Fluorometholone Modulates Gene Expression of Ocular Surface Mucins

Chapman University Chapman University Digital Commons

Pharmacy Faculty Articles and Research School of Pharmacy

4-9-2019

Jonathan Taniguchi Chapman University

Ajay Sharma Chapman University, [email protected]

Follow this and additional works at: https://digitalcommons.chapman.edu/pharmacy_articles

Part of the Genetic Processes Commons, Medical Biochemistry Commons, Medicinal and Pharmaceutical Chemistry Commons, Musculoskeletal, Neural, and Ocular Physiology Commons, Other Pharmacy and Pharmaceutical Sciences Commons, and the Sense Organs Commons

Recommended Citation Taniguchi J, Sharma A. Fluorometholone modulates gene expression of ocular surface mucins. Acta Ophthalmologica. 2019;97(8):e1082-e1088. https://doi.org/10.1111/aos.14113

This Article is brought to you for free and open access by the School of Pharmacy at Chapman University Digital Commons. It has been accepted for inclusion in Pharmacy Faculty Articles and Research by an authorized administrator of Chapman University Digital Commons. For more information, please contact [email protected]. Fluorometholone Modulates Gene Expression of Ocular Surface Mucins

Comments This is the accepted version of the following article:

Taniguchi J, Sharma A. Fluorometholone modulates gene expression of ocular surface mucins. Acta Ophthalmologica. 2019;97(8):e1082-e1088. https://doi.org/10.1111/aos.14113 which has been published in final form at https://doi.org/10.1111/aos.14113. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving.

This article is available at Chapman University Digital Commons: https://digitalcommons.chapman.edu/ pharmacy_articles/652 Page 1 of 26 Acta Ophthalmologica

1 2 3 4 5 Fluorometholone modulates gene expression of ocular surface mucins 6 7 8 Jonathan Taniguchi, Ajay Sharma* 9 10 Department of Biomedical and Pharmaceutical Sciences 11 12 Chapman University School of Pharmacy 13 14 Chapman University 15 Irvine, California 92618 16 17 18 19 For Peer Review 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 *Corresponding Author 42 43 email: [email protected] 44 45 Phone: 001-714-516-5498 46 Fax: 001-714-516-5481 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 2 of 26

1 2 3 Abstract: 4 5 6 Purpose: Mucins are vital to keep the ocular surface hydrated. Genes encoding for mucins 7 8 contain a glucocorticoid response element. The purpose of this study is to evaluate the 9 10 effect of fluorometholone, a glucocorticoid receptor agonist used in the management of 11 12 dry eye, on the gene expression of conjunctival and corneal epithelial cell mucins. 13 14 15 Methods: Stratified cultures of human conjunctival and corneal epithelial cells were 16 17 exposed to 25, 50 & 100 nM of fluorometholone alone or in presence of mifepristone, a 18 19 glucocorticoid receptorFor antagonist. Peer The mRNA Review was isolated from the cells and reverse 20 21 22 transcribed to cDNA. The cDNA was used for quantification of gene expression of mucin 23 24 (MUC) 1, 4, 16 &19 using real time PCR. 25 26 Results: Fluorometholone caused a dose and time dependent increase in the gene 27 28 29 expression of MUC1, MUC4, MUC16 and MUC19 in the conjunctival as well as corneal 30 31 epithelial cells. Mifepristone, a glucocorticoid receptor antagonist, inhibited 32 33 fluorometholone-mediated increase in the gene expression of conjunctival and corneal 34 35 mucins. At the tested concentration, neither fluorometholone nor mifepristone caused any 36 37 38 notable changes in the cellular phenotype or viability of conjunctival and corneal 39 40 epithelial cells. 41 42 Conclusion: Fluorometholone increases the gene expression of MUC1, MUC4, MUC16 43 44 45 and MUC 19 in the conjunctival and corneal epithelial cells through activation of 46 47 glucocorticoid receptors. The increased expression of mucins can be an additional 48 49 possible mechanism contributing to the beneficial effects of fluorometholone in dry eye 50 51 52 in addition to its well-known anti-inflammatory effects. 53 54 55 56 57 58 59 60 Page 3 of 26 Acta Ophthalmologica

