An In Vitro Model of Posterior Capsular Opacity: SPARC and TGF-2 Minimize Epithelial-to-Mesenchymal Transition in Lens Epithelium Norihito Gotoh,1,2 Nikole R. Perdue,3 Hiroyuki Matsushima,2 E. Helene Sage,1,4 Qi Yan,3,4 and John I. Clark3 PURPOSE. This report presents a novel model for studies of extracellular matrix (ECM) in posterior capsular opacification (PCO) in vitro. Lens epithelial cells (LEC) were cultured with an intraocular lens (IOL) on a surface of type IV collagen in an evaluation of the importance of the ECM– cell interaction in formation of PCO. Abnormal migration, proliferation, and expression of proteins associated with the epithelial-to-mesenchymal transition (EMT) that characterizes PCO were observed in the presence and absence of the matricellular protein SPARC (secreted protein, acidic and rich in cysteine), which regulates matrix– cell interactions. METHODS. The model for PCO in vitro consisted of an IOL placed on a membrane coated with collagen IV, a major constituent of the lens capsule. LECs from the lenses of wild-type (WT) and SPARC-null (SP-null) mice were cultured in the presence or absence of 10 ng/mL TGF-2 and 20 g/mL recombinant human SPARC (rhSP) for up to 6 days. The migration of LECs was quantified. Labeling with BrdU and the measurement of DNA synthesis were assays for cell proliferation. Expression of the EMT markers, collagen type I, fibronectin, and ␣-smooth muscle actin were assessed using immunocytochemistry or Western immunoblots. RESULTS. LEC migration, proliferation, and the synthesis of EMT markers were enhanced in SP-null compared with WT LECs. TGF-2 inhibited the migration and proliferation of both WT and SP-null LECs in the presence of rhSP. TGF-2 increased the production of collagen type I, fibronectin, and ␣-SMA. The responses of SP-null LECs were rescued by the addition of recombinant human (rh)SP. CONCLUSIONS. A simple IOL culture system was useful for the evaluation of the effects of SPARC and TGF-2 on PCO in vitro. The action of TGF-2 on LEC migration and proliferation is influenced by SPARC, a regulator of matrix– cell interactions. The results indicate a functional intersection between pathways activated by TGF-2 and SPARC in the formation of PCO.
From the Departments of 1Biological Structure and 3Ophthalmology, University of Washington School of Medicine, Seattle, Washington; the 2Department of Ophthalmology, Dokkyo University School of Medicine, Tochigi, Japan; and the 4Hope Heart Program, Benaroya Research Institute at Virginia Mason, Seattle, Washington. Supported by National Eye Institute Grants EY014150 and EY013180. Submitted for publication January 25, 2007; revised June 10, 2007; accepted August 20, 2007. Disclosure: N. Gotoh, Hoya Corp. (F); N.R. Perdue, None; H. Matsushima, None; E.H. Sage, None; Q. Yan, None; J.I. Clark, None The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact. Corresponding author: John I. Clark, Department of Biological Structure, Box 357420, University of Washington School of Medicine, 1959 NE Pacific Street, G-514 HSB, Seattle, WA 98195-7420;
[email protected]. Investigative Ophthalmology & Visual Science, October 2007, Vol. 48, No. 10 Copyright © Association for Research in Vision and Ophthalmology
(Invest Ophthalmol Vis Sci. 2007;48:4679 – 4687) DOI: 10.1167/iovs.07-0091
A
lthough posterior capsular opacification (PCO), known as secondary cataract or after-cataract, is the most common complication after intraocular lens (IOL) implantation,1–3 the cellular mechanisms responsible for PCO are yet to be understood. PCO is caused when lens epithelial cells (LECs) proliferate between the capsule and the IOL on the posterior surface of the lens capsule. The cells migrate along the posterior lens capsule and undergo an epithelial-to-mesenchymal transition (EMT), instead of differentiating into normal lens fiber cells attached to the surface of the posterior lens capsule.4 –7 The response of LECs can be considered a wound-healing reaction resulting from the activation of inflammatory cells and production of cytokines and growth factors after surgery, influenced by the extracellular matrix (ECM) of the lens capsule.8,9 Transforming growth factor (TGF)- is known to initiate myofibroblast formation, fibrosis, and cataract formation, and members of the TGF- family are abundant in the aqueous humor.10,11 TGF-1, -2, and -3 are present in mammalian aqueous humor with TGF-2 predominant.12–13 TGF- is pivotal in regulating proliferation, differentiation, and ECM expression by cells in a positive or negative manner,14,15 and LECs express TGF- isoforms and receptors.16,17 More specifically, human LECs and macrophages adhering to implanted IOLs express certain of the TGF- family members.18,19 Secreted protein, acidic and rich in cysteine (SPARC), also known as osteonectin and BM-40, is a highly-conserved prototypic matricellular glycoprotein that regulates interactions between ECM and LEC. SPARC exhibits complex ECM– cell activities that affect cell adhesion and proliferation, expression of ECM proteins and matrix metalloproteinases (MMPs), and growth factor production.20,21 SPARC functions in the development of angiogenesis, tumorigenesis, cataractogenesis, and wound repair22–25 and has the potential to influence PCO formation. In lenses of SPARC-null mice, fiber differentiation is abnormal, and progresses slowly over a period of weeks to a mature cataract. Rupture of the posterior capsule indicates the importance of SPARC as a regulator of capsule–LEC interactions.26 –29 Whereas SPARC regulates the expression of TGF- in several cell types including renal mesangial cells,30 periodontal ligament cells,31 pulmonary artery endothelial cells,32 and corneal epithelial cells,33 the effects of TGF- and SPARC in the formation of PCO are unknown. In this study, a novel in vitro PCO model was used to examine the effect of TGF-2 on LEC migration, proliferation, and EMT in cells from wild-type (WT) and SPARC-null (SP-null) mice. The effects of TGF-2 on migration and proliferation of LECs were influenced by the presence of SPARC. The pathways activated by TGF-2 are influenced by the ECM– cell interactions regulated by SPARC during formation of PCO. 4679
4680
Gotoh et al.
