Dr Maureen Boost, Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Hong Kong SAR, CHINA, E-mail: firstname.lastname@example.org
Purpose: The aim was to determine the cytotoxic effects of three multipurpose solutions (MPS) on human corneal epithelial cells (HCEC) and to assess the metabolic rates of recovering cells at different levels of cell membrane damage.
Method: The effects of one to 15 minutes exposure to multipurpose solutions containing polyquaternium-1 (MPS-A), polyaminopropyl biguanide (MPS-B) and polyhexanide (MPS-C) on HCEC were determined. Recovery rates at different levels of cell membrane damage were assessed after re-culture for two hours at 37°C. Cell viability and membrane integrity were assessed using Annexin V-FITC/7-AAD staining and flow cytometry. Effects of concentrations of 10 to 40 per cent multipurpose solutions on the metabolic rate of recovering HCEC were assessed using a Vybrant MTT cell proliferation assay kit.
Results: The highest percentage of late necrotic cells resulted after exposure to MPS-A compared with other solutions and the control (p < 0.001). The percentages of early necrotic cells after 10 and 15 minutes of soaking in MPS-B were significantly higher than the control and other multipurpose solutions (p < 0.001). Although MPS-C exposure also resulted in statistically significant higher percentages of early necrotic cells than the control (p < 0.005), these differences were small. No recovery was noted when HCEC treated with multipurpose solutions were re-cultured, with numbers of dead cells in MPS-B-treated cultures increasing fourfold. The MTT assay showed significant dose-response decreases of 500 nm absorbance for all MPS-treated cells. In 40 per cent MPS-A-treated HCEC, lack of activity indicated the cells were non-viable.
Conclusions: Multipurpose solutions induced varying levels of irreversible tissue sensitivity reactions, with MPS-A showing the greatest effects. The solutions damaged cell integrity and reduced metabolic rates suggesting delayed healing ability. The formulations of multipurpose solutions need to balance antimicrobial effectiveness with low cytotoxicity, which might not be currently possible to achieve. In light of our results, we suggest that contact lens wearers should be advised to rinse the soaked lenses with saline before lens insertion.
The corneal epithelium has an essential function, forming a defensive barrier to microbial infection between the external and internal ocular environments. Substances that interfere with normal corneal epithelial integrity and function might increase the likelihood of ocular infection. Contact lens wear has been associated with a variety of adverse ocular reactions,1–5 especially if they are not handled in accordance with practitioners' guidelines, including correct use of contact lens solutions.6,7 Contact lens solutions have an important role in safe contact lens wear, because they lubricate the lens surface, prevent lens dehydration, help remove deposits on the lenses and eliminate microbial development.8–10 The most popular products for disinfecting lenses are multipurpose solutions (MPS), which are designed for cleaning, disinfecting and storing contact lenses using a single formulation.8–10 These solutions contain disinfecting agents,8 including Polyquad (polyquaternium-1) or polyhexamethylene biguanide (PHMB), which have been shown to cause corneal staining,11,12 which is thought to be associated with cytotoxic effects.8–10 Although there is no direct evidence indicating that use of these MPS increases the risk of ocular infection, their ingredients might damage the ocular surface during wear and such damage could result in ocular inflammation.13,14 If the condition is not managed properly, inflammation might progress and the patient might need to cease contact lens wear.13–15 Therefore, MPS should maintain a balance between antimicrobial activity and biocompatibility, having a broad spectrum of antimicrobial activity and minimal toxicity.8,15
To determine adverse ocular reactions of MPS, many studies have adopted fluorescein examination for their assessment method,16–19 as this observation is routinely used to assess corneal damage/abrasions in contact lens practice. The results of these clinical studies vary considerably with most of the tested commercially available MPS being reported to cause corneal staining;9–12,16–19 however, most observations were considered clinically insignificant.12–15 Fluorescein assessment might not correlate well with cytotoxicity measured at a cellular level.15 Direct exposure to some MPS caused damage to porcine corneal epithelial cells as assessed using flow cytometry but this level of damage could not be identified using fluorescein examination.15 Therefore, it is possible that the effect of solution cytotoxicity is underestimated in clinical practice. More research is needed to determine the cytotoxicity of commercially available MPS. Studies at a cellular level are more sensitive and mechanism specific, allowing a better understanding of solution cytotoxicity.