1 2 3 Introduction 4 5 6 Mucins are high molecular weight glycoproteins that are expressed on the wet epithelial 7 8 surface of many organs including respiratory tract, gastrointestinal tract, genitourinary 9 10 tract, and the ocular surface (Corfield et al., 2001; Gipson 2001; Rose & Voynow, 2006; 11 12 Govindarajan & Gipson, 2010). Structurally, mucins consist of extensive glycan N-acetyl 13 14 15 galactosamine side chains attached to the amino acids of a protein core. The heavy 16 17 glycosylation imparts the mucins with hydrophilicity and a high negative charge. These 18 19 structural features accountFor their Peer two key physiological Review functions of repelling pathogens 20 21 22 and keeping the organ surface hydrated. Mucins can be tethered to the cell membrane of 23 24 epithelial cells or can be secreted. The ocular surface expresses 3 membrane-bound 25 26 mucins (MUC): MUC1, MUC4, MUC16, one secreted mucin MUC 19, and one gel- 27 28 29 forming secreted mucin, MUC5AC (Gipson & Argüeso, 2003; Gipson, 2004; 30 31 Ablamowicz & Nichols, 2016). The membrane-bound mucins are expressed on the apical 32 33 surface of corneal and conjunctival epithelial cells and the large gel-forming mucin is 34 35 primarily secreted by the goblet cells present in the conjunctiva (Gipson & Argüeso, 36 37 38 2003; Gipson, 2016). Besides ocular surface epithelial cells, lacrimal gland also releases 39 40 two secretory mucins into the tear film, gel-forming MUC5B and soluble MUC7. The 41 42 ocular surface mucins have been shown to play a critical role in keeping the eye surface 43 44 45 hydrated and lubricated. The ocular mucins have also been shown to perform janitorial 46 47 functions of removing allergens and pathogens from the eye surface (Gipson & Argüeso, 48 49 2003; Gipson, 2004; Govindarajan & Gipson, 2010; Ablamowicz & Nichols, 2016). 50 51 52 Given their role in keeping the eye surface hydrated, multiple studies have shown altered 53 54 levels or degree of glycosylation of various mucins in the patients suffering from dry eye 55 56 57 58 59 60 Acta Ophthalmologica Page 4 of 26

1 2 3 (Argüeso et al., 2002; Ramamoorthy & Nichols 2008; Gipson et al., 2013; Stephens et al., 4 5 6 2015). 7 8 Dry eye is one of the most frequently encountered conditions in ophthalmic 9 10 clinics. Current clinical management of dry eye includes the use of artificial tears, 11 12 punctual plugs, and anti-inflammatory drugs (Pflugfelder, 2004; Dogru & Tsubota, 13 14 15 2011). Topical cyclosporine and short-term corticosteroids are frequently prescribed anti- 16 17 inflammatory drugs in the management of moderate to the severe dry eye. Multiple 18 19 clinical studies have shownFor the Peerbeneficial effect Review of corticosteroids on the signs and 20 21 22 symptoms of dry eye disease (Avunduk et al., 2003; Pflugfelder et al., 2004; Pinto- 23 24 Fraga et al., 2016 Cutolo et al., 2017). Suppression of inflammation, a decrease in the 25 26 proinflammatory cytokines and chemokines, a reduction in the matrix metalloprotease 27 28 29 activity and an increase in the goblet cell density have been shown as the possible 30 31 mechanisms for the ameliorative effect of corticosteroids on dry eye. 32 33 Corticosteroids modulate transcription of a wide variety of genes via activation of 34 35 36 glucocorticoid receptors. Corticosteroids such as dexamethasone and budesonide have 37 38 been shown to modulate the gene expression of mucins in airway epithelial cells and 39 40 cancer cell lines (Imai et al., 2004; Bai et al., 2007; Milara et al., 2017). Corneal and 41 42 conjunctival epithelial cells express glucocorticoid receptors raising the possibility that 43 44 45 topical corticosteroids can modulate the gene expression of ocular mucins by activating 46 47 glucocorticoid receptors (Cavet et al., 2013; Kadmiel et al., 2016). Fluorometholone is a 48 49 fluorinated corticosteroid that has been shown to be efficacious in relieving the symptoms 50 51 52 of dry eye in non-Sjogren and Sjogren’s patients (Jee et al., 2015; Lin & Gong 2015, 53 54 Pinto-Fraga et al., 2016). Fluorometholone is a preferred and frequently prescribed 55 56 57 58 59 60 Page 5 of 26 Acta Ophthalmologica

1 2 3 topical ophthalmic corticosteroid because of its lower propensity to cause an increase in 4 5 6 intraocular pressure. Compared to other corticosteroids, fluorometholone is a less potent 7 8 glucocorticoid receptor agonist, and its effect on modulation of mucins gene expression 9 10 has not been evaluated. Therefore, the present study is designed to examine the effect of 11 12 fluorometholone on the gene expression of mucins in conjunctival and corneal epithelial 13 14 15 cells. 16 17 Materials and Methods 18 19 Conjunctival and cornealFor cell Peerculture Review 20 21 22 Telomerase-transformed human conjunctival and human corneal cells were used for the 23 24 experiments. Human conjunctival epithelial cells were plated at the density of 2X104/cm2 25 26 in keratinocyte serum-free medium supplemented with 25 μg/ml bovine pituitary extract, 27 28 29 0.2 ng/mL epidermal growth factor, and 0.4 mM CaCl2. Once the cultures achieved 50% 30 31 confluence, they were switched to a 1:1 mixture of supplemented keratinocyte serum-free 32 33 medium: low-calcium DMEM/F12. The cells were then allowed to reach 100% 34 35 confluence. The cells were then switched to DMEM/F12 medium supplemented with 36 37 38 1mM CaCl2, 10% calf serum and 10 ng/mL EGF for about 5 to 7 days to induce 39 40 stratification (Gipson et al., 2003). 41 42 The human corneal epithelial cells were grown and propagated in keratinocyte growth 43 44 45 medium supplemented with bovine pituitary extract, human epidermal growth factor, 46 47 insulin, hydrocortisone, epinephrine, transferrin and 0.15 mM calcium chloride. For 48 49 inducing stratifications, the cells were grown on collagen-coated 3.0µm PTFE transwell 50 51 52 membrane inserts in the supplemented keratinocyte growth medium containing 1.15 mM 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 6 of 26

