Abstract
purpose. The sclera has a collagen-rich extracellular matrix that undergoes significant biochemical and biomechanical remodeling during myopic eye growth. The integrin family of cell surface receptors play critical roles in extracellular matrix and biomechanical remodeling in connective tissues. This study identified the major collagen-binding integrin receptors in the mammalian sclera and investigated their mRNA expression during the development of and recovery from experimental myopia.
methods. The presence of the α1, α2, and β1 integrin subunits was examined by using tree-shrew–specific primers and RT-PCR. Scleral expression of α1β1 and α2β1 receptor proteins was further investigated by using Western blot analysis and immunocytochemistry. Myopia was induced monocularly by occluding pattern vision and scleral tissue collected after 24 hours and 5 days. In a subset of the 5-day treatment group, vision was restored for 24 hours before tissue was isolated. Total RNA was extracted, and integrin subunit expression levels were assessed with quantitative real-time PCR.
results. The presence of the major collagen-binding integrin subunits α1, α2, and β1 was confirmed by RT-PCR in both scleral tissue and cultured scleral fibroblasts. Both the α1 and α2 integrin subunit proteins were identified in tree shrew scleral tissues, and integrin receptor expression was localized to scleral fibroblast focal adhesions. After only 24 hours of myopia induction, a time when no structural elongation has occurred, significant decreases were observed in the expression of the α1 (−36%) and β1 (−44%) integrin subunits. After 5 days of myopia induction, α1 integrin expression had returned to baseline levels, whereas the α2 subunit showed a significant decrease in expression (−52%). The 5-day integrin profiles were maintained during recovery from the induced myopia, with only α2 integrin showing a statistically significant relative decrease in expression (−41%).
conclusions. The mammalian sclera expresses the major collagen-binding integrin subunits. The α1 and β1 subunit expression was decreased early during the development of myopia, whereas the regulation of α2 integrin occurred at a later time point. The differential regulation of α1β1 and α2β1 during the development of myopia may reflect specific roles for these receptors in the scleral extracellular matrix and biomechanical remodeling that accompanies myopic eye growth.
High myopia is characterized by excessive elongation of the eye, particularly in the axial dimension, and results in an increased risk of retinal and/or choroidal disease.
1 During the development of myopia the outer coat of the eye, the sclera, undergoes an active remodeling process, which results in a progressive thinning and weakening of the tissue. It appears to be this remodeling of the scleral extracellular matrix (ECM)
1 that facilitates the abnormal increase in eye size.
The scleral ECM is predominantly composed of type I collagen, contains several different proteoglycans, and is maintained by a population of fibroblast cells.
2 The remodeling that occurs as myopia develops, is characterized by changes in collagen synthesis, collagen degradation, and fibril diameter.
3 4 5 In addition, alterations have been observed in matrix metalloproteinase, glycoprotein, and growth factor levels (Norton TT et al.
IOVS 1995;36:ARVO Abstract 3517)
6 7 Although it is still not fully substantiated what initiates and regulates the scleral remodeling, studies in other fibroblast-maintained tissues undergoing ECM remodeling have highlighted the importance of cell-ECM signaling.
8 The integrin family of cell surface receptors are known to play a critical role in such communication.
Integrins are heterodimeric, transmembrane receptors, formed from the noncovalent association of an α and β subunit. Presently, there are 18 α and 8 β subunits that are known to form 24 distinct integrin receptors.
9 Each receptor binds a specific ECM ligand, although different integrin receptors can bind the same ligand.
10 This apparent redundancy is not reflected in vivo, indicating a further level of complexity beyond the receptor subunit composition.
11
Integrins have wide-ranging cellular effects. Initially discovered because of their role in linking the cell to the surrounding ECM,
10 12 integrins have subsequently been shown to regulate diverse cellular functions such as survival, proliferation, migration, and differentiation.
13 14 15 16 Their role in cellular adhesion also involves integrin receptors in mechanotransduction, the process by which cells convert mechanical forces into biochemical signals.
17 18 Such regulation is particularly important for those tissues that are under constant mechanical load and enables cells to respond to changes in the ECM stresses.
19 Integrins also share a close relationship with several growth factor receptors, such as the epidermal growth factor receptor (EGFR) and those of platelet-derived growth factor (PDGFR) and vascular endothelial growth factor (VEGFR-2),
20 21 22 with data even showing that growth factor signaling can be activated by integrins, independent of growth factor ligand.
