December 2012
Volume 53, Issue 13
Free
Glaucoma  |   December 2012
In Vitro and In Vivo Comparative Toxicological Study of a New Preservative-Free Latanoprost Formulation
Author Affiliations & Notes
  • Aude Pauly
    From the Insistut National de la Santé et de la Recherche Médicale, Paris, France; the
  • Christophe Roubeix
    From the Insistut National de la Santé et de la Recherche Médicale, Paris, France; the
  • Hong Liang
    From the Insistut National de la Santé et de la Recherche Médicale, Paris, France; the
    Centre Hospitalier National d'Ophtalmologie des Quinze-Vingts, Paris, France; the
  • Françoise Brignole-Baudouin
    From the Insistut National de la Santé et de la Recherche Médicale, Paris, France; the
    Faculté des Sciences Pharmaceutiques et Biologiques, Université Sorbonne Paris Cité, Paris, France; the
  • Christophe Baudouin
    From the Insistut National de la Santé et de la Recherche Médicale, Paris, France; the
    Université Versailles Saint-Quentin-en-Yvelines, Versailles, France; and
    Assistance Publique - Hôpitaux de Paris Hôpital Ambroise Paré, Service d'Ophtalmologie, Paris, France.
Investigative Ophthalmology & Visual Science December 2012, Vol.53, 8172-8180. doi:10.1167/iovs.12-10766
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Aude Pauly, Christophe Roubeix, Hong Liang, Françoise Brignole-Baudouin, Christophe Baudouin; In Vitro and In Vivo Comparative Toxicological Study of a New Preservative-Free Latanoprost Formulation. Invest. Ophthalmol. Vis. Sci. 2012;53(13):8172-8180. doi: 10.1167/iovs.12-10766.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To compare in vitro, on the human reconstituted corneal epithelial SkinEthics model, and in vivo, using an acute rabbit toxicological model, the effects of a benzalkonium chloride (BAK)-preserved solution of latanoprost and a preservative-free (PF) latanoprost solution.

Methods.: In vitro, the three-dimensional (3D) reconstituted human corneal epithelia (HCE) were treated with PBS, BAK-latanoprost, PF-latanoprost, or 0.02% BAK for 24 hours followed or not followed by a 24 hour post incubation recovery period. Cellular viability was evaluated using the 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) test at 24 hours and the apoptotic cells were counted using TUNEL labeling on frozen sections at 24 hours and 24 hours plus 24 hours. In vivo, rabbits received 50 μL of the same solutions, which were applied at 5 minute intervals a total of 15 times. Ocular surface toxicity was investigated using slit lamp biomicroscopy examination, conjunctival impression cytology (CIC), and corneal in vivo confocal microscopy (IVCM). Standard immunohistology also assessed inflammatory CD45-positive cells.

Results.: In vitro, BAK-latanoprost and 0.02% BAK induced significant apoptosis in the apical layers that correlated with the significant decrease of cell viability as assessed by the MTT test. PF-latanoprost slightly decreased cell viability and few apoptotic cells were found in the superficial layers, without reaching statistical significance compared with PBS. In vivo, clinical observation and IVCM images showed the lowest ocular surface toxicity with PBS and PF-latanoprost, while BAK-latanoprost and BAK induced abnormal corneoconjunctival aspects. PF-latanoprost showed the lowest CIC score, close to the PBS score and induced fewer CD45-positive cells in both the limbus and the conjunctiva compared with BAK and latanoprost, as assessed by immunohistology.

Conclusions.: We confirm that rabbit corneoconjunctival surfaces presented better tolerance when treated with PF-latanoprost compared with the standard BAK-latanoprost preparation or the BAK solution.

