*3.3. Surface Tension*

Gaap Ofteno® demonstrated a surface tension of 60.31 ± 0.35 mN/m, which was significantly higher compared to the other products (*<sup>p</sup>* ≤ 0.0001) (Figure 3). Xalmono® had the lowest surface tension of 39.00 ± 0.38 mN/m, and was significantly lower compared to the other products (*p* ≤ 0.0001). No significant differences were observed between Monoprost® (42.44 ± 0.75 mN/m) and Xaloptic® Free (42.76 ± 0.36 mN/m), or between Latanest® (43.15 ± 1.13 mN/m) and Xaloptic® Free.

**Figure 3.** Surface tension characterization of preservative-free 0.005% latanoprost products; Monoprost®, Latanest®, Gaap Ofteno®, Xalmono® and Xaloptic® Free. Values are listed as mean <sup>±</sup> SD, *<sup>n</sup>* = 3. A oneway ANOVA with a Tukey multiple-comparison test (*p* = 0.05) were performed. ns = not significant with *p* ≥ 0.05, \*\* *p* ≤ 0.01, \*\*\*\* *p* ≤ 0.0001.

#### *3.4. Cell Survival*

LDH and MTT analyses were performed on primary cultured human conjunctival goblet cells treated with diluted PF 0.005% latanoprost products. Cell survival was reported relative to the control, these being goblet cells treated with RPMI media. According to the results of the LDH assays, the mean cell survival after treatment with eye drops was 100.70 ± 12.8% (Monoprost®), 98.17 ± 9.4% (Latanest®), 101.40 ± 5.9% (Gaap Ofteno®), 99.36 ± 5.7% (Xalmono®) and 93.66 ± 6.0% (Xaloptic® Free). As seen in Figure 4a, the LDH assay showed no significant differences in cell survival between treatments. Similar results were obtained from the MTT analysis, showing no significant differences in cell survival between the treatments (Figure 4b). The mean cell survival compared to the control was 86.40 ± 9.1% (Monoprost®), 83.53 ± 9.9% (Latanest®*)*, 89.84 ± 18.2% (Gaap Ofteno®), 89.30 ± 12.7% (Xalmono®), and 89.51 ± 11.0% (Xaloptic® Free).

**Figure 4.** Mean cell survival analysis of human conjunctival goblet cells in % ± SD, relative to control, after 30 min of treatment with preservative-free 0.005% latanoprost eye drops. (**a**) Cell survival, examined with the LDH assay; (**b**) cell survival, examined with the MTT assay. Goblet cells were treated with diluted (1:7, *v*/*v*) preservative-free 0.005% latanoprost products: Monoprost®, Latanest®, Gaap Ofteno®, Xalmono®, and Xaloptic® Free. A one-way ANOVA with a Tukey multiplecomparison test was performed (*p* = 0.05), *n* ≥ 4; ns = not significant; *p* ≥ 0.05.

#### *3.5. Immunohistochemical Staining*

Immunohistochemical staining was used to visualize the cytoskeleton of the goblet cells, the nuclei, and mucin, as seen in Figure 5. Goblet cells treated with RPMI media showed mucin allocated around the nuclei (Figure 5A). Figure 5B–F demonstrated a similar pattern, showing mucin allocated near the nuclei when the cell cultures were treated with diluted PF 0.005% latanoprost eye drops: Monoprost®, Latanest®, Gaap Ofteno®, Xalmono®, and Xaloptic® Free.

**Figure 5.** *Cont*.

**Figure 5.** Immunohistochemical staining of human goblet cells, visualizing the cytoskeleton (column 1: Cytokeratin-7, green), mucin (column 2: mucin, red), the nucleus (column 3: DAPI, blue), and merged stainings (column 4: cytokeratin-7 (green), mucin (red) and DAPI (blue)). Cells were treated with diluted (1:7, *v*/*v*) 0.005% latanoprost preservative-free products. (**A**): RPMI media, (**B**): Monoprost ®, (**C**): Latanest®, (**D**): Gaap Ofteno®, (**E**): Xalmono®, and (**F**): Xaloptic® Free. *n* = 3.