1 2 3 CaCl2. Once the cells reached 100% confluence, the growth medium was added to the 4 5 6 bottom wells only, and the membrane inserts with the growing cells were exposed to air. 7 8 Creating this air-liquid interface resulted in stratification within one week (Robertson et 9 10 al., 2005). 11 12 Isolation of RNA and preparation of cDNA 13 14 15 The cells were harvested, and mRNA was isolated using QIAGEN’s FastLane Cell 16 17 cDNA kit. According to the kit protocol, the cells were washed with a washing buffer to 18 19 remove any remainingFor cell culture Peer medium andReview extracellular material. The cells were 20 21 22 then lysed with a hypotonic buffer that causes the release of the intracellular contents. 23 24 The lysate was then mixed with a gDNA wipeout buffer to remove the genomic DNA. At 25 26 this point, the lysate was mixed with a master mix containing reverse transcriptase, a 27 28 29 buffer, dNTPs, and a reverse transcription primer mix to create cDNA. 30 31 Real time PCR for quantification of mucins gene expression 32 33 The relative quantification of mucins 1, 4, 16 & 19 mRNA was performed using the real- 34 35 time PCR system (Biorad CFX thermocycler, Bio-Rad Laboratories). A 20 µl reaction 36 37 38 mixture containing 2 µl of cDNA, 2 µl of forward primer (200 nM), 2 µl of reverse 39 40 primer (200 nM), and 10 µl of 2X SYBR green super mix was run at a universal cycle 41 42 (95°C for 10 min, 40 cycles at 95°C for 15 s, and 55°C for 60 s) as described in our 43 44 45 previous papers [19-21], using the primers listed in Table 1. β-actin was used as 46 47 housekeeping genes. The relative change in gene expression was calculated using ΔΔCt 48 49 method. 50 51 52 53 54 55 56 57 58 59 60 Page 7 of 26 Acta Ophthalmologica

1 2 3 Experimental Design: 4 5 6 To test the effect of fluorometholone on human conjunctival and corneal epithelial cell 7 8 mucins gene expression, the cells were exposed to fluorometholone 25nM, 50nM, and 9 10 100nM for 12 hours and 24 hours. These doses were selected based on the previously 11 12 published reports demonstrating that fluorometholone activates glucocorticoid receptors 13 14 15 to bring about biological effects at these concentrations in the in vitro experiments (Hos 16 17 et al., 2011; Huynh et al., 2015; Yan et al., 2017). To test whether the effect of 18 19 fluorometholone on mucinsFor gene Peer expression Reviewwas glucocorticoid receptor-mediated, 20 21 22 human conjunctival and corneal epithelial cells were exposed to fluorometholone 23 24 (100nM) alone and in the presence of a glucocorticoid receptor antagonist, mifepristone 25 26 (10µM), for 24 hours. The mifepristone (10µM) alone treatment group was also included. 27 28 29 Pyridine (1µl/ml) was used to dissolve both fluorometholone and mifepristone. 30 31 Therefore, pyridine 1µl/ml was used to serve as the vehicle control. Each experiment was 32 33 conducted in quadruplicate. 34 35 Statistical Analysis: The data is presented as mean + standard error. The data was 36 37 38 checked for normal distribution prior to using one-way ANOVA for calculating the 39 40 statistical significance. The statistical analysis for comparing the effect of various doses 41 42 of fluorometholone to the vehicle control was performed using one-way ANOVA and 43 44 45 Dunnett’s test (Fig. 1 & Fig. 2). The statistical analysis for comparing the effect of 46 47 mifepristone to vehicle control and fluorometholone was performed using one-way 48 49 ANOVA and Tukey’s range test (Fig. 3 & Fig. 4). A p value of <0.05 was considered to 50 51 52 be statistically significant. 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 8 of 26