Introduction
Prostaglandin (PG) analogs have become first line treatments among glaucoma medications. Currently, five different PG analogs, isopropyl unoprostone, latanoprost, travoprost, bimatoprost, and tafluprost, are used for the treatment of glaucoma. These PGs are superior to beta-adrenoceptor antagonists in terms of lowering IOP, 1,2 and they induce no severe side effects during long term clinical use. 3 Among these PGs, latanoprost presents a highly advantageous balance in terms of IOP lowering efficacy and tolerance 2 and is still the mostly frequently used PG analog worldwide. However, a commercial solution of latanoprost is proposed in a preserved formulation, which raises a number of issues, especially in patients with an abnormal or sensitive ocular surface. Benzalkonium chloride (BAK) is the most commonly used preservative in eye drops. Indeed, it has already been shown to exhibit toxic and inflammatory effects in clinical, 4 in vivo, 5 and in vitro studies. 6 In cell culture, BAK induced oxidative effects, including mitochondrial activity and glutathione injury as well as caspase-dependent and -independent apoptosis counteracted by an autophagic process. 3 Recently, using in vitro 7 and in vivo approaches, 8 it has been shown that BAK induced corneal neurotoxicity. Moreover, chronic use of a preservative is responsible for apoptosis of conjunctival cells and conjunctival inflammation that have demonstrated negative effects (e.g., on glaucoma surgery efficacy). 9,10  
New preservative-free eye drops have therefore been developed. They have consistently demonstrated their good tolerance in vitro, 11 ex vivo, and in patients, who disclosed significantly fewer ocular symptoms and signs of irritation, such as pain, discomfort, or dry eye sensation. 12 Single-dose units, such as carteolol, 13 timolol, 14 or tafluprost, as well as Travatan Z, in which BAK has been replaced with the Sofzia self-preserved ionic buffer, are examples of new PG formulations without BAK or without preservative. 
We have developed new in vitro and in vivo tools for the assessment of tolerability and toxicity in a three-dimensional (3D) reconstituted corneal epithelial model, 15 and in an experimental model of acute toxicity in rabbits. 16 Our 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) test procedure, combined with immunohistologic analyses of human 3D tissue showed increased sensitivity levels and detected slight damage even in the most superficial layers. In vivo confocal microscopy (IVCM) and conjunctival impression cytology (CIC) are significant in vivo and ex vivo tools that are reliable for the observation, at a cellular level, of toxic and/or immunoallergic reactions on the ocular surface. They can analyze inflammation and apoptosis in the conjunctival epithelium without the need for sacrificing the test animal. In this study, we combined these two in vitro and in vivo models to evaluate the toxicological profile of PF-latanoprost, a new preservative-free (PF) latanoprost ophthalmic formulation, under clinical development, compared with commercially available BAK-latanoprost and 0.02% BAK, which is the concentration of BAK used in the commercial solution of BAK-latanoprost. 
Materials and Methods
Tissue Model and Treatments
The 3D-human corneal epithelial (HCE) model was supplied by SkinEthic Laboratories (Nice, France). It consists of immortalized HCE cells vertically grown on an inert permeable polycarbonate filter measuring 0.5 cm2 and cultivated for 5 days at the air–liquid interface in a supplemented chemically defined medium (modified MCDB 153; SkinEthic Laboratories). At reception, the 3D-HCEs were equilibrated for 4 hours in the defined maintenance medium supplied, before experiments were done. Thirty microliters of PBS, BAK 0.02%, BAK-latanoprost (Xalatan; Pfizer, New York, NY), and PF-latanoprost (Monoprost; Laboratoires Théa, Clermont-Ferrand, France) were applied on the apical surface of 3D-HCEs for 24 hours and 24 hours followed by 24 hour additional recovery time (Table). The post incubation time was chosen to allow the cells to recover if toxic effects were still reversible. This treatment procedure was expected to model the defense capacity of the tissue better. The apical surface was rinsed twice with 150 μL of PBS after treatment. Two series of 3D-HCEs were subjected to MTT testing at 24 hours and one series was used for inclusion in optimal cutting temperature medium (OCT) followed by immunofluorescence analyses of apoptosis on frozen sections at 24 hours and 24 hours plus 24 hours. 
Table. 
 
Test Items and Controls Used in This Study
Table. 
 
Test Items and Controls Used in This Study
Name Active Ingredient Batch Number Preservation System
PBS None Not applicable None
0.02% BAK solution - dilution of 0.1% BAK in PBS None FAB 07/05/2009 BAK
Monoprost eye drops Latanoprost 0.005% S879 None
Xalatan eye drops Latanoprost 0.005% R08267 0.02% BAK
MTT Test
Cellular viability was evaluated in duplicate at 24 hours with the MTT assay 1719 we had previously modified to improve the sensitivity of the test. 15 The 3D-HCEs were transferred to 24-well plates containing 300 μL of MTT solution diluted at 0.5 mg/mL in the culture medium, and 300 μL of the same MTT solution was applied on the apical surface of the 3D-HCEs. Reconstituted tissues were incubated for 3 hours at 37°C. Then they were transferred to 24-well plates containing 750-μL isopropanol, and 750-μL isopropanol was added to the apical surface of the 3D-HCEs. The plates were agitated for 2 hours at room temperature before reading the optical density (OD) at 570 nm versus OD 690 nm. The results were expressed as percentages of cell viability (mean ± SD) compared with the negative control (PBS). 
Animals and Eye Drop Treatments
Twenty four adult male New Zealand albino rabbits weighing 2.5 to 3.0 kg were treated according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Before all experiments, the rabbits were all anesthetized by subcutaneous injection of a 1:5 mixture of 100 mg/mL xylazine (Bayer, Puteau, France) and 100 mg/mL ketamine hydrogen chloride (Imalgène 500; Merial, Lyon, France). Fifty microliters of the test solutions were instilled in both eyes of the rabbits using a micropipette at 5 minute intervals a total of 15 times 16 (Table). The animals were divided into four groups of six rabbits each: five rabbits were used for clinical and IVCM observations at 4 hours (H4) and days (D1) after the administration of the treatment; and the remaining rabbit was killed for immunohistologic procedures at D1. H4 and D1 corresponded to the time course of inflammatory acute phase; H4 showing a significant inflammatory cell infiltration as hyperreflective points observed using HRT and D1 point corresponded to the time we observed the highest rate of inflammatory CD45-positive cells. 5,20  
Clinical Findings and Draize Test
The time was noted when obvious redness occurred on the ocular surface, calculated from the first instillation. The eyes were examined by slit lamp biomicroscopy and were scored according to a modified Draize test. 20  
IVCM Evaluation
The Heidelberg Retina Tomograph II/Rostock Cornea Module (Heidelberg Engineering GmbH, Heidelberg, Germany) laser scanning IVCM was used to examine the rabbits. IVCM scores were used to evaluate ocular surface toxicity profiles in four histologic zones. 20  
Cresyl Violet Staining in CIC
CIC specimens were collected as described in previous studies: two nitrocellulose membranes, 0.2-μL porosity (Millipore, Bedford, MA), were applied to the superior bulbar conjunctiva for the further cresyl violet cytology (1%, number 5235; Merck, Fontenay-sous-Bois, France) procedure. We evaluated cellular damage and inflammatory cell infiltration according to a published CIC score system. 21  
Immunofluorescence Analysis on Frozen Sections of 3D Epithelia and Rabbit Eyes
The 3D construct was transferred into a Petri dish containing 500 μL of PBS and was cut into two pieces under the binocular microscope (Nikon SMZ645; Nikon Instruments, Champigny sur Marne, France). Both pieces of tissue were fixed and embedded in OCT (Tissue-Tek; Miles Inc., Elkhart, IN) and frozen at −80°C. Vertical sections (10-μm thick) were cut with a cryotome (Leica CM 3050s; Leica Microsystems AG, Wetzlar, Germany), and then subjected to TUNEL for the detection of apoptotic cells (Roche Diagnostics, Meylan, France). 
Cryosections of enucleated eyes at D1 were prepared and incubated with anti-CD45 (1:50, CBL1412; Cymbus Biotechnology, Chandlers Ford, UK) to detect inflammatory cell infiltration. Apoptotic and inflammatory cells were counted under a fluorescent microscope (Olympus BX-UCB; Olympus, Melville, NY) equipped with a DP70 Olympus digital camera (Olympus). The results were expressed as the mean number of cells per field under the ×400 magnification and mean number of cells per square millimeter, respectively. 
Statistical Analysis
Statistical comparisons of control and treatment groups were performed using 2-way ANOVA, followed, if necessary, by multiple pairwise comparisons using Bonferroni adjustment for MTT data and quantification of TUNEL-positive cells and using the Fisher method for pair-wise comparisons for all in vivo data (Statview V; SAS Institute Inc., Cary, NC). 
Results
Modified MTT Test
After 24 hours of treatment, BAK and BAK-latanoprost induced a significant 48% and 45% decrease in cellular viability, respectively, as assessed by the MTT (P < 0.05 compared with control) (Fig. 1). PF-latanoprost slightly reduced the cell viability to 83%, without reaching significance compared with control. The PF-latanoprost–induced level of viability was significantly higher compared with those induced by BAK and BAK-latanoprost treatments (P < 0.05). 
Figure 1. 
 