23
Because of the important and wide-ranging effects of integrins, it is not surprising to find that these receptors are expressed within the eye. Several ocular tissues such as cornea, lens, retina, and choroid have been shown to express integrins.
24 25 26 By far the best characterized is the cornea, where integrins have roles in the maintenance of corneal integrity and epithelial wound healing.
24 27 In addition, these receptors have been implicated in neovascular retinal disease and retinal development.
28 29
The role of integrins in ocular growth, however, is less evident. As the sclera has a high ECM content, is under constant tension due to intraocular pressure and is regulated by growth factors, integrins are likely to be involved in the regulation of this tissue during eye growth. There are very limited data on integrin expression in the sclera, with only one study reporting immunohistochemical staining for the β1 subunit in
Xenopus scleral fibroblasts.
30 Recently, we reported the presence of 13 integrin receptors in the mammalian sclera, including the collagen-binding receptors, α1β1, α2β1, α10β1, and α11β1.
31 While there are limited data on the functional roles of the α10 and α11 collagen receptors, the α1β1 and α2β1 receptors are considered the major collagen-binding integrins and have been extensively studied.
32 33 The expression of these receptors in the sclera are of particular interest, because numerous reports have detailed alterations in collagen synthesis, degradation, and structure during the ocular elongation that results in myopia.
4 5 34
The present study addresses the lack of data on integrin expression within the mammalian sclera by identifying the major collagen-binding integrins α1β1 and α2β1 and characterizing changes in their expression during abnormal eye growth. As the sclera has a collagen-rich ECM and significant alterations in collagen turnover have been reported in animal models of myopia, alterations in these specific integrin receptors are highly likely to play a critical role in the scleral remodeling that accompanies myopia.
Maternally reared tree shrew (
Tupaia belangeri) pups were used 15 days after natural eye opening, a time when the tree shrew is particularly susceptible to induction of myopia.
35 Animals were divided into four experimental groups. Two groups of animals (
n= 6 and 5) were monocularly deprived of pattern vision for periods of 1 and 5 days by a translucent occluder placed in a head-mounted goggle.
36 A third group of animals (
n= 6) were monocularly deprived for 5 days, after which time they experienced 24 hours of unoccluded vision (recovery). The final experimental group underwent no visual manipulation (control group) and were age matched to the 5-day treatment group (20 days after eye opening). In those groups where the visual conditions were manipulated, right and left eye treatments were balanced whenever possible. Ocular refraction (retinoscopy) and axial ocular dimensions (A-scan ultrasonography) were collected for the 5-day and recovery groups, as previously described.
5 The sclerae used in primary fibroblast cell cultures and immunohistochemistry were taken from animals age-matched to those in the experimental groups (15–20 days after eye opening).
All animals used in the study were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.
After the specific treatment periods, animals were anesthetized (ketamine 90 mg/kg, xylazine 10 mg/kg), and a lethal dose of pentobarbital sodium (120 mg/kg) was administered before tissue collection. Left eyes were enucleated first, to randomize the processing of the treated eye. An incision was made posterior to the limbus, and the anterior segment containing the cornea and lens was removed. The eye cup was flatmounted, and the retina and choroid were dissected. For gene expression studies, a 7-mm scleral punch was isolated by using a surgical trephine and the optic nerve head was removed. Scleral tissue was immediately placed in liquid nitrogen and subsequently stored at −80°C.
For primary scleral fibroblast cultures, whole sclera was isolated and placed in a culture vessel (Nunc, Roskilde, Denmark) containing Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen, Carlsbad, CA) supplemented with 10% fetal calf serum (FCS), 25 mM HEPES, and 100 U/mL penicillin and streptomycin (JRH, Melbourne, Australia). Cell outgrowth from explants was observed after ∼1 week, and confluence was generally reached after 3 weeks. Cultures were passaged using 0.25% trypsin (Invitrogen), and cells between passages 2 and 5 were used. Primary skin fibroblast cultures were used as a positive control for integrin subunit expression and were established from a lateral skin flap that had been cleaned of excess fat and minced. Growth medium was identical with that described earlier.
Total RNA was isolated from scleral tissue by phenol-chloroform extraction.
37 All RNA samples were treated with DNase I (Promega Corp., Madison WI) and re-extracted before further use. For fibroblast cell cultures, total RNA was isolated by using commercial spin columns (RNeasy; Qiagen, Valencia, CA), as per the manufacturer’s instructions. The concentration and purity of the isolated RNA was assessed at 260 and 280 nm with a spectrophotometer (Shimadzu, Kyoto, Japan). RT-PCR was performed on 0.5 μg RNA with M-MLV reverse transcriptase (10 U) and an oligo dT
15 primer (Promega Corp.). Reverse transcription was allowed to proceed for 1 hour at 42°C, after which the reaction was heated to 94°C for 2 minutes, diluted, and stored at −20°C.
