Gelatinase A and TIMP-2 Expression in the Fibrous Sclera of Myopic and Recovering Chick Eyes

Abstract

purpose. Myopia, or nearsightedness, is characterized by excessive lengthening of the ocular globe and is associated with extracellular matrix remodeling in the posterior sclera. The activity of gelatinase A, a member of the matrix metalloproteinase family, has been shown to increase in the posterior sclera during the development of induced myopia in several species. In the present study, the distribution and relative expression of gelatinase A and its associated inhibitor, tissue inhibitor of metalloproteinases (TIMP)-2, were measured within the fibrous scleras of experimentally myopic (form-deprived) eyes, control eyes, and eyes recovering from form deprivation to better understand the mechanisms that regulate scleral remodeling and the rate of ocular elongation.

methods. Total RNA was extracted from the posterior scleras of form-deprived chick eyes, eyes recovering from deprivation myopia, and paired contralateral control eyes, and subjected to northern blot analysis analyses using cDNA probes to chicken gelatinase A and TIMP-2. The distribution of gelatinase A and TIMP-2 mRNAs was evaluated by in situ hybridization on frozen sections of chick scleras using ³³P-labeled RNA probes. Gelatinase A activity within the fibrous scleras of form-deprived eyes and paired contralateral recovering eyes was evaluated by gelatin zymography.

results. Northern blot analysis indicated that the relative expression of gelatinase A was increased by 128% in deprived eyes (P = 0.009), whereas after 1 day of recovery, levels were decreased by 80% in scleras from recovering eyes (P = 0.005). In contrast, TIMP-2 expression was significantly decreased (−53%, P = 0.027) in the posterior scleras of form-deprived eyes. No significant differences were detected in levels of TIMP-2 expression between recovering eyes and paired control eyes. In situ hybridization indicated that most of the gelatinase A transcripts were present in the fibrous layer of the posterior scleras from form-deprived and recovering eyes.

conclusions. Changes in the steady state levels of gelatinase A and TIMP-2 mRNA lead to changes in gelatinase activity within the fibrous sclera and mediate, at least in part, the process of visually regulated ocular growth and scleral remodeling.

High myopia in humans is characterized by excessive lengthening of the posterior portion of the ocular globe and is associated with scleral thinning in posterior regions and changes in the organization of scleral extracellular matrix. ¹ Although the cause of myopia in humans is complex, there is strong evidence from clinical and experimental studies that ocular elongation and myopia occur in response to alterations in the visual environment. ^{2 3 4 5 6 7} Myopia can be induced in a variety of animal species by rearing them under visual conditions that reduce visual contrast or by imposing hyperopic defocus on the retina with the use of negative lenses. ^{8 9} An avian model of experimental myopia, form-deprivation myopia in the chick, has been widely used as a model of myopia for reasons of convenience and economy because of the rapid rate of ocular growth and development of myopia. ^{10 11 12} In chicks, ocular elongation is associated with increased growth of the posterior sclera, as evidenced by increases in protein and proteoglycan synthesis and accumulation and by increases in total scleral mass. ^{13 14 15} After restoration of normal vision (recovery), the rate of proteoglycan synthesis in the posterior sclera rapidly decreases (within 24 hours) to levels below those of paired controls coincident with a temporary cessation of vitreous chamber elongation. ^{16 17}  

Unlike the sclera of most mammals, the sclera of a chick consists of an outer fibrous layer and inner cartilaginous layer the relative thicknesses of which vary inversely with each other. Although the overall thickness of the sclera remains unchanged during the development of myopia, the cartilaginous portion becomes thicker, and the fibrous portion becomes thinner than the same structures in control eyes. ^{18 19} Additionally, decreases in proteoglycan synthesis have been observed in the fibrous sclera of myopic chick eyes, relative to controls. ²⁰ The changes in the fibrous sclera of myopic chick eyes resemble those reported in the sclera of marmosets, ²¹ tree shrews, ²² and humans ²³ and have led to the hypothesis that the fibrous scleral layer of the chick is analogous to the fibrous sclera of most mammals. ¹⁹  

