Molecular Vision 2006; 12:1272-1282 <http://www.molvis.org/molvis/v12/a144/> Received 29 November 2005 | Accepted 15 September 2006 | Published 26 October 2006

Download Reprint

Decrease of cone opsin mRNA in experimental ocular hypertension

Heather R. Pelzel, Cassandra L. Schlamp, Gretchen L. Poulsen, James A. Ver Hoeve, T. Michael Nork, Robert W. Nickells
 
 

Departments of Ophthalmology and Visual Sciences and Biomolecular Chemistry, University of Wisconsin, Madison, WI

Correspondence to: Robert W. Nickells, Department of Ophthalmology and Visual Sciences, 6640 Medical Science Center, University of Wisconsin, 1300 University Ave., Madison, WI 53704; Phone: (608) 265-6037; FAX: (608) 262-1479; email: nickells@wisc.edu

Abstract

Purpose: This study was designed to test the hypothesis that photoreceptors are adversely affected in glaucoma. As a measure of this effect, we examined the levels of rod opsin, and red/green and blue cone opsin mRNAs in monkeys with experimental ocular hypertension and glaucoma and in human eyes from donors with diagnosed glaucoma.

Methods: Experimental ocular hypertension was induced in one eye of 19 cynomolgous and 2 rhesus monkeys by laser ablation of the trabecular meshwork. In 15 monkeys, the elevated IOP was reduced by trabeculectomy. When the animals had experienced prolonged elevations of IOP (128 to 260 days), they were killed and the eyes enucleated. Fresh retinal tissue from the macula, inferotemporal retina (mid-peripherpal), and far peripheral regions were harvested from some animals using a 3 mm trephine. The remaining retinas from these monkeys, and whole retinas from other animals were fixed. RNA isolated from each trephined sample was used for RNase Protection Analysis or real time PCR analysis to quantify opsin mRNA levels from different photoreceptor cell types. Fixed tissue was used for in situ hybridization studies. Human donor eyes (7 glaucoma and 4 control) were obtained from eye banks. All human specimens were used for in situ hybridization studies.

Results: Quantitative mRNA analysis and in situ hybridization studies both showed a reduction in the expression of red/green and blue cone opsin mRNAs in 6 monkey eyes with chronic ocular hypertension, relative to the contralateral eye. No loss of rod opsin mRNA was observed. The principal reduction occurred in cells of the mid-peripheral retina, a region of retina that often shows early and progressive damage in humans with glaucoma. In monkeys with ocular hypertension followed by trabeculectomy, there was a similar decrease in cone opsin mRNAs, but only in six out of fifteen (40%) of the monkeys. The decrease in these animals was correlated with a significantly elevated IOP at some time during the 2 weeks prior to euthanization and not with the extent of glaucomatous damage. Of the 7 human eyes with diagnosed glaucoma that were examined, 5 showed a decrease of cone opsin mRNA in the mid-peripheral retina, whereas none of the 4 normal eyes examined showed a decrease.

Conclusions: Ocular hypertension leading to glaucoma also affects the outer retina, particularly the cone photoreceptors. We speculate that these cells become stressed leading to a disruption in the expression of normal genes, such as that encoding opsin. There is some evidence that this effect is reversible, when IOP levels are reduced.

Introduction

Glaucoma is a major blinding disease that is principally characterized by the loss of retinal ganglion cells [1,2]. Recently, there has been growing interest in the effects of glaucoma on the outer retina, particularly photoreceptors. The earliest indication that glaucoma affected the outer retina originated from observations that this optic neuropathy behaves like an outer retinal disease with respect to Köllner's Law for color vision defects. Essentially, the law states that deficits in color vision arising from optic nerve disease result in red-green confusion, whereas outer retinal disease results in blue-yellow confusion. Glaucoma is an exception to this rule, as blue-yellow contrast sensitivity is compromised early in the disease [3,4]. Morphometric studies examining for outer retinal changes in glaucomatous eyes have yielded conflicting results. Panda and Jonas reported photoreceptor loss in eyes with trauma-induced secondary glaucoma [5]. Counts of nuclei in the outer nuclear layer of human retinas with primary open-angle glaucoma, however, failed to find any significant loss of cells relative to age-matched controls [6]. Similarly, counts of cone pedicles in the outer plexiform layer showed no significant cone loss in eyes from glaucomatous monkeys [7]. These studies all assumed that outer retinal effects would present as a loss of photoreceptors. Although photoreceptor cell loss may not be a consistent feature of glaucomatous eyes, these studies do not preclude the possibility that these cells exhibit physiological and potentially damaging effects related to the disease. A more recent study showed that abnormalities in the outer retina, including swollen photoreceptors, were common in human eyes diagnosed with glaucoma and in monkey eyes with experimental ocular hypertension [8]. Using a histochemical stain for carbonic anhydrase activity, the majority of these cells were identified as the red and green wavelength sensitive cones (red/green cones, also known as the long and medium-wavelength cones, respectively). It has been suggested that because equal stimulation of the red/green cones produces the sensation of "yellow" equal loss of both red and green cone function could lead to confusion along the blue-yellow axis [9]. Last, electrophysiological testing also supports the hypothesis that the outer retina is affected in experimental ocular hypertension in monkeys. Multifocal electroretinogram (mfERG) studies have found both changes in amplitude and an increase in latency that are produced by both inner and outer retinal components [10-12]. Another indication of outer retinal injury is an exaggerated or supranormal mfERG response, which has been observed in both acute [13] and chronic [14] experimental glaucoma in preliminary studies.

