Osmotically induced removal of lens epithelial cells to prevent PCO after pediatric cataract surgery: Pilot study to assess feasibility

1480

LABORATORY SCIENCE

Osmotically induced removal of lens epithelial cells to prevent PCO after pediatric cataract surgery: Pilot study to assess feasibility JinJun Zhang, PhD, MD, Ali Hussain, PhD, Sun Yue, PhD, MD, Tao Zhang, MD, John Marshall, PhD, FRCPath, FMedSci

Purpose: Development of an osmotic-shock technique to remove human lens epithelial cells (LECs) as a preventive measure to address posterior capsule opacification (PCO) after pediatric cataract surgery. Setting: Department of Genetics, UCL Institute of Ophthalmology, London, England, and Department of Ophthalmology, Ruijin Hospital, Jiao Tong University, Shanghai, China. Design: Laboratory study. Methods: Various tissue preparations of human LECs (cultured on coverslips/collagen-coated membrane inserts, human lens capsule biopsies, and lens organ cultured PCO models) were subjected to a single or incremental hyperosmotic shock (NaCl, 350–4000 mOsm/L) in the presence of inhibitors of the NaC-KC-2Cl cotransporter (NKCC) (to disable the regulatory volume increase [RVI] process). The integrity of the cell monolayer was determined by phase–contrast microscopy, viability assays, and measurement of transepithelial resistance.

P

ediatric cataract represents the major preventable cause of visual impairment and blindness in childhood.1 The only treatment for pediatric cataract is surgery. Unfortunately, cataract surgery in children results in the postsurgical complication of posterior capsule opacification (PCO), which occurs in nearly 100% of cases.2,3 In adults, PCO occurs in about 20% to 30% of eyes followed up for 3 to 5 years after cataract surgery.4–6 PCO is caused by the aggressive proliferation and migration of

Results: Hyperosmotic shock (400 mOsm/L) caused rapid cell shrinkage (<5 minutes) in all the LEC models studied. In the absence of the NKCC inhibitor, the shrunk cells gradually returned to their original cell volume and architecture over time, while still exposed to the hyperosmotic shock. However, inhibition of the RVI process disabled the ability for restoration of cell volume leading to persistent cell shrinkage, subsequently resulting in cell detachment from the underlying support medium.

Conclusion: Hyperosmotic shock in the presence of inhibitors of the RVI process was effective in rapidly detaching LECs from their basement membranes. This technique could potentially facilitate removal of residual LECs left on the lens capsule after cataract surgery, thus decreasing or eliminating the risk for aggressive cell proliferation and the development of PCO. J Cataract Refract Surg 2019; 45:1480–1489 Crown Copyright Q 2019 Published by Elsevier Inc. on behalf of ASCRS and ESCRS. All rights reserved.

lens epithelial cells (LECs), which remain in the capsular bag after surgery.7–9 The pathogenesis of PCO is well understood as a fibrotic condition initiated by an inflammatory response attributable to tissue trauma triggered by cataract surgery and then a foreign body reaction toward the implanted intraocular lens (IOL).10,11 In the normal lens, LECs are confined to the anterior capsule and do not proliferate and grow into the posterior capsule. However, in some patients and in

Submitted: January 24, 2019 | Final revision submitted: April 19, 2019 | Accepted: April 30, 2019 From the Department of Genetics (J. Zhang, Hussain, Yue, T. Zhang, Marshall), UCL Institute of Ophthalmology, London, England; the Department of Ophthalmology (Yue), Ruijin Hospital, Jiao Tong University, Shanghai, China. This research was supported by grants from the Medical Research Council (MRC: MR/L020300). Corresponding author: JinJun Zhang, PhD, MD, Department of Genetics, UCL Institute of Ophthalmology, 11-43 Bath Street, London, EC1V 9EL, London, England. Email: [email protected] or [email protected] Crown Copyright Q 2019 Published by Elsevier Inc. on behalf of ASCRS and ESCRS. All rights reserved.

0886-3350/$ - see frontmatter https://doi.org/10.1016/j.jcrs.2019.04.034

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

nearly all children undergoing cataract surgery, opacification results from proliferation of these LECs into the posterior capsule. These lenticular epithelial cells proliferate posteriorly and form an opaque membrane on the posterior capsule over time.10–13 This condition is characterized by the occurrence of visual axis opacification, which leads to rapid secondary vision damage. Attempts at reducing the incidence of PCO have been devised and include both additional surgical techniques, modifications of IOL materials and design, and pharmacological intervention. Posterior capsulorhexis has been designed to deprive LECs of a scaffold on which to grow, and this has considerably reduced the incidence of PCO; however, this procedure demands a higher level of surgical skill and is associated with additional problems.6,14 IOLs with a square-edged profile on the posterior surface appear to merely delay PCO rather than eliminating the problem.15,16 Pharmacological intervention using cytotoxic agents to inhibit proliferation of LECs runs the risk for toxic effects on surrounding ocular tissues.17–19 Unfortunately, even these procedures seldom facilitate a permanently clear visual axis and the incidence of pediatric PCO remains at an unacceptably high level.12–18 Whereas adult PCO is treated with Nd:YAG laser capsulotomy, the procedure is difficult to perform in children and requires repeated surgical intervention, most probably because of the aggressive wound-healing response in childhood.17,18 At present, there is no cure for pediatric PCO and therefore an urgent need exists for the development of a routine and simple clinical procedure that can prevent this secondary complication of cataract surgery in children. Because LECs are responsible for the opacification, their removal at the time of surgery should reduce or remove the risk for proliferation of cells and pediatric PCO. This assumption is therefore the basis for our approach for prevention of the condition. Maintenance of cell volume is a fundamental physiological process and a requirement for cell survival.20–22 Thus, cell death can be induced by both hyperosmotic and hypoosmotic stress.23–25 Both of these stresses have been assessed in regard to their usefulness for removing LECs and thereby curbing PCO. Using LEC cell lines, capsulotomy specimens, and sealed capsular bags, several studies26–28 have suggested that lysis of all LECs after exposure to distilled water (hypoosmotic insult) occurs between 3 minutes and 6 minutes. In vivo studies in rabbits29 have shown that irrigation of sealed capsular bags with distilled water for 3 minutes was effective for PCO prevention. A similar reduction in the PCO score was reported when rabbit eyes were irrigated with distilled water for 2 minutes.30 Human clinical trials with hypoosmotic shock have shown variable outcomes. Irrigation of a sealed capsule configuration with distilled water for 3 minutes resulted in a significant reduction in PCO over a long-term follow up but did not eliminate its occurrence.31 On the other hand, an irrigation period of 2 minutes with distilled water did not show any significant difference between PCO rates