We have previously shown that the development of myopia is associated with increased amounts of gelatinase A proenzyme and a decreased amount of tissue inhibitor of metalloproteinases (TIMP) in the posterior scleras of chick eyes, ²⁴ suggesting that after activation by an unknown mechanism, gelatinase A could mediate extracellular matrix turnover in the scleras of deprived eyes. This hypothesis is strengthened by the observation that the rate of proteoglycan turnover is higher in the posterior scleras of deprived eyes and that gelatinase A and/or stromelysin are involved in the degradation and removal of aggrecan from the cartilaginous scleral layer. ²⁵  

In the present study, we extend our previous studies to show that gelatinase A and TIMP-2 are primarily expressed in the fibrous scleral layer of the chick eye and that their steady state mRNA levels change depending on the visual condition and the rate of ocular elongation. These findings suggest that during ocular growth, gelatinase A is activated within the fibrous scleral layer and contributes to the scleral remodeling processes involved in regulating ocular size and refraction. 

Materials and Methods

Induction of Experimental Myopia

Form-deprivation myopia was induced in two-day old white leghorn chicks (Magic City Hatchery; Minot, ND) by applying translucent plastic goggles, as previously described. ^{13 16} Two experimental paradigms were used for the experiments is this study: 1) Scleras harvested for RNA isolation and purification were obtained from chicks that underwent monocular form deprivation in the right eye, with the left eye serving as a paired control. The goggles remained in place for 10 days, at which time one group of chicks was killed immediately, whereas others had the occluders removed and were allowed to recover from the form deprivation for 24 hours. From these visual manipulations scleras were obtained from form-deprived eyes and paired controls, as well as from recovering eyes and their paired controls. 2) Scleras harvested for in situ hybridization and gelatin zymography were obtained from chicks that underwent monocular form deprivation in the right eye for 7 days to induce myopia, at which time the occluder was removed and placed on the contralateral eye for 3 days. After the 10th day of treatment (the end of the 3rd day of form deprivation in the left eye), the birds were killed, and scleras were harvested for experimental analyses. This goggle-switching manipulation was conducted to obtain both form-deprived (left) eyes and recovering (right) eyes from the same birds, for paired analyses. Chicks were maintained on a 12:12-hour light–dark cycle and were checked daily. Birds that had goggles fall off were not included in the study. The chicks were maintained and treated in accordance with National Institutes of Health guidelines and with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

RNA Isolation

After 10 days of form deprivation or 10 days of form deprivation followed by 1 day of recovery from form deprivation, chicks were killed, and experimental and control eyes were enucleated. Each eye was divided into anterior and posterior hemispheres, and the posterior hemisphere was gently cleaned of all vitreous, retina, pigmented epithelium, pecten, choroid, and muscle. The posterior scleral hemispheres of control and experimental eyes were snap frozen in liquid nitrogen and stored at −80°C. Four or five scleras were pooled separately for each group and were pulverized using a cryogenic mill under liquid nitrogen (Spex, Metuchen, NJ), and the frozen powder was transferred to a test tube containing 6 ml reagent solution (Trizol; Gibco, Gaithersburg, MD). The pulverized scleras were then homogenized with a rotor–stator assembly (Virtis, Gardiner, NY) for 5 minutes at room temperature, and RNA was isolated from the homogenate using a standard protocol for the reagent. The RNA was quantified and its purity assessed by spectrophotometry at 260 and 280 nm and by denaturing gel electrophoresis. 

