BASIC INVESTIGATION
Effects of Androgen on Ultrastructure of Corneal Epithelium and Function of the Tear Film in BALB/c Mice Li Li, PhD, MD, Qianyan Kang, PhD, MD, Shuangmei Wang, MD, and Xuan Zheng, MD
Purpose: The aim of this study was to investigate the effects of androgen on the ultrastructure of corneal epithelium and function of the tear film in a mouse model.
Methods: Healthy adult male BALB/c mice were randomly apportioned to normal control, sham-operated, or orchiectomy groups. In the orchiectomy group, 4 subgroups with treatment [blank group, dihydrotestosterone (DHT) eye drop group, DHT injection group, and dimethyl sulfoxide (DMSO) eye drop group] were further established at 3 weeks after orchiectomy. Tear production, tear break-up time, and corneal fluorescein staining were evaluated in all groups at multiple time points. Serum androgen concentrations were measured by competitive radioimmunoassay, and ultrastructure of corneal epithelial cells was examined by transmission electron microscopy. Results: Before orchiectomy, the mean serum concentration of androgen was 43.4 ng/mL, which decreased to approximately 0 ng/mL at 1 week after orchiectomy. Significantly less tear production, shorter tear break-up time, higher corneal fluorescein staining score, shorter and flattened corneal epithelial microvilli, and looser intercellular desmosomes were observed in mice after orchiectomy. DHT supplements increased serum androgen levels, and some of the tear film functions and morphological features of microvilli and desmosomes were gradually close to those at baseline in the DHT injection group. Conclusions: Persistent corneal staining, decreased tear production, short break-up time, and unhealthy corneal ultrastructure were observed in mice that received orchiectomy for at least 8 weeks. A mouse that received orchiectomy could be used as a dry eye model. Exogenous DHT supplement improved corneal epithelial ultrastructure and tear film function in this model. Key Words: dry eye, androgen, ultrastructure of corneal epithelium (Cornea 2015;34:334–341)
D
ry eye is the most commonly encountered disease in ophthalmology, which affects 5%1 to more than 35%2
MATERIALS AND METHODS
Received for publication July 9, 2014; revision received September 15, 2014; accepted September 16, 2014. Published online ahead of print December 19, 2014. From the Department of Ophthalmology of First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China. Supported by a grant from the National Science Foundation of China. The authors have no conflicts of interest to disclose. Reprints: Li Li, PhD, MD, Department of Ophthalmology of First Affiliated Hosptial of Xi’an Jiaotong University, 277 West YanTa Road, Xi’an, Shaanxi Province, China 710061 (e-mail: [email protected]). Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
334
of people at various ages worldwide. Asian people have higher prevalence of dry eye than whites.2,3 Prevalence of dry eye disease rises dramatically with increasing age, with a significantly higher incidence in postmenopausal women.4–7 Not only perimenopausal stages in women but also hormonal diseases that can seriously result in dryness on the ocular surface.8–10 However, currently, all treatments of dry eye cannot completely alleviate symptoms and stop the progress of the disease. Better understanding of the relationship between sex hormones and dry eye should be favorable. Postmenopausal estrogen therapy has been proved to be one of the risk factor for dry eye4 and was found to be related with increased prevalence of dry eye.11 Moreover, decreased androgen levels in men and women with aging12 or congenital androgen insufficiency13 was associated with dry eye. Therefore, some researchers have speculated that androgen, rather than estrogen, is one protective factor for dry eye. Alterations of tear film are the most common pathological changes in dry eye.14 The tear film comprises 3 layers: lipid layer, aqueous layer, and glycocalyx or mucin layer.15 Many studies have shown that transmembrane mucins and integrity of the corneal epithelium ultrastructure are of great significance in maintaining tear film stability and reducing the occurrence of dry eye.16 Previous studies have demonstrated that androgen can regulate and control differentiation and secretion functions of tear ducts and meibomian gland,17,18 but few studies have specifically addressed the effect of androgen on the mucin layer and corneal epithelial ultrastructure. We conjectured that the decreased androgen level may influence the function of the tear film and the ultrastructure of the corneal epithelium, which leads to dry eye. If so, an exogenous androgen supplement may be a viable treatment for dry eye. We investigated this hypothesis by using a mouse model of dry eye induced through orchiectomy, to determine a theoretical basis for the treatment of dry eye using androgen.