1 2 3 Results 4 5 6 Dose and time dependent effect of fluorometholone on human conjunctival epithelial 7 8 cell mucins gene expression 9 10 To test the dose- and time-dependent effect of fluorometholone on mucins gene 11 12 expression, we exposed the human conjunctival epithelial cells to three different doses 13 14 15 (25, 50 & 100 nM) of fluorometholone for 12 and 24 hours. Stratified conjunctival cells 16 17 are reported to express membrane-tethered MUC1, MUC4, MUC16 and secreted 18 19 MUC19. FluorometholoneFor caused Peer a significant Review increase in the gene expression of all the 20 21 22 four mucins in the conjunctival epithelial cells. Moreover, the effect of fluorometholone 23 24 on the mucins gene expression was time-dependent and a notably greater increase in the 25 26 gene expression of all the four mucins was observed at 24 hours compared to 12 hours 27 28 29 (Fig. 1). Fluorometholone-mediated increase in MUC1 gene expression at 24 hours was 30 31 dose dependent. The 25 nM dose did not cause any notable increase in MUC1 gene 32 33 expression whereas 50nM and 100nM dose caused a 1.5-fold and 2.5-fold increase 34 35 respectively (Fig. 1A). For MUC4, no increase in gene expression was observed at 25 nM 36 37 38 dose whereas 1.5 folds increase was observed at both 50nM and 100nM dose (Fig. 1B). 39 40 For MUC 16 and MUC19, a 1.5- fold and 2-fold increase respectively, was noted with all 41 42 the three tested doses suggesting that effect of fluorometholone on these mucins may 43 44 45 have already reached a ceiling effect at 25 nM dose (Fig. 1C &D). 46 47 Effect of Fluorometholone on Human Corneal Epithelial Cell Mucins Gene 48 49 Expression 50 51 52 Based on the data obtained from conjunctival epithelial cells, fluorometholone- 53 54 mediated increase in mucins expression was significantly higher at 24-hour time point 55 56 57 58 59 60 Page 9 of 26 Acta Ophthalmologica

1 2 3 compared to the 12-hours. Therefore, the effect of fluorometholone on corneal epithelial 4 5 6 cell mucins gene expression was tested only at 24-hour time point. The effect of 7 8 fluorometholone on corneal epithelial cell mucin gene expression was dose-dependent, 9 10 with a significantly higher increase in gene expression observed at 100 nM dose 11 12 compared to the 50 nM dose. A statistically significant 1.4-fold, 1.8-fold and 2.5-fold 13 14 15 increase in the gene expression of MUC 1 & 4, MUC16 and 19 respectively, was 16 17 observed in the corneal epithelial cells treated with 100 nM fluorometholone (Fig. 2). 18 19 Effect of mifepristoneFor on fluorometholone-mediated Peer Review mucins gene expression 20 21 22 To test whether the fluorometholone-mediated increase in conjunctival and corneal 23 24 epithelial cell mucins gene expression was mediated through activation of glucocorticoid 25 26 receptors, we exposed human conjunctival and corneal epithelial cells to fluorometholone 27 28 29 in presence of mifepristone, a glucocorticoid receptor antagonist. As is evident from Fig. 30 31 3, mifepristone prevented fluorometholone-dependent increase in the gene expression of 32 33 mucins 1, 4, 16, and 19 in conjunctival epithelial cells suggesting that the 34 35 fluorometholone-mediated increase in mucins gene expression was indeed mediated 36 37 38 through activation of glucocorticoid receptors. Additionally, mifepristone treatment alone 39 40 did not affect the gene expression of conjunctival epithelial cell mucins (Fig. 3). 41 42 Mifepristone also markedly attenuated fluorometholone-mediated increase in mucins 43 44 45 1, 4, 16, and 19 gene expression in the corneal epithelial cells (Fig. 4). It is worthwhile to 46 47 note that mifepristone treatment alone also caused a significant decrease (0.5 to 0.75 fold) 48 49 in the gene expression of MUC1 & 4 compared to control cells. In vitro cultures of 50 51 52 human corneal epithelial cells require hydrocortisone, a glucocorticoid receptor agonist, 53 54 for their growth and survival. It is likely that the hydrocortisone may have caused an 55 56 57 58 59 60 Acta Ophthalmologica Page 10 of 26

1 2 3 increase in the baseline mucins 1& 4 gene expression in control corneal epithelial cells, 4 5 6 which is antagonized by mifepristone. 7 8 Effect of Fluorometholone and Mifepristone on Human Conjunctival and Corneal 9 10 Epithelial Cell Morphology and Viability 11 12 To examine the possibility of cellular toxicity or phenotypic changes by the tested 13 14 15 doses of the pharmacologic agents, we exposed the human conjunctival and corneal 16 17 epithelial cells to fluorometholone 50nM and 100nM, and mifepristone 10µM for 24 18 19 hours. No visible decreaseFor in cell Peer density was Review observed with the tested dose of 20 21 22 fluorometholone or mifepristone. Additionally, no apparent change in the morphology of 23 24 the conjunctival (Fig. 5) or corneal (Fig. 6) epithelial cells was noted with either 25 26 fluorometholone or mifepristone compared to the control cells. Cellular viability was 27 28 29 determined using trypan blue and neither fluorometholone nor mifepristone caused any 30 31 decrease in cell viability of the stratified human conjunctival and corneal epithelial cells 32 33 (data not shown). 34 35 Discussion 36 37 38 Artificial tears are the first-line therapy for the management of dry eye. However, a large 39 40 proportion of patients with moderate to severe dry eye do not respond adequately to tear 41 42 supplementation, thus necessitating the use of alternative therapeutic interventions 43 44 45 including the use of topical corticosteroids. Clinical studies have consistently shown the 46 47 beneficial effect of short-term topical corticosteroids on the signs and symptoms of 48 49 moderate to severe dry eye (Avunduk et al., 2003; Pflugfelder et al., 2004; Pinto-Fraga 50 51 52 et al., 2016 Cutolo et al., 2017). Corticosteroids are potent anti-inflammatory agents. 53 54 Inflammation is known to play a significant role in the pathogenesis of dry eye disease. 55 56 57 58 59 60 Page 11 of 26 Acta Ophthalmologica