Reverse Transcription–Polymerase Chain Reaction and Cloning

Two polymerase chain reaction (PCR) primers were selected for reverse transcription–polymerase chain reaction (RT-PCR) that flanked a region (−6–456 bp) of the published sequence of the chick 72-kDa matrix metalloproteinase (gelatinase A) gene. ²⁶ The forward primer (5′-TTAGCTGCACCGTCACCAATC-3′) corresponded to bases −6 to 14 of the chick gelatinase A sequence, whereas the backward primer (5′-AGCCATCACCATGTTCCCATC-3′) corresponded to bases 456 to 436 of the chick gelatinase A sequence. Two sets of primers were used to generate TIMP-2 cDNA clones, based on the published sequence of the TIMP-2 gene. ²⁷ One set (forward primer, 5′-ATGGGAACCCCATCAAGCGA-3′; backward primer, 5′-TTCTCCATCGCCCAGTCTGTCCAG-3′) was used to generate a 358-bp probe (TIMP-2 probe 1) between bases 185 and 542 of the TIMP-2 sequence, whereas a second set (forward primer, 5′-CGACGTAGTGATCCGAGCAAAG-3′; backward primer, 5′-TCACACAGCGTGATGTGCATC-3′) was used to generate a 261-bp probe (TIMP-2 probe 2) between bases 123 and 383 of the TIMP-2 DNA sequence. RT-PCR experiments were performed on 1 μg total chick scleral RNA, using an RT-PCR kit (GeneAmp, PE Applied Biosystems), according to standard protocol. Sequences of the amplified products were confirmed by nucleic acid sequencing using the dideoxy termination chain method (Sequenase V. 2.0; USB, Cleveland, OH). The amplified regions of the gelatinase A and TIMP-2 genes were cloned into the blunt-end vector (PCR-script; Strategene, La Jolla, CA), and the purified plasmids were used for the generation of cDNA and cRNA probes for northern and in situ hybridization experiments. 

Northern Blot Analysis

Total cellular RNA (10 μg/lane) isolated from the scleras of four or five form- deprived, recovering, and control eyes was resolved by formaldehyde-agarose denaturing gels, and transferred by capillary blotting to nylon membranes (Nytran, Schleicher and Schuell, Keene, NH). Membranes were prehybridized and hybridized using a buffer system (Rapid Hybe; Amersham Pharmacia Biotech, Piscataway, NJ), according to the standard protocol. Gelatinase A fragment was isolated from the PCR-script plasmid by restriction digestion and was labeled with ³²P-αdCTP using a random prime labeling system (Amersham Pharmacia Biotech) and was used as a probe for steady state levels of gelatinase A mRNA. An antisense control probe, an 80-bp fragment of the human 18S ribosomal RNA gene (Ambion, Austin, TX) was used to generate ³²P-RNA probes for the northern blot analysis to determine loading for each sample. Steady state levels of gelatinase A mRNA in form-deprived, recovering, and control scleras were measured by quantitative densitometry of specific bands on the developed autoradiogram. 

In Situ Hybridization

In situ hybridization was performed as previously described. ²⁸ After enucleation, eyes were opened by an equatorial incision, vitreous gel and liquid were removed, and posterior hemispheres of the eyes were fixed in 4% paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.4, 4°C) for 1 hour, followed by immersion in sterile 15% sucrose in phosphate-buffered saline (PBS) overnight at 4°C. The posterior hemispheres were then embedded in cutting compound (OCT; Miles, Elkhart, IN), and frozen sections (6μ m) of scleras were placed on slides (Superfrost Plus; Fisher Scientific, Pittsburgh, PA) and stored at −80°C. ³³P-UTP–labeled RNA transcripts were synthesized from cDNA clones in linearized PCR-script vector using T7 and T3 polymerase to generate sense and antisense RNA probes. Slides containing tissue sections were removed from the freezer and immediately placed in 90% EtOH for 5 minutes, followed by fixation with 4% paraformaldehyde in 0.1 M NaPO₄ (pH 7.4, 4°C) for 10 minutes at room temperature and then washed and processed for in situ hybridization, as previously described. ²⁸ Autoradiography was performed on ³³P-labeled sections using standard methods for nuclear track emulsion (Amersham Pharmacia Biotech). After exposure of slides to emulsion for 3 to 4 days at 4°C in a light-tight box containing desiccant, slides were developed, counterstained with hematoxylin and eosin, and photographed. 