Several in vitro studies have investigated the effects of contact lens solutions on immortalised human corneal epithelial cell (HCEC) lines and have focused on the level of membrane damage, with relatively less attention paid to cellular functions.8,20–26 New approaches have shown that MPS might reduce corneal epithelial cell turnover ability, impair corneal homeostasis, weaken barrier function and inhibit intercellular communication across gap junctions of corneal keratocytes.20–23,25 The fluorescein isocyanate (FITC) kit for flow cytometry, which can distinguish between early and late stages of apoptosis and necrosis, has not, to our knowledge, been used to investigate effects of cytotoxicity of MPS on human corneal cells.27 In addition, the threshold of cellular damage resulting in an inability to recover has not been determined.
The aims of the present study were to determine and compare the cytotoxic effects of three commercially available MPS on cultures of human corneal cells. The metabolic rates of recovering cells at different levels of cell membrane damage were also assessed.
MATERIALS AND METHODS
HCEC line ATCC CRL-11135 was seeded into 24-well cell culture plates, which had been pre-coated with 0.01 mg/ml fibronectin, 0.01 mg/ml bovine collagen type I and 0.01 mg/ml bovine serum albumin. The cells were cultured in keratinocyte serum-free medium (Gibco 17005-042; Life Technologies, Grand Island, NY, USA) supplemented with 0.05 mg/ml of bovine pituitary extract, 5 ηg/ml of epidermal growth factor, 500 ηg/ml of hydrocortisone and 0.005 mg/ml of human insulin at 37°C in a humidified, CO2-controlled (5%) incubator. All cell culture materials were purchased from Gibco/Invitrogen, Grand Island, NY, USA. Culture was continued until the formation of a healthy confluent monolayer of cells in the wells (Figure 1a). The cells appeared like normal corneal epithelial cells and expressed cornea-specific cytokeratin.
Cell integrity after MPS exposure and recovery
The culture medium in each well was removed and the cell layer washed with Dulbecco's phosphate-buffered saline (DPBS) (Sigma, St Louis, MO, USA). The epithelial culture was then exposed to DPBS (as a control) or one of the three MPS (Table 1) for periods of one, five, 10 and 15 minutes (16 samples in total). The MPS were the same brands previously tested against porcine eyes.15 A further four samples were exposed to DPBS (as a control) or MPS for 15 minutes and used to assess the ability of cell recovery. These samples were washed with DPBS after exposure and re-cultured with culture medium for two hours at 37°C before assessment. For assessment of cell damage, all samples were washed with DPBS and the cultures exposed to 0.25% trypsin/EDTA (Gibco/Invitrogen) for five minutes. The suspension of dissociated epithelial cells was carefully collected and centrifuged at 1500 rpm for five minutes at 4°C. The resulting cell pellet was re-suspended in 200 µl of DPBS. The suspension of approximately 1 x 106 cells was used for flow cytometry performed according to the standard procedure of the Annexin V-FITC/7-AAD kit (Beckman Coulter, Brea, CA, USA) using a Beckman Coulter flow cytometer.15 In brief, each cell suspension in DPBS was centrifuged at 1500 rpm for five minutes at 4°C. The cell pellet was re-suspended in 200 µl of binding buffer and 20 µl of Annexin V-FITC solution and 40 µl 7-AAD viability dye were added to each suspension. The suspension was kept on ice and incubated for 15 minutes in the dark. The suspension was then diluted with 800 µl binding buffer. A total of 1 x 105 cells from each suspension was assessed per treatment using flow cytometry. A computerised analysis system was used to determine the percentages of:
Table 1. Major ingredients in multipurpose contact lens solution tested in the present study (source of information: manufacturers' brochures or instruction leaflets)
Note: Coding of solutions is the same as used in the study on porcine eyes.15
1healthy cells (not stained with Annexin V-FITC or 7-ADD)
2apoptotic cells (stained with Annexin V-FITC only)
3early necrotic cells (stained with 7-ADD only)
4late necrotic cells (stained with both Annexin V-FITC and 7-ADD).
Metabolic rate of recovering cells after re-culture
Epithelial cultures were prepared as described above. A 96-well plate was used and each well for testing was seeded with 1 x 105 cells in 200 µl of medium. The cells were allowed to settle for 48 hours before testing in a 37°C incubator with 5% CO2. The culture medium was then removed and the cells washed with DPBS. Cells were exposed to 10, 20, 30 and 40% of DPBS or MPS for 12 hours and then re-cultured with culture medium for 96 hours (16 samples in total). The metabolic rate of recovering cells was assessed using a Vybrant MTT cell proliferation assay kit (V-13154) (Gibco/Invitrogen). According to the standard procedure, 100 µl of culture medium with 10% 12 mM of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was added to each well. The cells were incubated with MTT medium for four hours at 37°C with 5% CO2. The medium was removed and 100 µl of dimethylsulfoxide (DMSO) (Sigma) was added to each well to lyse the cells. The plate was scanned in the spectrometer at 570 nm with 120 seconds of medium shaking. The absorbance was recorded and the difference in absorbance analysed using statistical program Prism 2.0 (Graph Pad Software, San Diego, CA, USA).