Considering these morphometric and electrophysiologic findings, we hypothesized that the physiological levels of certain mRNAs would be altered in glaucoma. Opsin mRNA is a suitable marker for these experiments because (1) opsin gene expression is cell-type specific and therefore probes complimentary to the corresponding mRNAs act as cell-specific markers, (2) opsin is an abundant mRNA and therefore easy to detect in conventional quantitative and qualitative assays [15], and (3) opsin mRNA steady state levels are highly regulated and linked to a diurnal or circadian rhythm [16-20], where nearly 70-80% of the opsin mRNA is degraded and synthesized daily. In an injured cell, it is possible that the steady state mRNA levels will be affected due to the changes in the rate of synthesis or degradation. In addition, because opsin mRNA levels are turned over at a high rate, adverse effects on turnover will occur rapidly, making quantitative and qualitative examination opsin mRNAs a sensitive assay for damaged photoreceptors.

Methods

Handling of animals and the development of experimental glaucoma

All experimental methods and techniques adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and were approved by our institutional animal care and use committee. Two separate groups of monkeys were used in this study. The first group of animals consisted of 3 control (no glaucoma in either eye) cynomolgus monkeys (Macaca fascicularis), 4 cynomolgus monkeys with experimental glaucoma, and 2 rhesus monkeys (Macaca mulata) with experimental glaucoma. All the animals were approximately 8 years of age. Chronic experimental glaucoma was produced in one eye of each of the animals by laser destruction of the trabecular meshwork as described by Gaasterland and Kupfer [21]. A Kaufman-Wallow single mirror monkey gonioscopy lens (Ocular Instruments, Inc., Bellevue, WA) was used to deliver the laser to the trabecular meshwork over 1 to 3 treatment sessions spaced over a 2-4 week period [22]. During each treatment session, either 270° or 360° of the anterior (non-pigmented) meshwork was photocoagulated with a 532 nm diode laser (Oculight GL, Iridex Corp., Mountain View, CA) using 75 μm spots of 1.0 watt intensity and 0.5 s duration. Intraocular pressures were measured with applanation tonometry (Tono-Pen, Medtronic Solan Ophthalmic Products, Inc., Jacksonville, FL), while the animals were anesthetized with ketamine (10 mg/kg, intramuscular). The monkeys exhibited variable degrees of elevated IOPs over the course of the experiment (Table 1).

Monkeys enrolled in a glaucoma surgery study

In addition to the group of monkeys with chronic experimental ocular hypertension, retinal samples were collected from eyes of 16 cynomolgus monkeys previously used in a glaucoma surgery study [23]. In this study, ocular hypertension was induced in one eye of each animal using the methods indicated above. The elevated IOP was then reduced using a partial thickness trabeculectomy procedure. The purpose of the study was to test the efficacy of a novel antiproliferative agent (human p21 gene therapy transduced into cells of the conjunctiva and Tenon's layer using replication-deficient adenovirus) in preventing wound healing for a prolonged period of time after surgery (approximately 260 days). A complete description of the surgical protocol, dosing regiment of antiproliferative and control reagents, and follow-up procedures are described elsewhere [23]. Of note, this group of monkeys had IOPs measured under ketamine anesthesia every 7-10 days by 'minifield' Goldman applanation tonometry using a Haag-Streit slit lamp [24], rather than by a hand held Tonopen. Each animal in this study was killed by perfusion fixation of 4% paraformaldehyde in PBS under deep anesthesia. The eyes were enucleated and a region of the sclera and conjunctiva was dissected away for separate analysis. The remaining parts of the eye were stored in PBS at 4 °C.

Human eyes

The tenets of the Declaration of Helsinki were followed and approval by our institutional human subjects committee was granted for this study. Human donor eyes from individuals with no history of ocular disease were obtained as a gift from the Lions Eye Bank of Wisconsin and eyes from individuals with diagnosed glaucoma were obtained as a gift through the Glaucoma Research Foundation eye donor program. A more complete description of this collection is given elsewhere [8]. For this study, we screened glaucoma specimens for which we had been able to obtain clinical histories, particularly visual field results. Only in situ hybridization studies were conducted on the human specimens.

Harvesting of retinal tissue for analysis

To isolate fresh unfixed retinal tissue, animals were killed by overdose of intravenous anesthetic. The eyes were then removed and placed in 100 mM phosphate buffer (PB, pH 7.2) chilled to 4 °C. They were coronally sectioned at the level of the pars plana. Under a dissecting microscope, samples of retina were removed using disposable 3 mm trephines from the macula, mid-periphery, and far periphery (Figure 1). These samples were quickly frozen in dry ice for later processing. The remaining globes were then lightly fixed in 4% paraformaldehyde in PB containing 150 mM NaCl (PBS) at 22 °C for 1 h, followed by submersion in cold (4 °C) 0.4% paraformaldehyde in PBS for storage. After fixation, small pieces of retina near the mid-peripheral trephined region (Figure 1) were harvested for in situ hybridization studies (see below). Pieces of retina were also harvested from the 16 monkeys enrolled in a surgery study and human eyes obtained from eye banks. For these specimens, retinal pieces were harvested from previously fixed and dissected eyes stored in PBS at 4 °C. Retinal pieces were harvested in the same region as described for the animals used to isolate fresh unfrozen tissue.