Cellular viability of 3D-HCEs cultures for 24 hours in PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Figure 1. 
 
Cellular viability of 3D-HCEs cultures for 24 hours in PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Immunofluorescence Analysis and Quantification on Frozen Sections
Rare, homogenously disseminated, TUNEL-positive cells were found in the control 3D-HCE cultures at 24 hours (Fig. 2A) and at 24 hours plus 24 hours (Fig. 2B). BAK-containing solutions significantly increased the number of TUNEL-positive cells, with 44 cells per field and 43 cells per field found at 24 hours (Figs. 2A, 3) and 61 cells per field and 68 cells per field (Figs. 2B, 3) counted at 24 hours plus 24 hours after 0.02% BAK and BAK-latanoprost treatments, respectively. At 24 hours PF-latanoprost induced an effect that was comparable to that of PBS (Figs. 2A, 3). After the post incubation period, it slightly increased the number of apoptotic cells to 27 cells per field (Figs. 2B, 3), without reaching significance compared with control. Moreover, these effects were considered significantly different from those induced by 0.02% BAK and BAK-latanoprost (P < 0.05). 
Figure 2. 
 
Immunolocalization of TUNEL-positive cells on 3D-HCE cultures treated with (1) PBS, (2) PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost for (A) 24 hours and (B) 24 hours followed by a 24 hour post incubation period of time (×200 magnification).
Figure 2. 
 
Immunolocalization of TUNEL-positive cells on 3D-HCE cultures treated with (1) PBS, (2) PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost for (A) 24 hours and (B) 24 hours followed by a 24 hour post incubation period of time (×200 magnification).
Figure 3. 
 
Quantification of TUNEL-positive cells after 24 hour of treatment with PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost, followed or not by a 24 hour post incubation period of time.
Figure 3. 
 
Quantification of TUNEL-positive cells after 24 hour of treatment with PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost, followed or not by a 24 hour post incubation period of time.
Animal Clinical Findings
The PBS group (Fig. 4A) presented the same aspects as found in normal rabbit eyes without obvious abnormality. PF-latanoprost induced diffuse hyperemia in rabbits at H4 (Fig. 4B). Nevertheless, these clinical signs were limited compared with those observed after treatment with 0.02% BAK (Fig. 4C) and BAK-latanoprost (Fig. 4D). In these groups, diffuse hyperemia, chemosis, and secretions on the conjunctiva were observed, with conjunctival redness appearing very quickly after the first instillation. PF-latanoprost presented a slight and late redness. 
Figure 4. 
 
Photographs of rabbit eyes 4 hours after the 15 instillations of (A) PBS, (B) PF-latanoprost, (C) 0.02% BAK, and (D) BAK-latanoprost.
Figure 4. 
 