MATERIALS
AND
IOVS, October 2007, Vol. 48, No. 10
METHODS
Cell Culture All procedures were conducted in accordance with institutional guidelines for the use of animals in scientific research and adhered to the ARVO Statement for Use of Animals in Ophthalmic and Vision Research. LECs were cultured from the lenses of WT and SP-null C57Bl6/ Jx129SvJ mice at 1 to 2 months of age, as described previously.34 In brief, capsules with attached epithelial cells were cultured in 35-mm culture dishes coated with fibronectin and laminin (BD Biosciences, San Jose, CA). Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen Gibco, Grand Island, NY) containing 20% fetal bovine serum (FBS; Hyclone, Logan, UT), 100 U/mL penicillin G, 100 g/mL streptomycin SO4, and 2.5 g/mL amphotericin-B (Invitrogen-Gibco) was added to the dishes, and the lens capsular explants were incubated undisturbed for 7 days in a humidified atmosphere of 5% CO2 at 37°C. Primary cells were maintained and subcultured in 10% FBS/DMEM. Spontaneous immortalization occurred after passage six. Immortalization of longterm cultured LECs was confirmed by karyotype and by the number of population doublings. For long-term culture, cells were subcultured weekly in 5% FBS/DMEM.
Migration Assay Cells (6 ⫻ 105) were grown to confluence on 35-mm culture dishes in 10% FBS/DMEM before quiescence in serum-free DMEM over 48 hours. In the final 2 hours, TGF-2 (10ng/mL) and/or rhSPARC (rhSP; 20g/ mL) was added to the culture medium. The concentration of rhSP was based on previous studies of the effects of rhSP in culture.35,36 A pipette tip was used to scratch the LEC monolayer in a straight line to create a cell-free space. Cells were washed three times with serum-free DMEM and were incubated with serum-free DMEM in the absence or presence of TGF-2 and/or rhSP for 48 hours. During this time, three areas of migrating cells were photographed under a light microscope every 12 hours. Each area was divided into 10 sections for measurement of distance in the photographs (Photoshop; Adobe Systems, San Diego, CA). Migration distance was normalized to a control distance observed at 0 hours.
PCO Model In Vitro A square-edged acrylic IOL (VA60BB; Hoya Co., Saitama, Japan) was placed on a cell culture insert coated with 5 g/cm2 type IV collagen (BD Biosciences, San Jose, CA) in a 12-well culture plate to simulate the formation of PCO in cataract surgery and implantation of an IOL (Fig. 1). As seen in Figure 1A, the PCO forms posterior to the IOL where
Proliferation Assay Cells (4.5 ⫻ 104) were placed on a collagen IV membrane in 10% FBS/DMEM in the presence or absence of 10 ng/mL TGF-2 and/or 20 g/mL rhSP for 2 days and were incubated subsequently in serum-free DMEM for 1 day. The cells were labeled with 10 M 5-bromo-2⬘deoxyuridine (BrdU; Sigma-Aldrich, St. Louis, MO) for 2 hours. Cells on the collagen membrane were fixed in 0.4% paraformaldehyde for 20 minutes, washed with PBS, incubated in 2 N HCl/PBS for 30 minutes, and blocked with 0.1% Triton X-100/20% goat serum/PBS for 30 minutes. The cells were incubated with mouse anti-BrdU-fluorescein isothiocyanate (Roche, Indianapolis, IN) for 2 hours followed by two washes with PBS. The cells were subsequently incubated with the appropriate secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA) for 1 hour and were washed again with PBS. The cells were counterstained with Hoechst 33258 for 5 minutes, rinsed with PBS, and mounted on slides for microscopy. Immunofluorescence was detected with a microscope equipped for epifluorescence. Each image was photographed with a digital camera set for the same exposure time. The labeling index (%LI) was calculated by dividing the number of labeled nuclei by the total number of nuclei counted. For evaluation of DNA synthesis, the cells were pulse labeled with 2 Ci/mL [3H]-thymidine (6.70 Ci/mmol; Perkin Elmer Life Sciences, Inc., Torrance, CA) for 4 hours, rinsed with PBS three times, and fixed with ice-cold 10% trichloroacetic acid (TCA) for 30 minutes. The TCA-insoluble material was rinsed with 95% ethanol, air dried, and solubilized in 0.2 N NaOH for 45 minutes at 68°C. Incorporated [3H]-thymidine was measured in a liquid scintillation counter as counts per minute (cpm).
B
A LECs
contact is made with the lens capsule after cataract surgery. In the model for PCO, a tiny aluminum weight (0.85g) was placed on the IOL to maintain contact between the IOL and the collagen membrane. Cells (4.5 ⫻ 104) were placed into each culture well, and the cells were incubated in 700 L 10% FBS/DMEM in the presence or absence of porcine (Fig. 1B) TGF-2 (10 ng/mL) and/or rhSP (20 g/mL) within the upper chamber and in 1 mL of 10% FBS/DMEM in the lower chamber. After 6 days, the cells were stained with 0.1% crystal violet and photographed under the light microscope. The area beneath the IOL was selected, and the image contrast was adjusted to facilitate pixel counting with image-analysis software (Photoshop; Adobe Systems). The area of cell coverage was measured with a second computer program (Scion Image; Scion Corp., Frederick, MD). The percentage of migration was calculated and normalized to the control area of the entire IOL.
