would be expressed at far lower levels in the embryonic than in the neonatal eye. This may apply to other tissues since fatty acid oxidation is not a major source of energy until after birth and, as in the retina, catalase is found at low levels in fetal and new born rat kidney; again, activity rises in the first few weeks of post-natal life.45

In a study of the SODs of rat retina,46 it was found that Cu/Zn SOD could not be detected from birth to P5 but at P7, when the photoreceptors differentiated, this enzyme was immuno-histochemically-detectable in the nerve fiber layer, the ganglion cells and the RPE cells. No further change in levels was seen using this method up to 10 weeks. The use of immuno- histochemistry may not have been sensitive enough to detect SOD prior to P7 and low levels could be found at birth using enzyme assays. It is possible that SOD elevation correlates with the differentiation of the inner and outer plexiform layers and the photoreceptors and may play a role in the amelioration of damage by ROI generated as a result of elevated oxidative phosphorylation at the onset of synaptic transmission.

Thus, in the rat retina, expression of four antioxidant protective proteins follows similar patterns to those observed in other tissues in the few studies that have been performed. At E16 and E18, only MT and SOD were detectable and the retina would therefore be unprotected by HO-1 against heme-mediated ROI production or by catalase against H2O2. Since the blood retinal barrier does not develop until the first week of post-natal life in the rat and until the last month of pregnancy in the human, neonatal rats or premature infants may be particularly sensitive to heme-mediated ROI formation as a result of plasma leakage into the developing retina and the consequent presence of heme in the retina from hemolysis.

At birth, the rat retina shows a similar degree of maturation to the 4 to 5 month old human fetus47 and it is possible that anti-oxidant enzymes develop at an earlier stage of gestation in the human; nevertheless, if it is true that the need for some anti-oxidant proteins results from exposure to normoxic conditions, it is likely that, in the human infant too, their synthesis occurs late in development leaving the premature human infant particularly sensitive to ROI because of the lack of protective enzymes. Such sensitivity to ROI may also occur in the later stages of life since catalase, glutathione peroxidase and MT reduction is associated with age-related macular degeneration.26,48,49

 

Acknowledgments

We thank Indhira Handy for technical assistance and Stephen Haley for assistance with the color figures.

 

References

1. Palmer EA, Patz A, Phelps DL, Spencer T. Retinopathy of Prematurity. In: Ryan SJ, eds. Retina. 2nd ed. St, Louis, MO: Mosby: 1994: 1473-1498.

2. Hunt RC, Hunt DM, Gaur N, Smith A. Hemopexin in the human retina: protection of the retina against heme-mediated toxicity. J Cell Physiol. 1996; 168:71-80.

3. Chou Q, Russell M, Birch DE, Raymond J, Bloch W. Prevention of pre-PCR mis-priming and primer dimerization improves low-copy-number amplification. Nucl Acids Res. 1992; 7:1717-1723.

4. Zon LI, Dorfman DM, Orkin SH. The polymerase chain reaction colony miniprep. BioTechniques. 1989; 7:696-698.

5. Simon MM, Palmetshofer A, Schwarz T. From RNA to sequence clones within three days: a complete protocol. BioTechniques. 1994; 16:633-638.

6. Panoskaltsis-Mortari A, Bucy RP. In situ hybridization with digoxigenin-labeled RNA probes: facts and artifacts. BioTechniques. 1995; 18:300-307.

7. Schulte-Merker S. Whole-mount in situ hybridization. In: Westerfield M, ed. The Zebrafish Book. Eugene, OR: University of Oregon Press: 1994: 9.16-9.21.

8. Choudhury P, Chen WH, Hunt RC. Production of platelet-derived growth factor by interleukin-1b and transforming growth factor-b-stimulated retinal pigment epithelial cells leads to contraction of collagen gels. Invest Ophthalmol Vis Sci. 1997; 38:824-833.

9. Braekevelt CR, Hollenberg MJ. The development of the retina of the albino rat. Am J Anat. 1970; 127:281-302.

10. Reynaud X, Dorey CK. Epiretinal neovascularization induced by hypoxic episodes in the neonatal rat. Invest Ophthalmol Vis Sci. 1994; 35:3169-3177.