Photographs of rabbit eyes 4 hours after the 15 instillations of (A) PBS, (B) PF-latanoprost, (C) 0.02% BAK, and (D) BAK-latanoprost.
The PF-latanoprost group showed no statistical difference in ocular Draize scores compared with the PBS group (Fig. 5). At H4, 0.02% BAK and BAK-latanoprost presented higher Draize scores than PBS (P < 0.0001 for both) and PF-latanoprost (P = 0.0007 and P = 0.0008, respectively). At D1, the BAK-latanoprost group returned to a normal aspect, without statistical differences compared with the PBS or PF-latanoprost groups. However, ocular changes were still observed at D1 for BAK with higher scores than for PBS (P = 0.0257) and PF-latanoprost (P = 0.0350). 
Figure 5. 
 
Draize test scores at H4 and D1 after instillations of PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost, for 24 hours.
Figure 5. 
 
Draize test scores at H4 and D1 after instillations of PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost, for 24 hours.
IVCM Images and IVCM Score
Superficial Epithelium.
At H4 after instillation, PBS- and PF-latanoprost–instilled rabbits presented almost normal aspects of the corneal epithelium, with regular polygonal mosaic appearance, brightly reflective nuclei and no obvious desquamation or swelling (Fig. 6A). Preserved latanoprost and BAK solutions induced various pathological aspects of the corneal epithelium including partial desquamation of epithelial cells, irregular cell shapes, anisocytosis and loss of cell borders, abnormal reflectivity patterns, swollen cells, and inflammatory infiltration (Fig. 6A). 
Figure 6. 
 
In vivo confocal images of rabbit corneal superficial epithelium (Raw [A]), basal epithelium (Raw [B]), anterior stroma (Raw [C]), limbus (Raw [D]), and conjunctival substancia propria (Raw [E]) after instillations of (1) PBS, (2) preservative-free PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost at H4. Photograph size: 400 μm by 400 μm.
Figure 6. 
 
In vivo confocal images of rabbit corneal superficial epithelium (Raw [A]), basal epithelium (Raw [B]), anterior stroma (Raw [C]), limbus (Raw [D]), and conjunctival substancia propria (Raw [E]) after instillations of (1) PBS, (2) preservative-free PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost at H4. Photograph size: 400 μm by 400 μm.
Basal Epithelium.
PBS (Fig. 6B) did not induce any inflammation, while PF-latanoprost induced a slight infiltration of cells at H4 only with 23 ± 4.4 cells/mm2 (Fig. 6B), which was not considered significant compared with PBS. At H4, BAK 0.02% (Fig. 6B) induced inflammation with 79 ± 7 cells/mm2 and BAK-latanoprost (Fig. 6B) induced an increased inflammatory cell infiltration at significant levels (142 ± 28 cells/mm2; P = 0.0002 compared with PBS and P = 0.0004 compared with PF-latanoprost). 
Anterior Stroma.
We observed a slight inflammatory infiltration and disorganization only after instillations of BAK (Fig. 6C). No abnormalities of the posterior stroma or endothelium were observed with any treatment (data not shown). 
Limbus.
No obvious responses were seen in the PBS group (Fig. 6D), but in the PF-latanoprost group, sporadic cells were observed in the limbus area at H4 (Fig. 6D). In the 0.02% BAK (Fig. 6D) and BAK-latanoprost (Fig. 6D) groups, we observed obvious inflammatory infiltrations in the peripheral cornea and the limbus area. 
Conjunctival Stroma.
Both the PBS- and PF-latanoprost–instilled rabbits presented normal conjunctival blood vessels without any rolling inflammatory cells (Fig. 6E). In the 0.02% BAK (Fig. 6E) and BAK-latanoprost (Fig. 2D) groups, obvious rolling of inflammatory cells was recorded. 
IVCM score evaluation used to quantify abnormalities over the whole ocular surface showed no significant differences at H4 and D1 between PF-latanoprost and PBS (Fig. 7). BAK 0.02% and BAK-latanoprost showed higher IVCM scores compared with both PBS and PF-latanoprost (P < 0.0001 for all comparisons) at H4. At D1, IVCM scores were the highest after application of BAK alone (P < 0.03 compared with the three other groups), while BAK-latanoprost scores remained higher than the PBS and PF-latanoprost groups, without reaching significance. 
Figure 7. 
 
In vivo confocal microscopy scores at H4 and D1 after instillations of PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Figure 7. 
 
In vivo confocal microscopy scores at H4 and D1 after instillations of PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Cresyl Violet Staining on CIC
CIC specimens from rabbit eyes instilled with PBS showed a homogeneous cell sheet with a low mean CIC score of 0.75 ± 0.25 and 0.33 ± 0.33 at H4 and D1, respectively (Figs. 8A, 9): the epithelial cells were flat and regular, with a nucleocytoplasmic ratio ranging from 1:2 to 1:3. Goblet cells were clearly visible among or beside the epithelial cells. PF-latanoprost (Figs. 8B, 9) instillation induced a low, but increased CIC score (S = 6.25 ± 2.46) due to slight anisocytosis in the epithelium with normal nuclei and slight infiltration of inflammatory cells. Goblet cell number and morphology remained normal. Important inflammatory infiltration was observed after the instillation of 0.02% BAK (Figs. 8C, 9) and BAK-latanoprost (Figs. 8D, 9) leading to significantly increased CIC scores at H4 (S = 14 ± 1.8 and S = 12.4 ± 2.06, respectively, with P < 0.0001 compared with both PBS and PF-latanoprost treatments). In BAK-latanoprost–receiving eyes, a few conjunctival epithelial cells could still be observed, adjacent to inflammatory patches, with substantial anisocytosis and anisonucleosis. In BAK-receiving eyes, patches of inflammatory cell-containing secretions without any epithelial cells were observed. The number of goblet cells dramatically decreased in these conditions. 
Figure 8. 
 