1481

in treated versus control eyes.32 The work of Duncan et al.33 suggests that the 2-minute irrigation period in the latter study might have been too short to inhibit PCO to any significant degree. Hyperosmolar solutions have also been assessed in vitro for their ability to remove LECs. Exposure of a human cell line to 3M NaCl for 2 minutes resulted in the complete elimination of all cells, whereas a similar exposure to distilled water was ineffective.33 When sealed human capsular bags were used, a 2-minute exposure to 3M NaCl removed all cells and again distilled water was ineffective, reducing cell coverage by only 20%.33 These investigators also showed that the hyperosmotic NaCl induced a much higher apoptosis response in LECs compared with distilled water. The aforementioned studies indicate that the irrigation time is an important factor for the complete removal of viable LECs from their basement membrane. The times given in those studies were determined in tissue samples from adults; however, in children, they are likely to be prolonged because of the robust nature of the cells. These times will be dictated by the effectiveness of the compensatory mechanisms that serve to restore normal cellular volume. In the case of hyperosmotic shocks, the rate of cell shrinkage can also be modulated by the strength of the solution used. However, with high-strength osmotic solutions (eg, 3M NaCl), there is an additional risk for damaging neighboring tissues should there be an accidental leakage from the sealed capsular bag. Inhibition of the compensatory mechanisms offers one possibility for shortening the irrigation times so that they can be incorporated during the surgical procedure for cataracts. To our knowledge, this potential has not been assessed previously. Most animal cells have evolved an important defensive mechanism known as cell volume regulation to protect themselves from excessive swelling or shrinkage, mainly caused by osmotic changes in their environment. Cell volume constancy requires the continued operation of cell volume regulation mechanisms, including ion transport across the cell membrane as well as the accumulation or disposal of organic osmolytes and metabolites. Many cell types respond to osmotic swelling or shrinkage by activating transporters or metabolic pathways that are latent in cells with normal volume. The consequent cell solute changes and the resulting water movement tend to correct the volume disturbance. The mechanisms are commonly known as regulatory volume decrease and regulatory volume increase (RVI). These two processes operate primarily by altering the intracellular ion concentrations through the use of ion transport mechanisms located on the surface of the cells. For example, under a hypertonic challenge, vertebrate cells initially shrink by osmotic water loss but subsequently regulate their volume by a net gain of KCl or NaCl via NaC-KC-2Cl cotransport (NKCC) and NaC-HC and Cl–HC03 exchangers and a concomitant uptake of cell water to restore normal cell volume.21,22,34 If the stresses persist or are too powerful, or if the cells fail to operate their defense (RVI) Volume 45 Issue 10 October 2019

1482

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

processes adequately, the cells continue shrinking and subsequently disconnect from the extracellular matrix (ECM), and after this, apoptosis is initiated.22,34–38 In this report, we present a new cell removal procedure that facilitates the shrinkage process following the hyperosmotic insult by blocking the compensatory self-defensive RVI response. Inhibition of the RVI process is expected to reduce the irrigation time required for complete removal of LECs and therefore shorten the extension of the surgical procedure. Furthermore, RVI inhibition would allow the use of lower-strength hyperosmotic solutions, reducing the risk for potential damage to adjacent tissues in case of leakage from the sealed capsular bag. MATERIALS AND METHODS The ease and degree of detachment of human LECs might be dependent on the nature of the underlying substrate. For this reason, preparations employing different substrates were studied. For cultured cells, primary human LECs were subcultured onto plastic coverslips and onto collagen-coated membrane inserts. To assess the effects of osmotic shock on LECs attached to their natural substrate, capsulorhexis biopsies obtained within 15 minutes of cataract surgery were used together with biopsies obtained from donor human eyes. In addition, LECs present in a human PCO model were also assessed. After exposure of the cells to a hypertonic insult in the presence or absence of an RVI inhibitor (120 mM furosemide), the subsequent changes in layer morphology were monitored by phase–contrast microscopy. The degree of cell detachment was assessed by quantifying the MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide reaction product with the level being proportional to the quantity of viable cells remaining attached. The PCO model was used to assess the effectiveness of the new cell removal technique and whether residual cells (if any) had the potential to repopulate the posterior capsule. Establishment of Human Lens Epithelial Cell Cultures Preparation of Primary Human Lens Epithelial Cells Human eyes

were obtained from the eye banks in Bristol (England) and Shanghai (China) with the consent of the donors’ next of kin. Some of these had already undergone corneal button removal for corneal transplantation. The eyes (donor age range 34 to 75 years; postmortem time %48 hours) were dissected immediately upon arrival in the laboratory. Human LECs were cultured using the technique described previously.37,38 Briefly, the anterior lens capsule was excised from the underlying fiber cell mass, and 6 to10 capsules were usually placed in a centrifuge tube containing 2 to 4 mL of 0.05% trypsin–EDTA (ethylenediaminetetraacetic acid) (Sigma-Aldrich Co. LLC). The capsules were incubated for 15 minutes at 37 C and centrifuged at 200  g for 3 minutes at room temperature. The trypsin was neutralized by the addition of 1 mL of Dulbecco’s modified essential medium supplemented with 10% fetal bovine serum (FBS), 2 mmol/L l-glutamine, 25 mmol/L HEPES (pH, 7.4), 10 U/mL penicillin, and 10 mg/mL streptomycin (DMEM-FBS) (Sigma-Aldrich Co. LLC), and the tubes were centrifuged as above. Supernatant was removed, and the capsules were transferred to T25 or T75 cell culture flasks (based on the numbers of capsules used) containing DMEM-FBS. They were incubated at 37 C in a humidified atmosphere of 95% air:5% CO2. Outgrowth of cells from the capsules was observed after 5 to 7 days, and thereafter, the fresh medium was replaced every 2 to 3 days. After reaching confluence, the monolayer was passaged by the addition of 0.05% trypsin–EDTA with cells detaching after 5 to 10 minutes at 37 C. The trypsin was neutralized with DMEMFBS and the cell suspension was removed and centrifuged at Volume 45 Issue 10 October 2019