11. Penn JS, Tolman BL, Lowery LA. Variable oxygen exposure causes preretinal neovascularization in the newborn rat. Invest Ophthalmol Vis Sci. 1993; 34:576-585.

12. Penn JS, Henry MM, Wall PT, Tolman BL. The range of PaO2 variation determines the severity of oxygen-induced retinopathy in newborn rats. Invest Ophthalmol Vis Sci. 1998; 36:2063-2070.

13. Szabo ME, Droy-Lefaix MT, Doly M. Direct measurement of free radicals in ischemic/reperfused diabetic rat retina. Clin Neurosci. 1997; 4:240-245.

14. Namiki A, Brogi E, Kearney M, et al. Hypoxia induces vascular endothelial growth factor in cultured human endothelial cells. J Biol Chem. 1995; 270:31189-31195.

15. Stone J, Chan-Ling T, Pe'er J, Itin A, Gnessin H, Keshet E. Roles of vascular endothelial growth factor and astrocyte degeneration in the genesis of retinopathy of prematurity. Invest Ophthalmol Vis Sci. 1996; 37:290-299.

16. Shima DT, Adamis AP, Ferrara N, et al. Hypoxic induction of endothelial cell growth factors in retinal cells: identification and characterization of vascular endothelial growth factor (VEGF) as the mitogen. Molec Med. 1995; 1:182-193.

17. Miller JW, Adamis AP, Shima DT, et al. Vascular endothelial growth factor/vascular permeability factor is temporally and spatially correlated with ocular angiogenesis in a primate model. Am J Pathol. 1994; 145:574-584.

18. Kuroki M, Voest EE, Amano S, et al. Reactive oxygen intermediates increase vascular endothelial growth factor expression in vitro and in vivo. J Clin Invest. 1996; 98:1667-1675.

19. Adamis AP, Shima DT, Tolentino MJ, et al. Inhibition of vascular endothelial growth factor prevents retinal ischemia-associated iris neovascularization in a nonhuman primate. Arch Ophthalmol. 1996; 114:66-71.

20. Rozanowska M, Jarvis-Evans J, Korytowski W, Boulton ME, Burke JM, Sarna T. Blue light-induced reactivity of retinal age pigment. In vitro generation of oxygen-reactive species. J Biol Chem. 1995; 270:18825-18830.

21. Stone WL, Farnsworth CC, Dratz EA. A reinvestigation of the fatty acid content of bovine, rat and frog outer segments. Expt Eye Res. 1979; 28:387-397.

22. Anderson RE, Sperling L. Lipids of ocular tissues: VII Positional distribution of the fatty acids in the phospholipids of bovine retina rod outer segments. Arch Biochem Biophys. 1971; 144:673-677.

23. Laicine EM, Haddad A. Transferrin, one of the major vitreous proteins, is produced within the eye. Exp Eye Res. 1994; 59:441-446.

24. McGahan MC, Harned J, Goralska M, Sherry B, Fleisher LN. Transferrin secretion by lens epithelial cells in culture. Exp Eye Res. 1995; 60:667-673.

25. Davis AA, Hunt RC. Transferrin is made and bound by photoreceptor cells. J Cell Physiol. 1993; 156:280-285.

26. Newsome DA, Miceli MV, Liles MR, Tate DJ, Oliver PD. Antioxidants in the retinal pigment epithelium. Prog Ret Eye Res. 1994; 13 #1:101-123.

27. Oliver PD, Newsome DA. Mitochondrial superoxide dismutase in mature and developing retinal pigment epithelium. Invest Ophthalmol Vis Sci. 1992; 33:1909-1918.

28. Newsome DA, Dobard EP, Liles MR, Oliver PD. Human retinal pigment epithelium contains two distinct species of superoxide dismutase. Invest Ophthalmol Vis Sci. 1990; 31:2508-2513.

29. Naash MI, Nielsen JC, Anderson RE. The regional distribution of glutathione peroxidase and glutathione-S-transferase in adult and pre-term human retinas. Invest Ophthalmol Vis Sci. 1988; 29:22-26.

30. Naash MI, Anderson RE. Glutathione-dependent enzymes in intact rod outer segments. Expt Eye Res. 1989; 48:309-318.

31. Atalla L, Fernandez MA, Rao NA. Immunohistochemical localization of catalase in ocular tissue. Curr Eye Res. 1987; 6:1181-1187.