Because of the lack of tree shrew sequence information, primers were initially designed from human integrin sequences in areas of high interspecies identity. From these human primers, amplification products were obtained and sequenced (CEQ 8000; Beckman Coulter, Fullerton, CA), allowing tree shrew-specific primers to be designed for the respective integrin subunits
(Table 1) . PCR amplification (PCR Express; Hybaid, Ashford, UK) used
Taq polymerase (HotStarTaq; Qiagen) and 1 μM integrin primers. The amplification protocol consisted of 95°C (15 minutes), followed by 40 cycles of 95°C (45 seconds), 61°C (45 seconds), and 72°C (1 minute). Amplifications were also performed with the respective total RNA samples, to control for potential genomic contamination. Additional reactions to control for reverse transcription and PCR contaminations were also performed.
The regulation of integrin subunit mRNA was assessed in the various treatment groups by real-time PCR (LightCycler; Roche, Mannheim, Germany) and a double-stranded DNA binding dye (SYBR green 1; Roche). Total RNA (0.5 μg) was reverse transcribed as described earlier, and amplifications were performed with a commercial PCR mix (FastStart DNA Master Mix; Roche). The amplification protocol for each gene was optimized
(Table 1)to ensure that primer-dimer formation was minimized and only the specific product was produced (as judged by product melting curves and agarose gel electrophoresis). Integrin gene copies were quantified in triplicate with reference to an external standard, which consisted of known amounts of the respective PCR fragments. To correct for variations in the amount of template, the copy number of integrin subunits were calculated relative to the housekeeping gene hypoxanthine phosphoribosyl transferase (HPRT). Primer efficiencies for the amplification of the β1, α1, and α2 integrins and HPRT gene fragments were calculated to be 1.72, 1.83, 1.70, 1.70, respectively.
Sclerae from age-matched animals were homogenized in a lysis buffer (20 mM Tris-HCl, 137 mM NaCl, 10% glycerol, 1% NP-40; pH 8) and insoluble protein was removed after centrifugation (15,000 rpm for 10 minutes). Protein concentrations were estimated (DC protein assay; Bio-Rad, Hercules, CA) and 40 μg of protein was added to the SDS sample buffer, boiled for 5 minutes, and separated on a 12% acrylamide gel according to the method of Laemmli.
38 Western blot analysis were performed as described by Towbin and Gordon
39 and the monoclonal α1 integrin and polyclonal α2 integrin antibodies (FB12, AB1944; Chemicon International, Temecula, CA) were each used at a dilution of 1:500. Mouse and rabbit HRP-conjugated secondary antibodies (1:1000; Chemicon) were used, and the signal detected using a luminescence reaction (ECL detection; GE Healthcare, Uppsala, Sweden).
For dual labeling, tree shrew scleral fibroblasts were grown on collagen type I–coated slides, fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) and processed as in Rout et al.
40 The monoclonal α2β1 antibody (clone BHA2.1; Chemicon) was used in conjunction with a rabbit polyclonal vinculin antibody (H-300; Chemicon), and staining was visualized using fluorescently conjugated antibodies (Alexa Fluor 488 and 594, respectively; Invitrogen-Molecular Probes, Eugene, OR).
Ocular biometric data are presented as mean absolute values or as the mean of the interocular differences between eyes (treated − control) ± SE. Quantitative gene expression data were obtained from triplicate samples after comparison to the external standard curves, using the fit points method (LightCycler software ver. 3.5; Roche). Group mean data were expressed either as absolute gene copies per 1000 copies of HPRT (±SEM) or as the percentage difference between treated and contralateral fellow control eyes (±SEM). Differences between treated and control copies were assessed with paired t-tests, whereas the absolute copy numbers obtained for the different treatment groups were compared by one-way analysis of variance (ANOVA).
Despite the extensive scleral collagen remodeling that occurs during myopia, there is no previous evidence of the major collagen binding integrin receptors, α1β1 and α2β1, in the sclera. Using RT-PCR the scleral expression of α1, α2 and β1 integrin subunits was assessed. As observed in
Figure 1 , the use of tree-shrew–specific primers enabled fragments (α1 integrin, 198 bp; α2 integrin, 179 bp; and β1 integrin, 184 bp) to be amplified from the mRNA of the major collagen-binding integrin subunits in both scleral tissue and in primary cultures of tree shrew scleral fibroblasts. Tree shrew skin fibroblasts were included as a positive control, since α1, α2, and β1 integrin expression has been reported in these cells.
