Lens Cell Survival after Exposure to Stress in the Closed Capsular Bag

Abstract

purpose. Despite recent improvements in IOL design posterior capsule opacification (PCO) remains a significant clinical problem after cataract surgery. The Perfect Capsule device (Milvella, Ltd., Epping, Australia) permits the introduction and subsequent removal of potentially toxic agents into the closed capsular bag. The present purpose was to compare the relative effectiveness of exposing cells within the human capsular bag with a range of stresses with clinical potential and to compare the response to the same agents when applied to rabbit capsular bags and to cultured human lens cells.

methods. Human capsular bags were prepared from donor eyes and sealed with the Perfect Capsule device. Distilled water, 3 M NaCl, 250 μg/mL, 25 mg/mL 5-fluorouracil and 100 μM thapsigargin (Tg) were introduced for 2 minutes. The bags were then perfused with Eagle’s minimum essential medium (EMEM) and an IOL inserted before the bags were dissected and pinned to the base of plastic culture dishes. The bags were maintained in EMEM for 28 days and phase images were acquired throughout. Rabbit eyes were prepared and cultured in a similar manner, although tests were limited to 5-FU (25 mg/mL) and Tg (30–300 μM). FHL124 cells were cultured on plastic (EMEM supplemented with 5% FCS), and serum was removed for 24 hours before exposure to the same agents as human bags. Cell survival was assessed by quantifying Coomassie blue staining after 4 days.

results. Initially, NaCl induced by far the most obvious signs of cell death, especially of anterior cells, followed by 5-FU>water>Tg. However, by 2 weeks, cell death became more apparent in the Tg-exposed bags, and, at the end of 4 weeks, there were no cells surviving. Cells on the posterior capsule were confluent in water-exposed bags (similar to unexposed controls), whereas after NaCl exposure, coverage was incomplete but greater than after 5-FU. In the rabbit bags, exposure to 25 mg/mL 5-FU totally eliminated cells, but 100 μM Tg was ineffective. At the end of a 4-day culture of FHL124 cells exposed for 2 minutes to NaCl, 5-FU or Tg, there were no cells surviving, whereas there was 50% cell survival compared with control cells after water treatment.

conclusions. Tissue-cultured human lens cells are much more sensitive to short-term hyperosmotic than hyposmotic stress, with a rapid onset of cell death of cultured cells exposed to 3 M NaCl. This finding was confirmed in human capsular bags although adherent fibers appear to offer additional protection to 5-FU which can be overcome by the very hydrophobic Tg. The application of the Perfect Capsule system in concert with thapsigargin provides a promising means of preventing PCO.

Modern cataract surgery involves extracapsular extraction to remove opaque lens fibers followed by IOL implantation, which normally restores excellent visual acuity. However, a secondary loss of acuity occurs in a significant number of patients, due to the robust growth of residual lens cells within the lens capsular bag—a condition termed PCO (posterior capsular opacification) that occurs in approximately 20% of eyes followed up for 3 to 5 years after senile cataract surgery. ^{1 2} In children, PCO can occur in almost all cases within the same period. In pediatric surgery it is common to perform an additional posterior rhexis to deprive lens cells of a scaffold on which to grow, and this, allied to a radical approach to IOL design, has greatly reduced the incidence of PCO. ^{3 4} It is unlikely, however, that this radical bag-in-the-lens approach will be widely adopted for adult cataract operations, because not only does it demand a higher degree of skill, but there are additional associated complications. ^{3 4} Recent advances in conventional IOL design, and in particular, a square edge profile for the posterior surface of the IOL, have served merely to delay PCO rather than to eliminate the problem. ^{5 6} Currently, the only method available to treat PCO is to perform a posterior capsulotomy (usually by Nd-YAG laser) to clear the visual axis, but this procedure is relatively expensive and not without risk as retinal detachment, and cystoid macular edema can occur. ⁷  