Cloning of monkey opsin cDNAs

Partial cDNA clones for rod, blue, and red/green opsin were synthesized by RT-PCR using template cDNA generated from monkey retina (M. fascicularis) total RNA. Primers were designed using either monkey or human mRNA sequences (Accession numbers: rod: S76579/U49742/K02281; blue: M13295-M13299; red/green: M13300-M13306/K03490-K03494). The sequences were: rod: 5'-CTC TAC ACC TCT CTG CAT GG (forward) and 5'-GCA ACG CCC ATG ATG GCA TG (reverse); blue: 5'-TCT TCG TCG CCA GCT GTA AC (forward) and 5'-GGA GAC GCC AAT ACC AAT GG (reverse); and red/green: 5'-ACC CTT TGA TGG CTG CCC TG (forward) and 5'-TCA TGC AGG CGA TAC CGA GG (reverse). Because of their close similarity, a single probe was generated that recognized both red and green opsin. The rod probe corresponds to nts 378-568; blue to nts 278-502; and red/green to nts 898-1095. All cDNAs spanned at least one intron. The cDNAs were blunt end cloned into the SmaI site of pBK-CMV (Stratagene, La Jolla, CA) and sequenced. This vector allows for the synthesis of both sense and antisense RNA probes using T3 and T7 promoters, respectively.

RNase protection assays

Total RNA was extracted from frozen pieces of trephined retina using an acid phenol extraction procedure described previously [25]. In separate control experiments using non-experimental eye tissue, we estimated that each trephined piece yielded approximately 1 μg of total RNA. To enhance yield during the extraction, all samples were spiked with 2.5 μg of yeast tRNA, which acted as a carrier during the extraction process. The pelleted RNA was then resuspended in 20 μl of water. Two μl of each sample was used for Rod opsin analysis and 13 μl of each sample was used for red/green opsin analysis. The remaining RNA was used to make first strand cDNA for real time RT-PCR (qPCR) analysis (see real time RT-PCR below). Previous experiments had determined that Blue cone opsin message was too rare to detect in these small isolates using RPA. Each sample was reacted with ³²P-UTP labeled antisense RNA probes (5x10⁵ cpm/sample) made from the previously described opsin clones. RPA was conducted as previously described [26,27]. Protected probes were run on 8 M urea, 6% polyacrylamide gels, fixed, dried, and exposed to X-ray film. Protected bands were cut out and quantified in a liquid scintillation counter. After correction for the difference in input, the ratio of rod:cone mRNA was determined in each trephined region.

In situ hybridization

In situ hybridization was carried out using a retinal wholemount procedure previously described in the literature [25,28] and digoxigenin-labeled antisense RNA probes. Hybridized probe was detected using an alkaline phosphatase processed substrate (Nitro-Blue Tetrazolium and BCIP). For en face sectioning through the inner segment regions of the photoreceptors, the retina pieces were first glued to small squares of Whatman No. 4 filter paper using cyanomethacrylate glue before undergoing the in situ hybridization procedure. Once hybridized and stained, the retina pieces were cleared by dehydration in a graded methanol series and then embedded in glycolmethacrylate (JB4 Plus, Polysciences, Inc., Warrington, PA) in an orientation that allowed for cutting of either en face or transverse sections. Five micron sections were examined and photographed using Nomarski interference microscopy.

Qualitative assessment of opsin staining of paired monkey specimens (both eyes of each animal) was made by two observers (H.R.P. and R.W.N.), who were masked to the procedure performed on each monkey (control animal or experimental glaucoma). The eyes were scored relative to each other as having equal staining intensity, or one eye having greater staining relative to the other, for each probe examined. Human eyes were scored by a single observer (T.M.N.). Since a comparative analysis could not be made on human eyes, staining intensity was scored on a 4-point scale 0 showing no stain, 1 showing mild staining, 2 showing moderate staining, and 3 showing marked or intense staining.

Real time RT-PCR

For quantitative real-time reverse transcription polymerase chain reaction (qPCR), first strand cDNA was first made from the remaining 5 μl of total RNA extracted from each piece of trephined retina, using reverse transcriptase and oligo(dT) as a primer [26,29]. Each qPCR reaction was set up using the opsin specific primers described in the above section entitled, "Cloning of monkey opsin cDNAs" with SYBR Green PCR master mix containing AmpliTaq Gold (Applied Biosystems, Foster City, CA), and 1 μl of the first strand cDNA template (containing approximately 100 pg of cDNA). Each primer set was replicated three times for each sample. PCR was run on an ABI 7300 Real Time PCR system (Applied Biosystems). The amplification conditions were 95 °C (15 s) and 55 °C (60 s) for 40 cycles for the rod and red/green primers and 95 °C (15 s) and 63.5 °C (60 s) for 40 cycles for the blue primers. Specificity of product synthesis was confirmed by conducting a melting curve of the samples after each run. In addition, agarose gel electrophoresis of the PCR products was conducted on the initial reactions for each primer set.

To quantify the abundance of each opsin cDNA, every PCR run was conducted with a standard curve comprising a dilution series of cloned opsin cDNA templates ranging from 10² to 10⁹ copies. Data were collected from threshold values using the automatic function of the Sequence Detection System software program. The amount of opsin cDNA in each sample was then calculated from the standard curve data in each respective PCR run. The total amount of cDNA for each opsin in each sample was calculated as the mean of the triplicate samples.