41 Because genomic contamination can act as a template during PCR amplification, total RNA was used as a negative control. The lack of amplification products in these controls confirmed the specificity of the integrin subunits in scleral samples. Additional negative controls for contamination due to reverse transcription and PCR amplification were also performed and similarly showed no fragment amplification (data not shown).
To confirm the gene expression data, Western blot analysis for the α1 and α2 subunit proteins were performed on tree shrew scleral samples. As observed in
Figure 2 , both receptors are expressed in the tree shrew sclera. The 200-kDa α1 product and the 180 kDa α2 protein have been identified in human cells.
42 43 The higher bands (α1, 230 kDa; α2, 200 kDa) may represent posttranslational modifications, with the 230-kDa product described in mouse tissue.
42 To confirm further the presence of integrin protein, scleral fibroblasts were colabeled with the α2β1 integrin antibody and an antibody to vinculin, a structural protein present in focal adhesions.
44 Figure 3shows that α2β1 integrin and vinculin
(Figs. 3A 3B)can be colocalized (
Fig. 3C , arrows), consistent with clustering of these receptors at focal adhesions on the cell surface.
10
To assess the effect of excessive eye growth on α1, α2, and β1 subunit expression, myopia was induced by using monocular deprivation of pattern vision, and ocular measures were taken
(Table 2) . After 5 days of occluder wear, myopia was induced in the treated compared with the contralateral fellow control eye (−8.4 ± 1.0 D), with 0.16 ± 0.02 mm of vitreous chamber elongation resulting in an overall increase in axial length (0.15 ± 0.01 mm). This reflects previously reported data indicating that increased vitreous chamber depth is responsible for the excessive axial eye size observed in human myopes.
45 The same trend was observed in the recovery group after 5 days of occluder wear, while removal of the occluder resulted in no statistically significant alteration in either vitreous chamber depth or axial length (
P = 0.60, 0.73 respectively). Measurements from contralateral fellow control eyes were not significantly different from those obtained from normal animals.
Quantitative real-time PCR was used to estimate integrin copy numbers at the two time points during myopia development (24 hours and 5 days) and in the group of animals that were allowed to recover for 24 hours from the induced myopia. Because of the numerous experimental variables such as interanimal differences, RNA extraction, and reverse transcription efficiency, the determination of absolute number of gene copies between individual animals can be problematic. This inherent variability can be reduced through the use of the contralateral fellow control (unoccluded) eye, which serves as a within-animal, genetic control. We have previously reported that by expressing data relative to this control, highly repeatable and sensitive gene expression data can be obtained.
46
The use of intereye, within-animal comparison is valid, however, only if the number of genes in the untreated (contralateral control) eye are similar to those found in the visually unmanipulated or normal eye.
Figure 4shows the number of copies for β1, α1, and α2 integrins in the contralateral fellow control eye compared with the normal group of animals. As is evident, each subunit is present at different levels, with each of the four groups showing β1 > α1 > α2 integrin expression. Despite some mean variation in the absolute number of copies within genes, there was no significant difference in the amount of the respective subunits in the contralateral fellow control eyes of the treatment groups and the eyes from the untreated animals (ANOVA; β1,
P = 0.19; α1,
P = 0.07; and α2,
P = 0.24). Despite this result, differences in gene expression between control and normal eyes (control eye effect) have been reported elsewhere.
34 47 This discrepancy may reflect differences in gene quantification technique.
Figure 5shows the effect of myopia induction on integrin subunit expression. After 24 hours of visual deprivation
(Fig. 5A)significant decreases in β1 (−44%,
P < 0.001) and α1 (−36%,
P < 0.05) subunit expression were observed when compared with the contralateral control eye of the same animal. The α2 integrin exhibited no significant change in expression after 24 hours of induced myopia (−12%,
P = 0.34). The absolute number of gene copies for treated and contralateral fellow control eyes are presented in the inset of
Figure 5Aand show β1 integrin to have the highest mean subunit expression (5000 copies/1000 copies HPRT), followed by α1 and α2 (2950 and 1100 copies/1000 copies HPRT, respectively).