200  g for 5 to 8 minutes at 9 C in a swing-out-bucket centrifuge. The supernatant was aspirated and the pellet resuspended in fresh DMEM-FBS and cultured either on collagen-coated membrane inserts (Corning Transwell-COL, 6.5 mm, Sigma-Aldrich Co. LLC) at a density of 1  105 cells/cm2 or on 13.0 mm culture coverslips (Nune Thermanox, Thermo Fisher Scientific, Inc.) at a density of 1  104 cells/cm2. The cells were maintained in culture with the medium being replaced every 2 days until they reached 90% to 100% confluence, whereupon they were considered ready for experimentation. Use of Continuous Curvilinear Capsulorhexis Biopsies To select LECs retaining their original cellular polarity and architecture, human lens capsule biopsies of continuous curvilinear capsulorhexis (LEC-CCC) samples were freshly obtained from cataract surgery (Department of Ophthalmology, Ruijin Hospital, Shanghai, China). The biopsy samples were collected from the operating theaters (stored in a balanced salt solution [Hanks’ Balanced Salt Solution, Sigma-Aldrich Co.]) and transferred to the laboratory within 10 to 15 minutes of surgery. In some of the quantification experiments, the LEC-CCC biopsies from donor eyes were used (24 hours postmortem). In brief, the intact crystalline lens–zonule–ciliary body complex was dissected from the globe and pinned with 6 to 8 entomological pins through the ciliary body to a soft silicone ring. The specimen was iridectomized, and then LEC-CCC biopsies of 8.0 mm diameter were obtained and stored in the balanced salt solution ready for the experiments. Establishment of Human Lens Organ Culture Posterior Capsule Opacification Model An in vitro PCO model, initially developed

by Liu et al.39 and modified by Cleary et al.40 was used. In brief, the intact crystalline lens–zonule–ciliary body complex was dissected from the globe in one piece and pinned with 6 to 8 entomological pins through the ciliary body to a soft silicone ring with an internal diameter of 12.7 mm. The specimen was iridectomized, the lens nucleus removed, and the cortex cleaned using a conventional manual surgical technique. The posterior and equatorial capsules were left intact as an envelope (bags), stored in a plastic culture dish, and covered with DMEM-FBS. The capsule was supported by the native zonules, and the entire capsular bag was visible, allowing observation of LEC growth. As expected, LECs grew and spread to cover the anterior and posterior capsules, replicating the formation of secondary cataracts after normal cataract surgery. The posterior capsule was covered by a confluent monolayer of LECs within a mean duration of 10.4 days G 1.4 (SD).40 An increase in capsular tension and capsular wrinkling occurred in the subsequent 4 to 5 days. Hyperosmotic Shock and Cell Detachment Assays

Treatment of Lens Epithelial Cells Grown on Coverslips Once the LECs had reached 80% to 90% confluence, the coverslips were rinsed in a balanced salt solution and then exposed to a hyperosmotic NaCl solution (400 mOsm/L) in the presence or absence of an NKCC inhibitor, furosemide (120 mM), for a period of 10 minutes. At the termination of the experiment, the coverslips were washed twice with a balanced salt solution medium. Phase– contrast microscopy was used to evaluate the morphological status of these cells. Treatment of Lens Epithelial Cells Grown on Collagen-Coated Membrane Inserts LECs on collagen-coated membranes were

exposed to varying concentrations of NaCl (170 to 2000 mM) in the presence or absence of 120 mM furosemide for 10 minutes, and the percentage of cells remaining on the inserts was determined by the MTT assay. Serum-free DMEM containing 1 mg/ mL MTT was added to the cells followed by an incubation of 4 hours at 37 C. The MTT solution was then aspirated and the blue formazan reaction product within the cells was extracted over 12 hours by the addition of 150 mL of dimethyl sulfoxide

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

(DMSO). The supernatant containing the blue formazan reaction product was then collected and transferred to a 96-well plate for measuring the absorbance on an ELISA (enzyme-linked immunosorbent assay) plate reader with test and reference wavelengths of 570 nm and 630 nm, respectively (OD570–OD630). Absorbance values for the controls were designated as 100%, and the cell viability levels calculated according to the manufacturer’s instructions. The result reported for each condition is the mean of 5 determinations. In a separate experiment, the integrity of the monolayer after exposure to 400 mOsm/L NaCl was assessed by continuously monitoring the transepithelial resistance (TER) for 60 minutes. Treatment of Continuous Curvilinear Capsulorhexis Biopsies

The LEC-CCC biopsies were washed in a balanced salt solution and exposed to the hypertonic NaCl solution in the presence or absence of 120 mM furosemide for 10 minutes. The capsules were then rinsed in a balanced salt solution and examined by phase-contrast microscopy. In other experiments, the capsules were removed from the test solutions at various time periods (2 to 60 minutes), quickly washed in a balanced salt solution, and examined by phase–contrast microscopy. Treatment of Lens Organ Posterior Capsule Opacification Culture Model Time Course of Cell Detachment The time course of cellular detachment in the PCO model was determined after exposure to a 400 mOsm/L NaCl hyperosmotic solution containing 120 mM furosemide. The hyperosmotic solution was removed at 2, 5, 10, 20, 30, and 60 minutes after exposure, and the capsular bags were processed for the MTT assay. The organ cultures were washed with a balanced salt solution and incubated with serumfree DMEM containing 1 mg/mL MTT for 4 hours at 37 C. The lens capsular bag was then carefully isolated from the native zonules and the MTT solution was aspirated. The capsular bags were then transferred to a multiwell plate and 150 mL DMSO added to extract the blue formazan reaction product from the LECs. The capsular bags were then removed, and the absorbance of the formazan product was measured using an ELISA plate reader as described earlier. Efficiency of the Cell Removal Technique Having determined the exposure time required for complete cell removal, further incubations were undertaken to confirm the efficiency of cell removal. Control organ cultures (n Z 10) were exposed to a balanced salt solution, while the experimental group (n Z 20) was treated with the hyperosmotic NaCl solution for a period determined to remove over 95% to 100% of cells (from above). Organ cultures were then processed for the MTT assay. Some of the preparations were examined (before the MTT assay) with an inverted microscope (IX71, Olympus Corp.) and photographed using a digital camera. To determine whether the remaining cellular population after treatment with furosemide (!5%) was capable of repopulating the capsular bag, a further experiment was undertaken with 16 organ cultures with full PCO development. Eight of these were designated as controls and the remaining 8 were treated to remove cells. Serum containing DMEM was then added to these bags, and all cultures were returned to the incubator for 14 days. After the incubation period, the MTT assay was again used to determine cell viability in each group. Measurement of Transepithelial Electrical Resistance Subcultured primary LECs on collagen-coated membrane inserts were allowed to reach confluence and then maintained for another 24 to 48 hours. All cultures were examined by inverted microscopy and only inserts that appeared adequate (with no obvious areas of missing cells) were used, amounting to approximately 75% of