Zymography

Five-millimeter punches were isolated from the posterior scleras of chicks in which the right eye was form deprived for 7 days, and then the left eye was form deprived for 3 days (experimental paradigm 2, described earlier) to obtain tissue from paired recovering (right) and form-deprived (left) eyes. Scleral punches were frozen at −80°C for later use. After thawing, scleras were immersed in Hanks’ balanced salt solution (Ca²⁺-Mg²⁺–free), and the fibrous layer of the sclera was gently dissected from the cartilaginous layer of sclera with the aid of a dissecting microscope. Soluble proteins were extracted from the fibrous scleras with a solution of 2.0% sodium dodecyl sulfate (SDS), followed by vortexing for 10 minutes as previously described. ^{24 29} Protein concentrations were determined immediately after extraction using a protein assay (microBCA, Pierce, Rockford, IL), and equal amounts of protein (4.6 μg) were applied to an 11% SDS-acrylamide minigel containing 0.2% gelatin. After sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), gelatin zymography was performed as described previously. ²⁴ Digitized images of the wet gels were obtained using a flatbed scanner, and densitometry was performed on negative images of the zymograms using image analysis software (NIH Image ver. 6.4; National Institutes of Health, Bethesda, MD). 

Statistical Analyses

The probability distribution of mRNA-to-rRNA ratios obtained from northern blot analysis was log-normal, and Pearson correlation coefficients based on this assumption ranged from 0.917 to 0.960. Two-tailed t-tests were used to test the null hypothesis that the ratio of the experimental sample to control was 1 (the log of the experimental sample to control was 0). 

Student’s two-tailed t-tests for matched pairs were used to compare densitometry data from zymograms from paired form-deprived and recovering eyes. 

Results

RT-PCR and Cloning

Using RT-PCR on total scleral RNA, we have obtained a 463-bp PCR product corresponding to the N-terminal region within the propeptide of gelatinase A, based on the published gelatinase A sequence from chick embryo fibroblasts. ²⁶ Two TIMP-2 cDNA probes were generated using RT-PCR from total chick scleral RNA, corresponding to the complete TIMP-2 sequence as shown in Figure 1 . These PCR products were cloned, sequenced, and verified by comparison of their nucleotide sequences with those published for embryonic chicken fibroblasts. ^{26 27}  

Northern Blot Analyses

Gelatinase A and TIMP-2 mRNA were analyzed by northern blot analysis and quantified by scanning densitometry of the bands on autoradiographs. Gelatinase A mRNA from control, myopic, and recovering scleras was present as a 3.2-kb band, in agreement with other studies on gelatinase A mRNA (Fig. 2A ). When expressed relative to levels of 18S rRNA, steady state levels of gelatinase A mRNA were found to be significantly increased in the scleras of form-deprived eyes (128.30% ± 23.051%, P = 0.009), compared with paired controls, and significantly decreased in the scleras of recovering eyes (−80.18% ± 6.438%; P = 0.005; Fig. 2B ). 

TIMP-2 mRNA from control, myopic, and recovering scleras was present as a 4.7-kb message (Fig. 2C) , larger than previously reported message sizes for TIMP-2, which ranged from 1.0 kb to 3.8 kb. ^{27 30 31} To confirm that the 4.7-kb band represented the TIMP-2 message, a second RT-PCR product was generated from a different region of the TIMP-2 gene, which was cloned and used as a probe for northern blot analysis. This second TIMP-2 cDNA clone also hybridized to the 4.7-kb band, indicating that the 4.7-kb band represents the size of the TIMP-2 message in chick sclera (data not shown). TIMP-2 expression was compared in control, recovering, and form-deprived eyes and determined to be significantly decreased in the scleras from form- deprived eyes (−53.11% ± 9.605%; P = 0.027; Fig. 2D ). No significant differences were detected in levels of TIMP-2 mRNA in scleras from recovering and paired control eyes. When contralateral control eyes of form-deprived chicks were compared with contralateral control eyes of chicks recovering from form deprivation, no significant differences were detected in gelatinase A or TIMP-2 expression (P = 0.355 and P = 0.532, respectively). 

In Situ Hybridization

The distribution of gelatinase A and TIMP-2 mRNA within the chick sclera was determined using in situ hybridization analyses with ³³P-labeled antisense cRNA probes on sections of form-deprived and recovering scleras. Gelatinase A expression was most intense in the fibrous layer of the recovering and myopic chick sclera, with only a very weak, diffuse signal in the cartilaginous layer (Fig. 3) . The distribution of TIMP-2 was very similar to that of gelatinase A, with highest levels of expression in the fibrous scleral layer in both recovering and myopic eyes (Fig. 4) . 