| www.corneajrnl.com
Animals The Committee for Ethical Animal Experiments of the First Affiliated Hospital of Xi’an Jiaotong University approved the experimental protocol. The research was performed in strict accordance with the Association of Research in Vision and Ophthalmology statement for the Use of Animals in Ophthalmic and Vision Research. Healthy male BALB/c mice weighing 20 6 2 g and 6 to 8 weeks old were obtained from the animal center at Medical Cornea Volume 34, Number 3, March 2015
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
School of the Xi’an Jiaotong University. Mice were reared under standard laboratory conditions.
Experimental Models We divided healthy male BALB/c mice into 3 groups: normal control group, sham-operated group, and orchiectomy group (ORCH). According to treatment options, ORCH group was randomly separated into 4 subgroups as follows: blank ORCH, dihydrotestosterone (DHT) eye drop group (ORCHDHTdrop), DHT injection group (ORCH-DHTinj), and dimethyl sulfoxide (DMSO) eye drop control group (ORCH-DMSOcontrol). DHT or DMSO treatment was started at 3 weeks after orchiectomy. In mice receiving orchiectomy, the animals were anesthetized by intraperitoneal injection of 10% chloral hydrate (0.4 mL/kg). The skin of the mice was cut, bilateral testes were resected, and incisions were sewn up. After surgery, 0.1 mL penicillin G was injected once a day for 3 days to prevent infection. In the sham-operated group, mice received the same procedure as in the ORCH group, except for resection of the bilateral testes. In the ORCH-DHTdrop group, 5 mL of 0.03% DHT eye drops (DHT dry powder dissolved in 0.1% DMSO) was used for each eye 4 times per day. In the ORCH-DHTinj group, 50 mL of DHT solution (0.04%, DHT dry powder dissolved in 0.1% DMSO) was injected subcutaneously once daily. In the ORCH-DMSOcontrol group, 5 mL of 0.1% DMSO eye drops was used 4 times daily for each eye as vehicle control.
Serum Androgen Concentration After death, eyes of mice were removed and more than 500 mL of blood was collected. The serum was transferred to a 1-mL centrifuge tube and stored in a 220°C refrigerator. Testosterone levels were detected by competition radioimmunoassay (Testo, Tianjin, China).
Effects of Androgen on the Cornea and Tear Film
fluorescein. Corneal fluorescein staining was classified using a grading system that was modified by Yang et al,19 based on the size of corneal staining. The cornea was divided into 4 quadrants, the extent of staining in each quadrant was classified as: grade 0, no punctuate staining; grade 1, ,5 staining points; grade 2, staining points more than grade 1 but less than grade 3; and grade 3, massive staining. The total grade for corneal fluorescein staining for 1 eye was the sum of all 4 quadrants and could be potentially any number from 0 to 12.
Transmission Electron Microscopy
Corneas were fixed in 4% glutaraldehyde solution for 2 to 4 hours, washed with 1/15 M phosphate buffer, postfixed in osmium tetroxide for 1 to 2 hours, washed again, and dehydrated in an acetone series. Specimens of the cornea were embedded, in accordance with the standard method. The embedding blocks were sliced to 50 nm. After baking and dyeing, the ultrastructure of corneal epithelial cells was observed under an electron microscope (Hitachi H7650; Hitachi, Tokyo, Japan).
Statistical Analysis Statistical analyses were performed using SPSS for Windows version 13.0. Measurements were presented as mean 6 SD. One-way analysis of variance was used for androgen level, tear production, TBUT, and corneal scores analysis, followed by either Bonferroni or Dunnett t test to compare data at several time points within the same group. P , 0.05 was considered statistically significant.