Lens capsule
PCO Upper chamber
Collagen IV membrane
Lower chamber
FIGURE 1. Schematic for the PCO formation (A) and the PCO model in vitro (B). An acrylic IOL was placed on a cell culture insert coated with type IV collagen in a 12-well culture plate. The insert has 1.0-m pores in the supporting PET (polyethylene terephthalate) membrane for diffusion of medium between culture chambers. A tiny aluminum weight (w) placed on the IOL equalized the contact pressure of IOL with the collagen surface of the insert membrane. LECs (4.5 ⫻ 104) were placed into the culture well and were incubated in 700 L 10% FBS/DMEM in the presence and absence of 10 ng/mL TGF-2 and/or 20 g/mL rhSP in the upper chamber. The lower chamber contained 1 mL 10% FBS/DMEM. LECs were cultured for as long as 6 days to measure proliferation and migration in this IOL model for PCO formation.
Novel Model of Posterior Capsular Opacity
IOVS, October 2007, Vol. 48, No. 10
Immunocytochemistry Residual LECs that migrate onto the posterior lens capsule can undergo a process of EMT and transdifferentiate into myofibroblast-like cells that contribute significantly to the formation of capsular fibrosis. These LECs produce EMT markers including ␣-smooth muscle actin (␣-SMA), collagen type I, and fibronectin (FN).37 The expression of ␣-SMA is an important marker for EMT in lens capsules in vivo and expression of ␣-SMA is a common feature of cultured mammalian LECs.38 Whereas ␣-SMA is a useful marker for EMT in immunoblots and tissue sections, in cell culture there can be considerable variation with cell density, culture medium, and substrate conditions.38 Variability was a complication of the immunocytochemical localization of ␣-SMA in the IOL
4681
model for PCO. The migrating cells shown in Figure 1B were fixed in 10% formalin for 20 minutes, washed three times with PBS, and blocked (20% Aquablock; East Coast Biologics, Inc., Berwick, ME) for 45 minutes. The cells were incubated with mouse anti-fibronectin (Sigma-Aldrich) or rabbit anti-collagen type I IgG (Calbiochem, San Diego, CA) for 2 hours. The cells were washed 4 times with PBS and incubated with the appropriate secondary antibody conjugated to Texas red (Zymed Laboratories, South San Francisco, CA) or fluorescein isothiocyanate (Jackson ImmunoResearch Laboratories) for 1 hour. The cells were washed four times with PBS, counterstained with Hoechst 33258 for 5 minutes, rinsed with PBS, and mounted on slides. Immunofluorescence was detected with a microscope equipped for epifluorescence. Each image was photographed with a digital camera using the same exposure time.
Western Immunoblot The cells on the collagen membrane (Fig. 1B) were homogenized with mammalian protein extraction reagent (M-PER; Pierce, Rockford, IL) containing a complete protease inhibitor cocktail (Roche, Indianapolis, IL). The cell lysates were collected according to the manufacturer’s instructions. Proteins (8 g/lane) were separated under reducing conditions by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 3% to 8% Tris-acetate gels (Invitrogen, Carlsbad, CA). Next, the proteins were transferred to a polyvinylidene difluoride (PVDF) membrane (Invitrogen). After a blocking step in Aquablock/ PBS-Tween for 1 hour, the membranes were incubated with primary antibodies overnight at 4°C. Proteins were detected with mouse antifibronectin (Sigma-Aldrich) and mouse anti-GAPDH IgG (Ambion, Inc., Austin, TX). The membranes were washed four times with 0.2% PBSTween and incubated with the respective HRP-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, Inc.) for 1 hour. Immunoreactive bands were revealed by enhanced chemiluminescence (Pierce).
Statistical Analyses Data were expressed as the mean ⫾ SD. Statistical comparisons were made according to Student’s t-test; P ⬍ 0.05 was considered to be statistically significant.
RESULTS Cell Migration on a Plastic Tissue Culture Plate LEC migration into the cell-free region created by the scratch was observed under a light microscope every 12 hours (Fig. 2A). The cell migration was observed in photographs at each time point with the image-analysis software (Photoshop; Š
FIGURE 2. Effect of TGF-2 on LEC migration in the presence and absence of SPARC. A linear cell-free zone was created in confluent LEC cultures. Cells were incubated in serum-free medium in the presence or absence of TGF-2 and/or rhSP for 48 hours. The migration of cells into the cell-free zone was photographed under a light microscope every 12 hours (A). The migration distance was measured (in micrometers) in each image and recorded as ⌬Dt (migration distance) ⴝ Dt (distance at each time) ⴚ D0 (distance at 0 hours). The fastest migration was observed in the SP-null (‚) and SP-null⫹TGF[b]-2 (Œ) cultures (B). The slowest migration was observed in WT⫹TGF-2 (F) and SP-null⫹rhSP⫹TGF-2 (f) cultures. Moderate migration was observed in WT (E) and SP-null⫹rhSP (䡺) cultures. SP-null LECs migrated faster than WT LECs and SP-null⫹rhSP LECs after 48 hours (C). In the presence of TGF-2, migration was inhibited. Data are expressed as the mean ⫾ SD. The differences between groups were significant to P ⬍ 0.05 except for those labeled NS which include WT versus WT⫹TGF2, SP-null versus SP-null⫹TGF-2, and WT⫹TGF-2 versus SPnull⫹rhSP⫹TGF-2. Migration was enhanced in the absence of SPARC, and TGF2 inhibition was greater in the presence than in the absence of SPARC.