For each experiment, 10 replicates were performed on the same batch of cells and the mean and standard deviation (SD) for each experimental condition were calculated. The data from the cellular analysis were verified as being normally distributed by one sample Kolmogorov–Smirnov D test before performance of MPS parametric tests for analysis. Repeated measures analysis of variance (ANOVA) was used to investigate the differences in the percentage of healthy cells, apoptotic cells and necrotic cells and the differences in the absorbance of MTT between the control group and each of the three treatment groups. A post-hoc test (Tukey–Kramer multiple comparisons test) was performed if a significant difference (p < 0.05) was detected in ANOVA.
Figure 1 shows the morphology of cultured human corneal cells after 15 minutes exposure in DPBS or MPS. Exposure to MPS-A showed a strong cytopathic effect and disruption of the monolayer (Figure 1b) and minor changes were observed in cells exposed to MPS-B (Figure 1c), but exposure to MPS-C resulted in no apparent change (Figure 1d) compared with the healthy control cells (Figure 1a). When epithelial cells were exposed to DPBS (control group) for up to 15 minutes, less than five per cent were in an apoptotic stage (stained with Annexin V-FITC), early necrotic stage (stained with 7-AAD) or late necrotic stage (stained with both), indicating that over 96 per cent of cells remained healthy. Exposure to MPS-A resulted in the highest percentage of late necrotic cells for each exposure time and these percentages were also significantly higher than those for the other two MPS tested (p < 0.001). The percentages of early necrotic cells after 10 and 15 minutes of soaking in MPS-B were significantly higher than those of the control and other MPS (p < 0.001). Table 2 shows a summary of the results obtained after various exposure times for each MPS. There was a significant correlation between the percentage of dead cells and the time of exposure in both MPS-A (r = 0.9993; p < 0.0001) and MPS-B (r = 0.9972; p < 0.0001). Although exposure to MPS-C resulted in a statistically significant increase of 7-AAD-stained cells compared with those observed for cells in DPBS (p < 0.005), these differences were small and clinically insignificant. When the cells treated for 15 minutes with MPS were re-cultured for two hours, a significant increase in dead cells was found for all MPS (p < 0.0001). After two hours of re-culture, the percentage of dead cells in cultures treated with MPS-B was fourfold higher than before re-culture. Figure 2 shows the percentage of healthy cells at the various exposure times for each MPS before and after two hours of re-culture.
Table 2. Percentages of cells displaying cytotoxic effects after exposure to undiluted test solutions using flow cytometry (mean ± SD; n = 10)
No staining (no evidence of cell damage)
Annexin V-FITC staining (apoptotic cells)
7-AAD staining (cells in early necrosis)
Annexin V-FITC + 7-AAD staining (cells in late necrosis)
DPBS: Dulbecco's phosphate-buffered saline (as control); MPS-A: multipurpose solution with polyquaternium-1; MPS-B: multipurpose solution with polyaminopropyl biguanide; MPS-C: multipurpose solution with polyhexanide.
Significantly lower than those in DPBS and MPS-C at each exposure time (p < 0.005).
Significantly higher than those in DPBS at each exposure time (p < 0.005).
Significantly higher than those in DPBS and MPS-C at each exposure time (p < 0.005).
The MTT assay showed a significant dose-response decrease of 500 nm absorbance for all MPS-treated cells. Figure 3 shows the absorbance at the various concentrations of each MPS after 12 hours of exposure and following 96 hours of re-culture. While the period of recovery did allow for metabolism to resurge in the cells (as can be seen by the 10-fold increase in absorbance), differences between the effects of the MPS remain clear. In the period of re-culture, all cells exposed to MPS had reduced metabolic rates compared with the control and MPS-C-treated cells deviated further from the control than in the initially exposed cultures. In the 40 per cent MPS-A treated group, no absorbance was noted, indicating that the treated cells were non-viable.