1483

those surveyed. TER across the monolayer was monitored over a period of 60 minutes using STX-2 chopstick electrodes connected to an epithelial volt ohmmeter (EVOM, World Precision Instruments, LLC) when the cells were exposed to hyperosmotic NaCl (400 mOsm/L). Net TER values were calculated by subtracting the mean resistance of 10 empty inserts from the value recorded for each monolayer grown on the collagen-coated membrane inserts. Final resistance–area products (U.cm2) were obtained by multiplication with the effective growth area. Measurement of Solution Osmolarity The osmolarities of the media were checked with a vapor pressure osmometer (Wescor, Inc.). The normal growth medium was approximately 300 Osmol/kg H2O. Preparation of NaD-KD-2ClL Cotransporter Inhibitor The treatment solution was made by dissolving the furosemide powder in DMSO to make a stock solution of 100 mM. An appropriate volume of the stock solution was then added to a phosphate-buffered saline (control solution) or the hyperosmotic NaCl solution to obtain the final desired concentrations used in the experiments.

Statistical Analysis Data are expressed as means G standard error of the mean (SEM), resulting from 3 to 10 independent experiments performed in duplicate. Statistical analysis was performed using analysis of variance combined with the Student t test for all comparison pairs. A P value less than 0.05 was considered statistically significant unless otherwise stated.

RESULTS Cell Layer Morphology and Hyperosmotic Shock

LECs exposed to the hyperosmotic NaCl solution containing furosemide showed rapid cell shrinkage (!1 minute) and gradual detachment from their respective substrates or capsules. Figure 1, A, B, and C, show the typical response of LECs grown on coverslips to a hyperosmotic insult. Whereas control preparations displayed an intact monolayer (Figure 1, A), exposure to the hyperosmotic insult showed cell shrinkage, clustering, separation, and detachment from the underlying support structure (Figure 1, B). This resulted in almost complete cell removal, leaving behind a clear support medium (Figure 1, C). The same test was repeated on LEC-CCC biopsies removed during cataract surgery to evaluate the impact of the hypertonic insult on fresh in situ LECs (Figure 1, D, E, and F). The CCC biopsies, which were incubated in the control balanced salt solution, were characterized by a continuous intact epithelial sheet (Figure 1, D). Exposure to the hyperosmotic insult resulted in cell shrinkage with an enlargement of the extracellular space around these cells (Figure 1, E). Rinsing these preparations in the balanced salt solution resulted in extensive denudation of the previously confluent monolayer leaving behind some dead cells and their debris (Figure 1, F). The human lens organ culture PCO model was chosen to see whether this new osmotic shock approach would have sufficient efficacy to remove proliferating LECs, especially those germinating cells located in the equatorial zone, which was the suggested primary source of migrating cells leading to PCO. Figure 1, G, H, and I, show the morphological changes in the LEC monolayer in response to the Volume 45 Issue 10 October 2019

1484

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

Figure 1. Phase–contrast photographs of cell detachment caused by hyperosmotic shocks. A: Intact monolayer (control). B: Clustering and separation/detachment of the cells from the monolayer and coverslips after hyperosmotic shock. C: Substantially clear substrate after the cells have been washed away (original magnification 10). D: Control in which the LECs are tightly packed and appear as a confluent smooth sheet with a few groups of dead cells and cell debris. E: Treated LEC-CCC biopsy after the hyperosmotic shock, showing enlarged extracellular spaces around cells. F: No residual cells left after thorough washing (original magnification 20). G: Control, showing proliferating LECs on both the anterior and posterior capsule with discernable multilayers. The arrows point to CCC. H: Side view of organ cultures showing partial cellular denudement of both anterior and posterior capsules with some cells beginning to detach in the equatorial zone. The arrows indicate site of the CCC and the equatorial edge. I: Side view of the equatorial region, showing complete cell removal after the hyperosmotic treatment and washing (original magnification 4). (CCC Z continuous curvilinear capsulorhexis; LECs Z lens epithelial cells).

osmotic insult. The control preparations (Figure 1, G) clearly show the presence of proliferating LECs on both the anterior and posterior capsule with multilayers being discernable. Hyperosmotic insult resulted in partial cellular denudement of both anterior and posterior capsules with some cells beginning to detach in the equatorial zone (Figure 1, H), Rinsing these capsule bags with the balanced salt solution showed complete removal of LECs leaving behind a clear capsular structure (Figure 1, I).