1 2 3 Therefore, the therapeutic effect of corticosteroids on dry eye is largely attributed to their 4 5 6 anti-inflammatory properties. Results of the present study demonstrate that 7 8 fluorometholone increased the gene expression of mucins in cultured human conjunctival 9 10 and corneal epithelial cells. Mucins are vital for ocular surface hydration and lubrication. 11 12 Theoretically, an increase in ocular surface mucins will retard the evaporation of tear 13 14 15 film, thus alleviating the symptoms of dry eye. Therefore, our results suggest that an 16 17 increase in ocular surface mucins can be an additional possible mechanism contributing 18 19 to the therapeutic effectsFor of corticosteroids Peer inReview dry eye. To our knowledge, this is the first 20 21 22 report demonstrating that fluorometholone, a weak glucocorticoid receptor agonist, can 23 24 cause an increase in the gene expression of mucins in the conjunctival and corneal 25 26 epithelial cells. Our data is further supported by a recent case study of two patients 27 28 29 suffering from dry eye who upon topical fluorometholone instillation showed an increase 30 31 in ocular gene expression of MUC16 (Machida et al., 2016). Previous studies have tested 32 33 the effect of corticosteroids on mucins gene expression in the airway epithelial cells and 34 35 36 cancer cells (Imai et al., 2004; Bai et al., 2007; Milara et al., 2017). The corticosteroid- 37 38 mediated changes in mucin gene expression have been shown to differ from one cell type 39 40 to another. The modulation has also been shown to be differential for various mucins. For 41 42 example, corticosteroids have been shown to cause an increase in MUC1 expression in 43 44 45 human prostate cancer cells, but an opposite effect is observed in the gastric mucosa 46 47 (Okazaki et al., 1998; Imai et al., 2004). Further, corticosteroid treatment caused an 48 49 increase in expression of membrane-tethered mucins but decreased the expression of 50 51 52 secreted mucins in nasal polyp epithelium (Martínez-Antón et al., 2008). In the present 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 12 of 26

1 2 3 study, an increase in all the ocular surface mucins was observed with fluorometholone in 4 5 6 both the corneal as well as conjunctival epithelial cells. 7 8 Glucocorticoids exert their effects by binding to glucocorticoid receptors that are 9 10 ubiquitously present in most mammalian cells, including corneal and conjunctival 11 12 epithelial cells. The glucocorticoid receptors exist as a complex regulated by chaperones 13 14 15 such as heat shock proteins 90 and 70. Activation of the glucocorticoid receptor by ligand 16 17 initiates a conformational change, releasing the chaperone molecules and creating a 18 19 glucocorticoid-glucocorticoidFor receptorPeer (GC-GCR) Review complex. The GC-GCR complex 20 21 22 translocates to the nucleus and can bring about gene expression changes by directly 23 24 binding to glucocorticoid response elements located on the promoter regions of 25 26 corticosteroid-regulated genes. Mucins genes have been shown to contain a 27 28 29 glucocorticoid response element in their promoter (Chen et al., 2006; Chen et al., 2012). 30 31 Our results demonstrate that the fluorometholone-mediated modulation of ocular mucins 32 33 is glucocorticoid receptor mediated because the effects are completely blocked by 34 35 mifepristone, a glucocorticoid receptor antagonist. Our results are further supported by 36 37 38 the earlier reports demonstrating that fluorometholone activates glucocorticoid receptors 39 40 to mediate its biological effects at the concentrations that were used for this study (Hos et 41 42 al., 2011; Huynh et al., 2015; Yan et al., 2017). Fluorometholone for ophthalmic use is 43 44 45 available as 0.1% suspension which is equivalent to 2600 M concertation. No published 46 47 data is available regarding the concertation of fluorometholone that is achieved in the tear 48 49 50 film of patients after the topical instillation of 0.1% (or 2600M) suspension. Just like 51 52 any other eye drops, fluorometholone ocular suspension will undergo dilution and rapid 53 54 drainage upon instillation. However, it can be contemplated that despite dilution or 55 56 57 58 59 60 Page 13 of 26 Acta Ophthalmologica