#### **4. Discussion**

Anti-glaucomatous treatment is lifelong, and the importance of minimizing adverse effects is crucial to increasing patients' adherence to the treatment regimen and for healthrelated quality of life [26]. Glaucoma patients experience side effects from the treatment, such as dry eye disease (DED), a foreign-body sensation, and a stinging or burning sensation in the eye [27]. The cause of DED is either reduced lacrimation or increased evaporation from the ocular surface. Changes in the ocular surface may lead to a hyperosmolar condition, which is a recognized risk factor for inducing proinflammatory stress [8,28,29]. The instability of the tear film can cause ocular discomfort or irritation, while pH value, osmolality, and surface tension all play a role in achieving a healthy ocular surface. The current study revealed significant differences in the physicochemical properties of pH value, osmolality, and surface tension. Some of the variations may be due to the product formulation, as Monoprost®, Xaloptic® Free, and Xalmono® were dispensed from single-dose units, whereas Gaap Ofteno® and Latanest® were dispensed from multi-dose bottles. Osmolality, pH value, and surface tension may vary due to variations in the solvents or stabilizers caused by the removal of preservatives. The physicochemical properties of the undiluted eye drops illustrated the properties of the eye drops immediately after application to the ocular surface in patients.

The pH value of tear fluid is 7.4–7.6 [30–32]. To optimize ocular comfort, the pH value of the ophthalmic formulations should match the tear film or at least be in the ocular range of 6.6–7.8, as an acidic or alkaline pH value induces lacrimation, ocular pain, and discomfort [23,32]. In this study, we found that Latanest® (pH value 6.33 ± 0.003) and Gaap Ofteno® (pH value 6.34 ± 0.004) were acidic, with pH values below the ocular range. The remaining three products, Monoprost® (pH value 6.84 ± 0.032), Xalmono® (6.70 ± 0.003), and Xaloptic® Free (6.71 ± 0.000), had pH values within the ocular range, with Monoprost® closest to the pH value of the tear film. An acidic pH value may cause side effects, such as ocular discomfort and increased lacrimation upon instillation, whereas a pH within the recommended range should not provide any discomfort related to the pH value. When the products were diluted with RPMI media, the pH values of all products were within the recommended pH range. Our findings imply that it would be of great interest to investigate the pH value of PF PGA products upon dilution with tear fluid in patients, to elucidate if differences in pH value may cause prolonged ocular discomfort.

The osmolality of the tear fluid varies from 310 to 350 mosmol/kg [31]. All products except Gaap Ofteno® (325.9 ± 2.9 mosmol/kg) were hypo-osmolar, compared to the tear fluid. As previously mentioned, a hyperosmolar tear film is related to ocular irritation and DED. Based on our findings, the osmolality of the tested products should not be of particular concern, as they demonstrated iso- or hypo-osmolar properties.

The surface tension of the tear fluid varies from 40 to 46 mN/m and ensures a stable tear film and tear film break-up time [33,34]. Furthermore, the surface tension influences the eye drops' ability to spread and adhere to the cornea, once applied [34]. We found that Monoprost® (42.44 ± 0.75 mN/m), Xaloptic® Free (42.76 ± 0.36 mN/m), and Latanest® (43.15 ± 1.13 mN/m) had a surface tension in the physiological range of the tear fluid. Gaap Ofteno® (60.31 ± 0.35 mN/m) had a surface tension well above the physiological range, while Xalmono® had a surface tension just below 40 (39.0 ± 0.38 mN/m). Surface tension exceeding the physiological range may cause instability of the tear film and is associated with dry eyes [34]. In addition, higher surface tension will increase the drop volume released from the bottle [35]. The drop volume will also affect the amount of latanoprost released [23]. A greater drop volume will increase the washout and may result in less uptake of latanoprost and reduced efficacy in terms of lowering IOP. Thus, increased surface tension in ophthalmic solutions may lead to both the ocular adverse effects caused by a destabilized tear film and potentially reduce the efficacy of the eye drops.

Keeping goblet cells unharmed and unstressed is essential for maintaining a stable tear film for a healthy ocular surface [9,10]. Diluted PF 0.005% latanoprost eye drops showed no negative effects on either cell survival or mucin release after treatment, according to the LDH assay, MTT assay, and immunohistochemical staining. Other studies have investigated the differences between preserved and preservative-free PGAs. Treatment with PF tafluprost showed reduced pro-apoptotic and pro-oxidative stress in a conjunctival epithelial cell line, compared to preserved PGAs [36]. In patients, the goblet cell density was significantly increased after six months of treatment with PF tafluprost, compared to the baseline. In comparison, patients treated with preserved tafluprost showed increased goblet cell density after one month of treatment, but no significant long-term effects were reported [37].