Quantification scheme for mRNA levels

Since only minute amounts of RNA were isolated from each trephined sample, we decided to express the amount of opsin mRNA detected in each sample as a ratio of rod opsin to cone (red/green or blue, respectively) opsin. With this approach, we did not have the limitation of first trying to measure the RNA concentration in each sample in order to standardize the amount of RNA used for RPA or qPCR experiments. The ratio of rod to cone opsin should be consistent from trephined samples of identical pieces of retina from different eyes, regardless of the efficiency of RNA extraction for each sample.

Results

In situ hybridization of opsin mRNAs: studies in monkeys with ocular hypertension and experimental glaucoma

Digoxingenin-labeled RNA probes were used for in situ hybridization experiments on retinal pieces harvested from the mid-peripheral region of the inferotemporal retina (see Figure 1). This region was evaluated because it exhibits the earliest and greatest amount of damage (other than areas immediately adjacent to the optic nerve) in glaucoma visual field tests in humans [30,31]. In addition, monkeys with experimental glaucoma, who have been trained to perform visual field tests, show a similar pattern of scotoma formation [32,33]. Probe hybridization was evaluated in sections cut either tangentially (en face) through the photoreceptor inner segment region, or radially through the entire retina. Figure 2 shows en face histology of specimens from the control and experimental eyes of the same monkey. The intensity of rod labeling was similar between control and glaucomatous eyes. In these images, the unlabeled cone inner segments in the glaucoma eye had a larger mean diameter (approximately 45% larger) than those in the control eye, consistent with previous observations of cone swelling. Red/green opsin was detected in these inner segment regions in control eyes, but not in glaucoma. Nomarski contrast microscopy of these specimens indicated the presence of cones in the glaucomatous eyes, suggesting that loss of mRNA was not owing to cell loss. Similarly, blue cone opsin was detected only in the control eye and not the experimental eye. In total, all four monkeys with chronic elevated IOP used for these experiments showed a decrease of both red/green and blue cone opsin mRNAs in the ocular hypertensive eye. None of the monkeys in this cohort exhibited a change in rod opsin mRNA staining. Two monkeys with no experimental glaucoma showed no remarkable difference in the opsin labeling between their left and right eyes.

Quantitative analysis of opsin mRNAs

RNase protection assays: Figure 3 shows the raw data for one cynomolgus experimental monkey. The amount of radioactivity in each protected fragment was quantified by liquid scintillation counting of the excised bands and the ratio of rod to red/green opsin mRNA in each trephined region was determined (Table 2). A larger ratio reflects a greater amount of rod mRNA than red/green cone mRNA in each trephined sample. An increase in the ratios in the experimental eye may occur if there were an increase in rod mRNA, or a decrease in red/green mRNA, or both. The results from in situ hybridization studies, (Figure 2) suggested that the ratio increases were most likely due to a decrease of red/green opsin mRNA. To compare the ratio of mRNAs in trephined samples taken from the control and experimental retinas, we also calculated the fold change in the ratio of the experimental eye (R_E) to the ratio in the control eye (R_C). The macular region showed a 4.3 fold increase in the ratio, while the mid-peripheral region showed an increase greater than seven-fold (Table 2). The peripheral retina of this animal showed a modest decrease in the ratio in the experimental eye.

Real time RT-PCR: qPCR was also conducted on RNA samples extracted from trephined specimens taken from two nonexperimental cynomolgus monkeys and five monkeys (two rhesus and three cynomolgus) with experimental ocular hypertension in one eye. In some cases, tissue was designated for other studies and was not available for all the target regions of the retina for all animals. As in the RPA data shown in Table 2, we calculated the ratio of rod to cone opsin mRNA in each piece of trephined retina. For red/green opsin, the mean ratio was higher in the eyes with experimental glaucoma compared to control eyes, in all regions of the retina, but was highest in the mid-peripheral retina (Table 3). For blue opsin, the mean ratios were higher in the mid-peripheral and peripheral trephined pieces of retina of the eyes with experimental glaucoma, but not in the macula (Table 3). Although there was a trend for the mean ratios to be higher in the experimental eyes, a comparison of the ratios for each probe in each region showed no statistical difference between control and experimental eyes in this group comparison. A possible reason for the lack of a statistical correlation was the high variability in the ratios from individual to individual. To account for this, we also evaluated the fold changes in the ratios (using the calculation of R_E/R_C) between the two eyes of each monkey. Table 4 summarizes the mean fold changes in normal monkeys and animals with experimental ocular hypertension. In comparisons of two normal eyes of the same monkey we detected a range of 0.9 to 2.3 fold difference in the R_C1/R_C2 for both red/green and blue opsin mRNAs in all regions of the retina examined. For comparisons made between the control and glaucomatous eyes of each experimental monkey, the mean R_E/R_C for red/green opsin was within this range in both the macula and periphery of these monkeys, suggesting that there was minimal change in cone mRNA levels in these regions between the two eyes. Conversely, the mean R_E/R_C for red/green opsin mRNA was significantly increased in the mid-periphery of experimental monkeys (Student's t test, p=0.046). The mean R_E/R_C for blue opsin mRNA was also increased in the macula and mid-periphery of glaucoma monkeys, but was also highly variable and not statistically significant (p=0.12 for the two regions combined).