Theoretically, the most efficient way to prevent PCO is to eliminate all the cells within the capsular bag at surgery, and a range of cytotoxic agents have been tested in several human and animal model systems. Promising results have been obtained in vitro from lens cell tissue culture and capsular bag studies and also from animal model investigations. For example, both saporin ⁸ and thapsigargin (Tg) ⁹ have been successfully targeted to capsular bag cells by coating PMMA lenses with them, but the risk of toxic effects on surrounding ocular tissues has inhibited their potential for human clinical trials. Practically, therefore, elimination of PCO requires cytotoxic agents specifically targeted to the capsular bag with a sufficiently low subsequent leakage rate that surrounding tissues are not affected. The Perfect Capsule device (Fig. 1)recently developed by Maloof et al. ¹⁰ permits the capsular bag to be sealed off from the environment, permitting cytotoxic agents to be both delivered selectively to the bag and, equally importantly, to be removed during the course of the operation. This method has been used to deliver demineralized water and the detergent Triton X-100 to rabbit capsular bags in vivo (Maloof A et al. IOVS 2003;44:ARVO E-Abstract 284) and although there was no damage to surrounding tissues, both agents reduced the population of epithelial cells remaining in the bags after surgery. Preliminary studies have begun with this device in patients undergoing cataract surgery (Genge JR et al. IOVS 2002;43:ARVO E-Abstract 434) and, on perfusing closed bags with a trypan blue solution, no contamination of surrounding tissues with irrigation fluid was observed. 

So that the surgical time is not greatly prolonged compared with present methodology, a 2-minute window is an appropriate time for irrigation (Maloof A et al. IOVS 2003;44:ARVO E-Abstract 284; Genge JR et al. IOVS 2002;43:ARVO E-Abstract 434), ¹⁰ and it is important to establish which cytotoxic agent produces the most efficient inhibition of PCO within this time frame. This may not in fact be seen as the immediate induction of cell death, as some promising agents suppress protein synthesis and cell growth leading to reduced cell viability and apoptosis. ^{9 11} We first tested four agents (distilled water, 3 M NaCl, 25 mg/mL 5-FU, and 100 μM Tg) on a lens cell line (FHL124) that expresses 99% of the genes of native lens cells ¹² and found that all were relatively effective within the 2-minute time period. We also report data from experiments in which the Perfect Capsule system has been attached to capsular bags generated within human donor eyes. ^{13 14} The resultant capsular bags were irrigated with the same four agents for 2 minutes, and the data revealed that Tg was the only agent capable of inducing total cell death within the bag. We argue that this is due to the very hydrophobic nature of the molecule. As the rabbit eye is the main animal model of choice in investigations of PCO, ^{15 16} we performed a limited study of the relative sensitivity of human and rabbit bags to two of the agents tested and revealed important species differences. 

Materials and Methods

Cell Culture

FHL124 cells ^{12 13 17 18} were routinely cultured in EMEM supplemented with 5% FCS (Invitrogen-Life Technologies, Ltd., Paisley, UK) and seeded on 24-well plates for a growth assay. 

Growth Assays

FHL124 cells were seeded on 24-well plates and maintained in EMEM supplemented with 5% FCS. The medium was replaced with nonsupplemented EMEM and cultured for a further 24 hours. ¹² The cells were exposed to experimental conditions for 2 minutes and maintained in medium for 4 days. At the end of the experiments, the cells were fixed for 30 minutes with 4% formaldehyde at room temperature. The cells were then washed in PBS, stained with Coomassie brilliant blue for 10 minutes, and washed several times to remove excess dye. For an analysis of total dye uptake (related to total cell protein), the PBS was replaced with 1 mL of 70% ethanol, and the dishes were agitated for 1 hour until all dye was removed. A portion (200 μL) of the ethanol-dye mixture from each well was added to a clear plastic 96-well microtiter plate and the dye content measured (Wallac Victor2 1420 multilabel counter with Workout software, ver. 1.5; Perkin Elmer, Wellesley, MA). 

Preparation and Culture of Capsular Bags

Human eye tissue donated for research was obtained from the East Anglian Eye Bank, and usage was in accordance with the tenets of the Declaration of Helsinki. As previously described, ¹³ a sham cataract operation was performed on donor eyes, and the window on the anterior capsule was sealed with a grooved silicone disc (Fig. 1)from the Perfect Capsule device ¹⁹ (Milvella, Ltd., Sydney, Australia). The control (EMEM) or experimental conditions were injected through a port in the silicone arm extension into the sealed capsular bag continuously for 2 minutes (2 mL/min), and the capsular bags were then rinsed with EMEM. A PMMA IOL (Rayner Opticians, Chesham, UK) was implanted into each bag, and the bags were maintained in EMEM for 28 days. ¹⁴ Ongoing cell observations were performed with a phase-contrast microscope (Nikon, Tokyo, Japan), and images were captured with a digital camera (Coolpix 950; Nikon) with associated software. 

Rabbits (New Zealand White) were obtained from a registered supplier (Harlan UK Ltd., Oxon, UK), and the animals were killed by barbiturate overdose. The eyes were enucleated and the corneas removed. A sham cataract operation was performed, and the capsular bags prepared as just described. 