RESULTS Serum Androgen Concentration
Tear break-up time (TBUT) was evaluated under cobalt blue light after application of 5 mL of 1% sodium fluorescein (Sigma-Aldrich, St Louis, MO) by a micropipette into both eyes of the nonanesthetized mouse. The average of 3 observations of the time required for dry spots to appear on the corneal surface after blinking was recorded.
Serum androgen concentrations (ng/mL) in the normal control group, sham-operated group, and ORCH group (baseline, before orchiectomy) were 43.0 6 6.7, 41.1 6 6.1, and 43.4 6 5.1, respectively. There were no significant differences among them (P . 0.05). There were also no significant differences between androgen concentrations of normal control and sham-operated groups at each time point. However, 1 week after orchiectomy in the ORCH group, mean serum androgen concentration decreased nearly to 0 ng/mL, a statistically significant change compared with baseline (43.4 ng/mL; P , 0.05). After 1 week of DHT treatments of castrated mice, by either eye drops (ORCH-DHTdrop) or injection (ORCHDHTinj), serum androgen concentrations were much higher compared with those of untreated ORCH mice, and those of ORCH-DHTinj group were higher than baseline values (Table 1). Androgen levels of ORCH-DHTdrop group remained lower than those at baseline. Serum androgen concentrations in the ORCH-DMSOcontrol group were only slightly higher (0–3 ng/ mL) and significantly lower than those of the ORCH-DHTtreated groups at each time point (P , 0.05).
Corneal Fluorescein Staining
Tear Function
Tear Production Tear production was measured with phenol redimpregnated cotton thread (no systemic or topical anesthesia). The thread (PCOT test; Testo) was held with jeweler forceps and applied to the ocular surface in the lateral canthus for 60 seconds. The wet length of the thread was measured in millimeters using the scale on the cotton thread. In this study, we observed the comprehensive and integrated effect on tear production, but we did not measure porphyrin levels in the tear samples.
Tear Break-up Time
Corneal fluorescein staining was evaluated under cobalt blue light and imaged after application of 5 mL of 1% sodium
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
Significant differences in tear production and TBUT were not detected among normal control group, sham-operated www.corneajrnl.com |
335
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
Li et al
TABLE 1. Serum Androgen Concentrations in the ORCH group ORCH-DHTdrop
Baseline Postorchiectomy Treatment duration
ORCH-DHTinj
Weeks
Serum Androgen Concentration, ng/mL
P*
— 1 2 1 3 5 9
43.4 6 5.1 0 0 28.5 6 5.0*† 32.1 6 4.3*† 31.1 6 6.3*† 31.1 6 5.4*†
— — — 0.020 0.014 0.016 0.014
P†
Serum Androgen Concentration, ng/mL
P*
P†
— — — ,0.001 ,0.001 ,0.001 ,0.001
43.7 6 5.1 0 0 53.1 6 9.6† 79.9 6 4.2*† 57.6 6 7.1*† 56.9 6 3.8*†
— — — 0.061 ,0.001 0.038 0.042
— — — ,0.001 ,0.001 ,0.001 ,0.001
*Compared with baseline. †Compared with 2 weeks after castration.
group, and baseline of orchiectomy group. No corneal staining was observed in these groups (Table 2). Two weeks after orchiectomy, tear production and TBUT in the ORCH group were significantly lower than those at baseline (Table 2). Corneal fluorescein staining scores were significantly higher after 2 weeks in these mice and continued to increase in severity until at 8 weeks when massive fluorescein staining was observed in each quadrant of the cornea (P , 0.05; Fig. 1). To determine the effect of exogenous DHT on tear function, we used different treatments for 9 weeks starting 3 weeks after orchiectomy. In the ORCH-DHTdrop group, phenol red line length at 3, 5, and 9 weeks was almost that of the baseline value (Table 3). TBUT and corneal fluorescein scores (Table 3 and Fig. 1) recovered only slightly and remained significantly worse than at baseline. In the ORCH-DHTinj group, treatment was associated with near recovery of baseline tear production at 3, 5, and 9 weeks and corneal staining scores at 5 and 9 weeks, but TBUT remained significantly lower throughout the experimental period. Tear functions in the ORCH-DMSOcontrol group did not improve (Table 3 and Fig. 1).