1 2 3 drainage, a tear film concentration of ≥ 100 nM that was used in this study, is very likely 4 5 6 to be present in the tear film upon instillation of 2600 M fluorometholone suspension. 7 8 Therefore, it may be suggested that ocular instillation of fluorometholone will achieve 9 10 enough tear film concentration to cause modulation of ocular surface mucins in patients 11 12 13 as was observed in the cultures of conjunctival and corneal epithelial cells exposed to 100 14 15 nM dose of fluorometholone used in this study. Our assumption is directly supported by a 16 17 recent report showing that ocular instillation of 0.1 % fluorometholone caused an 18 19 For Peer Review 20 increase in ocular gene expression of MUC16 in the dry eye patients (Machida et al., 21 22 2016). 23 24 In conclusion, our data shows that fluorometholone increases the gene expression of 25 26 27 MUC1, MUC4, MUC16 and MUC 19 in the conjunctival and corneal epithelial cells 28 29 through activation of glucocorticoid receptors. The increased expression of mucins can be 30 31 an additional possible mechanism contributing to the beneficial effects of 32 33 fluorometholone in dry eye in addition to its well-known anti-inflammatory effects. 34 35 36 37 38 Acknowledgements: Authors thank Dr. Ilene Gipson, Professor of Ophthalmology, 39 40 Harvard Medical School, Boston, USA, for providing human conjunctival epithelial cells 41 42 43 and Dr. James V Jester, Professor, Ophthalmology, School of Medicine, University of 44 45 California, Irvine, USA, for providing human corneal epithelial cells. 46 47 References: 48 49 50 Ablamowicz AF, Nichols JJ (2016): Ocular surface membrane-associated mucins. Ocul 51 52 Surf 14: 331-341. 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 14 of 26

1 2 3 Argüeso P, Balaram M, Spurr-Michaud S, Keutmann HT, Dana MR, Gipson IK (2002): 4 5 6 Decreased levels of the goblet cell mucin MUC5AC in tears of patients with Sjögren 7 8 syndrome. Invest Ophthalmol Vis Sci. 43: 1004-1011. 9 10 Avunduk AM, Avunduk MC, Varnell ED, Kaufman HE (2003): The comparison of 11 12 efficacies of topical corticosteroids and non-steroidal anti-inflammatory drops on dry eye 13 14 15 patients: a clinical and immunocytochemical study. Am J Ophthalmol 136: 593-602. 16 17 Bai CH, Song SY, Kim YD (2007): Effect of glucocorticoid on the MUC4 gene in nasal 18 19 polyps. LaryngoscopeFor 117: 2169-2173. Peer Review 20 21 22 Cavet ME, Volhejn S, Harrington KL, Zhang JZ (2013): Anti-allergic effects of 23 24 mapracorat, a novel selective glucocorticoid receptor agonist, in human conjunctival 25 26 fibroblasts and epithelial cells. Mol Vis 19: 1515-1525. 27 28 29 Chen Y, Nickola TJ, DiFronzo NL, Colberg-Poley AM, Rose MC (2006): 30 31 Dexamethasone-mediated repression of MUC5AC gene expression in human lung 32 33 epithelial cells. Am J Respir Cell Mol Biol 34: 338-347. 34 35 Chen Y, Watson AM, Williamson CD et al (2012): Glucocorticoid receptor and histone 36 37 38 deacetylase-2 mediate dexamethasone-induced repression of MUC5AC gene expression. 39 40 Am J Respir Cell Mol Biol 47: 637-644. 41 42 Corfield AP, Carroll D, Myerscough N, Probert CS (2001): Mucins in the gastrointestinal 43 44 45 tract in health and disease. Front Biosci 6: D1321-D357. 46 47 Cutolo CA, Barabino S, Bonzano C, Traverso CE (2017). The use of topical 48 49 corticosteroids for treatment of dry eye syndrome. Ocul Immunol Inflamm: 1-10. 50 51 52 Dogru M, Tsubota K (2011): Pharmacotherapy of dry eye. Expert Opin Pharmacother 12: 53 54 325-334. 55 56 57 58 59 60 Page 15 of 26 Acta Ophthalmologica

1 2 3 Gipson IK (2001): Mucins of the human endocervix. Front Biosci 6: D1245-D1255. 4 5 6 Gipson IK (2004): Distribution of mucins at the ocular surface. Exp Eye Res 78: 379- 7 8 388. 9 10 Gipson IK (2016): Goblet cells of the conjunctiva: A review of recent findings. Prog 11 12 Retin Eye Res 54: 49-63. 13 14 15 Gipson IK, Argüeso P (2003): Role of mucins in the function of the corneal and 16 17 conjunctival epithelia. Int Rev Cytol 231: 1-49. 18 19 Gipson IK, Spurr-MichaudFor S, ArgüesoPeer P, Tisdale Review A, Ng TF, Russo CL (2003): Mucin 20 21 22 gene expression in immortalized human corneal-limbal and conjunctival epithelial cell 23 24 lines. Invest Ophthalmol Vis Sci 44: 2496-2506. 25 26 Gipson IK, Spurr-Michaud SJ, Senchyna M, Ritter R 3rd, Schaumberg D (2011): 27 28 29 Comparison of mucin levels at the ocular surface of postmenopausal women with and 30 31 without a history of dry eye. Cornea 30: 1346-1352. 32 33 Govindarajan B, Gipson IK (2010): Membrane-tethered mucins have multiple functions 34 35 on the ocular surface. Exp Eye Res 90: 655-663. 36 37 38 Hos D, Saban DR, Bock F, Regenfuss B, Onderka J, Masli S, Cursiefen C (2011). 39 40 Suppression of inflammatory corneal lymphangiogenesis by application of topical 41 42 corticosteroids. Arch Ophthalmol 129: 445-452. 43 44 45 Huynh TP, Barwe SP, Lee SJ, et al. (2015): Glucocorticoids suppress renal cell 46 47 carcinoma progression by enhancing Na,K-ATPase beta-1 subunit expression. PLoS One 48 49 10: e0122442. 50 51 52 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 16 of 26