The present study showed that exposure to MPS led to cell membrane damage in the HCEC line. Similar cell lines to that in the current study have been used by several other researchers in attempts to determine cytotoxic effects (Table 3).20–23,25,26 Although originally designed to be used as an air-interface model, this cell line has been used as a monolayer. It has been suggested that HCEC in a monolayer might not be representative of normal epithelial cells and display such features as giant cells and enlarged vesicles; however, examination of our monolayers revealed typical healthy epithelial cells as shown in Figure 1a. Production of stratified layers of corneal cells requires the use of high levels of growth factor, about 10-fold more than for monolayers. The presence of this high concentration of growth factor might enhance cell recovery after exposure to toxic factors and disguise the toxic effects. Studies by Pauly and colleagues28 and Baudouin and colleagues29 have demonstrated that stratified epithelium and a monolayer gave similar results in investigations of cytotoxicity.
Table 3. Comparison of studies of cytotoxicity of multipurpose solutions (MPS)
ATP: adenosine triphosphate, B: ReNu Multiplus, C1: Complete Moisture Plus, C2: Complete with Aquify, dUTP: 2′-Deoxyuridine, 5′-Triphosphate, EGF: epidermal growth factor, HCEC: human corneal epithelial cell, HCGF: haemopoietic cell growth factor, HCLE: human corneal limbal epithelial cells, O1: Optifree Express, O2: Optifree Replenish, S: Solocare Plus with aqualube, TUNEL: terminal-deoxynuceotidyl transferase-mediated dUTP nick endlabelling
Immortalised HCLE with 0.2 ng/ml EGF for monolayer and 10 ng/ml EGF for multilayer
Microscopical observation (cells in monolayer)
10 minutes O2, B and C1 exposure—swelling of cells
10 minutes O1 exposure—immediate disruption of most cell junctions but cells recovered on re-culture
Flow cytometry analysis (live/dead assay)
10, 20 or 60 minutes O1 and 20 or 60 minutes O2 exposure—significant decrease in cell viability compared with B, C1 and control
Cell proliferation assay (cells in multilayer)
20 minutes O1, O2, B and C1 exposure—reduction in proliferation rate and cell densities at day 5. 10, 20 or 60 minutes exposure to all MPS—no significant change in cell morphology or decrease in cell viability
8 hour exposure O1 and B soaked Lotrafilcon A lens—decrease in cell viability; however, both C1 soaked lens—no effect
Flow cytometry analysis (for integrin expression and caspase activation)
24 hour exposure—more downregulation of integrin expression for all MPS than at 8 hour exposure. 24 hour exposure O1 soaked Lotrafilcon A lens—increase in caspase activation
The results are in accordance with our previous study of commercially available MPS, in which tissue sensitivity reactions were detected with Annexin V-FITC/7-AAD staining in porcine epithelial cells.15 The ability of this kit to distinguish between early and late necrosis27 allowed for differences between solutions, even those containing the same antimicrobial agents. These differences cannot be differentiated using simpler staining methods described in other studies.23,25 The current study showed that direct exposure to undiluted MPS-A for 10 minutes resulted in 30 per cent of HCEC displaying cell membrane damage (combination of early and late necrosis). Although the damage observed was less, approximately half of that in our previous report using porcine corneal cells, it was still significantly higher than the levels observed after exposure to the other two MPS and DPBS. When the exposure to MPS was increased to 15 minutes, MPS-B showed a similar level of cell membrane damage compared with MPS-A. Almost 60 per cent of damaged cells exposed to MPS-A were in the late necrotic stage, while less than 20 per cent of late necrotic cells were observed in cultures exposed to MPS-B. These findings agree with other reports,30,31 which suggested that MPS containing polyhexamethylene biguanide cause less cytotoxic effects than solutions with Polyquad. In contrast, a recent clinical study32 showed that MPS-B produced more corneal staining than either MPS-A or MPS-C. The results of the present study showed that more than 80 per cent of damaged cells in MPS-B were in the early necrosis stage; however, recovery of this damage did not occur after two hours of re-culture and when re-tested the cells were in the late stage of necrosis. This suggests that MPS-B has a slower cytotoxic effect but exposure to the solution leads to a similar final outcome to that of MPS-A. Oriowo31 suggested that the differences in cytotoxicity are due to the antimicrobial agents used in MPS. The effects on the cells observed in the present study involved mainly early and late necrotic changes, with few cells remaining in the apoptotic stage. This is in contrast to the findings of Tanti, Jones and Gorbet26 using a capsase assay, which demonstrated capsase activation and apoptosis, especially after exposure to Polyquad-based solution. Both studies showed that Polyquad-based MPS and MPS-B had stronger adverse effects on viability than MPS-C.