Immunocytochemistry of the Capsular Bag

All reagents were from Sigma-Aldrich (Poole, UK) unless otherwise stated. Three washes were performed briefly in phosphate-buffered saline (PBS)/bovine serum albumin (BSA) and Nonidet (0.02% and 0.05%, respectively). The pinned capsules were fixed for 30 minutes in 4% formaldehyde in PBS and permeabilized in PBS containing 0.5% Triton X-100, also for 30 minutes. Nonspecific sites were blocked with appropriate serum (1:50 in 1% BSA/PBS). Anti-vimentin (clone V9) was diluted 1:100 and applied for 60 minutes at 35°C, followed by washing. It was visualized with fluorescein isothiocyanate (FITC)-conjugated secondary antibodies. The F-actin cytoskeleton was stained with Texas red X-phalloidin (Invitrogen-Molecular Probes, Leiden, The Netherlands) for 30 minutes, and cell nuclei with 4′,6′-diamidino-2-phenylindole-2 HCl (DAPI) at 1 mg/mL for 15 minutes, all at room temperature. The stained preparations were again washed extensively, floated onto microscope slides, and mounted (Vectashield; Vector Laboratories, Peterborough, UK). Images were viewed with either a fluorescence microscope (Eclipse E800; Nikon) or a confocal microscope (Viewscan DVC-250; Bio-Rad, Richmond, CA) with cooled charge-coupled device (CCD) camera (Princeton Instruments, Marlow, UK) and software (MetaMorph; Universal Imaging, West Chester, PA). 

Statistical Analyses

A t-test analysis was performed (Excel; Microsoft, Redmond, WA), to determine any statistically significant difference between groups. A 95% confidence interval (CI) was used to assess significance. 

Results

The four toxic agents to be tested were distilled water, 3 M NaCl, 5-FU, and Tg. The concentrations of the first two were defined by the minimum and maximum osmotic stress to be delivered, ^{15 20} whereas two clinical concentrations for 5-FU are possible. Lower concentrations (100–250 μg/mL) are used to test anticancer protocols, ^{21 22} whereas higher levels (25–50 mg/mL) are applied externally to the eye during glaucoma surgery. ^{23 24} We had shown in prior work that Tg was effective as a coating on IOLs inserted into the in vitro bag, ⁹ but it was not possible to extract from these results an effective concentration to be applied in experiments lasting for only 2 minutes. FHL124 cells were therefore exposed to a wide range of Tg concentrations (100 nM–100 μM), and after exposure the cells were cultured for 4 days, either in serum-free EMEM or in EMEM supplemented with viscoelastic or serum. The presence of viscoelastic was tested, as is routine practice during cataract surgery, to protect corneal endothelial cells. A possible protective effect of serum was also tested, as serum proteins are not only normally present at low levels in the aqueous, but they can increase as a result of a weakening of the blood-ocular barrier during surgery. ²⁵ The data in Figure 2Ain fact show a protective effect of both serum and viscoelastic at all Tg concentrations except 100 μM, which was totally effective in eliminating cells by the end of the 4-day culture period. 

The other three agents were also tested in the presence of serum and viscoelastic, and distilled water was surprisingly ineffective under all conditions (Fig. 2B) . The lower concentration of 5-FU (250 μg/mL) was more effective than water, but failed to eliminate the cells. The 25-mg/mL formulation was much more effective, although there was a small protective effect from serum. NaCl (3 M) totally eliminated the cells under all conditions. Although it is estimated that 2 minutes represents a reasonable exposure during surgery, it would be of great value if the time could be shortened. In fact, 100 μM Tg is totally effective after an exposure of only 30 seconds (Fig. 3A) . However, when the exposure time is reduced still further, then Tg is no more effective than either 5-FU or NaCl (Fig. 3B) . 

It is a relatively simple matter to quantify cell growth on the posterior capsule in human bags after sham cataract surgery by noninvasive, phase microscopy as, initially, there are no epithelial cells residing there. ¹³ In the present experiments, cell coverage was assessed from phase images taken after 28 days of culture, as control cells were confluent (100% cover) by this time (Figs. 4A 4B) . ^{13 14} There was no significant effect after exposure to the lower concentration of 5-FU and while distilled water reduced coverage by only 20%, the hyperosmotic shock induced by 3 M NaCl was much greater. 5-FU and Tg were by far the most effective agents in inhibiting cell migration across the posterior capsule (Fig. 4B) . Although it was immediately apparent from the phase images (Fig. 4A)that there were cells residing on the anterior capsule of the distilled water– and NaCl-treated bags, this observation was not obvious in the 5-FU- and Tg-treated bags. Furthermore, in a phase study, it is not always possible to discern whether the cells observed are alive or dead. A further immunocytochemical study was therefore performed to determine the actin, vimentin, and chromatin distribution within cells residing on either the posterior or anterior capsule. 