1 2 3 Imai M, Hwang HY, Norris JS, Tomlinson S (2004): The effect of dexamethasone on 4 5 6 human mucin 1 expression and antibody-dependent complement sensitivity in a prostate 7 8 cancer cell line in vitro and in vivo. Immunology 111: 291-297. 9 10 Jee D, Park M, Lee HJ, Kim MS, Kim EC (2015): Comparison of treatment with 11 12 preservative-free versus preserved sodium hyaluronate 0.1% and fluorometholone 0.1% 13 14 15 eyedrops after cataract surgery in patients with preexisting dry-eye syndrome. J Cataract 16 17 Refract Surg. 41: 756-763. 18 19 Kadmiel M, JanoshaziFor A, Xu X, Peer Cidlowski JAReview (2016): Glucocorticoid action in human 20 21 22 corneal epithelial cells establishes roles for corticosteroids in wound healing and barrier 23 24 function of the eye. Exp Eye Res 152: 10-33. 25 26 Lin T, Gong L (2015): Topical fluorometholone treatment for ocular dryness in patients 27 28 29 with Sjögren syndrome: a randomized clinical trial in China. Medicine 94: e551. 30 31 Machida Y, Shoji J, Harada N, Inada N (2016): Two patients with dry eye disease 32 33 followed up using an expression assay of ocular surface mucin. Case Rep Ophthalmol 7: 34 35 208-215. 36 37 38 Martínez-Antón A, de Bolós C, Alobid I, et al. (2008): Corticosteroid therapy increases 39 40 membrane-tethered while decreases secreted mucin expression in nasal polyps. Allergy 41 42 63: 1368-1376. 43 44 45 Milara J, Morell A, Ballester B, Armengot M, Morcillo E, Cortijo J (2017): MUC4 46 47 impairs the anti-inflammatory effects of corticosteroids in patients with chronic 48 49 rhinosinusitis with nasal polyps. J Allergy Clin Immunol 139: 855-862. 50 51 52 53 54 55 56 57 58 59 60 Page 17 of 26 Acta Ophthalmologica

1 2 3 Okazaki K, Chiba T, Hajiro K (1998): Downregulation of gastric mucin gene expression 4 5 6 and its biosynthesis by dexamethasone in the human. J Clin Gastroenterol 27 Suppl 1: 7 8 S91-S96. 9 10 Pflugfelder SC (2004): Antiinflammatory therapy for dry eye. Am J Ophthalmol 137: 11 12 337-342. 13 14 15 Pflugfelder SC, Maskin SL, Anderson B, et al. (2004): A randomized, double-masked, 16 17 placebo-controlled, multicenter comparison of loteprednol etabonate ophthalmic 18 19 suspension, 0.5%, andFor placebo forPeer treatment Reviewof keratoconjunctivitis sicca in patients with 20 21 22 delayed tear clearance. Am J Ophthalmol 138: 444–457. 23 24 Pinto-Fraga J, López-Miguel A, González-García MJ, et al. (2016): Topical 25 26 fluorometholone protects the ocular surface of dry eye patients from desiccating stress: A 27 28 29 randomized controlled clinical trial. Ophthalmology 123: 141-153. 30 31 Ramamoorthy P, Nichols JJ (2008): Mucins in contact lens wear and dry eye conditions. 32 33 Optom Vis Sci. 85: 631-642. 34 35 Robertson DM, Li L, Fisher S, Pearce VP et al. (2005): Characterization of growth and 36 37 38 differentiation in a telomerase-immortalized human corneal epithelial cell line. Invest 39 40 Ophthalmol Vis Sci. 46: 470-478. 41 42 Rose MC, Voynow JA (2006): Respiratory tract mucin genes and mucin glycoproteins in 43 44 45 health and disease. Physiol Rev 86: 245-278. 46 47 Stephens DN, McNamara NA (2015): Altered mucin and glycoprotein expression in dry 48 49 eye disease. Optom Vis Sci. 2015 92: 931-938. 50 51 52 Yan H, Wang Y, Shen S, Wu Z, Wan P (2017): Corticosteroids effects on LPS-induced 53 54 rat inflammatory keratocyte cell model. PLoS One 12: e0176639. 55 56 57 58 59 60 Acta Ophthalmologica Page 18 of 26