(Fig. 2) were long and thick; the intercellular desmosomes anchored tightly (Fig. 3). After orchiectomy, the epithelial microvilli gradually became shorter, flattened, and fewer. Eight weeks after orchiectomy, some fissure zones were found among corneal epithelial layers. Until 6 weeks after orchiectomy, quantity and morphological characteristics of intercellular desmosomes remained normal. However, at 8 weeks after orchiectomy, separation of desmosomes was observed in the superficial layer of corneal epithelial cells (Fig. 3). After 1 week of treatment, corneal epithelial microvilli in mice of ORCH-DHTdrop and ORCH-DMSO groups were not significantly different compared with the untreated ORCH group 2 weeks after orchiectomy (Fig. 2). In mice of ORCHDHTinj group, after 1 week of treatment, quantity of corneal epithelial microvilli increased slightly compared with that of nontreated ORCH group. With continued treatment, corneal epithelial microvilli increased gradually in the ORCH-DHTinj and ORCH-DHTdrop groups, and near recovered to baseline levels in the ORCH-DHTinj group at 9 weeks of treatment. The epithelial microvilli in mice of the ORCH-DMSOcontrol group remained shorter and looser, and fissure zones were observed (Fig. 2). Corneal epithelial intercellular desmosomes remained in regular arrangement and normal density in mice of ORCHDHTinj and ORCH-DHTdrop groups. However, separation of some desmosomes at 9 weeks after treatment was obvious in the superficial layer of corneal epithelial cells in the ORCHDMSOcontrol group (Fig. 3).
Transmission Electron Microscopy In the normal control group, sham-operated group, and baseline of orchiectomy group, corneal epithelial microvilli TABLE 2. Tear Functions in Different Groups
Phenol Red Line, mm P
Weeks Normal control Sham operated Baseline of orchiectomy group Postorchiectomy without treatment
— — — 1 2 4 6 8
TBUT, s
5.0 4.9 5.3 4.6 4.3 3.4 2.6 2.1
6 6 6 6 6 6 6 6
0.6 0.4 1.0 0.8 0.6* 0.6* 0.3* 0.4*
Corneal Staining, Score P
0.928 0.963 — 0.157 0.034 0.001 ,0.001 ,0.001
68.2 62.5 68.3 58.7 47.2 39.5 32.7 31.0
6 6 6 6 6 6 6 6
2.4 3.8 12.9 10.0 7.6* 3.4* 3.9* 2.4*
0.389 0.682 — 0.087 0.041 0.002 0.001 ,0.001
P
0.7 2.2 2.7 4.7 7.8
0 0 0 6 0.8 6 1.5* 6 0.5* 6 0.8* 6 0.8*
— — — 0.354 0.004 ,0.001 ,0.001 ,0.001
*Compared with baseline.
336
| www.corneajrnl.com
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
Effects of Androgen on the Cornea and Tear Film
FIGURE 1. Corneal fluorescein staining in different groups at several time points. The green color on the cornea shows fluorescein staining. A, Baseline. B, Orchiectomy 1 week. C, Orchiectomy 2 weeks. D, Orchiectomy 4 weeks. E, Orchiectomy 6 weeks. F, Orchiectomy 8 weeks. G–I, ORCHDHTdrop 1 week, ORCH-DHTinj 1 week, ORCH-DMSOcontrol 1 week. J–L, ORCHDHTdrop 3 weeks, ORCH-DHTinj 3 weeks, ORCH-DMSOcontrol 3 weeks. M–O, ORCHDHTdrop 5 weeks, ORCH-DHTinj 5 weeks, ORCH-DMSOcontrol 5 weeks. P–R, ORCHDHTdrop 9 weeks, ORCH-DHTinj 9 weeks, ORCH-DMSOcontrol 9 weeks.