32. Oliver PD, Tate DJ, Newsome DA. Metallothionein in human retinal pigment epithelial cells: expression, induction and zinc uptake. Curr Eye Res. 1992; 11:183-188.

33. Newsome DA, Oliver PD, Deupree DM, Miceli MV, Diamond JG. Zinc uptake by primate retinal pigment epithelium and choroid. Curr Eye Res. 1992; 11:213-217.

34. Woodford BJ, Tso MO, Lam K-K. Reduced and oxidized ascorbates in guinea pig retina under normal and light-exposed conditions. Invest Ophthalmol Vis Sci. 1983; 24:862-867.

35. Ebadi M, Iversen PL, Hao R, et al. Expression and regulation of brain metallothionein. Neurochem Int. 1995; 27:1-22.

36. Erickson JC, Masters BA, Kelly EJ, Brinster RL, Palmiter RD. Expression of human metallothionein-III in transgenic mice. Neurochem Int. 1995; 27:35-41.

37. Vallee BL. The function of metallothionein. Neurochem Int. 1995; 27:23-33.

38. Ebadi M. Biochemical alteration of a metallothionein-like protein in developing rat brain. Biol Trace Elem Res. 1986; 11:117-128.

39. Suzuki K, Nakajima K, Otaki N, Kimura M. Metallothionein in developing human brain. Biol Signals. 1998; 3:188-192.

40. Heine PA, Di S, Ross LR, Anderson LL, Jacobson CD. Relaxin-induced expression of Fos in the forebrain of the late pregnant rat. Neuroendocrinol. 1997; 66:38-46.

41. Choi AM, Alam J. Heme oxygenase-1: function, regulation, and implication of a novel stress-inducible protein in oxidant-induced lung injury. Am J Resp Cell Molec Biol. 1996; 15:9-19.

42. Lin JH, Villalon P, Nelson JC, Abraham NG. Expression of rat liver heme oxygenase gene during development. Arch Biochem Biophys. 1989; 270:623-629.

43. Abraham NG, Lin JH, Mitrione SM, et al. Expression of heme oxygenase gene in rat and human liver. Biochem Biophys Res Commun. 1988; 150:717-722.

44. Thaler MM, Gemes DL, Bakken AF. Enzymatic conversion of heme to bilirubin in normal and starved fetuses and newborn rats. Pediatric Res. 1972; 6:197-201.

45. Stefanini S, Serafini B, Cimini A, Sartori C. Differentiation of kidney cortex peroxisomes in fetal and newborn rats. Biol Cell. 1994; 82:185-193.

46. Oguni M, Tanaka O, Tamura H, Shinohara H, Kato K, Setogawa T. Ontogeny of copper-zinc and manganese superoxide dismutase in the developing rat retina: immunohistochemical and immunochemical study. Ophthal Res. 1995; 27:227-233.

47. Tripathi BJ. Embryology. In: Tripathi RJ, ed. Basic and Clinical Science Course: 1997-1999 Section 2: Fundamentals and Principles of Ophthalmology. San Francisco, CA: American Academy of Ophthalmology: 1997: 124-179.

48. Nicolas MG, Fujiki K, Murayama K, et al. Studies on the mechanism of early onset macular degeneration in cynomolgus monkeys. II. Suppression of metallothionein synthesis in the retina in oxidative stress. Expt Eye Res. 1996; 62:399-408.

49. Liles MR, Newsome DA, Oliver PD. Antioxidant enzymes in the aging human retinal pigment epithelium. Arch Ophthalmol. 1991; 109:1285-1288.

 

 LEGENDS TO FIGURES

 Hosodcatfn.TIF (249924 bytes) click for enlargement

 Figure 1. Expression of Anti-Oxidant Protein mRNAs in Embryonic and Neonatal Rat Eyes.

RNA was extracted from embryonic and neonatal rat eyes at days E16 (lane 1), E18 (lane 2), E20 (lane 3), immediately after birth (lane 4), P1 (lane 5) and P4 (lane 6). The RNAs were converted to cDNA using reverse transcriptase and the cDNAs were amplified by PCR under conditions of linear amplification using primers for HO-1 (panel A), MT-1 (panel B), Cu/Zn SOD (panel C), catalase (panel D) and b-actin (panel E). These PCR products were analyzed by gel electrophoresis along with marker PCR fragments (lane M) and transferred to nitrocellulose. The cDNA was hybridized to 32P-labeled probes for the appropriate genes. The RNAs used at each time point were pooled from the eyes of six rats.