Role of Regulatory Volume Increase in Osmotically Induced Detachment of Lens Epithelial Cells

The effect of osmolarity on the degree of detachment of LECs was studied in the presence and absence of the RVI inhibitor, furosemide. Primary LECs cultured on collagen-coated membrane inserts were exposed to various hypertonic NaCl solutions (170 to 2000 mM, corresponding to final osmolarities of 350 to 4000 mOsm/L) in the presence or absence of furosemide, and the proportion of cells remaining attached was quantified. With an operational RVI process in place (ie, in the absence of furosemide), hyperosmotic shocks resulted in LEC detachment but required very high levels of NaCl to elicit substantial cell loss (Figure 2). Thus, over a concentration range of 800 to 1200 mM NaCl, only 40% to 60% of LECs were removed. Although hyperosmotic shock alone can shrink and detach LECs, a very high level of hyperosmotic stress is required to completely remove these cells, and therefore, it is not suitable for clinical implementation. Volume 45 Issue 10 October 2019

However, the tolerance of these cells to the hyperosmotic NaCl was dramatically reduced after RVI was disabled (Figure 2). Nearly 100% of the cells were removed at approximately 40 0mM NaCl. The graphs clearly show a significant shift in the cell survival curve when the RVI process is disabled allowing the use of lower levels of hypertonic stress to induce LEC removal from the capsules. Figure 3 shows the role played by the RVI process in combating hyperosmotic insults. To illustrate the recovery mechanism, LEC-CCC biopsies were chosen because of their native cellular characteristics. The biopsies were exposed to a hyperosmotic NaCl solution (400 mOsm/L), and the cellular responses were photographed after 0, 2, 5, 30, and 60 minutes of exposure. The control biopsies, which were maintained in a balanced salt solution, showed a tightly packed and confluent cellular sheet (Figure 3, A). After 1 to 2 minutes of exposure to the hypertonic solution, the extracellular space around these cells was enlarged, which was indicative of cellular shrinkage (Figure 3, B). This space continued to enlarge for up to the 5 minutes of observation (Figure 3, C). This was then followed by a gradual reduction in the extracellular space with restoration of cell volume during between 30 and 60 minutes of exposure (Figure 3, D and E). To further demonstrate the RVI process in response to hyperosmotic shocks, changes in TER were examined. TER was measured using STX-2 chopstick electrodes connected to an epithelial volt ohmmeter and net TER values were obtained by subtracting the mean resistance of empty inserts. LEC-CCC preparations could not be used for this

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

Figure 2. Effect of furosemide on osmotically induced LEC detachment. Datapoints (Mean G SEM, 4 independent experiments) represent the percentage of cells remaining attached to the membrane insert after the given hyperosmotic shock. The right curve represents the response of LECs to hyperosmotic shock in the presence of an intact RVI process. LECs show considerable resistance to shrinkage and detachment at lower hyperosmotic insults and require much higher levels of shock (w2000 mM NaCl) for complete removal, a level that is not clinically viable. However, the presence of furosemide that compromises the RVI process, leads to efficient cell detachment at much lower hyperosmotic insults (left curve) (LEC Z lens epithelial cell; RVI Z regulatory volume increase).

1485

investigation because of the likelihood of edge leakage when these samples are clamped in an Ussing chamber. Instead, primary LECs grown on collagen-coated membrane inserts were used. The culture inserts were exposed to a hyperosmotic solution of NaCl (400 mOsm/L) for 60 minutes and the TER changes were monitored continuously. The range of the TER of the LEC monolayer was 187 to 302 U/cm2, with a mean value of 228.3 G 63.4 U/cm2 before the application of hyperosmotic shock. Exposure to the hyperosmotic solution resulted in a rapid drop in TER, and this is compatible with morphological shrinkage of LECs. By 5 minutes, the TER had dropped to values observed with empty inserts (25.41 G 7.61 U/cm2 versus 22.2 G 4.68 U/cm2 for the empty insert), suggesting that cellular junctions and architecture had been disturbed or completely disconnected. These findings were consistent with the morphological changes shown in Figure 3, B and C. Thirty minutes after the hyperosmotic challenge, TER gradually recovered and paralleled the morphological recovery shown in Figure 3, D and E, reaching a value of 144.3 G 42.1 U/cm2 after 60 minutes. The TER values in Figure 3, F, are expressed as changes in TER compared with those for time-matched controls, and they were calculated as the means G SEM of 4 independent experiments, each performed in triplicate.

Figure 3. Regulatory volume increase-dependent recovery after a hyperosmotic challenge. Continuous curvilinear capsulorhexis biopsies were exposed to hyperosmotic shock and morphological changes recorded with phase– contrast microscopy at various time intervals. A: Control biopsies showed the confluent and intact presence of the LEC layer. B: After 2 minutes of exposure to the hyperosmotic solution, the extracellular space around these cells enlarged, indicating cell shrinkage. C: Extracellular space continued to increase at the 5-minute observation point. D and E: These changes were followed by gradual contraction of the extracellular space as cells regained their volume over 30 to 60 minutes. All micrographs are original magnification 20. F: Changes in TER compared with those for time-matched controls. The effect of the hyperosmotic insult on the integrity of the cellular monolayer was also followed by monitoring TER on LECs grown on collagencoated membrane inserts over zero to 60 minutes. Data are given as the means G SEM of 5 independent experiments (LEC Z lens epithelial cell; TER Z transepithelial electrical resistance).

Volume 45 Issue 10 October 2019

1486

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

Figure 4. Time dependence of the cell removal procedure. LECs in the organ cultural posterior capsule opacification model were exposed to hyperosmotic NaCl solution and capsules were removed at various times to assess the quantity of cells remaining attached by the MTT assay. The datapoints (Mean G SEM of 3 experiments) were fitted by an exponential function of the type OD Z Aexp( kt), giving A Z 1.19 and k Z 0.065 with a half-life (ln2/k) as 10.7 minutes. Thus, virtually all LECs were detached within approximately 30 minutes (LECs Z lens epithelial cells).

Timecourse of Cell Detachment

Unlike LEC culture preparations and anterior capsule biopsies that are characterized by a single monolayer of cells, the PCO organ culture model also has multilayered regions of proliferating cells. The latter model is therefore more likely to be resistant to attempts at cell detachment, which presents an opportunity to experimentally map the time course of cell removal. After the organ cultures developed PCO characteristics, they were exposed to the hyperosmotic solution (400 mOsm/L) containing 120 mM furosemide for 2, 5, 10, 20, 30, 60, and 70 minutes. At each timepoint, the organ cultures were quickly rinsed with the balanced salt solution and processed for the MTT assay to quantify the cells remaining attached. Figure 4 shows the time course of cell loss after exposure to the hyperosmotic solution. The data