In the present study, we tested the effects of undiluted MPS, but as these had fairly limited effects up to a 10-minute exposure, it seemed unlikely that diluted solutions would cause damage. Hence, effects of undiluted MPS were examined for longer-term exposures and effects on recovery.
Investigation of the cytotoxic effects of other constituents of MPS revealed that they might also contribute to differences in effects among solutions;8 however, this study did not evaluate the effects of individual constituents, because contact lens users have no control over the constituents, but can choose the MPS that provides the most comfort.
Investigation of the ability of HCEC to recover after exposure to diluted MPS showed that both MPS-A and MPS-B inhibited cell metabolism as determined by MTT activity, even at concentrations as low as 20 per cent. At 40 per cent dilution, activity was severely affected after prior exposure to these solutions and some effects were noted for MPS-C. These results indicate that long-term exposure to diluted solutions, while not leading to immediate cell death, might interfere with cell metabolism causing changes over a longer time; however, in real life, exposure to a 40 per cent concentration for an extended period is unlikely. We chose a 12-hour exposure to diluted solutions to represent leaching from a contact lens during day wear. In Hong Kong, most people wear their contact lenses for at least 12 hours per day. Several dilutions were included for this phase of the study to represent different uptake/release patterns of various lens types and to take into account that the concentration in the eye over the period of wear will change. Other workers have shown damage to HCEC as a result of leaching of solutions to cells in contact with a pre-soaked contact lens.26 Washing the cells with DPBS in this study might not thoroughly remove all PHMB attached to the epithelial cell surface. This might also explain why MPS-C has relatively little effect on the cells. In contrast to MPS-A and MPS-B, MPS-C has hydroxypropyl methylcellulose as a lubricant in its formulation, which might rapidly bind to PHMB lowering its chemical availability and bioactivity.31 Lack of cytotoxicity of MPS-C was also noted in a study of effects of leaching MPS from lenses onto a HCEC monolayer.26 Thus, chemical variations that exist between the solutions could yield differential sensitivity reactions.
Our flow cytometry findings showed that exposure to undiluted MPS-A and MPS-B resulted in irreversible cytotoxic damage to HCEC. These damaged cells displaying necrotic changes had lost their mitotic ability. If exposure to MPS leads to a decrease in the cell proliferation rate, recovery of the cornea after damage might be adversely affected. In the present study, HCEC re-cultured for 96 hours following exposure to MPS-A and MPS-B had significantly reduced metabolic rates, indicating reduced cell proliferation compared with those exposed to MPS-C, although cells exposed to MPS-C also showed a significant reduction in cell growth compared with the control. This cytotoxic effect appeared to be dose related as exposure to 40 per cent MPS-A had no demonstrable effect on cell growth or viability. It can be argued that the experimental design does not reflect real-life conditions as blinking and dilution by tears will rapidly reduce the concentration of cytotoxic agents coming into contact with the cornea following lens insertion. If lenses were inserted directly from the MPS, there would be a short period of high concentration of the solution in the eye, although this would differ depending on the solution, lens type and patient insertion technique. Analysis of silicone hydrogel lenses presoaked with Polyquad and PHMB has indicated that these compounds bind to the lens surface or deposits on the lens and leach out slowly during wear.33,34 Willcox and colleagues34 found that Polyquad was released quicker than PHMB in lotrafilcon B lenses, suggesting that reflex blinking and lacrimation could wash out the remaining Polyquad more rapidly than PHMB-based MPS. This difference might explain why there are more reports of discomfort in patients using PHMB-soaked lenses than those using Polyquad formulations.11,17,34 Some practitioners recommend rinsing the lens with saline before insertion and this would reduce and dilute residual MPS on the lens. The present study showed that a dilution of 40 per cent resulted in no toxic events occurring in the cells.
In conclusion, the results of the present investigation revealed that MPS for soft contact lenses induced varying levels of irreversible cytotoxic effects, with MPS-A exposure leading to the most changes. The solutions damaged cell integrity and might affect wound healing ability. The results might explain why some contact lens wearers report discomfort and irritation with contact lenses, leading some users to cease lens wear completely. There is obviously a need to balance antimicrobial effectiveness with low cytotoxicity and it might not be possible to achieve both goals at this time. In light of these results, we would suggest that contact lens wearers should be advised to rinse the soaked lenses with saline before lens insertion.
GRANTS AND FINANCIAL SUPPORT
This work was financially supported by research grants G YD-48, U517, JBB7P and B-Q20J from The Hong Kong Polytechnic University.