As previously mentioned, it is of great interest to minimize the adverse effects of treatment, since glaucoma is a chronic condition. A frequently reported side effect of PGAs is conjunctival hyperemia [38,39]. PF PGAs, e.g., Monoprost® and Gaap Ofteno®, have previously been examined and compared to preserved PGAs. A meta-analysis of 21 studies found that conjunctival hyperemia was significantly reduced in Monoprost® compared to preserved PGAs, with no significant differences in IOP between treatments [40]. In addition, Rouland et al. showed noninferiority in IOP efficacy and a significantly reduced conjunctival hyperemia after treatment with Monoprost®, compared to treatment with Xalatan® [41].

Gonzales et al. compared the stability, efficacy, and adverse effects of Gaap Ofteno® with the preserved brand product, Xalatan® [42]. As Gaap Ofteno® is dispensed from a multidose bottle, the stability of the drug was also examined. The study showed that the two products were comparable in terms of IOP reduction and safety, evaluated as conjunctival hyperemia, with a similar prevalence of 11.3% (Gaap Ofteno®) and 11.9% (Xalatan®). The above-mentioned studies support our findings that PF 0.005% latanoprost products are non-toxic to human conjunctival goblet cells under experimental conditions.

When patients instill ophthalmic solutions in the eye, the tear film dilutes the ophthalmic solutions and only 1–7% of the instilled drug reaches the aqueous humor [43]. We chose to treat goblet cell cultures with diluted eye drops at a constant concentration for 30 min. In patients, the tear film dilutes the eye drops gradually upon administration, with a reduced expected exposure time (less than 30 min) as the fluid turnover rate of the conjunctival cul-de-sac in the eye is 0.5–2.2 μL/min. With this turnover rate, the drug remains in the conjunctival cul-de-sac for approximately 3–5 min [43]. The RPMI media used in this study was not identical to the tear film and will have different capacities, for instance, in buffering. Therefore, the physicochemical properties of eye drops diluted with tear film may affect the goblet cells in other ways than those identified in this study.

The in vitro model used in this study has some limitations since it cannot be directly compared to the conditions in patients. We did not quantify the amount of mucin released from goblet cells; the immunohistochemical staining that was performed only detected the presence of mucin. In addition, we did not evaluate the long-term effects of treating human conjunctival goblet cells with PF products. Thus, clinical trials would be desirable to examine whether the physicochemical differences addressed in this study may influence the long-term efficacy and safety profile of PF eye drops.

#### **5. Conclusions**

In conclusion, this study identified significant differences in pH value, osmolality, and surface tension among five PF 0.005% latanoprost products. The variations in physicochemical properties, such as acidic pH values or high surface tension, may potentially destabilize the tear film and reduce the tolerability of eye drops on the ocular surface. However, the variations in the physicochemical properties of the PF eye drops had no negative effects on either cell survival or mucin release. Since the efficacy was not examined in this in vitro experiment, clinical studies would be of great interest in elucidating the potential differences among PF PGA treatments, in terms of efficacy, tolerability, and the side effects related to long-term treatment with PF 0.005% latanoprost products.

**Author Contributions:** Conceptualization, M.K., J.C.F., A.H., S.H., G.P. and J.J.; methodology, M.K. and J.J.; investigation and data curation, J.C.F.; formal analysis, J.C.F., A.H. and M.K.; writing original draft preparation, J.C.F.; writing—review and editing, J.C.F., A.H., J.J., G.P., S.H., B.C. and M.K.; supervision, M.K., J.J., A.H.; funding acquisition, M.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Laboratoires Théa (France).

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki, and approved by the Danish National Committee on Health Research (H-17007902) and the Norwegian Regional Committees for Medical and Health Research Ethics (REK: 2013/803).

**Informed Consent Statement:** Patient consent for publication not applicable.

**Data Availability Statement:** Datasets are available on reasonable request.

**Acknowledgments:** Special thanks to laboratory technician Charlotte Taul for her laboratory assistance. We acknowledge the Hørslev-Fonden for their equipment grant for a tensiometer. We also acknowledge the Vissing-Fonden for their project support grant to meet the running costs. The study at the Center for Eye Research, Oslo, Norway, was partially supported by funding from the Norwegian Association of the Blind and Partially Sighted.

**Conflicts of Interest:** M.K. is speaker and in advisory boards for Abbvie, Santen and Thea. M.K. has a collaborative grant from Thea and serve as a consultant for Abbvie and Thea.