The higher-power fluorescence microscope images (Fig. 5A)clearly show live cells on the posterior capsule in control, distilled water, NaCl, and 5-FU (250 μg/mL) treated bags, whereas no live cells were seen on the central posterior capsule of the 5-FU- and Tg-treated bags. Near the rhexis region, highly fluorescent cells were also seen on the outer surface of the anterior capsule in the corresponding lower-power images (Fig. 5B) . The underlying cells on the juxtaposed anterior and posterior capsule regions are less clearly visible, mainly because of a poorer penetration of antibodies and fluorescent labels (see also Ref. ¹³ ). However, it is quite clear that live cells remained in the outer regions of the 5-FU-treated bags, whereas none remained in the Tg bags. 

Cell growth in the rabbit capsular bags was more rapid and the control cells were confluent after 2 weeks (Fig. 6A) . In the phase images, no cells were apparent on the posterior capsule of the 5-FU-treated bags—a finding that was confirmed in the immunofluorescence study (Fig. 6B) . On the contrary, the posterior capsules of Tg-treated bags were confluent by this time, even when the concentration was increased to 300 μM. In contrast, 100 μM Tg totally inhibited anterior and posterior capsule cell cover in human bags (Fig. 5A) . 

Discussion

In the past few years, there has been a rapid increase in new technologies to provide optimal visual acuity after cataract surgery. ^{26 27 28 29 30} Despite these advances, PCO remains to a greater or lesser extent a possible risk of a secondary loss of visual acuity in a significant proportion of patients, and recent studies have indicated that approximately 30% of patients require laser surgery within 4 years of surgery. ⁴ In fact, it is likely that, in the more novel accommodative approaches, this problem is exacerbated compared with conventional technologies. ²⁸ It is more important therefore, if these new technologies are to be incorporated, that the possibility of PCO be eliminated as much as possible. It is unlikely that this solution will come in the form of IOLs coated with cell-arresting compounds, as the new technologies will demand a choice of matrix elements from which to form the lens and, in the case of refilling the bag with deformable substances, a coating simply could not be formed. The simplest and most efficient means of preventing PCO will therefore be to treat the capsular bag before the IOL is implanted and if a toxic agent is to be used, some means must be found of preventing contamination of surrounding ocular tissues. The Perfect Capsule system ¹⁹ provides a means of not only introducing a toxic agent into the bag, but also of removing it afterward. The crucial decision to make is therefore to find an agent or manipulation that will effectively lead to the elimination of PCO within the 2-minute window during the surgical procedure. A vast range of agents incorporating a large number of experimental models have been used in different approaches in the past to predict an approach that shows promise for translation to the clinic (see for example, Refs. ^{1 31} for reviews). For this study we chose to compare the effect of four agents in three model systems. Human tissue cultured cells, in vitro capsular bags and in vivo rabbit models have all been used in the past, but there has been no comparative study of the relative predictive efficiency of them. Similarly, there have been few comparative studies of the relative effectiveness of different agents. We have compared distilled water, 3 M NaCl, 5-FU, and Tg. 

Distilled water has in fact been used in an in vitro human lens cell study (Crowston JG et al. IOVS 2003;44:ARVO E-Abstract 278) in which it was found that a 2-minute exposure of isolated anterior capsules (rhexis specimens) was sufficient to eliminate all live cells. In a small follow-up clinical investigation (Pandey SK et al. IOVS 2004;45:ARVO E-Abstract 318) distilled water was applied in the perfect capsule system where it was found to be safe in terms of a lack of any effect on other ocular tissues. The authors also reported that there appeared to be less anterior capsule opacification 6 months later in the treated eyes than in control eyes that had undergone routine cataract surgery. Injection of distilled water for a 3-minute period into rabbit capsular bags in vivo has also been reported as a very efficient means of preventing PCO. ¹⁵ The present data from human tissue cultured cells and capsular bags (Figs. 2 3 4 5)would predict that a 2-minute exposure to distilled water would not be sufficient to inhibit PCO to any significant degree. However, we noted that, after exposure of capsular bags, there was a high degree of cell death induced in the anterior epithelial cells (data not shown) indicating that they are more sensitive to hypoosmotic stress than were the equatorial cells. To this extent, anterior rhexis samples do not represent a good predictive model in which to test anti-PCO technologies. 

