Article Text

Download PDFPDF

Original article
Effects of glaucoma medications and preservatives on cultured human trabecular meshwork and non-pigmented ciliary epithelial cell lines
  1. David A Ammar,
  2. Malik Y Kahook
  1. Rocky Mountain Lions Eye Institute, Department of Ophthalmology, School of Medicine, University of Colorado Denver, Aurora, Colorado, USA
  1. Correspondence to Dr Malik Y Kahook, Rocky Mountain Lions Eye Institute, Department of Ophthalmology, University of Colorado Denver, 1675 Aurora Court, Mail Stop F-731, Aurora, CO 80045, USA; malik.kahook{at}


Aims We investigated the potential cytotoxicity of various topical ophthalmic glaucoma formulations containing different preservatives in cultured human trabecular meshwork (TM) and non-pigmented ciliary epithelial (NPCE) cell lines.

Methods We tested 0.004% travoprost preserved with either 0.015% benzalkonium chloride (BAK), sofZia or 0.001% Polyquad (PQ); and 0.005% latanoprost preserved with 0.020% BAK. We also tested a range of BAK concentrations in balanced salt solution (BSS). TM cells were treated for 10 min at 37°C with solutions diluted 1:10 to mimic the reduced penetration of topical preparations to the anterior chamber. Viability was determined by the uptake of the fluorescent vital dye calcein-AM (n=6).

Results BAK solutions (diluted 1:10) demonstrated a dose-dependent reduction in cell viability in both cell types (TM and NPCE). With a 1:10 dilution of 0.020% BAK, there were significantly more living NPCE cells (89±6%) than TM cells (57±6%; p<0.001). In TM cells, travoprost + BAK had statistically fewer live cells (83±5%) than both travoprost + sofZia (97±5%) and travoprost + PQ (97±6%; p<0.05). Compared with BSS-treated NPCE cells, travoprost had statistically fewer live cells (p<0.05) when preserved with BAK (85±16%), sofZia (91±6%) or PQ (94±2%).

Conclusions These results demonstrate that substitution of BAK from topical ophthalmic drugs results in greater viability of cultured TM cells, the cells involved in the conventional outflow pathway. Cultured NPCE, responsible for aqueous inflow, appear more resilient to BAK.

  • Benzalkonium chloride
  • ciliary epithelium
  • glaucoma
  • pharmaceutical preservatives
  • synthetic prostaglandins
  • trabecular meshwork
  • biochemistry
  • imaging, glaucoma
  • physiology
  • angiogenesis
  • optic nerve
  • angle
  • clinical trial
  • treatment medical
View Full Text

Statistics from


Topical medications to decrease intraocular pressure are the first line of treatment for glaucoma in the USA and across the world. One of the commonly prescribed classes of hypotensive agents are prostaglandin analogues (PGAs). PGAs act primarily by enhancing uveoscleral outflow of aqueous humour; however, PGAs also appear to act on the trabecular meshwork (TM) to facilitate outflow through the conventional outflow pathway as well.1–3 Unfortunately, studies have shown that the side effects of daily topical glaucoma therapy include alterations in tear film, corneal and conjunctival cell death, loss of goblet cells, and an increase in inflammatory cells and cytokines.4–8 These ocular surface changes have typically been blamed on the preservative commonly used in multi-dose bottles of topical medication, benzalkonium chloride (BAK). In addition, the report from the International Dry Eye Workshop identified BAK as an important extrinsic cause of imperfect ocular surface wetting, early tear film breakup and dry eye.9

It has long been known that BAK at antiseptic concentrations increases corneal permeability to hydrophilic agents.10 While this can potentially increase the delivery of topically applied drugs to the aqueous humour (AH) and ultimately the sites of outflow, increased corneal permeability would also increase delivery of BAK itself into the AH. Little is known about the effects of BAK on the TM endothelial cells that populate this conventional outflow pathway. The aim of this study was to compare the in vitro effects of two PGAs in four formulations preserved with either BAK (varying concentrations), the cationic polymer compound Polyquad (PQ; 0.001%), or the oxidising preservative sofZia. We assayed the effects of these agents diluted 1:10 on cultured human non-pigmented ciliary epithelial cells (responsible for AH production) and human TM cells (involved in AH outflow).