4682
Gotoh et al.
IOVS, October 2007, Vol. 48, No. 10
FIGURE 3. Area of LEC migration in the PCO model in vitro. The migration of LECs between the collagen membrane and the IOL was observed in cells stained with 0.1% crystal violet and photographed for quantification. In each image, the circular edge of the IOL is delineated in black (A). The image contrast was adjusted to facilitate counting of black pixels beneath the IOL and the collagen membrane (B). The cell density and size remained approximately the same in each culture, and cell migration, rather than cell spreading, accounted for the observed changes in the area covered by the cells (C). The area covered with SP-null LECs was greater than that covered by WT LECs and SP-null⫹rhSP LECs (D). The percentages of LEC coverage (%Acell) ⫽ [Acell (area covered by LECs)/AIOL (area of the entire IOL: encircled)] ⫻ 100. In the presence of TGF-2, migration was inhibited, and the presence of SPARC enhanced the inhibition significantly. Data are expressed as the mean ⫾ SD. The differences between groups were significant at P ⬍ 0.05 except for those labeled NS which include SP-null versus SP-null⫹TGF-2. Similar to the scratch assay, migration was enhanced in the absence of SPARC, and TGF2 inhibited migration in the presence or absence of SPARC.
Adobe Systems), and the migration distance was plotted in Figure 2B. The cells in all cultures migrated during the 48 hours of observation. The migration distance in the cultures containing SP-null cells was much greater than the migration distance in cultures containing WT cells. For SP-null cells, migration was 375 ⫾ 48 m at 24 hours and 575 ⫾ 53 m at 48 hours, and for WT cells the migration was only 188 ⫾ 10 m at 24 hours and 349 ⫾ 5 m at 48 hours. The migration distances for SP-null cells incubated with rhSP was 243 ⫾ 50 m at 24 hours and 450 ⫾ 63 m at 48 hours. Although the effect of TGF-2 was inhibitory, the effect was greater in the presence of than in the absence of SPARC: 163 ⫾ 28 m at 24 hours and 279 ⫾ 52 m at 48 hours in WT cells; 374 ⫾ 50 m at 24 hours and 524 ⫾ 41 m at 48 hours in SP-null cells; and 187 ⫾ 25 m at 24 hours and 310 ⫾ 29 m at 48 hours in SP-null cells containing rhSP (Fig. 2B). In the scratch assay, the effect of TGF-2 was strongest in the presence of SPARC and least in the absence of SPARC (Fig. 2C).
Cell Migration in the PCO Model In Vitro LEC migration between the collagen membrane and the IOL was measured in culture (Fig. 3), and the migrating cells were photographed after being stained with 0.1% crystal violet (Fig. 3A). The image contrast was adjusted to facilitate pixel counting and quantification of the change in area resulting from cell migration (Fig. 3B). The area of cell migration was measured in the digital images and was compared in the histogram shown in Figure 3C. Areas were 25 ⫾ 4 mm2 for WT cells, 32 ⫾ 6 mm2 for SP-null cells, and 17 ⫾ 2 mm2 for SP-null cells containing rhSP. The effect of TGF-2 on cell migration was evaluated: 8 ⫾ 1 mm2 for WT cells, 29 ⫾ 4 mm2 for SP-null cells, and only 10 ⫾ 2 mm2for SP-null cells containing rhSP. The migration of SP-null cells in the IOL cultures was greater than that of WT cells (Fig. 3C), and the addition of rhSP decreased migration in SP-null cultures. The results of these experiments confirmed the use of IOLs in the PCO model for migration studies (Fig.
Novel Model of Posterior Capsular Opacity
IOVS, October 2007, Vol. 48, No. 10 2C). Similar to the scratch migration assays, TGF-2 had the strongest inhibitory effect on WT cells and SP-null cells containing rhSP, and minimal effect on SP-null cells without rhSP in the PCO system.
Cell Proliferation in the PCO Model In Vitro The proliferation of WT and SP-null cells was observed in the presence or absence of TGF-2 and rhSP in Figure 4. LECs labeled with BrdU were immunostained for BrdU (Fig. 4A, red) and nuclei were stained with 4⬘,6⬘-diamino-2-phenylindole
4683
(DAPI; Fig. 4A, blue). The labeling index (LI) of SP-null cells was 38% ⫾ 10%, and that of WT cells was 27% ⫾ 9%. The LI for SP-null cells in the presence of rhSP was 29% ⫾ 9%, similar to WT cells. In the presence of TGF-2, the LI was reduced to 15% ⫾ 5% in WT cells and 15% ⫾ 6% in SP-null cells cultured with rhSP. [3H]thymidine incorporation by WT cells was 7,465 ⫾ 810 cpm compared with 21,733 ⫾ 1,650 cpm for SP-null cells (Fig. 4C). Consistent with the results from Fig. 4B, high levels of DNA synthesis were observed in SP-null cells relative to WT cells. In cultures of SP-null cells containing rhSP, [3H]thymidine incorporation decreased to 9493 ⫾ 1074 cpm and resembled WT. [3H]thymidine incorporation in the presence of TGF-2 was only 848 ⫾ 10 cpm in WT cells, increased to 15,206 ⫾ 262 cpm in SP-null cells and was reduced to 7,258 ⫾ 935 cpm in SP-null cells containing rhSP.