Overall, the data obtained by quantitative mRNA and in situ hybridization studies suggested that a loss of red/green opsin mRNA consistently occurred in the mid-peripheral retinas of monkeys with experimental ocular hypertension, while blue opsin mRNA showed a more variable pattern of loss.

Evaluation of opsin expression in monkeys after glaucoma surgery

We also performed in situ hybridization studies for opsin mRNA on the retinas harvested from monkeys involved in a large glaucoma surgery study [23]. These samples provided a unique opportunity to examine the changes in opsin expression in retinas exposed to elevated IOP with subsequent IOP lowering therapy. Of the sixteen animals initially enrolled in this study, we were able to collect in situ data on 15 specimens. Nine monkeys showed no change in opsin mRNAs in the control and experimental eyes, while a loss of at least red/green cone mRNA was detected in the experimental eyes of six animals. For each of these two groups, we examined the amount of glaucomatous damage (as a function in the change in Cup:Disc ratio of the experimental eye) and the change in cumulative IOP ((IOP_experimental-IOP_control) x Days). The group of animals exhibiting a loss of cone mRNA had higher levels of cumulative IOP and more damage to their optic nerve, but these differences were not statistically significant (Table 5). Examination of the weekly IOP history of the eyes of these animals, however, showed that monkeys exhibiting loss of cone opsin mRNA had significantly higher peak IOPs measured in the experimental eye sometime within the two-week period before they were killed (Table 5). There was no correlation between the change in opsin mRNA levels and the antiproliferative treatment used during surgery, with the exception that these animals were more likely to have been treated with negative control reagents (balanced saline solution or adenovirus with no transgene-5 out of 6 monkeys), which would be expected for animals with reduced IOP control following failure of the filtration surgery.

In situ hybridization studies on human retinas

The data shown indicate that cone opsin mRNA is decreased in monkey eyes with ocular hypertension. To test if this phenomenon was also present in human eyes, we performed in situ hybridization analyses on post mortem retinas of normal donors and patients who had been diagnosed as having glaucoma. We obtained results for four eyes with no history of ocular disease and seven eyes diagnosed with glaucoma. A summary of the subjective scoring of the in situ results is shown in Table 6. Three of the four normal eyes showed strong labeling for all opsin probes, while one eye showed moderate staining for rod and red/green opsin probes (blue was not tested in this eye). All seven of the glaucoma specimens exhibited strong staining for rod opsin mRNA, but five of these eyes had weak or no staining for red/green opsin mRNA. Blue opsin mRNA was relatively weak or absent in three of these five eyes, and moderate in a fourth (blue opsin was not tested for in the fifth eye). Three of these five individuals had defects on their visual field exams, one had a normal exam, and no exam was available for the last one. Two of the glaucoma eyes showed strong staining for red/green and blue opsin mRNAs. One of these eyes (GL13) showed no defects on a visual field exam (Figure 4), while the other individual did have a visual field defect. Examples of two of the glaucoma specimens, compared to age-matched healthy eyes from human donors, are shown in Figure 4.

Discussion

Both qualitative (in situ hybridization) and quantitative studies on rod, red/green, and blue cone opsin mRNA levels in monkeys indicate that experimental ocular hypertension is associated with a decrease in the steady state level of cone opsin mRNAs, principally in the mid-peripheral (inferotemporal) retina and to a lesser extent in the macula. There was no evidence that this loss of mRNA was correlated to cone cell loss: microscopic evaluation of the tissues used for our in situ experiments showed the presence of nonstained cells, rather than the absence of cells. In situ hybridization assessment of opsin mRNAs was also made on a cohort of monkeys that had ocular hypertension, but were then treated to lower their IOP. In these animals, we also found a decrease in cone opsin expression, but only in those individuals that had significantly elevated IOPs within two weeks of being killed. There was no significant association between a decreased level of cone mRNA and the amount of glaucomatous damage or the overall level of elevated IOP present in each experimental eye. Last, in situ hybridization studies using human donor eyes showed that loss of red/green cone opsin mRNA was also potentially a feature of human retinas with glaucoma. One diagnostic feature of glaucoma, however, defects in a patient's visual field exam, only loosely correlated with the loss of cone opsin transcripts in the retinas (three out of four patients with defects exhibited cone opsin mRNA loss, while one patient with cone opsin mRNA loss had a normal visual field). The lack of a stronger correlation may be related to both timing and anatomy, since visual field exams were not available immediately before death and are principally a function of inner retinal defects. Validation of this observation clearly requires more eyes for study.

Collectively, these results are consistent with previous morphometric studies showing adverse effects on the outer retina, particularly the red and green cone population, in monkeys with experimental glaucoma and possibly the human disease as well. Our data suggest that the adverse effects on photoreceptor cells are principally mediated by ocular hypertension.

The primary trigger causing a reduction in cone opsin mRNA levels

As previously noted above, the decrease of cone opsin mRNA levels is likely not caused by the loss of cells themselves. Although photoreceptor cell dropout has been reported in glaucoma [5,8], cone loss, per se, is probably not a major feature of the disease process. Two studies, that looked at radial sections of glaucomatous eyes found no photoreceptor cell loss to within a 5% error of estimation [6,7], while one other study examined tangential sections and was able to detect some red/green cone loss in some human eyes with glaucoma, but not in rhesus monkeys with experimental glaucoma [8]. Instead, Nork et al. [8] found that photoreceptor swelling was common in both human eyes with diagnosed glaucoma and the eyes of monkeys with experimental ocular hypertension and glaucoma.