Cresyl violet-stained conjunctival impression cytology from rabbits instilled with (A) PBS, (B) PF-latanoprost, (C) 0.02% BAK, and (D) BAK-latanoprost at H4 (×400 magnification).
Figure 8. 
 
Cresyl violet-stained conjunctival impression cytology from rabbits instilled with (A) PBS, (B) PF-latanoprost, (C) 0.02% BAK, and (D) BAK-latanoprost at H4 (×400 magnification).
Figure 9. 
 
Conjunctival impression cytology scores at H4 and D1.
Figure 9. 
 
Conjunctival impression cytology scores at H4 and D1.
Immunohistology of CD45-Positive Cells in Rabbit Cryosections
PF-latanoprost–instilled rabbit eyes showed no infiltration of CD45+ inflammatory cells in the peripheral cornea (Fig. 10A) but a slight infiltration in the limbus (675 ± 217 cells/mm2; Figs. 10B, 11) and conjunctiva (650 ± 87 cells/mm2; Figs. 10C, 11), with no statistically significant differences with the PBS eye (Figs. 10, 11). BAK 0.02% and BAK-latanoprost induced numerous CD45+ inflammatory cells in the peripheral cornea, limbal, and conjunctival areas: in the limbus 1250 ± 232 cells/mm2 were found for 0.02% BAK (Figs. 10B, 11); and 1925 ± 368 cells/mm2 for BAK-latanoprost (Figs. 10B, 11), with no differences. In the conjunctiva, 1425 ± 322 cells/mm2 were counted for 0.02% BAK (Figs. 10C, 11), and 1525 ± 242 cells/mm2 for BAK-latanoprost (Figs. 10C, 11). The numbers of CD45+ cells were similar between BAK-latanoprost and BAK, and were significantly higher in both groups compared with the PBS (P < 0.0001) and PF-latanoprost (P < 0.0001) groups. 
Figure 10. 
 
Immunohistology of CD45+ cells (green) in (A) peripheral cornea, (B) limbus, and (C) conjunctiva at D1 after instillations of (1) PBS, (2) PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost.
Figure 10. 
 
Immunohistology of CD45+ cells (green) in (A) peripheral cornea, (B) limbus, and (C) conjunctiva at D1 after instillations of (1) PBS, (2) PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost.
Figure 11. 
 
Cell counts of CD45+ cells in limbus and conjunctiva of rabbit cryosections at D1 after treatment with PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Figure 11. 
 