Expression of EMT Markers Produced by Cells on the PCO Model In Vitro The effect of TGF-2 on the expression of EMT proteins in the presence and absence of SPARC was evaluated in the PCO model in vitro (Fig. 5). LECs produced type I collagen (Fig. 5A) and fibronectin (Fig. 5B) and assembled fibronectin and collagen into the ECM. The staining intensities of fibronectin and collagen type I appeared to be greater in SP-null cells than in WT cells or SP-null cells containing rhSP. In the presence of TGF-2, the staining intensities of both fibronectin and collagen type I increased (Figs. 5A, 5B). There are limitations to immunocytochemistry, which is descriptive, and Western immunoblots were used to quantify expression of EMT markers. Intracellular fibronectin in the LECs between the collagen membrane and the IOL was observed by immunoblot analysis (Fig. 5C). SP-null cells appeared to produce more fibronectin per cell than did WT cells, and the expression was reduced on addition of rhSP. In the presence of TGF-2, the expression of fibronectin in WT LECs increased threefold relative to WT LECs without TGF-2. In the presence of TGF-2, the expression of fibronectin in SP-null LECs increased approximately 1.3-fold relative to SP-null LECs without TGF-2. The effect of TGF-2 on SP-null LECs containing rhSP was similar to the effect on WT LECs and fibronectin expression increased approximately 1.7fold (Fig. 5D), as quantified by densitometric scanning of fibronectin bands and normalization for cell number with the GAPDH signal. Although the immunohistochemistry indicated that more type I collagen was present in cultured SP-null LECs relative to WT and SP-null LEC containing rhSP (Fig. 5A), quantitative analysis of collagen type I by immunoblot analysis was not possible, because poor antibody reactivity to denaŠ
FIGURE 4. Effect of TGF-2 on the proliferation of lens epithelial cells under the IOL. LECs cultured on collagen IV and labeled with BrdU were immunostained for BrdU (red) and counterstained for nuclei (blue) (A). Labeled and unlabeled LECs were counted under the microscope (40⫻). (B) The % BrdU-labeled index was calculated as LI ⫽ [NBrdU (number of labeled nuclei)/Ntotal (number of total nuclei)] ⫻ 100. The LI for SP-null LECs was greater than that of WT LECs and SP-null⫹rhSP LECs. In the presence of TGF-2, the LI decreased. Data are expressed as the mean ⫾ SD. The differences between groups were significant to P ⬍ 0.05 except for those labeled NS, which include WT versus SPnull⫹rhSP, WT⫹TGF2 versus SPnull⫹rhSP⫹TGF2, and SPnull versus SPnull⫹TGF2. (C) [3H]thymidine incorporation was highest in SP-null relative to WT LEC and SP-null⫹rhSP LEC. TGF-2 inhibited [3H]thymidine incorporation in all cultures. Data are expressed as the mean ⫾ SD. The differences between groups were significant to P ⬍ 0.05 except for bars labeled NS, which include WT versus SPnull⫹rhSP; and WT versus SPnull⫹rhSP⫹TGF2. In the absence of SPARC, proliferation increased and TGF2 inhibited proliferation in the presence or absence of SPARC.
4684
Gotoh et al.
IOVS, October 2007, Vol. 48, No. 10
IOVS, October 2007, Vol. 48, No. 10 tured collagen type I was observed in gels. Taken together, the results of the immunocytochemistry and Western immunoblots were consistent with increased expression of markers for EMT in the absence of SPARC that was influenced by TGF-2.
DISCUSSION The effect of TGF-2 on LEC migration, proliferation, and EMT is influenced by SPARC, a matricellular protein, in an in vitro IOL model for PCO. Both migration and proliferation of SP-null LEC increased relative to WT cells and were subsequently diminished by the addition of rhSP to the culture medium. We concluded that SPARC inhibits the migration and proliferation in LEC in vitro. Counteradhesion and antiproliferation are two major functions of SPARC defined in vitro.39 The counteradhesive function of SPARC is achieved in part by the dissolution of focal adhesion complexes at the ECM– cell interface and reorganization of actin stress fibers.29 Cell adhesion and subsequent spreading on compatible substrates are essential requirements for proper growth and survival of differentiating cells. Disruption of the interactions between the ECM and epithelial cells is expected to lead to cell rounding and, in some instances, apoptosis.40 The observed inhibition of LEC migration in the presence of SPARC is consistent with the counteradhesive function of SPARC. Previous studies also indicated that SPARC is a potent cell-cycle inhibitor that arrests cells in mid-G1, independent of discernible changes in cell shape.41– 43 Fibroblasts, mesangial cells, and smooth muscle cells isolated from SP-null mice exhibited a higher rate of proliferation relative to their WT counterparts. SP-null cells were more sensitive than WT cells to the inhibition of cell cycle progression in the presence of rhSPARC.35 These earlier reports are consistent with the observed inhibition of LEC proliferation by SPARC in the present study. The conclusions provide support for the hypothesis that SPARC inhibits PCO formation. In the PCO model, TGF-2 inhibited migration and proliferation of LECs from lenses of SPARC-null mice and stimulated the production of fibronectin and type I collagen relative to WT controls. The results confirmed that the effect of TGF-2 on migration, proliferation, and the production of two key matrix proteins is influenced by SPARC. TGF- is associated with the rapid remodeling of connective tissues and has been shown to regulate the expression of ECM proteins,44 to inhibit proliferation of cultured LECs,45 and to influence lens fiber differentiation. The effects of SPARC in the PCO model are in some respects similar to the effects of SPARC on EMT when inappropriate TGF- signaling in the anterior LECs results in an EMT resembling forms of human cataract.46 TGF- may be more important in the formation of anterior subcapsular cataract than in the formation of PCO. Numerous studies indicate that TGF--induced EMT is part of a wound-healing response in Š
FIGURE 5. Effect of TGF-2 on the expression of fibronectin, collagen type I, and ␣-SMA by LECs in an IOL model culture. Cells cultured with an IOL on collagen IV were immunostained for (A) collagen type I (green) and (B) fibronectin (FN; red) and were counterstained for nuclei with DAPI (blue). The staining intensities of FN and collagen I appeared greater in SP-null LECs than in WT LECs and SP-null⫹rhSP LECs. Proteins produced by cultured LECs were isolated for immunoblot analysis of FN, ␣-SMA, and GAPDH (C). The histogram summarizes the results of scanning densitometry of FN (D) and ␣-SMA (E) in the immunoblots. SP-null LECs produced more FN and ␣-SMA than WT LECs. In the presence of TGF[b]-2, production of FN and ␣-SMA was stimulated. Data were normalized to the internal loading control (GAPDH). All differences were statistically significant, except ␣-SMA in WT cell cultures. Expression of markers for EMT was enhanced in the absence of SPARC and increased in the presence of TGF-2.