After 5 days of induced myopia
(Fig. 5B) , the β1 subunit still showed a mean decrease in expression; however, this was not significant (
P = 0.13), whereas α1 integrin expression returned to baseline. Unlike the 24-hour data, the α2 subunit showed a significant decrease in expression (−52%
P < 0.05) when compared with the contralateral fellow control eye. Data from animals that underwent no visual manipulation showed no interocular differences in integrin subunit mRNA expression, and these results have been included in
Figure 5A and 6for comparison. Again, the inset shows the absolute number of copies of the integrin subunits in treated and contralateral fellow control eyes, with β1-integrin expression highest (3290 copies/1000 copies HPRT), followed by the α1 and α2 subunits (1330 and 950 copies/1000 copies HPRT, respectively).
The expression data obtained from those animals that were allowed to recover for 24 hours, after 5 days of myopia induction, essentially reflect the gene regulation observed after 5 days of deprivation. As observed in
Figure 6 , there was a mean decrease in β1 subunit expression (−30%,
P = 0.19), whereas α1 integrin also showed a slight decrease (−12%,
P = 0.32). As with the 5-day data, α2 integrin was the only subunit to exhibit a significant decrease in expression (−41%
P < 0.05). The number of copies of the β1, α1, and α2 integrin subunits in the treated and contralateral fellow control eyes are shown in the inset (8300, 2020, and 490 copies/1000 copies HPRT, respectively).
Myopia is characterized by an increase in the axial dimension of the eye, which is facilitated by an active remodeling of the sclera.
2 Studies in other tissue systems have highlighted the critical roles of integrin receptors during ECM remodeling. As collagen is the major ECM protein in the sclera and undergoes significant changes during the development of myopia, the collagen-binding integrin receptors are likely to be involved. This study is the first to detail the presence of the major collagen-binding integrin receptors α1β1 and α2β1 in the sclera. Furthermore, data showed that the α1, α2, and β1 integrin subunits underwent a time- and subunit-dependent decrease in expression during induction of, and recovery from, myopia.
As collagen constitutes 90% of scleral dry weight, it is perhaps unsurprising that the major collagen-binding integrins, α1β1 and α2β1, are present in the sclera (Norton TT et al.
IOVS 1995;36:ARVO Abstract 3517). The data show that scleral fibroblasts actively synthesize subunit mRNA and express α1β1 and α2β1 receptors at focal adhesions on their cell surface. The fact that the β1 subunit is present in both receptors is reflected in the higher expression levels observed in the quantitative PCR data. Assuming linearity between mRNA content and protein expression, the absolute number of the genes suggests that most of the β1 subunit is associated with the α1 and α2 subunits. However, because the β1 integrin can associate with an additional 10 α subunits, other β1-containing receptors could be present within the sclera, albeit at a lower level. Recent work from this laboratory has found that 10 α and 4 β integrin subunits are expressed in the sclera.
31 Further work is necessary to quantify relative expression of these scleral integrin receptors.
The collagen-binding integrins, α1β1, α2β1, α10β1, and α11β1 all contain the α1 domain which is responsible for collagen recognition. Although integrins can recognize multiple collagen types, they have higher affinities for their preferred ligands. The α1β1 integrin is known to bind basement membrane collagens preferentially, such as collagen type IV and XIII, whereas fibrillar collagens such as types I, III, and V are the preferential ligands for α2β1 integrin.
48 Previous work has demonstrated the presence of these collagens in the mammalian sclera.
4 49 As type I collagen represents most of the scleral collagen content (>99%; Norton TT et al.
IOVS 1995;36:ARVO Abstract 3517) and types III and V are thought to be critical in collagen fibril association in the sclera,
4 the α2β1 receptor may be a key scleral fibroblast receptor. However, due to the redundancy in the integrin-ligand interactions, specific functional knockout experiments would have to be performed on scleral fibroblasts, before the importance of α2β1 can be confirmed.
Integrin gene expression was assessed at two different time points during induction of myopia: one reflecting early signaling before structural elongation of the eye and the other assessing changes during excessive eye growth causing myopia.
34 The major collagen-binding integrin subunits show distinct receptor-specific and time-dependent changes in expression. The α1 and β1 subunits show significant mRNA downregulation after 24 hours, whereas the α2 subunit expression is reduced after 5 days. As the α1 and α2 subunits only associate with β1 integrin and the both receptors were identified in vivo, one may speculate that the α1β1 receptor is regulated in the early stages of myopia development, whereas the α2β1 receptor is involved in the consequent ocular enlargement underlying myopia.