 

 

Figure 2. Relative Expression of Anti-Oxidant Protein mRNAs in Embryonic and Neonatal Rat Eyes.

The relative amounts of radio-labeled probe hybridized to the PCR fragments in the autoradiograms shown in figure 1 were determined by densitometry and the levels for each of the anti-oxidant protein mRNAs were normalized relative to the amount of actin mRNA (Panel E) to correct for any differences in the amount of mRNA in each preparation. Panel A: HO-1 mRNA. Panel B: MT-1 mRNA. Panel C: Cu/Zn SOD mRNA. Panel D: catalase mRNA.

 

 98hofn.TIF (3042164 bytes) click for enlargement

 

Figure 3. Detection of HO-1 mRNA in rat ocular tissue by in situ hybridization.

Sections of rat eyes at E16 (panel A and B), E18 (panel C), E21 (panel D), P0 (panel E), P1 (panel F) and P4 (panel G) were hybridized to an antisense probe for HO-1 mRNA. No signal is seen in the retina at E16 or E18 (panels A-C). The dark area of the lens in panels A and C are the result of the optics and do not represent a positive signal. An intense signal is apparent at E21 particularly in the ganglion cell layer (to the left of the section in panel D) and the neuroblastic layer. The signal remains but is less intense on the day of birth (panel E) and then is lost by post-natal days 1 and 4 (panels F and G). No signal is given by the sense probe for HO-1 at E16 (panel H), P0 (panel I), P1 (Panel J) or P4 (panel K). In all higher power sections, the vitreous side of the retina is to the left and the choroidal side to the right. These results are typical of three separate experiments. gc, ganglion cell layer; ipl, inner plexiform layer; nbl, neuroblastic layer; nfl, nerve fiber layer.

 

98fnmt2.TIF (3041790 bytes) click for enlargement

Figure 4. Detection of MT mRNA in rat ocular tissue by in situ hybridization.

Sections of rat eyes at E16 (panel A and B), E18 (panel C and D), E21 (panel E), P0 (panel F), P1 (panel G) and P4 (panel H) were hybridized to an antisense probe for MT mRNA. A strong signal is seen at E16 in both the lens and the neuroblastic layer (panels A and B). The lens is still strongly labeled at E18 but the label in the retina has diminished (panels C and D). MT mRNA remains apparent in the ganglion cell layer at E21 (panel E) but diminishes thereafter (panels F-H). No signal is given by the sense probe for MT at E16 (panel I), P0 (panel J), P1 (Panel K) or P4 (panel L). In all higher power sections, the vitreous side of the retina is to the left and the choroidal side to the right. The similarity of the various forms of MT means that the probes used in these in situ hybridization studies will detect all forms of MT known to be present in the retina. These results are typical of three separate experiments. gc, ganglion cell layer; nbl, neuroblastic layer; nfl, nerve fiber layer.

 

98catsodf2.TIF (3042386 bytes) click for enlargement

Figure 5. Detection of catalase and Cu/Zn SOD mRNAs by in situ hybridization.

Sections of rat eyes at E18 (panel A and B) and P1 (Panel D) were hybridized to an anti-sense probe for catalase mRNA. No signal is seen at E18 (panels A and B) but a strong signal, particularly in the ganglion cell layer (top edge of section), is seen at P1 (panel D). No signal is seen with a sense probe at E18 (panel C) or at P1 (panel E). Similar sections were hybridized with an anti-sense probe for Cu/Zn SOD at E18 (panels F and G) and P1 (panel I). Strong signals are observed at both times in the ganglion cell layer and the neuroblastic cell layer (panels F, G and I). No signal is seen with a sense probe at E18 (panel H) or P1 (panel J). In higher power sections, the vitreous side of the retina is to the top and the choroidal side to the bottom. These results are typical of those in two separate experiments. gc, ganglion cell layer; nbl, neuroblastic layer; nfl, nerve fiber layer.