Cell counts of CD45+ cells in limbus and conjunctiva of rabbit cryosections at D1 after treatment with PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Discussion
Given its similar structure to the living HCE, 22 the 3D-HCE was thought to be a good surrogate of in vivo testing in the field of ocular irritation and ocular safety assessment. It is considered a quality controlled and easy to perform test system, with a high level of reproducibility. 2325 In a previous study, we optimized the MTT test, adapted to 3D models, 15 so as to improve the sensitivity of the test by taking into account the toxic effects not only in the basal layer, but also in the most superficial layers. With this new procedure, we demonstrated that 0.02% BAK and BAK-latanoprost decreased cell viability at levels close to 50%, while PF-latanoprost slightly decreased cell viability to 83%, with no significant difference compared with PBS, but a significant difference with 0.02% BAK and BAK-latanoprost. Moreover, this result was correlated with the quantification of the apoptotic, TUNEL-positive cells on epithelial tissue cryosections at 24 hours and at 24 hours plus 24 hours, and this study showed that PF-latanoprost induced relatively few apoptotic cells in the superficial layer of the epithelium. On the contrary, 0.02% BAK and BAK-latanoprost induced higher levels of apoptosis in the apical layers of the epithelium, with a significant effect compared with PBS and PF-latanoprost treatments. 
We used the well established in vivo toxicological model of acute instillations, which is a good indicator of further absence of ocular toxicity in a more conventional use over the long term for glaucoma patients. 1620 Using this model, the clinical Draize score of PF-latanoprost was slightly increased, although it did not reach significance. Compared with 0.02% BAK and BAK-latanoprost, which quickly induced redness and swelling, PF-latanoprost seemed to be better tolerated, with limited hyperemia observed at H4 that disappeared at D1. 
These results were confirmed at the cellular level, using high-resolution IVCM to follow the morphological changes in the animals' cornea, limbus, and conjunctiva, with an outstanding level of sensitivity. Using the scoring scale previously published by our group, we were able to attribute a score to each product tested that corresponded to the IVCM analysis of the whole ocular surface at a microscopic level. 20 These scores were the highest with 0.02% BAK and BAK-latanoprost treatments at H4, while PF-latanoprost induced a mild score that was not considered significantly different from that of PBS. In particular, only a few inflammatory cells were detected in the basal epithelium and/or in the limbal area at H4, while 0.02% BAK and BAK-latanoprost induced marked inflammation in the same areas. All IVCM scores decreased at D1, with the 0.02% BAK alone score remaining significantly higher compared with PBS. The cytological analysis of conjunctival imprints revealed that PF-latanoprost was the best-tolerated treatment among PF-latanoprost, 0.02% BAK and BAK-latanoprost, as it only induced slight anisocytosis and rare infiltration of inflammatory cells. The other two solutions induced marked infiltration of inflammatory cells and dramatic destruction of the epithelial conjunctival cells and goblet cells. Nevertheless, the recovery at D1 was remarkable with BAK-latanoprost, with a significantly reduced CIC score. The quantification of CD45+ cells on cryosections showed the highest levels of inflammation with 0.02% BAK and BAK-latanoprost. PF-latanoprost-treated eyes disclosed a moderate increase of CD45+ cells in the conjunctiva and the limbus that was not significant compared to PBS. Finally, we observed that the BAK-latanoprost did not exhibit additive toxic effect of the mild PF-latanoprost and the 0.02% BAK cytotoxicity observed at H4 and D1. These results highlight the possible protective effect of the prostaglandin over the preservative toxicity already reported by some studies. 25  
These results are in accordance with many other studies that demonstrated the pro-inflammatory and pro-apoptotic effect of BAK on ocular surface epithelia, beginning with Swan et al. in 1944. 26 Despite the number of articles supporting the deleterious effect of BAK on the ocular surface, 3,57 debate continues on the benefit of BAK in eye drop formulations. The first argument advanced to maintain BAK arises from its capacity to increase corneal permeability, acting as a penetration enhancer for active compounds improving their pharmacologic effects. 2729 In contrast, in a prospective clinical study Gross et al. demonstrated that BAK-free travoprost presents similar IOP-lowering efficacy as BAK-containing travoprost. 30 Furthermore, a comparative pharmacokinetic study did not report any significant difference in the tafluprost concentration found in rabbit aqueous humor between the BAK-associated and BAK-free formulations. 31 Furthermore, a clinical study showed a similar IOP-lowering effect with both formulations of latanoprost (Rouland J, et al. IOVS 2012;A240:ARVO E-Abstract 5099). These previous studies suggested that BAK is not necessary to provide drug penetration. 2731  
Glaucoma is a chronic degenerative pathology that requires permanent pharmacological management. The current treatments are not efficient enough to stop the progress of the disease. More than 50% of glaucomatous patients require the addition of more than one drug, increasing ocular surface BAK accumulation and, thus, the risk of side effect occurrence. Moreover, after many years of treatments, patients could undergo surgery when IOP becomes unresponsive to eye drop medications; therefore, the quality of the ocular surface is an important factor of efficient filtration surgery. 32 The new glaucoma eye drops favor the patient's compliance and, thus, potentiate treatment efficiency by alleviating progressive side effects. Non preserved latanoprost has proved to be better tolerated than BAK-containing formulations and may therefore improve the therapeutic observance and quality of life of glaucoma patients. 
References
Ishida N Odani-Kawabata N Shimazaki A Hara H. Prostanoids in the therapy of glaucoma. Cardiovasc Drug Rev . 2006;24:1–10. [CrossRef] [PubMed]
Orme M Collins S Dakin H Kelly S Loftus J. Mixed treatment comparison and meta-regression of the efficacy and safety of prostaglandin analogues and comparators for primary open-angle glaucoma and ocular hypertension. Curr Med Res Opin . 2010;26:511–528. [CrossRef] [PubMed]