Material and methods

Cell culture

TM42, a primary TM cell line, was grown in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Carlsbad, California, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen), 50 U/ml penicillin and 50 U/ml streptomycin (Invitrogen). 59HCE-SV, an SV40 transformed cell line derived from human non-pigmented ciliary epithelial cells (NPCE), was grown in DMEM supplemented with 10% FBS, 50 mg/ml gentamicin sulphate (Sigma-Aldrich Corp., St Louis, Missouri, USA), 50 U/ml penicillin and 50 U/ml streptomycin. Both TM and NPCE cell lines were cultured in flasks and plates previously coated with 0.1% gelatine (Sigma-Aldrich) for 2 h at 37°C.

Test and control solutions

Commercially available topical ophthalmic preparations included 0.004% travoprost preserved with 0.015% BAK (Travatan; Alcon Laboratories, Inc., Fort Worth, Texas, USA), 0.004% travoprost preserved with sofZia (Travatan Z; Alcon), 0.004% travoprost preserved with 0.001% Polyquad (Travatan PQ; Alcon), and 0.005% latanoprost preserved with 0.020% BAK (Xalatan; Pfizer Inc., New York, New York, USA). We also tested a range of BAK solutions (0.001%, 0.005%, 0.010%, 0.015% and 0.020%) in balanced salt solution (BSS; Alcon). BSS contained the following: 6.5 g/l sodium chloride (NaCl), 0.75 g/l potassium chloride (KCl), 0.48 g/l calcium chloride dihydrate (CaCl2•2H2O), 0.3 g/l magnesium chloride hexahydrate (MgCl2•6H2O), 3.9 g/l sodium acetate trihydrate (C2H3NaO2•3H2O), 1.7 g/l sodium citrate dihydrate (C6H5Na3O7•2H2O), and sodium hydroxide and/or hydrochloric acid to adjust pH to approximately 7.5. For viability assays, the live control solution was BSS and the dead control solution was a fixative solution containing 70% methanol and 0.2% saponin in BSS.

Viability assays

The LIVE/DEAD Viability/Cytotoxicity Kit for mammalian cells (Invitrogen) was used to assay cell death immediately after exposure to test solutions. The kit contained stock solutions of ethidium homodimer (Eth-1) and Calcein-AM. In this vital stain system, the Calcein-AM ‘LIVE’ stain accumulates in healthy cells due to non-specific cellular esterase activity that cleaves the cell-permeable ‘AM’ moiety from the dye. The Eth-1 stain will fluorescently label nuclear DNA unless excluded by an intact plasma membrane. This system therefore measures both plasma membrane integrity as well as cellular enzyme activity.

Approximately 5×104 TM cells or 1.5×104 NPCE cells were each plated into wells of a 96-well tray (Cellstar black, Greiner Bio-One North America Inc., Monroe, North Carolina, USA) in the appropriate culture media. Cells were assayed upon reaching confluence, usually between 2 and 3 days post-plating. Culture media was removed by aspiration and replaced with 100 μl of test or control solutions diluted 1:10 in serum-free DMEM. After exposure at 37°C and 5% CO2 for 10 min, test and control dilutions were aspirated and replaced with a 100 μl solution of 2 μM Calcein-AM and 5 μM Eth-1 in Dulbecco's phosphate-buffered saline without calcium or magnesium (D-PBS; Invitrogen). D-PBS has an approximate pH of 7.4 and contains the following: 0.2 g/l potassium chloride (KCl), 0.2 g/l potassium phosphate monobasic (KH2PO4), 8 g/l sodium chloride (NaCl), 2.16 g/l sodium phosphate dibasic heptahydrate (Na2HPO4•7H2O). After 20 min, uptake of fluorescent dye was quantified in a Synergy 4 Multi-Mode Microplate Reader using the Gen5 Reader Control and Data Analysis Software using the band-pass filter setting (BioTek, Winooski, Vermont, USA). Live cells were quantified by determining the Calcein fluorescence (F528) collected at 528±20 nm from an excitation of 485±20 nm.