Novel Model of Posterior Capsular Opacity
4685
LECs characterized by induced expression of numerous ECM proteins.47– 49 It has been demonstrated that SPARC can enhance TGF-1 expression (mRNA and protein) in cultured mouse mesangial cells.36 In SP-null mesangial cells, significant reduction in synthesis of TGF-1 mRNA was reported, and addition of SPARC to SP-null cells in culture restored the expression of TGF-1 to levels typical of WT cells.36 Our results indicate a functional intersection between pathways activated by TGF-2 and those involving the matricellular protein SPARC. Further studies are needed to determine the importance of these pathways in ECM– cell interactions and the process of PCO. PCO is a common complication after cataract surgery and IOL implantation. Studies that considered the inhibition of PCO are based on two possible mechanisms of PCO prevention: physical alteration of the design and material properties of IOL.50,51 It has been reported in clinical studies that a sharpedged acrylic IOL prevented PCO formation.52,53 Although it may be more relevant to study PCO clinically, details of LEC migration and proliferation are difficult to determine in patients, because PCO in patients and experimental animals are influenced by a variety of factors, including age, surgical procedure, and postoperative inflammation.54,55 Given the limitations of studies in vivo, investigators have developed human or animal models in vitro.56,57 Although the culture systems simulate only the postoperative condition, the proportion of viable cells remaining on the anterior capsule varied widely among preparations, from 30% to 80%.58,59 Therefore, the design of the in vitro model of PCO used in this study included a cell culture insert with a 1.0-m pore polyethylene terephthalate (PET) membrane coated with collagen type IV (Fig. 1). Although the model was developed independently, there are similarities to that of Kurosaka et al.60 After cataract surgery, a square-edged IOL can form a sharp capsular bend, and the migration of LECs can be inhibited at the edges.61 Therefore, maintenance of contact pressure was important. In our PCO model, LEC proliferated beyond the IOL edge, similar to PCO formation in human IOL implants. LECs on the collagen coated membranes expressed fibronectin and type I collagen (Fig. 5), indicating that cells in the PCO model underwent an EMT. Although these experiments were conducted in vitro, the PCO model mimics the behavior of LECs during PCO formation. Our results indicate that regulation of the ECM-cell interaction by the matricellular protein SPARC is a key factor in PCO. Current experiments with the IOL– collagen insert model for the PCO model are evaluating of SPARC inhibitors for the prevention of PCO.
Acknowledgments The authors thank Gail Workman for providing rhSPARC, Hidetoshi Iwamoto (Hoya Corp. Medical Division) for providing acrylic IOLs, and Joy Ghosh and Shawna Goldstein for assistance with the manuscript.
References 1. Apple DJ, Solomon KD, Tetz MR, et al. Posterior capsule opacification. Surv Ophthalmol. 1992;37(2):73–116. 2. Bertelmann E, Kojetinsky C. Posterior capsule opacification and anterior capsule opacification. Curr Opin Ophthalmol. 2001; 12(1):35– 40. 3. Schaumberg DA, Dana MR, Christen WG, Glynn RJ. A systematic overview of the incidence of posterior capsule opacification. Ophthalmology. 1998;105(7):1213–1221. 4. Novotny GE, Pau H. Myofibroblast-like cells in human anterior capsular cataract. Virchows Arch A Pathol Anat Histopathol. 1984;404(4):393– 401. 5. McDonnell PJ, Zarbin MA, Green WR. Posterior capsule opacification in pseudophakic eyes. Ophthalmology. 1983;90(12):1548 – 1553.
4686
Gotoh et al.