Despite apparent ligand-binding redundancy, each receptor has specific signaling roles. The use of knockout animals has shown that the α1β1 receptor positively regulates cell proliferation through the Shc-mediated growth pathway.
50 This growth response appears to be cell-type specific, with fibroblast cells exhibiting a positive regulation, whereas mesangial cells show a reduction in proliferative rates when α1β1 is overexpressed.
50 51 It therefore follows that the decreased α1β1 levels observed early in the development of myopia would result in decreased scleral fibroblast proliferation. Such decreases in fibroblast proliferation have been observed in animals during developing myopia.
52
The α1β1 receptor also performs a negative-feedback role in collagen synthesis, such that stimulation of the receptor results in suppression of collagen synthesis.
53 This has been supported in vitro and in specific α1 integrin knock out animals that show increased collagen synthesis in the dermis.
53 54 This role of α1β1 appears inconsistent with the 24-hour data presented in the present study, because decreases rather than increases in collagen synthesis occur after induction of myopia.
4 However, a previous study has reported that changes in collagen expression occur later in the development of myopia (>4 days).
34 It must also be noted that the regulation of collagen by α1β1 can be complicated by the overriding effects of other integrin receptors, such as α2β1 and cytokines such as TGF-β.
55 56
As mentioned, the specific decrease in the α2 subunit after 5 days of induced myopia may indicate a role for the α2β1 receptor in ocular growth mechanisms, rather than initial signaling. After 5 days of induced myopia, one of the most obvious ocular changes is an increase in the axial length of the eye, which is ultimately governed by the biomechanical properties of the sclera. Studies on the α2β1 integrin receptor have shown that it induces fibroblast cell contraction of collagenous matrices.
57 Thus, the decreases in the α2 subunit may result in a decreased ability of the scleral fibroblasts to maintain a highly contractile nature. This deficiency would result in a reduction in the strength of the sclera, which has been observed in biomechanical studies of scleral tissue from myopic eyes and results in the characteristic increase in eye size.
58
As with the α1β1 receptor, the α2β1 receptor is also known to regulate collagen synthesis. However, unlike α1β1, α2β1 is a positive regulator of collagen synthesis. Thus, the decrease observed in the 5-day data correlates well with studies reporting myopia-induced decreases in collagen synthesis at this time point.
4 34 In addition, α2β1 integrin activates matrix metalloproteinases (MMPs), with ligand binding upregulating the expression of MMP-1, MMP-13, and MT1-MMP.
53 59 60 Although MT1-MMP expression has been investigated during myopia induction, it is unclear whether expression levels are altered (Siegwart JT et al.
IOVS 2004;45:ARVO E-Abstract 1232).
61
The experimental paradigm in which the occluder is removed typically results in the eye’s modifying its growth over time to recover from the induced refractive error. In this study we decided to assess changes in gene expression in the first 24 hours of recovery to determine how early in the process integrin signaling might be involved. The fact that we found no significant changes in gene expression of integrins at this time point, compared with the 5-day myopia group, was a little surprising, although it should be noted that there was also no significant reduction in axial elongation of the eye (or vitreous chamber depth) in this 24-hour recovery period (see
Table 2 ). Thus, this time point may be too early in the recovery process to observe gene expression changes, especially if there is a lag in scleral response to a stop-growth signal. This notion is supported by scleral thickness data showing that the tissue remains thinned after a 1-day recovery.
4 However, sulfate incorporation into scleral glycosaminoglycans does show significant differences at this same time point in recovery, suggesting that there is some initial response in scleral remodeling to the removal of the occluder.
62 Obviously, further investigation during this early period in recovery is required.
It is also possible that, early in the recovery process, alternate mechanisms to that used for myopic eye growth are activated. Potential evidence for alternate mechanisms comes from a recent study into the response of the tree shrew eye to increased intraocular pressure.
63 Although the tree shrew eye initially increased in size in response to increased pressure, active shortening occurred within 1 hour. Because of the short time period, it is highly unlikely that gene expression changes were involved in this response. Rather, it was suggested that the presence of active myofibroblasts resulted in the shortening. The early stages of the recovery process (≤1day) may involve an initial activation of these highly contractile myofibroblasts, with gene regulation only occurring at the later stages.
Although these data indicate that the expression of major collagen-binding integrin subunits is regulated at distinct times during the development of myopia, further work is needed to elucidate the functional significance of such changes. However, as α1β1 and α2β1 receptors are involved in extracellular matrix and cellular biomechanical remodeling in other tissues, changes in their expression are certain to play a role in the initiation and ongoing scleral response to myopic eye growth.