Data and statistical analysis

Fluorescent data were analysed as outlined in the manufacturer's instructions. Briefly, all F528 fluorescence was corrected by first subtracting the cross-over Eth-1 fluorescence, determined as the F528 fluorescence of dead cells incubated with 5 μM Eth-1 only. F528 fluorescence was then normalised to the corrected F528 fluorescence of the BSS-treated live controls (100% Live). Data were reported as the mean±SD from two independent experiments performed in triplicate (n=6). Mean values for each concentration were analysed by t test (Excel, Microsoft; Redmond, Washington, USA); the level of significance was set at p<0.05.


Toxicity of BAK alone in cultured human ocular cell lines

Following a 10 min exposure to test solutions diluted 1:10, cultured TM cells had no statistically significant decrease in viability between the 0.001% BAK (75±5%) and 0.015% BAK (67±8% live cells) exposures. However, TM cells did show a statistically significant decrease in cell viability when comparing 1:10 dilutions of 0.015% BAK (67±8%) and 0.020% BAK (57±6% live cells, p<0.05). Compared with BSS controls, cultured NPCE cells showed a slight but significant reduction (p<0.05) in the number of live cells with a 1:10 dilution of 0.001% BAK when assayed after a 10 min exposure (figure 1). However, there was no statistically significant decrease in viability between 1:10 dilutions of the 0.001% BAK (90±3%) and 0.020% BAK (89±6%) exposures. At all concentrations of BAK tested, there were statistically fewer live TM cells compared with NPCE cells (figure 1, *).

Figure 1

Per cent of live cultured trabecular meshwork (TM) and non-pigmented epithelial (NPCE) cells after a 10 min exposure to a range of concentrations of benzalkonium chloride (BAK). The number of live cells was normalised to the number of live cells in balance salt solution (BSS)-treated controls. Original concentrations of BAK shown on the y-axis were diluted 1:10 in serum-free media before exposure to cultured TM or NPCE cells. Data are reported as the mean±SD of six replicates. All concentrations of BAK in both cell types were significantly different (p<0.05) from their respective BSS-treated controls. *p<0.05 for TM versus NPCE.

Cell survival with PGA preserved with BAK

TM cells exposed to 1:10 dilutions of PGAs preserved with BAK had significantly higher numbers of live cells than their respective concentrations of BAK (figure 2, ‡). In TM cells, diluted travoprost with 0.015% BAK had 83±5% live TM cells while diluted 0.015% BAK resulted in 67±8% live cells (p<0.001); latanoprost with 0.020% BAK had 66±5% live cells while 0.020% BAK resulted in 57±6% live cells (p<0.05). In NPCE cells (figure 3, ‡), a 1:10 dilution of latanoprost with 0.020% BAK (96±3% live cells) performed significantly better than 0.020% BAK (89±6% live NPCE cells; p<0.05). However, a 1:10 dilution of travoprost with 0.015% BAK (85±16% live NPCE cells) performed similarly to diluted 0.015% BAK (89±12%) in the number of live NPCE cells. When compared against one another, we found no difference between the two PGA formulations with BAK in the number of live NPCE cells (figure 3, p=0.1). However, cultures exposed to travoprost with 0.015% BAK (diluted 1:10) had significantly more live TM cells (figure 2, 83±5%) than latanoprost with 0.020% BAK (66±5%, p<0.001).

Figure 2

Per cent of live cultured trabecular meshwork (TM) cells after a 10 min exposure to various commercially available prostaglandin analogues (diluted 1:10 in serum-free media). The number of live cells was normalised to the number of live cells in balance salt solution (BSS)-treated controls. Data are reported as the mean±SD of six replicates. *p<0.05 versus BSS control; †p<0.05 versus travoprost + Polyquad (PQ); ‡p<0.05 versus benzalkonium chloride (BAK) vehicle.

Figure 3

Per cent of live cultured non-pigmented epithelial cells (NPCE) after a 10 min exposure to various commercially available prostaglandin analogues (diluted 1:10 in serum-free media). The number of live cells was normalised to the number of live cells in balanced salt solution (BSS)-treated controls. Data are reported as the mean±SD of six replicates. *p<0.05 versus BSS control; ‡p<0.05 versus benzalkonium chloride (BAK) vehicle. PQ, Polyquad.