Baudouin C Labbe A Liang H Preservatives in eyedrops: the good, the bad and the ugly. Prog Retin Eye Res . 2010;29:312–334. [CrossRef] [PubMed]
Zhivov A Kraak R Bergter H Kundt G Beck R Guthoff RF. Influence of benzalkonium chloride on langerhans cells in corneal epithelium and development of dry eye in healthy volunteers. Curr Eye Res . 2010;35:762–769. [CrossRef] [PubMed]
Liang H Baudouin C Labbe A Riancho L Brignole-Baudouin F. Conjunctiva-associated lymphoid tissue (CALT) reactions to antiglaucoma prostaglandins with or without BAK-preservative in rabbit acute toxicity study. PLoS One . 2012;7:33913. [CrossRef]
Galletti JG Gabelloni ML Morande PE Benzalkonium chloride breaks down conjunctival immunological tolerance in a murine model [published online ahead of print June 13, 2012]. Mucosal Immunol . doi: 10.1038/mi.2012.44.
Sarkar J Chaudhary S Namavari A Corneal neurotoxicity due to topical benzalkonium chloride. Invest Ophthalmol Vis Sci . 2012;53:1792–1802. [CrossRef] [PubMed]
Martone G Frezzotti P Tosi GM An in vivo confocal microscopy analysis of effects of topical antiglaucoma therapy with preservative on corneal innervation and morphology. Am J Ophthalmol . 2009;147:725–735. [CrossRef] [PubMed]
Calonge M Diebold Y Saez V Impression cytology of the ocular surface: a review. Exp Eye Res . 2004;78:457–472. [CrossRef] [PubMed]
Schwab IR Linberg JV Gioia VM Benson WH Chao GM. Foreshortening of the inferior conjunctival fornix associated with chronic glaucoma medications. Ophthalmology . 1992;99:197–202. [CrossRef] [PubMed]
Brasnu E Brignole-Baudouin F Riancho L Guenoun JM Warnet JM Baudouin C. In vitro effects of preservative-free tafluprost and preserved latanoprost, travoprost, and bimatoprost in a conjunctival epithelial cell line. Curr Eye Res . 2008;33:303–312. [CrossRef] [PubMed]
Jaenen N Baudouin C Pouliquen P Manni G Figueiredo A Zeyen T. Ocular symptoms and signs with preserved and preservative-free glaucoma medications. Eur J Ophthalmol . 2007;17:341–349. [PubMed]
Baudouin C de Lunardo C. Short-term comparative study of topical 2% carteolol with and without benzalkonium chloride in healthy volunteers. Br J Ophthalmol . 1998;82:39–42. [CrossRef] [PubMed]
Chabi A Varma R Tsai JC Randomized clinical trial of the efficacy and safety of preservative-free tafluprost and timolol in patients with open-angle glaucoma or ocular hypertension. Am J Ophthalmol . 2012;153:1187–1196. [CrossRef] [PubMed]
Pauly A Meloni M Brignole-Baudouin F Warnet JM Baudouin C. Multiple endpoint analysis of the 3D- reconstituted corneal epithelium after treatment with benzalkonium chloride: early detection of toxic damage. Invest Ophthalmol Vis Sci . 2009;50:1644–1677. [CrossRef] [PubMed]
Ichijima H Petroll WM Jester JV Cavanagh HD. Confocal microscopic studies of living rabbit cornea treated with benzalkonium chloride. Cornea . 1992;11:221–225. [CrossRef] [PubMed]
Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods . 1983;65:55–63. [CrossRef] [PubMed]
Doucet O Robert C Zastrow L. Use of a serum-free reconstituted epidermis as a skin pharmacological model. Toxicol In Vitro . 1996;10:305–313. [CrossRef] [PubMed]
Doucet O Lanvin M Zastrow L. A new in vitro human epithelial model for assessing eye irritating potential of formulated cosmetic products. Toxicol In Vitro . 1998;11:273–283.
Liang H Brignole-Baudouin F Rabinovich-Guilatt L Reduction of quaternary ammonium-induced ocular surface toxicity by emulsions: an in vivo study in rabbits. Mol Vis . 2008a;14:204–216.
Haller-Schober EM Schwantzer G Berghold A Fischl M Theisl A Horwath-Winter J. Evaluating an impression cytology grading system (IC score) in patients with dry eye syndrome. Eye . 2006;20:927–933. [CrossRef] [PubMed]
Nguyen DH Beuerman RW De Wever B Rosdy M. Three-dimensional construct of the human corneal epithelium for in vitro toxicology. In: Salem H Katz S eds. Alternative Toxicological Methods . Boca Raton, FL; CRC Press; 2003:147–159.
Doucet O Lanvin M Thillou C Reconstituted human corneal epithelium: a new alternative to the Draize eye test for the assessment of the eye irritation potential of chemicals and cosmetic products. Toxicol In Vitro . 2006;20:499–512. [CrossRef] [PubMed]
Van Goethem F Adriaens E Alepee N Prevalidation of a new in vitro reconstituted human cornea model to assess the eye irritating potential of chemicals. Toxicol In Vitro . 2006;20:1–17. [CrossRef] [PubMed]
Guenoun JM Baudouin C Rat P Pauly A Warnet JM Brignole-Baudouin F. In vitro comparison of cytoprotective and antioxidative effects of latanoprost, travoprost, and bimatoprost on conjunctiva-derived epithelial cells. Invest Ophthalmol Vis Sci . 2005;46:4594–4599. [CrossRef] [PubMed]
Swan KC. Reactivity of the ocular tissues to wetting agents. Am J Ophthalmol . 1944;27:1118–1122. [CrossRef]
Burstein NL. Preservative alteration of corneal permeability in humans and rabbits. Invest Ophthalmol Vis Sci . 1984;25:1453–1457. [PubMed]
Majumdar S Hippalgaonkar K Repka MA. Effect of chitosan, benzalkonium chloride and ethylenediaminetetraacetic acid on permeation of acyclovir across isolated rabbit cornea. Int J Pharm . 2008;348:175–178. [CrossRef] [PubMed]
Okabe K Kimura H Okabe J Effect of benzalkonium chloride on transscleral drug delivery. Invest Ophthalmol Vis Sci . 2005;46:703–708. [CrossRef] [PubMed]
Gross RL Peace JH Smith SE Duration of IOP reduction with travoprost BAK-free solution. J Glaucoma . 2008;17:217–222. [CrossRef] [PubMed]
Pellinen P Lokkila J. Corneal penetration into rabbit aqueous humor is comparable between preserved and preservative-free tafluprost. Ophthalmic Res . 2009;41:118–122. [CrossRef] [PubMed]
Gedde SJ Schiffman JC Feuer WJ Herndon LW Brandt JD Budenz DL, Tube Versus Trabeculectomy Study Group. Three-year follow-up of the tube versus trabeculectomy study. Am J Ophthalmol . 2009;148:670–684. [CrossRef] [PubMed]
Footnotes
 Supported by grants from Laboratoires Théa and the Institut National pour la Santé et la Recherche Médicale (INSERM).
Footnotes
4  These authors contributed equally to the work presented here and should therefore be regarded as equivalent authors.
Footnotes
 Disclosure: A. Pauly, Laboratoires Théa (F); C. Roubeix, Laboratoires Théa (F); H. Liang, None; F. Brignole-Baudouin, Laboratoires Théa (C); C. Baudouin, Alcon (C), Allergan (C), Merck (C), Santen (C), Laboratoires Théa (C)
Figure 1. 
 