6. Cobo LM, Ohsawa E, Chandler D, et al. Pathogenesis of capsular opacification after extracapsular cataract extraction: an animal model. Ophthalmology. 1984;91(7):857– 863. 7. Wormstone IM. Posterior capsule opacification: a cell biological perspective. Exp Eye Res. 2002;74(3):337–347. 8. Meacock WR, Spalton DJ, Stanford MR. Role of cytokines in the pathogenesis of posterior capsule opacification. Br J Ophthalmol. 2000;84(3):332–336. 9. Saika S, Ohmi S, Kanagawa R, et al. Lens epithelial cell outgrowth and matrix formation on intraocular lenses in rabbit eyes. J Cataract Refract Surg. 1996;22(suppl 1):835– 840. 10. Lee EH, Seomun Y, Hwang KH, et al. Overexpression of the transforming growth factor-beta-inducible gene betaig-h3 in anterior polar cataracts. Invest Ophthalmol Vis Sci. 2000;41(7):1840 – 1845. 11. Lee EH, Joo CK. Role of transforming growth factor-beta in transdifferentiation and fibrosis of lens epithelial cells. Invest Ophthalmol Vis Sci. 1999;40(9):2025–2032. 12. Jampel HD, Roche N, Stark WJ, Roberts AB. Transforming growth factor-beta in human aqueous humor. Curr Eye Res. 1990;9(10): 963–969. 13. Inatani M, Tanihara H, Katsuta H, et al. Transforming growth factor-beta 2 levels in aqueous humor of glaucomatous eyes. Graefes Arch Clin Exp Ophthalmol. 2001;239(2):109 –113. 14. Massague J. The transforming growth factor-beta family. Annu Rev Cell Biol. 1990;6:597– 641. 15. Miller DA, Pelton RW, Derynck R, Moses HL. Transforming growth factor-beta: a family of growth regulatory peptides. Ann N Y Acad Sci. 1990;593:208 –217. 16. Saika S, Okada Y, Miyamoto T, et al. Smad translocation and growth suppression in lens epithelial cells by endogenous TGFbeta2 during wound repair. Exp Eye Res. 2001;72(6):679 – 686. 17. Saika S, Miyamoto T, Kawashima Y, et al. Immunolocalization of TGF-beta1, -beta2, and -beta3, and TGF-beta receptors in human lens capsules with lens implants. Graefes Arch Clin Exp Ophthalmol. 2000;238(3):283–293. 18. Saika S, Miyamoto T, Okada Y, et al. Transforming growth factorbeta isoform proteins in cell and matrix deposits on intraocular lenses. J Cataract Refract Surg. 2000;26(5):709 –715. 19. Saika S, Miyamoto T, Tanaka T, et al. Latent TGFbeta binding protein-1 and fibrillin-1 in human capsular opacification and in cultured lens epithelial cells. Br J Ophthalmol. 2001;85(11):1362– 1366. 20. Yan Q, Sage EH. SPARC, a matricellular glycoprotein with important biological functions. J Histochem Cytochem. 1999;47(12): 1495–1506. 21. Brekken RA, Sage EH. SPARC, a matricellular protein: at the crossroads of cell-matrix communication. Matrix Biol. 2001;19(8):816 – 827. 22. Bornstein P, Sage EH. Matricellular proteins: extracellular modulators of cell function. Curr Opin Cell Biol. 2002;14(5):608 – 616. 23. Maillard C, Malaval L, Delmas PD. Immunological screening of SPARC/Osteonectin in nonmineralized tissues. Bone. 1992;13(3): 257–264. 24. Lane TF, Sage EH. The biology of SPARC, a protein that modulates cell-matrix interactions. FASEB J. 1994;8(2):163–173. 25. Raines EW, Lane TF, Iruela-Arispe ML, et al. The extracellular glycoprotein SPARC interacts with platelet-derived growth factor (PDGF)-AB and -BB and inhibits the binding of PDGF to its receptors. Proc Natl Acad Sci USA. 1992;89(4):1281–1285. 26. Gilmour DT, Lyon GJ, Carlton MB, et al. Mice deficient for the secreted glycoprotein SPARC/osteonectin/BM40 develop normally but show severe age-onset cataract formation and disruption of the lens. EMBO J. 1998;17(7):1860 –1870. 27. Norose K, Clark JI, Syed NA, et al. SPARC deficiency leads to early-onset cataractogenesis. Invest Ophthalmol Vis Sci. 1998; 39(13):2674 –2680. 28. Bassuk JA, Birkebak T, Rothmier JD, et al. Disruption of the Sparc locus in mice alters the differentiation of lenticular epithelial cells and leads to cataract formation. Exp Eye Res. 1999;68(3):321–331. 29. Yan Q, Clark JI, Wight TN, Sage EH. Alterations in the lens capsule contribute to cataractogenesis in SPARC-null mice. J Cell Sci. 2002; 115:2747–2756.