Zymography

To carry out a paired analysis of gelatinase activity in fibrous sclera from form-deprived and recovering eyes, a goggle-switching paradigm was used (see Materials and Methods section). Gelatinolytic proteins present in the fibrous layer of the scleras from paired recovering and form-deprived eyes were identified and quantified on gelatin substrate gels. The proenzyme and active forms of gelatinase A were present in the fibrous scleras of recovering and form-deprived eyes, as indicated by the presence of the 65- and 58-kDa bands on gelatin zymograms, respectively (Fig. 5A ). Scanning densitometry of zymograms indicated that the fibrous scleras of form-deprived eyes contained significantly higher levels of the active form of gelatinase A than contralateral recovering scleras (77.98% ± 26.81%; P = 0.041; Fig. 5B ). No significant differences were detected in the levels of the gelatinase A proenzyme in the fibrous scleras from form-deprived and recovering eyes. These results indicate that increased gelatinase A expression and decreased TIMP-2 expression in the fibrous sclera results in significantly higher levels of activated gelatinase A in form-deprived eyes. 

Discussion

In the present study, we show that gelatinase A and TIMP-2 were expressed primarily in the fibrous sclera of the chick eye and the steady state levels of mRNA for these genes were modulated by the visual environment. Results from northern blot analysis indicated that the message size for gelatinase A in the chick sclera was similar to that in other published reports, whereas the TIMP-2 message in chick sclera was found to be larger (4.7 kb) than that reported for chick embryo fibroblasts (2.4 kb). Although there are no published reports of the message size of TIMP-2 in posthatch chick tissues, transcript sizes ranging from 1.0 to 3.8 kb have been reported for the TIMP-2 gene in human and murine normal and tumor tissues. ^{30 31} It has been suggested that alternative 5′ untranslated regions or alternative polyadenylation signals could account for different transcript sizes observed for mammalian TIMP-2, ³⁰ and it is possible that a similar mechanism may be responsible for the differences in observed in the chick TIMP-2 transcript size. 

Results from the present study show that during the development of form-deprivation myopia, steady state levels of gelatinase A mRNA were increased, and levels of TIMP-2 mRNA were decreased in the posterior sclera. After restoration of normal-form vision, gelatinase A mRNA levels decreased in the posterior sclera. In situ hybridization experiments indicated that gelatinase A and TIMP-2 were expressed primarily in the outer fibrous layer of the chick scleras from form-deprived and recovering eyes. We have previously shown that the rate of thickening of the fibrous sclera in the posterior pole of form-deprived eyes is significantly slower than that of control eyes and that this results in a significantly thinner fibrous sclera than in the same region of contralateral control eyes. ¹⁹ After restoration of normal form vision, the rate of scleral thickening increases significantly in the recovering eye. The changes in gelatinase A and TIMP-2 expression observed during the induction and recovery from experimental myopia in the present study suggest that gelatinase A activity in the posterior fibrous sclera is involved in the modulation of fibrous scleral thickness in chicks. Interestingly, gelatinase A and TIMP-2 have been shown to be strongly expressed in the perichondrium of the developing mouse mandible ³² and in the human fetal limb, ³³ where it has been suggested that they may participate in the regulated breakdown and “controlled sliding” of the perichondrium presumed necessary for the developing cartilage to expand and grow properly. The fibrous layer of the chick sclera is adjacent to a cartilaginous layer and may be analogous to the perichondrium of developing cartilage. Therefore, changes observed in the fibrous layer of the chick sclera during ocular growth may be similar to those that occur in the perichondrium during the remodeling of other cartilaginous structures. 