were best described by an exponential function of the type OD Z A  e kt, where A is the type OD at zero time, k is the decay constant, and t is the time (in minutes). The half-life of the decay process was about 10 minutes, and therefore, in this PCO model, nearly all LECs could be removed by about 30 minutes of exposure to the hypertonic solution. A more extensive analysis was performed to confirm the aforementioned results. Altogether, 30 organ cultures were prepared, which allow for full establishment of PCO. All these cultures showed the typical confluent monolayer or multilayer of proliferated LECs on both anterior and posterior capsules (Figure 5, A). Ten of these cultures were washed with the balanced salt solution and incubated with the balanced salt solution for 30 minutes at 37oC. After an additional wash in the balanced salt solution, they were processed for the MTT assay to determine the viability of the remaining cells. The experimental group comprised 20 cultures, and these were exposed to a hypertonic NaCl solution (400 mOsm/L) containing 120 mM furosemide and incubated for 30 minutes at 37oC. They were then rinsed in the balanced salt solution, photographed on an inverted microscope, and processed for the MTT assay. A representative phase–contrast photograph (Figure 5, B) shows the capsular bag to be clear with no evidence of residual LECs on the anterior or posterior capsules. MTT data (Figure 5, C) shows that the hyperosmotic treatment resulted in virtual removal of all LECs (P ! .001). The potential for repopulation after treatment (by residual LECs, if any) was also assessed. Altogether, 24 PCO model preparations were obtained, and 8 of these were designated as controls. The other 16 preparations were treated with furosemide for 30 minutes and washed with the balanced salt solution. Of these, 8 were processed for the MTT assay to determine the extent of cell removal. The remaining 8 preparations, together with the control samples, were then incubated for an additional 14 days Figure 5. Effect of a 30-minute exposure to hyperosmotic stress in the presence of furosemide on detachment of LECs in the posterior capsule opacification organ culture model. A: Phase–contrast photograph of the organ culture model before hyperosmotic shock showing confluent monolayered or multilayered proliferated LECs on the anterior and posterior capsules (arrows point to continuous curvilinear capsulorhexis). B: After exposure to the hyperosmotic insult for 30 minutes, the micrograph shows a clear capsular bag and no evidence of residual LECs on the anterior or posterior capsules. C: The MTT assay confirms the results of the phase–contrast micrographs in that the hyperosmotic insult removed all LECs from the lens bags (P ! .001) (controls, n Z 10 lenses; hyperosmotic treated, n Z 20 lenses) (LECs Z lens epithelial cells).

Volume 45 Issue 10 October 2019

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

Figure 6. Recovery potential of LECs after furosemide treatment. After furosemide treatment and washing, the capsular bags (with residual cells, if any) were incubated with serum containing Dulbecco’s modified essential medium for 14 days and the LECs were quantified with the MTT assay. There was no evidence that residual cells remaining after the furosemide treatment had repopulated the capsular surface (P ! .001).

with DMEM-FBS, with the medium being changed every 2 to 3 days. At the end of the incubation period, all preparations were processed with the MTT assay. The furosemide treatment removed virtually all LECs, and the subsequent 14-day incubation did not result in repopulation of the capsular bag (Figure 6). DISCUSSION PCO is caused by the rapid and aggressive proliferation and migration of residual LECs after cataract surgery and is more aggressive in children. Theoretically, because these residual cells are responsible for postsurgical opacification, their removal at the time of surgery should decrease or eliminate the risk for aggressive LEC proliferation and consequent development of PCO. In this study, a simple method was evaluated to facilitate the complete removal of LECs from the lens capsular bag by disabling the RVI mechanism that normally protects cells from hyperosmotic damage. The effects of hyperosmotic shock on cell shrinkage and detachment have been well studied and understood.34–36 The survival of most normal epithelial cells requires adhesion to their basement membranes (ECM). Loss of the cell– ECM contact can be elicited by cell shrinkage, which consequently triggers adherent cells to rearrange their focal adhesion contacts and their cytoskeletal attachments to those contacts, leading to apoptosis.40–42 It is well known that hyperosmotic stress leads to transient cell shrinkage and loss of adhesion to the basement membrane ECM.41–43 If the stress persists or is too powerful, or if the cells fail to operate their defense (RVI) processes adequately, continuous shrinkage occurs followed by disconnection from the ECM and initiation of apoptosis.34–36,41–43 Malek et al.42 demonstrated that bovine aortic endothelial cells clustered and detached from the monolayer after exposure to additions of mannitol (400 or 600 mOsm/L).42 Another study44 found that porcine pulmonary endothelial cells exposed to hyperosmotic media (500 to 700 mOsm/L) underwent clustering and detachment from the culture plates, displaying

1487

apoptotic morphology as shown by the binding of the DNA dye (Hoechst 33342). The present study has shown that LECs respond in a similar fashion to hyperosmotic stress. In addition, disabling the RVI mechanism increases the sensitivity to hyperosmotic stress such that low levels of hyperosmotic shock can cause rapid cell shrinkage and detachment from the lens capsule. Earlier studies35,42 have shown that the time taken to detach the majority of hyperosmotic shock-exposed cells was over 6 hours. This long duration is probably attributable to the intervention of RVI defense mechanisms.34 It is well understood that when a cell shrinks, the shrinkage activates transport pathways on their surface, which results in the net influx of Cl, Na, and K. These ions are brought into the cell primarily by the activation of Na/Cl or Na/K/Cl cotransporters and functionally coupled Na/H and Cl/HCO3 exchangers, leading to the inflow of water into the cells and restoration of cell volume.35,36 These adaptive responses serve to either restore the near-normal cell volume or promote the reinforcement of cell structure by cytoskeleton remodeling to survive osmotic stress. The inability of cells to regulate their volume in response to hyperosmotic shock results in cell shrinkage and in the cases where cells grow on a substrate, detachment from that substrate and finally death through the activation of death-signaling processes.41–43 For example, hyperosmotic shock in S49 neothymocytes that lack RVI mechanisms was sufficient in itself to cause cell shrinkage, detachment, and apoptotic intervention.43–44 In contrast, hyperosmotic shock did not cause persistent cell detachment and therefore apoptosis in other cell types (COS, HeLa, GH3) that possess RVI activity.41–44 In agreement with these studies, we found that when RVI mechanisms were sustained in human LECs, more than 50% of these cells survived a hyperosmotic shock of approximately 1000 mOsm/L for 10 minutes (Figure 2). However, the inhibition of their RVI mechanisms significantly sensitized these cells to hyperosmotic shocks and led to persistent cell shrinkage which subsequently resulted in their detachment, facilitating their easy removal. Two factors are important if the technique is to be transferred to a clinical setting. First, the exposure time to the hyperosmotic should be as short as possible (preferably ! 3 minutes) and yet allow complete detachment of residual LECs. With LECs grown on plastic–membrane inserts or with monolayers present on biopsied anterior capsules, cell removal after furosemide treatment occurred within the designated 10 minutes of examination. However, shrinkage of LECs was observable within the first 1 to 2 minutes, and further work is required to map cell removal profiles less than 10 minutes (Figure 3). Using the PCO model preparation, the time for complete removal of LECs was about 30 minutes. However, this model is the extreme case with the presence of multilayers of proliferating and attached cells that are especially difficult to remove. In pediatric surgery and in the absence of PCO, the time for complete removal of LECs is expected to be much shorter. The main purpose for using this PCO model was to assess the Volume 45 Issue 10 October 2019