IOVS, October 2007, Vol. 48, No. 10 30. Francki A, McClure TD, Brekken RA, et al. SPARC regulates TGFbeta1-dependent signaling in primary glomerular mesangial cells. J Cell Biochem. 2004;91(5):915–925. 31. Fujita T, Shiba H, Van Dyke TE, Kurihara H. Differential effects of growth factors and cytokines on the synthesis of SPARC, DNA, fibronectin and alkaline phosphatase activity in human periodontal ligament cells. Cell Biol Int. 2004;28(4):281–286. 32. Shanker G, Olson D, Bone R, Sawhney R. Regulation of extracellular matrix proteins by transforming growth factor beta1 in cultured pulmonary endothelial cells. Cell Biol Int. 1999;23(1): 61–72. 33. Abe K, Hibino T, Mishima H, Shimomura Y. The cytokine regulation of SPARC production by rabbit corneal epithelial cells and fibroblasts in vitro. Cornea. 2004;23(2):172–179. 34. Yan Q, Weaver M, Perdue N, Sage EH. Matricellular protein SPARC is translocated to the nuclei of immortalized murine lens epithelial cells. J Cell Physiol. 2005;203(1):286 –294. 35. Emery J. Capsular opacification after cataract surgery and capsule. Curr Opin Ophthalmol. 1998;9(1):60 – 65. 36. Nagamoto T, Eguchi G, Beebe DC. Alpha-smooth muscle actin expression in cultured lens epithelial cells. Invest Ophthalmol Vis Sci. 2000;41(5):1122–1129. 37. Motamed K, Sage EH. SPARC inhibits endothelial cell adhesion but not proliferation through a tyrosine phosphorylation-dependent pathway. J Cell Biochem. 1998;70(4):543–552. 38. Chen CS, Mrksich M, Huang S, et al. Geometric control of cell life and death. Science. 1997;276(5317):1425–1428. 39. Funk SE, Sage EH. The Ca2(⫹)-binding glycoprotein SPARC modulates cell cycle progression in bovine aortic endothelial cells. Proc Natl Acad Sci USA. 1991;88(7):2648 –2652. 40. Funk SE, Sage EH. Differential effects of SPARC and cationic SPARC peptides on DNA synthesis by endothelial cells and fibroblasts. J Cell Physiol. 1993;154(1):53– 63. 41. Sage EH, Vernon RB. Regulation of angiogenesis by extracellular matrix: the growth and the glue. J Hypertens Suppl. 1994;12(10): S145–S152. 42. Bradshaw AD, Francki A, Motamed K, et al. Primary mesenchymal cells isolated from SPARC-null mice exhibit altered morphology and rates of proliferation. Mol Biol Cell. 1999;10(5): 1569 –1579. 43. Kingsley DM. The TGF-beta superfamily: new members, new receptors, and new genetic tests of function in different organisms. Genes Dev. 1994;8(2):133–146. 44. Kurosaka D, Nagamoto T. Inhibitory effect of TGF-beta 2 in human aqueous humor on bovine lens epithelial cell proliferation. Invest Ophthalmol Vis Sci. 1994;35(9):3408 –3412. 45. de Iongh RU, Wederell E, Lovicu FJ, McAvoy JW. Transforming growth factor-beta-induced epithelial-mesenchymal transition in the lens: a model for cataract formation. Cells Tissues Organs. 2005;179(1–2):43–55. 46. Liu J, Hales AM, Chamberlain CG, McAvoy JW. Induction of cataract-like changes in rat lens epithelial explants by transforming growth factor beta. Invest Ophthalmol Vis Sci. 1994;35(2):388 – 401. 47. de Iongh RU, Gordon-Thomson C, Chamberlain CG, et al. Tgfbeta receptor expression in lens: implications for differentiation and cataractogenesis. Exp Eye Res. 2001;72(6):649 – 659. 48. Hales AM, Chamberlain CG, McAvoy JW. Cataract induction in lenses cultured with transforming growth factor-beta. Invest Ophthalmol Vis Sci. 1995;36(8):1709 –1713. 49. Francki A, Bradshaw AD, Bassuk JA, et al. SPARC regulates the expression of collagen type I and transforming growth factor-beta1 in mesangial cells. J Biol Chem. 1999;274(45):32145–32152. 50. Nishi O, Nishi K. Preventing posterior capsule opacification by creating a discontinuous sharp bend in the capsule. J Cataract Refract Surg. 1999;25(4):521–526. 51. Tobari I, Iwaki Y, Miyake K. Effect of tranilast eyedrops in preventing posterior capsule opacification: preliminary report. J Cataract Refract Surg. 1999;25(10):1394 –1399. 52. Hayashi K, Hayashi H, Nakao F, Hayashi F. Changes in posterior capsule opacification after poly(methyl methacrylate), silicone, and acrylic intraocular lens implantation. J Cataract Refract Surg. 2001;27(6):817– 824.
Novel Model of Posterior Capsular Opacity
IOVS, October 2007, Vol. 48, No. 10 53. Oner FH, Gunenc U, Ferliel ST. Posterior capsule opacification after phacoemulsification: foldable acrylic versus poly(methyl methacrylate) intraocular lenses. J Cataract Refract Surg. 2000; 26(5):722–726. 54. Liu CS, Wormstone IM, Duncan G, et al. A study of human lens cell growth in vitro: a model for posterior capsule opacification. Invest Ophthalmol Vis Sci. 1996;37(5):906 –914. 55. Nagamoto T, Eguchi G. Effect of intraocular lens design on migration of lens epithelial cells onto the posterior capsule. J Cataract Refract Surg. 1997;23(6):866 – 872. 56. Wormstone IM, Liu CS, Rakic JM, et al. Human lens epithelial cell proliferation in a protein-free medium. Invest Ophthalmol Vis Sci. 1997;38(2):396 – 404. 57. Quinlan M, Wormstone IM, Duncan G, Davies PD. Phacoemulsification versus extracapsular cataract extraction: a comparative
58.
59.
60.
61.
4687
study of cell survival and growth on the human capsular bag in vitro. Br J Ophthalmol. 1997;81(10):907–910. Nagamoto T, Hara E. Lens epithelial cell migration onto the posterior capsule in vitro. J Cataract Refract Surg. 1996;22(suppl 1):841– 846. Nagamoto T, Bissen-Miyajima H. A ring to support the capsular bag after continuous curvilinear capsulorrhexis. J Cataract Refract Surg. 1994;20(4):417– 420. Kurosaka D, Obasawa M, Kurosaka H, Nakamura K. Inhibition of lens epithelial cell migration by an acrylic intraocular lens in vitro. Ophthalmic Res. 2002;34(1):29 –37. Nishi O. Influence of intraocular lens material and design on the development of posterior capsule opacification [in German]. Ophthalmologe. 2005;102(6):572–578.