Our molecular studies are consistent with the hypothesis that cone photoreceptors respond adversely to elevated IOP. Reduced choroidal blood flow, resulting in a reduction in oxygenation and ischemia to the photoreceptor cell layer, may be one possible mechanism for this. Several studies have documented decreased choroidal blood flow in humans with glaucoma [34-40]. Since this is the principal blood supply feeding the outer retina, it is possible that both rods and cones are exposed to lower oxygen levels associated with reduced blood flow, but that cones are more susceptible. Studies have also documented an inverse relationship between the rate of choroidal blood flow and the intraocular pressure [41], indicating that a possible relationship between reduced blood flow, elevated IOP, and cone dysfunction may exist.

It is also possible that this effect on cone photoreceptors is reversible, so that they resume normal levels of opsin expression (and possibly function) some time after the IOP levels subside. This would account for why we did not detect the loss of cone opsin even in monkeys who had high IOPs well before, but not within the two weeks prior to, the time they were sacrificed in this surgery study. A similar relationship to elevated IOP and the reversibility phenomenon has also been observed in mfERG recordings in rhesus monkeys with acutely elevated IOP [13]. The hypotheses of cone susceptibility to elevated IOP and reversibility of the opsin expression pattern is testable in future studies.

The significance of a decrease in cone opsin mRNA

Opsin proteins are constantly being turned over in vertebrate photoreceptors [42]. The process of renewal is a highly coordinated process involving the cyclic transcription of both rod and cone opsin genes, followed by the active translation of mRNAs and the accumulation of protein [16,19,43]. In all vertebrate photoreceptors analyzed, these two events precede a burst of outer segment disc shedding and phagocytosis by the retinal pigmented epithelial cells, which is in turn followed by a period of disc renewal [20,44]. In most photoreceptors, this coordinated process is governed by either a diurnal rhythm regulated by the animal's light-dark cycle, or by the organism's circadian clock, or both [16-18]. Estimates of the renewal rate in mammals, indicates that cone outer segments are completely renewed in 5.4 days, while rods outer segments are renewed at a slightly slower rate of 7.75 days [44]. Thus, a normal cone photoreceptor, is turning over its opsin protein complement an average of about 20% per day. It stands to reason that a decrease in opsin mRNA would compromise the ability of a cone cell to maintain this high rate of protein renewal and result in a reduction of the normal steady state levels of protein.

How might the reduction of cone photopigment affect the physiology of the retina in a glaucomatous eye? Cones are less sensitive to light than rods [45], so it is conceivable that a reduction in photopigment associated with the loss of its mRNA, would further desensitize cones to photon capture and impair their ability to achieve a threshold level of phototransduction. Cone desensitization could account for several psychophysical and electrophysiological phenomena associated with glaucoma. Blue-yellow contrast sensitivity (short-wavelength automated perimetry or SWAP) changes are often able to detect early glaucomatous damage in humans before conventional white on white (achromatic) automated perimetry. Since an equal mixture of perception of red and green light produces the physiological sensation of yellow, the loss of red/green cone function would impair a patient's ability to detect yellow light in a SWAP exam. Essentially a brighter yellow light would be required to generate more photons to elicit a signal in cones with reduced photopigment. Interestingly, studies have also shown that SWAP exhibits a higher degree of variability in subjects who have been tested on multiple occasions compared to standard perimetry [46]. The variability factor may be explained by the potential ability of cones to eventually recover when IOP is reduced and express relatively normal levels of opsin again. In this sense, SWAP testing may be confounded by both the reversibility of red/green cone sensitivity and the loss of retinal ganglion cells subserving these regions of the retina.

In addition to perimetry testing, outer retinal changes detected in preliminary mfERG studies show that the macular and mid-peripheral responses to light stimulation of the retina are delayed in glaucoma patients [47] and in monkeys with experimental glaucoma [11]. The delay could also reflect a lowered sensitivity of cone photoreceptors, the dominant photon-sensing cell type in this region of the retina. Currently, these tests are not specific enough to distinguish between a decrease in cone function an, or impairment of retinal circuitry associated with light capture. However, the evidence showing abnormalities in cone mRNA expression for their respective photopigments may be a factor when considering the explanation for glaucoma-related changes in tests of outer retinal function.

In summary, a variety of studies have suggested that glaucoma involves changes in the outer retina. Our data build on these past studies by showing that cone photoreceptors are negatively affected by ocular hypertension because they exhibit a decrease in the normal levels of their opsin mRNAs. These data should reinforce the concept that although the critical pathology of glaucoma is localized to the inner retina (ganglion cell layer), it is a disease with a profound impact on the retina as a whole.

Acknowledgements

This work was supported by grants from the National Eye Institute (R03 EY1790 and R01 EY12223 to R.W.N., R01 EY14041 to T.M.N., and CORE grant P30 EY016665 to the Department of Ophthalmology and Visual Sciences), a Grant-in-Aid from the Fight for Sight, Prevent Blindness America (to R.W.N.) the Glaucoma Research Foundation (to T.M.N.), an unrestricted research gift from Allergan, Inc., and an unrestricted research grant from Research to Prevent Blindness. Support for the monkeys used for the glaucoma surgery study was provided from Canji, Inc., to R.W.N. The authors would also like to thank Dr. Paul Kaufman for the donation of retinal tissue from pairs of normal monkey eyes and Ms. Cassandra L. Miller for her assistance in summarizing the IOP data.