1488

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

likelihood of repopulating the capsule after cell removal with the hyperosmotic treatment. In addition, consideration has to be given to the strength of the hyperosmotic shock to avoid collateral damage to adjacent tissues in the case of accidental leakage of the solution. Using an RVI inhibitor allows the strength of the hyperosmotic shock to be reduced considerably. Plasma osmolarities are in the range of 275 to 299 mOsm/L, and in the present study, a 400 mOsm/L solution with the RVI inhibitor was sufficient to remove all LECs. Factors dictating the time for complete removal of LECs are the efficacy of the RVI inhibitor, the concentration of the inhibitor, and the strength of the osmotic shock. Further studies are planned to evaluate a clinically useful RVI inhibitor, and to optimize the aforementioned relevant parameters so as to reduce the exposure period and thereby provide a viable protocol for clinical assessment. In summary, we have developed and reported a simple LEC cell removal procedure that rapidly detaches these cells from their ECM (the lens capsular bag), allowing their complete removal during the surgical procedure for cataract removal. This should decrease or eliminate the risk for aggressive cell proliferation and PCO development after pediatric cataract surgery.

WHAT WAS KNOWN  Pediatric posterior capsule opacification (PCO) is caused by the aggressive proliferation and migration of lens epithelial cells (LECs) that remain in the lens capsule after cataract surgery. This occurs in almost every case. So far, there is no cure for this condition, and therefore, there is an urgent need for a solution.

WHAT THIS PAPER ADDS  This new method removed human LECs from their extracellular matrix (the lens capsule) by interfering with their selfdefensive regulatory volume increase process while exposing them to hyperosmotic challenge, thus promoting cell detachment. This method provided a viable solution for pediatric PCO.

REFERENCES 1. World Health Organization. State of the World’s Sight: VISION 2020: The Right to Sight: 1999–2005. World Health Organization, 2005; Available at: http://www.who.int/iris/handle/10665/43300. Accessed June 13, 2019 2. Jensen AA, Basti S, Greenwald MJ, Mets MB. When may the posterior capsule be preserved in pediatric intraocular lens surgery? Ophthalmology 2002; 109:324–327 3. Trivedi RH, Wilson ME Jr. Posterior capsulotomy and anterior vitrectomy for the management of pediatric cataracts. In: Wilson ME Jr, Trivedi RH, Pandey SK, eds, Pediatric Cataract Surgery: Techniques, Complications, and Management. Philadelphia, PA, Williams & Wilkins, Lippincott, 2005; 83–92 4. Nishi O. Posterior capsule opacification. Part 1: Experimental investigations. J Cataract Refract Surg 1999; 25:106–117 5. Wormstone IM. Posterior capsule opacification: a cell biological perspective. Exp Eye Res 2002; 74:337–347 6. Leysen I, Coeckelbergh T, Gobin I, Smet H, Daniel Y, De Groot V, Tassignon MJ. Cumulative neodymium:YAG laser rates after bag-in-thelens and lens-in-the-bag intraocular lens implantation: comparative study. J Cataract Refract Surg 2006; 32:2085–2090