The matrix of the fibrous scleral layer is composed of irregularly arranged lamellae consisting of collagen type I fibers interspersed with the small chondroitin-dermatan sulfate proteoglycans biglycan and decorin. ^{14 34} The cartilage-associated macromolecules collagen types II and IX and the proteoglycan aggrecan are absent or considerably reduced in amount in the fibrous scleral layer. ³⁴ Although denatured collagen has been considered the primary substrate for gelatinase activity, ³⁵ it has been shown that human and chicken gelatinase A, when free of TIMP, are capable of cleaving fibrillar collagen to an equal or greater extent than interstitial collagenase (matrix metalloproteinase[ MMP]-1). ³⁶ A related gelatinase, gelatinase B (MMP-9) is unable to cleave fibrillar collagen, indicating that the specific collagenolytic activity of gelatinase A is not a general property of gelatinases. Therefore, it is possible that gelatinase A activity alone could be responsible for the changes in fibrous scleral thickness observed in deprived and recovering chick eyes. 

Gelatinase A is secreted as a latent proenzyme and must be activated extracellularly. However, unlike other members of the MMP family, progelatinase A and TIMP-2 bind selectively through ionic interactions between their C-terminal domains. ^{37 38 39 40} The result is decreased efficiency of activation of the proenzyme and lower specific activity of the subsequently activated proteinase. ^{41 42 43} After cleavage of the profragment, the N-terminal domain of active gelatinase A becomes available for binding to all TIMPs, which act to inhibit the action of the gelatinase. The results of our previous studies indicated that under conditions of visual deprivation, an increased amount of free proenzyme, together with a decrease in amount of TIMP, resulted in a significant increase in availability of gelatinase A proenzyme for activation. These earlier findings are supported by the results of the present study, which showed that the increased rate of ocular growth resulting from visual deprivation was associated with increased steady state levels of gelatinase A mRNA and decreased steady state levels of TIMP-2 mRNA in the posterior sclera. These alterations in levels of mRNA resulted, at least in part, in an increased amount of the gelatinase A proenzyme and decreased levels of TIMP-2 in the scleras of experimentally myopic eyes, which led to increased levels of active gelatinase and increased collagenolytic activity within the posterior fibrous sclera. In contrast to the increase in gelatinase A mRNA observed during the development of myopia, a decrease in gelatinase A mRNA was observed in the posterior sclera during recovery from experimental myopia, when the rate of ocular elongation is slower than that of control eyes. We speculate that the lower levels of gelatinase A mRNA in the scleras of recovering eyes, together with normal levels of TIMP-2 mRNA, shift the balance between synthesis and degradation back to an anabolic state, with lower levels of collagenolytic activity within the posterior fibrous sclera. 

In general, regulation of MMPs may occur at the levels of transcription, activation of the latent proenzyme, and inhibition by specific inhibitors such as TIMPs. ⁴⁴ Our results indicate that scleral gelatinase A and TIMP-2 were regulated at least partly at the level of steady state mRNA, possibly by transcriptional control. Unlike other MMPs, progelatinase A is constitutively expressed by many cell types and is not readily induced by agents such as tissue plasminogen activator (TPA) or interleukin (IL)-1α, which have been shown to increase transcription of other MMPs. ³⁷ Unlike other MMPs, the gelatinase A promoter also has no an upstream transforming growth factor (TGF)-β inhibitory element (TIE) ⁴⁵ and is slightly upregulated rather than inhibited by TGF-β1. ^{40 46} Interestingly, TGF-β1 has also been shown to decrease TIMP-2 mRNA transcript levels in tumor cell lines, whereas it increases TIMP-1 transcript levels. ³⁰ Thus, TGF-β1 induces the same changes in gelatinase A and TIMP-2 expression in tumor cells that we have observed in the sclera during the development of myopia. Although several studies have examined TGF-β in experimentally myopic eyes, ^{47 48 49} there is no consensus about whether this growth factor has a stimulatory or an inhibitory role in the regulation of ocular growth, but on the basis of the results reported here, we predict that TGF-β1 can stimulate scleral remodeling and ocular elongation, as suggested by Rohrer and Stell ⁴⁷ and Seko et al. ⁴⁸  

It has been hypothesized that the fibrous sclera controls the remodeling of extracellular matrix in the cartilaginous sclera. ²⁰ This hypothesis is supported by our results, which show that the expression of gelatinase A and TIMP-2 in the fibrous sclera was modulated by the visual environment. These changes in gene expression were responsible, in part, for increased levels of gelatinase activity in the scleras of form-deprived eyes. This increased gelatinase activity may act to thin the fibrous sclera and participate in the turnover of proteoglycans from the cartilaginous sclera. The mechanisms that regulate transcription of gelatinase A and TIMP-2 in the sclera have not been characterized, but further investigations are under way. 