References

1. Quigley HA, Nickells RW, Kerrigan LA, Pease ME, Thibault DJ, Zack DJ. Retinal ganglion cell death in experimental glaucoma and after axotomy occurs by apoptosis. Invest Ophthalmol Vis Sci 1995; 36:774-86.

2. Nickells RW. Apoptosis of retinal ganglion cells in glaucoma: an update of the molecular pathways involved in cell death. Surv Ophthalmol 1999; 43:S151-61.

3. Poinoosawmy D, Nagasubramanian S, Gloster J. Colour vision in patients with chronic simple glaucoma and ocular hypertension. Br J Ophthalmol 1980; 64:852-7.

4. Lakowski R, Drance SM. Acquired dyschromatopsias: the earliest functional losses in glaucoma. Doc Ophthalmol Proc Ser 1979; 19:159-65.

5. Panda S, Jonas JB. Decreased photoreceptor count in human eyes with secondary angle-closure glaucoma. Invest Ophthalmol Vis Sci 1992; 33:2532-6.

6. Kendell KR, Quigley HA, Kerrigan LA, Pease ME, Quigley EN. Primary open-angle glaucoma is not associated with photoreceptor loss. Invest Ophthalmol Vis Sci 1995; 36:200-5.

7. Wygnanski T, Desatnik H, Quigley HA, Glovinsky Y. Comparison of ganglion cell loss and cone loss in experimental glaucoma. Am J Ophthalmol 1995; 120:184-9.

8. Nork TM, Ver Hoeve JN, Poulsen GL, Nickells RW, Davis MD, Weber AJ, Vaegan, Sarks SH, Lemley HL, Millecchia LL. Swelling and loss of photoreceptors in chronic human and experimental glaucomas. Arch Ophthalmol 2000; 118:235-45.

9. Nork TM. Acquired color vision loss and a possible mechanism of ganglion cell death in glaucoma. Trans Am Ophthalmol Soc 2000; 98:331-63.

10. Raz D, Seeliger MW, Geva AB, Percicot CL, Lambrou GN, Ofri R. The effect of contrast and luminance on mfERG responses in a monkey model of glaucoma. Invest Ophthalmol Vis Sci 2002; 43:2027-35.

11. Ver Hoeve JN, Kim CBY, Heatley GA, Kaufman PL, Kiland JA, Nickells RW, Nork TM. Delay of an early multifocal ERG wave feature in experimental glaucoma. Invest Ophthalmol Vis Sci 2001; 42:S148.

12. Ver Hoeve JN, Kim CBY, Nickells RW, Kaufman PL, Heatley GA, Faha, B, Nork TM. Recovery of late multifocal ERG waveform complex in epxerimental ocular hypertension following IOP reduction. ARVO Annual Meeting; 2002 May 5-10; Fort Lauderdale (FL).

13. Wilkerson CL, Kim CBY, Kaufman PL, Gabelt BT, Seeman JL, Nork TM, Ver Hoeve JN. Acute elevation of intraocular pressure increases multifocal ERG amplitude. Invest. Ophthalmol Vis Sci 2001; 42:S147.

14. Nork TM, Kim CBY, Lucarelli MJ, Levin LA, Kaufman PL, Ver Hoeve JN. Multifocal electroretinographic (mfERG) evidence for persistent outer retinal injury in chronic ocular hypertension in non-human primates. ARVO Annual Meeting; 2005 May 1-5; Fort Lauderdale (FL).

15. Nickells RW, Burgoyne CF, Quigley HA, Zack DJ. Cloning and characterization of rod opsin cDNA from the Old World monkey, Macaca fascicularis. Invest Ophthalmol Vis Sci 1995; 36:72-82.

16. von Schantz M, Lucas RJ, Foster RG. Circadian oscillation of photopigment transcript levels in the mouse retina. Brain Res Mol Brain Res 1999; 72:108-14.

17. Pierce ME, Sheshberadaran H, Zhang Z, Fox LE, Applebury ML, Takahashi JS. Circadian regulation of iodopsin gene expression in embryonic photoreceptors in retinal cell culture. Neuron 1993; 10:579-84.

18. Halstenberg S, Lindgren KM, Samagh SP, Nadal-Vicens M, Balt S, Fernald RD. Diurnal rhythm of cone opsin expression in the teleost fish Haplochromis burtoni. Vis Neurosci 2005; 22:135-41.

19. McGinnis JF, Whelan JP, Donoso LA. Transient, cyclic changes in mouse visual cell gene products during the light-dark cycle. J Neurosci Res 1992; 31:584-90.

20. Korenbrot JI, Fernald RD. Circadian rhythm and light regulate opsin mRNA in rod photoreceptors. Nature 1989; 337:454-7.

21. Gaasterland D, Kupfer C. Experimental glaucoma in the rhesus monkey. Invest Ophthalmol 1974; 13:455-7.

22. Kaufman PL, Wallow IH. Minified diagnostic contact lenses for biomicroscopic examination and photocoagulation of the anterior and posterior segment in small primates. Exp Eye Res 1985; 40:883-5.