Cellular viability of 3D-HCEs cultures for 24 hours in PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Figure 1. 
 
Cellular viability of 3D-HCEs cultures for 24 hours in PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Figure 2. 
 
Immunolocalization of TUNEL-positive cells on 3D-HCE cultures treated with (1) PBS, (2) PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost for (A) 24 hours and (B) 24 hours followed by a 24 hour post incubation period of time (×200 magnification).
Figure 2. 
 
Immunolocalization of TUNEL-positive cells on 3D-HCE cultures treated with (1) PBS, (2) PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost for (A) 24 hours and (B) 24 hours followed by a 24 hour post incubation period of time (×200 magnification).
Figure 3. 
 
Quantification of TUNEL-positive cells after 24 hour of treatment with PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost, followed or not by a 24 hour post incubation period of time.
Figure 3. 
 
Quantification of TUNEL-positive cells after 24 hour of treatment with PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost, followed or not by a 24 hour post incubation period of time.
Figure 4. 
 
Photographs of rabbit eyes 4 hours after the 15 instillations of (A) PBS, (B) PF-latanoprost, (C) 0.02% BAK, and (D) BAK-latanoprost.
Figure 4. 
 
Photographs of rabbit eyes 4 hours after the 15 instillations of (A) PBS, (B) PF-latanoprost, (C) 0.02% BAK, and (D) BAK-latanoprost.
Figure 5. 
 
Draize test scores at H4 and D1 after instillations of PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost, for 24 hours.
Figure 5. 
 
Draize test scores at H4 and D1 after instillations of PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost, for 24 hours.
Figure 6. 
 
In vivo confocal images of rabbit corneal superficial epithelium (Raw [A]), basal epithelium (Raw [B]), anterior stroma (Raw [C]), limbus (Raw [D]), and conjunctival substancia propria (Raw [E]) after instillations of (1) PBS, (2) preservative-free PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost at H4. Photograph size: 400 μm by 400 μm.
Figure 6. 
 
In vivo confocal images of rabbit corneal superficial epithelium (Raw [A]), basal epithelium (Raw [B]), anterior stroma (Raw [C]), limbus (Raw [D]), and conjunctival substancia propria (Raw [E]) after instillations of (1) PBS, (2) preservative-free PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost at H4. Photograph size: 400 μm by 400 μm.
Figure 7. 
 
In vivo confocal microscopy scores at H4 and D1 after instillations of PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Figure 7. 
 
In vivo confocal microscopy scores at H4 and D1 after instillations of PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Figure 8. 
 
Cresyl violet-stained conjunctival impression cytology from rabbits instilled with (A) PBS, (B) PF-latanoprost, (C) 0.02% BAK, and (D) BAK-latanoprost at H4 (×400 magnification).
Figure 8. 
 
Cresyl violet-stained conjunctival impression cytology from rabbits instilled with (A) PBS, (B) PF-latanoprost, (C) 0.02% BAK, and (D) BAK-latanoprost at H4 (×400 magnification).
Figure 9. 
 
Conjunctival impression cytology scores at H4 and D1.
Figure 9. 
 
Conjunctival impression cytology scores at H4 and D1.
Figure 10. 
 
Immunohistology of CD45+ cells (green) in (A) peripheral cornea, (B) limbus, and (C) conjunctiva at D1 after instillations of (1) PBS, (2) PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost.
Figure 10. 
 
Immunohistology of CD45+ cells (green) in (A) peripheral cornea, (B) limbus, and (C) conjunctiva at D1 after instillations of (1) PBS, (2) PF-latanoprost, (3) 0.02% BAK, and (4) BAK-latanoprost.
Figure 11. 
 
Cell counts of CD45+ cells in limbus and conjunctiva of rabbit cryosections at D1 after treatment with PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Figure 11. 
 
Cell counts of CD45+ cells in limbus and conjunctiva of rabbit cryosections at D1 after treatment with PBS, PF-latanoprost, 0.02% BAK, and BAK-latanoprost.
Table. 
 
Test Items and Controls Used in This Study
Table. 
 
Test Items and Controls Used in This Study
Name Active Ingredient Batch Number Preservation System
PBS None Not applicable None
0.02% BAK solution - dilution of 0.1% BAK in PBS None FAB 07/05/2009 BAK
Monoprost eye drops Latanoprost 0.005% S879 None
Xalatan eye drops Latanoprost 0.005% R08267 0.02% BAK
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×