1 2 3 Fig. 1. Real-time PCR quantification of MUC1 (A), MUC4 (B), MUC16 (C) and MUC19 4 5 6 (D) gene expression in human conjunctival epithelial cells treated with fluorometholone 7 8 (25, 50, and 100nM) for 12 hours and 24 hours. 9 10 * p < 0.05 compared to control 11 12 13 14 15 Fig. 2. Real-time PCR quantification of MUC1 (A), MUC4 (B), MUC16 (C) and MUC19 16 17 (D) gene expression in human corneal epithelial cells treated with fluorometholone (50 18 19 and 100nM) for 24 hours.For Peer Review 20 21 22 * p < 0.05 compared to control 23 24 25 26 Fig. 3. Effect of mifepristone (10µM), a glucocorticoid receptor antagonist, on 27 28 29 fluorometholone (100nM)-mediated changes in the gene expression of MUC1 (A), 30 31 MUC4 (B), MUC16 (C) and MUC19 (D) in human conjunctival epithelial cells 32 33 quantified using real-time PCR at 24 hours after the treatment. 34 35 * p < 0.05 compared to control; θ p < 0.05 compared to fluorometholone 36 37 38 39 40 Fig. 4. Effect of mifepristone (10µM), a glucocorticoid receptor antagonist, on 41 42 fluorometholone (100nM)-mediated changes in the gene expression of MUC1 (A), 43 44 45 MUC4 (B), MUC16 (C) and MUC19 (D) in human corneal epithelial cells quantified 46 47 using real-time PCR at 24 hours after the treatment. 48 49 * p < 0.05 compared to control; θ p < 0.05 compared to fluorometholone 50 51 52 53 54 55 56 57 58 59 60 Page 19 of 26 Acta Ophthalmologica

1 2 3 Fig. 5. Representative phase contrast microscopy images of human conjunctival epithelial 4 5 6 cells treated with mifepristone (10µM) or fluorometholone (50nM and 100nM) for 24 7 8 hours. 9 10 11 12 Fig. 6. Representative phase contrast microscopy images of human corneal epithelial 13 14 15 cells treated with mifepristone (10µM) or fluorometholone (50nM and 100nM) for 24 16 17 hours. 18 19 For Peer Review 20 21 22 23 24 Table 1: Sequence and accession number of primers used for the gene expression 25 26 quantification of human mucins using real time PCR 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 20 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Peer Review 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Fig. 1. Real-time PCR quantification of MUC1 (A), MUC4 (B), MUC16 (C) and MUC19 (D) gene expression in 38 human conjunctival epithelial cells treated with fluorometholone (25, 50, and 100nM) for 12 hours and 24 39 hours. 40 * p < 0.05 compared to control 41 42 43 199x191mm (300 x 300 DPI) 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 21 of 26 Acta Ophthalmologica

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Peer Review 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Fig. 2. Real-time PCR quantification of MUC1 (A), MUC4 (B), MUC16 (C) and MUC19 (D) gene expression in human corneal epithelial cells treated with fluorometholone (50 and 100nM) for 24 hours. 40 * p < 0.05 compared to control 41 42 43 199x199mm (300 x 300 DPI) 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 22 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Peer Review 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Fig. 3. Effect of mifepristone (10µM), a glucocorticoid receptor antagonist, on fluorometholone (100nM)- 41 mediated changes in the gene expression of MUC1 (A), MUC4 (B), MUC16 (C) and MUC19 (D) in human 42 conjunctival epithelial cells quantified using real-time PCR at 24 hours after the treatment. 43 * p < 0.05 compared to control; θ p < 0.05 compared to fluorometholone 44 45 46 199x210mm (300 x 300 DPI) 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 23 of 26 Acta Ophthalmologica

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Peer Review 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Fig. 4. Effect of mifepristone (10µM), a glucocorticoid receptor antagonist, on fluorometholone (100nM)- 42 mediated changes in the gene expression of MUC1 (A), MUC4 (B), MUC16 (C) and MUC19 (D) in human 43 corneal epithelial cells quantified using real-time PCR at 24 hours after the treatment. 44 * p < 0.05 compared to control; θ p < 0.05 compared to fluorometholone 45 46 47 199x217mm (300 x 300 DPI) 48 49 50 51 52 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 24 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Peer Review 20 21 22 23 24 25 26 27 28 Fig. 5. Representative phase contrast microscopy images of human conjunctival epithelial cells treated with 29 mifepristone (10µM) or fluorometholone (50nM and 100nM) for 24 hours. 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 25 of 26 Acta Ophthalmologica

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Peer Review 20 21 22 23 24 25 26 27 28 Fig. 6. Representative phase contrast microscopy images of human corneal epithelial cells treated with 29 mifepristone (10µM) or fluorometholone (50nM and 100nM) for 24 hours. 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Acta Ophthalmologica Page 26 of 26

1 2 Mucin Forward Primer Reverse Primer Accession # 3 MUC1 5’AGCACCGACTACTACCA 5’CAGATCCTGGCCTGAACTT NM_002456.5 4 AGA3’ AAT3’ 5 MUC4 5’GTGACCTCAGCAGTCTC 5’AGAAGCTGATGTGTCTTGG NM_018406 6 AATAA3’ ATAG3’ 7 MUC16 5’GAAGCCTCCAGAGATAC 5’TATGGCCTGGGATAGGAGA NM_024690.2 8 AAAGG3’ ATA3’ 9 MUC19 5’CCTCTCTGGATGTGGGA 5’CAGATCCACTGCCTCTAGT HM801842.1 10 TTTG3’ TTC3’ 11 β-actin 5’GGACCTGACTGACTACC 5’CGTAGCACAGCTTCTCCTT NM_001101.3 12 TCAT3’ AAT3’ 13 14 15 16 17 18 19 For Peer Review 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60