23. Heatley G, Kiland J, Faha B, Seeman J, Schlamp CL, Dawson DG, Gleiser J, Maneval D, Kaufman PL, Nickells RW. Gene therapy using p21WAF-1/Cip-1 to modulate wound healing after glaucoma trabeculectomy surgery in a primate model of ocular hypertension. Gene Ther 2004; 11:949-55.

24. Kaufman PL, Davis GE. "Minified" Goldmann applanating prism for tonometry in monkeys and humans. Arch Ophthalmol 1980; 98:542-6.

25. Schlamp CL, Nickells RW. Light and dark cause a shift in the spatial expression of a neuropeptide-processing enzyme in the rat retina. J Neurosci 1996; 16:2164-71.

26. Schlamp CL, Johnson EC, Li Y, Morrison JC, Nickells RW. Changes in Thy1 gene expression associated with damaged retinal ganglion cells. Mol Vis 2001; 7:192-201 <http://www.molvis.org/molvis/v7/a27/>.

27. Levin LA, Schlamp CL, Spieldoch RL, Geszvain KM, Nickells RW. Identification of the bcl-2 family of genes in the rat retina. Invest Ophthalmol Vis Sci 1997; 38:2545-53.

28. Li HS, Yang JM, Jacobson RD, Pasko D, Sundin O. Pax-6 is first expressed in a region of ectoderm anterior to the early neural plate: implications for stepwise determination of the lens. Dev Biol 1994; 162:181-94.

29. Schlamp CL, Poulsen GL, Nork TM, Nickells RW. Nuclear exclusion of wild-type p53 in immortalized human retinoblastoma cells. J Natl Cancer Inst 1997; 89:1530-6.

30. Hart WM Jr, Becker B. The onset and evolution of glaucomatous visual field defects. Ophthalmology 1982; 89:268-79.

31. Phelps CD, Hayreh SS, Montague PR. Comparison of visual field defects in the low-tension glaucomas with those in the high-tension glaucomas. Am J Ophthalmol 1984; 98:823-5.

32. Harwerth RS, Smith EL 3rd, DeSantis L. Experimental glaucoma: perimetric field defects and intraocular pressure. J Glaucoma 1997; 6:390-401.

33. Harwerth RS, Carter-Dawson L, Shen F, Smith EL 3rd, Crawford ML. Ganglion cell losses underlying visual field defects from experimental glaucoma. Invest Ophthalmol Vis Sci 1999; 40:2242-50.

34. Rojanapongpun P, Drance SM, Morrison BJ. Ophthalmic artery flow velocity in glaucomatous and normal subjects. Br J Ophthalmol 1993; 77:25-9.

35. Michelson G, Groh MJ, Groh ME, Grundler A. Advanced primary open-angle glaucoma is associated with decreased ophthalmic artery blood-flow velocity. Ger J Ophthalmol 1995; 4:21-4.

36. Cellini M, Possati GL, Caramazza N, Caramazza R. Colour Doppler analysis of the choroidal circulation in chronic open-angle glaucoma. Ophthalmologica 1996; 210:200-2.

37. Yamazaki Y, Drance SM. The relationship between progression of visual field defects and retrobulbar circulation in patients with glaucoma. Am J Ophthalmol 1997; 124:287-95.

38. Butt Z, O'Brien C, McKillop G, Aspinall P, Allan P. Color Doppler imaging in untreated high- and normal-pressure open-angle glaucoma. Invest Ophthalmol Vis Sci 1997; 38:690-6.

39. Kaiser HJ, Schoetzau A, Stumpfig D, Flammer J. Blood-flow velocities of the extraocular vessels in patients with high-tension and normal-tension primary open-angle glaucoma. Am J Ophthalmol 1997; 123:320-7.

40. Fuchsjager-Mayrl G, Wally B, Georgopoulos M, Rainer G, Kircher K, Buehl W, Amoako-Mensah T, Eichler HG, Vass C, Schmetterer L. Ocular blood flow and systemic blood pressure in patients with primary open-angle glaucoma and ocular hypertension. Invest Ophthalmol Vis Sci 2004; 45:834-9.

41. Findl O, Strenn K, Wolzt M, Menapace R, Vass C, Eichler HG, Schmetterer L. Effects of changes in intraocular pressure on human ocular haemodynamics. Curr Eye Res 1997; 16:1024-9.

42. Fernald RD. Principles of sensory regeneration. In: Bock GR, Marsh J, Whelan J, editors. Regeneration of Vertebrate Sensory Receptor Cells. New York: Wiley; 1991. p. 318-329.

43. Young RW, Droz B. The renewal of protein in retinal rods and cones. J Cell Biol 1968; 39:169-84.

44. Fisher SK, Pfeffer BA, Anderson DH. Both rod and cone disc shedding are related to light onset in the cat. Invest Ophthalmol Vis Sci 1983; 24:844-56.

45. Rodieck RW. The First Steps in Seeing. Sunderland (MA): Sinauer Associates; 1998.

46. Blumenthal EZ, Sample PA, Zangwill L, Lee AC, Kono Y, Weinreb RN. Comparison of long-term variability for standard and short-wavelength automated perimetry in stable glaucoma patients. Am J Ophthalmol 2000; 129:309-13.

47. Ver Hoeve JN, Murdock TJ, Heatley GA, Nork TM. Increased latency of early multifocal ERG response in glaucoma. Invest Ophthalmol Vis Sci 2000; 41:S520.