Volume 45 Issue 10 October 2019

7. Apple DJ, Solomon KD, Tetz MR, Assia EI, Holland EY, Legler UF, Tsai JC, Castaneda VE, Hoggatt JP, Kostick AM. Posterior capsule opacification. Surv Ophthalmol 1992; 37:73–116 8. Marcantonio JM, Vrensen GF. Cell biology of posterior capsule opacification. Eye (Lond) 1999; 13:484–488 9. Raj SM, Vasavada AR, Kaid JS, Vasavada VA, Vasavada VA. Post-operative capsular opacification. Nepal J Ophthalmol 2009; 1:43–59 10. Nibourg LM, Gelens E, Kuijer R, Hooymans JMM, van Kooten TG, Koopmans SA. Prevention of posterior capsular opacification. Exp Eye Res 2005; 136:100–115 11. Bertelmann E, Kojetinsky C. Posterior capsule opacification and anterior capsule opacification. Curr Opin Ophthalmol 2001; 12:35–40 12. Koch DD, Kohnen T. Retrospective comparison of techniques to prevent secondary cataract formation after posterior chamber intraocular lens implantation in infants and children. J Cataract Refract Surg 1997; 23:657– 663 13. Ram J, Brar GS, Kaushik S, Gupta A, Gupta A. Role of posterior capsulotomy with vitrectomy and intraocular lens design and material in reducing posterior capsule opacification after pediatric cataract surgery. J Cataract Refract Surg 2003; 29:1579–1584 14. De Groot V, Leysen I, Neuhann T, Gobin I, Tassignon MJ. One-year followup of bag-in-the-lens intraocular lens implantation in 60 eyes. J Cataract Refract Surg 2006; 32:1632–1637 15. Nagamoto T, Eguchi G. Effect of intraocular lens design on migration of lens epithelial cells onto the posterior capsule. J Cataract Refract Surg 1997; 23:866–872 16. Peng Q, Visessook N, Aple DJ, Pandey SK, Werner L, Escobar-Gomez M, Schoderbek R, Solomon KD, Guindi A. Surgical prevention of posterior capsule opacification. Part 3: Intraocular lens optic barrier effect as a second line of defense. J Cataract Refract Surg 2000; 26:198–213 17. Oliver M, Milstein A, Pollack A. Posterior chamber lens implantation in infants and juveniles. Eur J Implant Refract Surg 1990; 2:309–314 18. Pandey SK, Apple DJ, Werner L, Maloof AJ, Milverton EJ. Posterior capsule opacification: a review of the aetiopathogenesis, experimental and clinical studies and factors for prevention. Indian J Ophthalmol 2004; 52:99–112 19. James C, Collison DJ, Duncan G. Characterization and functional activity of thrombin receptors in the human lens. Invest Ophthalmol Vis Sci 2005; 46:925–932 20. Hoffmann EK, Simonsen LO. Membrane mechanisms in volume and pH regulation in vertebrate cells. Physiol Rev 1989; 69:315–382 21. Russell JM. Sodium-potassium-chloride cotransport. Physiol Rev 2000; 80:211–276
rijard B, Wu IH, Davis RJ. An osmosensing signal 22. Galcheva-Gargova Z1, De transduction pathway in mammalian cells. Science 1994; 265:806–808 23. Huang Z, Tunnacliffe A. Response of human cells to desiccation: comparison with hyperosmotic stress response. J Physiol 2004; 558:181–191 24. Jackle T, Hasel C, Melzner I, Bruderlein S, Jehle PM, Moller P. Sustained hypoosmotic stress induces cell death: apoptosis by defeat. Am J Physiol 2001; 281:C1716–C1726 €ltz D. Gadd45 proteins induce G2/M arrest and modulate 25. Mak SK, Ku apoptosis in kidney cells exposed to hyperosmotic stress. J Biol Chem 2004; 279:39075–39084 26. Crowston JG, Maloof A, Healey PR, Neilson G, Milverton EJ. Water modulation of lens epithelial cells during cataract surgery. J Cataract Refract Surg 2003; 29:2464–2465 27. Crowston JG, Healey PR, Hopley C, Neilson G, Milverton EJ, Maloof A. Water-mediated lysis of lens epithelial cells attached to lens capsule. J Cataract Refract Surg 2004; 30:1102–1106
zycki R, Kozlowski W, Kosatka M. Lens 28. Re˛ kas M, Klu
s A, Stankiewicz A, Ro epithelial cell damage after exposure to deionized water to prevent posterior capsule opacification after phacoemulsification: an in vitro study. Med Sci Monit 2010; 16:BR102–106 29. Fernandez V, Fragos MA, Billotte C, Lamar P, Orozco MA, Dubovy S, Willcox M, Parel JM. Efficacy of various drugs in the prevention of posterior capsule opacification: experimental study of rabbit eyes. J Cataract Refract Surg 2004; 30:2598–2605 30. Kim SY, Kim JH, Choi JS, Joo CK. Comparison of posterior capsule opacification in rabbits receiving either mitomycin-C or distilled water for sealed-capsule irrigation during cataract surgery. Clin Exp Ophthalmol 2007; 35:755–758 31. Re˛ kas M, Klu
s A, Kosatka M. Sealed-capsule irrigation with distilled deionized water to prevent posterior capsule opacificationdprospective, randomized clinical trial. Curr Eye Res 2013; 38:363–370 32. Rabsilber TM, Limberger IJ, Reuland AJ, Holzer MP, Auffarth GU. Longterm results of sealed capsule irrigation using distilled water to prevent posterior capsule opacification: a prospective clinical randomised trial. Br. J Ophthalmol 2007; 91:912–915

LABORATORY SCIENCE: REMOVAL OF LECS TO PREVENT PCO AFTER PEDIATRIC CATARACT SURGERY

33. Duncan G, Wang L, Neilson GJ, Wormstone IM. Lens cell survival after exposure to stress in the closed capsular bag. Invest Ophthalmol Vis Sci 2007; 48:2701–2707 34. Alvarez LJ1, Candia OA, Turner HC, Polikoff LA. Localization of a NaC-KC2Cl cotransporter in the rabbit lens. Exp Eye Res 2001; 73:669–680 35. Cammarata PR, Xu GT, Huang L, Zhou C, Martin M. Inducible expression of NaC/myo-inositol cotransporter mRNA in anterior epithelium of bovine lens: affiliation with hypertonicity and cell proliferation. Exp Eye Res 1997; 64:745–757 36. Frisch SM. Evidence for a function of death-receptor-related, deathdomain-containing proteins in anoikis. Curr Biol 1999; 23:1047–1049 €maa M, Martins LM, Downward J. Involvement of FADD and 37. Ryto caspase-8 signalling in detachment-induced apoptosis. Curr Biol 1999; 23:1043–1046 €maa M, Lehmann K, Downward J. Matrix detachment induces 38. Ryto caspase-dependent cytochrome c release from mitochondria: inhibition by PKB/Akt but not Raf signaling. Oncogene 2000; 14:4461–4468 39. Liu CS, Wormstone IM, Duncan G, Marcantonio JM, Webb SF, Davies PD. A study of human lens cell growth in vitro: a model for posterior capsule opacification. Invest Ophthalmol Vis Sci 1996; 37:906–914

1489

40. Cleary G, Spalton DJ, Zhang J-J, Marshal J. In vitro lens capsule model for investigation of posterior capsule opacification. J Cataract Refract Surg 2010; 36:1249–1252 €ussinger D. Role of p38MAPK in cell volume 41. vom Dahl S, Schliess F, Graf D, Ha regulation of perfused rat liver. Cell Physiol Biochem 2001; 11:285–294 42. Malek AM, Goss GG, Jiang L, Izumo S, Alper SL, Hsu CY. Mannitol at clinical concentrations activates multiple signaling pathways and induces apoptosis in endothelial cells. Stroke 1998; 29: 2631–2640 43. Yancey PH, Siebenaller JF. Co-evolution of proteins and solutions: protein adaptation versus cytoprotective micromolecules and their roles in marine organisms. J Exp Biol 2015; 218:1880–1896 44. Alfieri RR, Cavazzoni A, Petronini PG, Bonelli MA, Caccamo AE, Borghetti AF, Wheeler KP. Compatible osmolytes modulate the response of porcine endothelial cells to hypertonicity and protect them from apoptosis. J Physiol 2001; 549:499–508

Disclosures: None of the authors have any financial or proprietary interest in the materials or methods mentioned in the paper.

Volume 45 Issue 10 October 2019