Hyperosmotic stresses are more efficient in reducing cell growth in both tissue-cultured cells and capsular bags (Figs. 2B 4) , and it has been pointed out that exposure of the intact lens to hyperosmotic stresses induces much greater opacification than does exposure to equivalent hypoosmotic solutions. ³² Cell death can be induced by both hyperosmotic and hypo-osmotic stress, ^{33 34 35} and in either case specific cell death pathways are induced. The lower sensitivity to hyposmotic stress in this case may arise from these signaling differences or simply through differences in water and ion diffusion through the unstirred layers surrounding the less-exposed equatorial cells. A further and perhaps more attractive possibility is that hyperosmotic stresses have been reported as less reversible than hyposmotically induced effects, perhaps because of the greater protein-protein interaction involved when cells shrink. ³²  

The antimetabolite 5-FU has a long history of clinical use both in cancer chemotherapy ^{21 22} and glaucoma surgery. ^{23 36} It is worth noting that whereas 5-FU is applied at effective plasma concentrations on the order of 200 μg/mL in cancer chemotherapy, the corresponding concentration used in glaucoma surgery is much greater (25–50 mg/mL). Both concentration ranges were tested in the present study. There have been several investigations of the possible application of 5-FU in preventing PCO. McDonnell et al. ³⁷ first demonstrated that prolonged exposure of human lens cells to the low concentration of 30 μg/mL induced a 50% reduction in growth. As the present study has shown, shorter exposure times demand higher concentrations to be effective (Fig. 2B)and, in an in vivo rabbit investigation in which a mini capsulorrhexis valve was used to close the capsule, Fernandez et al. ¹⁵ reported that 33 mg/mL 5-FU significantly retarded the appearance of posterior capsule fibrosis. In that study, it was not as effective as distilled water. However, it should be noted that in that particular study, all capsular bags had evidence of lens regeneration. In an in vivo rabbit study where 5-FU was delivered in a slow-release form from an intracapsular ring (0.25 μg/h), it was quite ineffective. ¹⁶ 5-FU (25 mg/mL) appears to have been particularly effective in the present study (Fig. 4) , and the major reason for its effectiveness probably lies in the fact that the bag closed by the Perfect Capsule system can be continuously perfused with solution (10 mL/min) during the 2-minute period. Unstirred layers can therefore be disturbed and the effective concentration at the cell interface is much higher. This effect cannot be achieved by either simple injection into the open capsule or through the minivalve system into the closed capsular bag (Orozco MA et al. IOVS 2003;44:ARVO E-Abstract 280). ¹⁵  

Thapsigargin provided the most efficient means of preventing PCO in the human capsular bag system and the underlying mechanism of action is quite different from the other stresses applied. For the first few days, cell appearance was unaltered (see also Ref. ⁹ ), but then cell death became progressively more and more apparent. The reason lies in the mechanism of action of Tg, which is initially to disable calcium signaling by depleting the ER calcium store to induce first cell cycle arrest and later apoptosis. ¹¹ 5-FU has a much more rapid toxic effect, and this is partly because it is only soluble at high concentrations in alkaline solutions (pH 9.2). The effective concentration of 25 mg/mL approximates to 150 mM, and this will also induce an osmotic effect on cells. It is also likely that Tg is effective because of its hydrophobic nature, ⁹ and so it can cross cell boundaries rapidly, rendering the buffering nature of residual fibers less effective. The FHL124 cells serve as a valuable tool to identify which cytotoxic agents are likely to be successful when applied to native human lens epithelial cells. However, the capsular bag system provides a more complete model of the in vivo scenario, because the role of residual fiber cells and unstirred layers can be assessed. 

The in vitro rabbit capsular bag system does not appear to represent a good predictive model for the events that follow exposure of human lens cells to 5-FU or Tg, despite its having the same spatial organization. The reason that rabbit capsular bags are more sensitive to 5-FU may be that the rabbit lenses used in the present experiments (and indeed in all such studies) are much younger than their human counterparts. There is a considerable body of literature demonstrating that lenses of young animals are more sensitive to a range of stresses. ^{38 39} Furthermore, the rabbit cells grow faster, and 5-FU targets dividing cells. The reason that the rabbit capsular bag is less sensitive to Tg may be the different nature of the rabbit eye. ⁴⁰ It is much more open to attack by foreign substances and so may well possess an efficient multidrug transporting system to remove offending molecules. ⁴¹ Tg has in fact been identified as a target for multi-drug-resistance transporters. ⁴²  

In summary, the data derived from this in vitro investigation of a range of stresses applied to human lens cells predicts that the application of the Perfect Capsule system in concert with Tg is a most promising means of preventing PCO. 