 Supported by National Institutes of Health Grant EY09391 (JAR).

 Submitted for publication March 12, 1999; revised May 27 and July 9, 1999; accepted July 19, 1999.

 Commercial relationships policy: N.

 

Figure 1.

 

Schematic illustration of the chicken TIMP-2 gene and the regions where cDNA clones (TIMP-2 probes 1 and 2) were generated. The chick TIMP-2 sequence consists of a 663-bp open reading frame, from which TIMP-2 probe 1 (358 bp) and TIMP-2 probe 2 (261 bp) were cloned.

Figure 2.

 

Expression of gelatinase A and TIMP-2 in the sclera of control (CO), form-deprived (FD), and recovering (R) eyes. Northern blot analysis of gelatinase A (A) and TIMP-2 (C) expression in the posterior sclera from eyes with the indicated visual conditions. rRNA (18S) levels were used to standardize for loading differences among the samples. Steady state gelatinase A mRNA levels (B) and TIMP-2 mRNA levels (D) in posterior sclera from form-deprived and recovering eyes, relative to that of paired control eyes. Gelatinase A and TIMP-2 levels were determined by scanning densitometry of northern blot analysis autoradiograms and expressed relative to the level of 18S rRNA. ∗∗P < 0.01;∗ P < 0.05, for n = 3 or 4 separate northern blot analysis of total scleral RNA from paired control and form-deprived and control and recovering eyes (two-tailed t-test).

Figure 3.

 

In situ hybridization of gelatinase A mRNA in 6-μm frozen sections using a ³³P-labeled antisense probe. Corresponding hematoxylin-eosin–stained bright-field (A, C, and F) and dark-field (B, D, and E) images of the sclera from a form-deprived eye (A, B) and from an eye recovering from form-deprivation myopia (C, D). Sense-labeled sections (E, F) contained minimal background labeling. Note that within the sclera of form-deprived and recovering eyes, gelatinase A expression is most intense within the fibrous scleral layer (FL), whereas minimal staining is seen in the cartilaginous layer (CL). EOM, extraocular muscle. Bars, 100 μm.

Figure 4.

 

In situ hybridization of TIMP-2 mRNA in 6-μm frozen sections using ³³P-labeled antisense TIMP-2 probe 1. Corresponding hematoxylin-eosin–stained bright-field (A, C, and F) and dark-field (B, D, and E) images of the sclera from a form-deprived eye (A, B) and from an eye recovering from form-deprivation myopia (C, D). Sense-labeled sections (E, F) contained minimal background labeling. Similar to the distribution of gelatinase A, TIMP-2 expression is most intense within the fibrous scleral layer (FL), whereas minimal staining is seen in the cartilaginous layer (CL). EMO, extraocular muscle; RPE, retinal pigment epithelium. Bars, 100 μm.

Figure 5.

 

Gelatin zymography of fibrous scleral extracts. (A) Negative image of a gelatin zymogram of extracts from the fibrous sclera isolated from paired recovering (R) and form-deprived (FD) eyes from chicks 1 through 4. Bands migrating at approximately 65 kDa and 58 kDa were the major species in extracts of fibrous sclera from recovering and form-deprived eyes, representing the proenzyme (pro) and active forms of gelatinase A, respectively. (B) Gelatinase A activity within the fibrous sclera. Gelatinase A activity was quantified by densitometric analyses of the zymograms, as described in the Materials and Methods section. Values represent mean ± SEM, for n = 4 birds. ∗P = 0.041, (two-tailed t-test for matched pairs).

The authors thank Mark Olson for his assistance with the microscopy used in this study and Gayle Streier and Wanda Weber for their skilled help with photographic work.