Cell survival with PGAs preserved with PQ or sofZia

For TM cells exposed to a 1:10 dilution of the PGA travoprost, replacement of BAK with PQ (97±6% live TM cells) or sofZia (97%±5% live TM cells) resulted in survival levels statistically indistinguishable from BSS controls (figure 3). Exposure of TM cells to travoprost preserved with PQ or sofZia resulted in a significant increase in TM cell survival (figure 2, †) compared with either travoprost with 0.015% BAK (83±5%, p<0.001) or latanoprost with 0.020% BAK (66±5%, p<0.001). For NPCE cells exposed to the PGA travoprost, replacement of BAK with PQ or sofZia had no significant effect on cell survival (figure 3, p>0.05).


BAK, a quaternary ammonium compound, is one of the earliest preservatives used for topical ophthalmic medications and is the most commonly used preservative in topical glaucoma medications today.11 BAK is an antiseptic with detergent properties that disrupts bacterial cell membranes. Another quaternary ammonium compound, polidronium chloride (polyquaternium-1, PQ) has been used by Alcon Laboratories Inc. as a preservative in artificial tears since 1987. As the name implies, PQ is a cationic polymer of many quaternary ammonium structures with a 27-fold higher molecular weight than BAK. PQ lacks a hydrophobic region and, as a result, has no surfactant properties. Due to these two chemical properties, PQ is unable to penetrate mammalian cells and cause cytotoxic effects.12 A third preservative, sofZia, causes oxidative damage and subsequent death to organisms that lack the enzymes cytotochrome oxidase or catalase. These enzymes are lacking in most species of bacteria but are present in mammalian cells.

We have found that cultured human TM and NPCE cell lines, which would both be in direct contact to the AH, have different sensitivities to concentrations of BAK. NPCE cells appear to have little sensitivity to BAK; TM cells in contrast show a significant cell death with a 10 min exposure with as little as 0.0001% BAK. These findings have great clinical relevance, since this amount of BAK represents ∼1/100th of the concentration of BAK used in many topical ophthalmic preparations. Other studies have shown that brief exposure of cultured TM cells to BAK can increase apoptotic cell markers and significantly decrease cell growth at levels 1/100th of that used as a preservative.13 14 We know of no studies that have quantified the amount of BAK that can penetrate into the anterior chamber; however, BAK increases corneal permeability to fluorescein,10 which has a similar molecular weight to the water-soluble BAK, and it is expected that BAK will also penetrate to the AH in a similar manner. Studies in rabbits and humans demonstrate a peak concentration of PGAs in AH to be less than 1/1000th of the concentration applied.15 16 However, since BAK is absorbed and accumulates into ocular tissues,17 the concentration of BAK in the AH may be much higher than the PGA levels.

In conclusion, our findings indicate that replacement of BAK with other preservatives (PQ, sofZia) could potentially be beneficial for TM cell viability. The deleterious effect of BAK on TM cells may be an underappreciated concern. Since the number of live TM cells within the meshwork was found to be statistically lower in patients with primary open-angle glaucoma,18 maintaining TM cell number should be as important a metric of disease treatment as intraocular pressure. Our brief exposure model is a single step in understanding the role of glaucoma therapy in TM cell physiology and further in vivo studies are needed prior to making any clinically relevant conclusions based on our findings.


The authors would like to thank Dr Doug Rhee (Massachusetts Eye and Ear Infirmary, Boston, Massachusetts, USA) for the kind gift of the TM42 cell line and Dr Miguel Coca-Prados (Yale Medical School, New Haven, Connecticut, USA) for the kind gift of the NPCE cell line.


View Abstract


  • Funding The current study was funded by Alcon.

  • Competing interests MYK has received research support from Alcon, Genentech, The American Glaucoma Society, Pfizer, Allergan and Merck. DAA has received research support from Alcon.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • At a glance
    Harminder S Dua Arun D Singh