²  Deceased January 17, 2007.

 Supported by The Humane Research Trust and Milvella, Ltd.

 Submitted for publication November 9, 2006; revised December 13, 2006; accepted April 20, 2007.

 Disclosure: G. Duncan, Milvella, Ltd. (F, P); L. Wang, Milvella, Ltd. (F, P); G. J. Neilson, Milvella, Ltd. (E); I.M. Wormstone, Milvella, Ltd. (F, P)

 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.

 

Figure 1.

 

The Perfect Capsule system. This is a sterile, single-use, vacuum-sealed delivery system designed to remove lens epithelial cells from inside the capsular bag after capsulorrhexis; in a procedure called SCI (sealed capsule irrigation). It consists of a grooved silicone disc (arrow and inset) which is placed over the capsulorrhexis window to form a seal, and the extension arm permits solutions to be introduced into the bag. In addition, solutions can be withdrawn through the aspiration port (arrowhead).

Figure 2.

 

(A) Effect of exposing FHL124 cells for 2 minutes to a range of Tg concentrations under three conditions: in serum-free EMEM (SF); EMEM supplemented with 5% FCS or with a 5% viscoelastic (Healon, AMO, Santa Ana, CA). The cell population was measured after 4 days of growth in the three different conditions and is expressed as a percentage of the final estimated growth in the respective control medium without exposure to Tg. (B) Effect of exposing FHL124 cells to distilled water, 5-FU (250 μg/mL and 25 mg/mL) under the three conditions described in (A). The experiments repeated with different batches of cells with fresh solutions, and very similar results were obtained. *Significantly different from the respective control data. P < 0.05, n = 4.

Figure 3.

 

The relative effectiveness of exposing FHL124 cells to distilled water, 5-FU (25 mg/mL), NaCl (3 M), and Tg (100 μM) for either 1 minute or 30 seconds (A) and 15 seconds or 7.5 seconds (B). The cell population was assessed after 4 days’ growth in EMEM, and control data were generated by simply exposing the cells for the respective times to fresh EMEM. The experiments were repeated with a different batch of cells with fresh solutions and very similar results were obtained. *Significantly different from the respective control data. P < 0.05, n = 4.

Figure 4.

 

(A) Modified dark-field images of four capsular bag quarters showing the posterior capsule (central area) and outer anterior capsule captured after 28 days of culture. Bags had been exposed (with Perfect Capsule attached) for 2 minutes to distilled water, 3 M NaCl, 25 mg/mL 5-FU, or 100 μM Tg. They were then maintained in EMEM for the remainder of the period. (B) Influence of a 2-minute exposure to the different agents on the growth of cells on the posterior capsule; 100% represents complete cover (control), whereas exposure to Tg led to total cell death (0%). Data are expressed as the mean ± SEM (n = 4). The experiments were performed in a matched-pair format. *Significant difference between test and control data (P < 0.05).

Figure 5.

 

(A) Epifluorescence micrographs showing actin, vimentin, and chromatin distribution within human lens epithelial cells on the central area of the posterior capsule. (B) Low-power micrographs of cells on the anterior (AC) and posterior (PC) areas of the lens capsule. Note that cells grow on the upper and lower surfaces of the anterior capsule but because of a differential penetration of antibodies, the cells on the upper surface are more clearly defined.

Figure 6.

 

(A) Modified dark-field images of four rabbit capsular bag quarters showing the posterior capsule (central area) and outer anterior capsule captured after 14 days of culture. Bags had been exposed (with Perfect Capsule attached) for 2 minutes to distilled water, 3 M NaCl, 25 mg/mL 5-FU, or 100 μM Tg. They were then maintained in EMEM for the remainder of the period. (B) Epifluorescence micrographs showing actin, vimentin, and chromatin distribution within rabbit lens epithelial cells on the central area of the posterior capsule.

The authors thank John Reddan for providing the FHL124 cell line, Pamela Keeley of the East Anglian Eye Bank for providing human lens donor material, and Diane Alden for invaluable tissue culture assistance.