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.corneajrnl.com |
337
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
Li et al
TABLE 3. Tear Functions in ORCH Group With DHT or DMSO Treatments Postorchiectomy, wk Phenol red line length, mm/min ORCH-DHTdrop P* P† ORCH-DHTinj P* P† DMSO control P* P† TBUT, s ORCH-DHTdrop P* P† ORCH-DHTinj P* P† DMSO control P* P† Corneal staining, score ORCH-DHTdrop P* P† ORCH-DHTinj P* P† DMSO control P* P†
Treatment Time, wk
Baseline
1
2
1
3
5
9
5.3 6 1.0 — — 5.3 6 1.0 — — 5.3 6 1.0 — —
4.6 6 0.9 0.185 0.468 4.6 6 0.8 0.185 0.468 4.6 6 0.8 0.185 0.468
4.3 6 0.6* 0.042 1.000 4.3 6 0.6* 0.042 1.000 4.3 6 0.6* 0.042 1.000
4.2 6 1.0* 0.025 0.826 4.3 6 0.9* 0.024 0.897 3.6 6 0.7*† ,0.001 0.030
4.4 6 1.1 0.081 0.765 4.6 6 0.8 0.122 0.390 3.2 6 0.7*† ,0.001 ,0.001
4.4 6 1.8 0.073 0.803 5.2 6 1.2† 0.867 0.027 3.1 6 0.6*† ,0.001 ,0.001
4.6 6 1.4 0.177 0.483 5.4 6 1.3† 0.819 0.010 2.3 6 0.4*† ,0.001 ,0.001
68.3 6 12.9 — — 68.3 6 12.9 — — 68.3 6 12.9 — —
58.7 6 10.0 0.086 0.089 58.7 6 10.0 0.086 0.089 58.7 6 10.0 0.086 0.089
47.2 6 7.6* 0.044 1.000 47.2 6 7.6* 0.044 1.000 47.2 6 7.6* 0.044 1.000
32.6 6 8.7*† ,0.001 ,0.001 36.8 6 10.9*† ,0.001 0.012 31.6 6 6.5*† ,0.001 ,0.001
31.1 6 5.6*† ,0.001 ,0.001 33.3 6 5.5*† ,0.001 0.001 27.7 6 7.8*† ,0.001 ,0.001
34.4 6 4.4*† ,0.001 0.001 41.3 6 6.8* ,0.001 0.151 27.6 6 6.1*† ,0.001 ,0.001
35.8 6 9.7*† ,0.001 0.002 53.9 6 12.8* 0.001 0.098 21.5 6 5.4*† ,0.001 ,0.001
0 — — 0 — — 0 — —
0.7 6 0.8† 0.335 0.032 0.7 6 0.8† 0.335 0.032 0.7 6 0.8† 0.335 0.032
2.2 6 1.5* 0.002 1.000 2.2 6 1.5* 0.002 1.000 2.2 6 1.5* 0.002 1.000
2.9 6 2.2* ,0.001 0.278 2.7 6 2.3* ,0.001 0.379 3.2 6 1.9* ,0.001 0.179
3.6 6 2.5* ,0.001 0.056 1.8 6 1.7* 0.003 0.464 3.2 6 1.9* ,0.001 0.179
2.5 6 1.8* ,0.001 0.629 0.9 6 0.9† 0.109 0.030 3.9 6 2.1*† ,0.001 0.020
1.7 6 1.7* 0.018 0.469 0.8 6 1.3† 0.145 0.021 4.6 6 2.8*† ,0.001 0.002
*Compared with baseline. †Compared with 2 weeks after castration.
DISCUSSION An ideal animal model is very helpful for understanding the mechanism of a disease and making progress toward diagnosis or treatment. In this study, we performed orchiectomy for BALB/c mice to mimic human androgen deficiency and successfully set up a dry eye mouse model. Similar dry eye animal models through orchiectomy were established in rat20 and rabbit.21 Recently, 5a-reductase inhibitor was reported to be used to mimic androgen deficiency and to establish a dry eye rat model.22 In this study, we observed a significant reduction in tear production, shorter break-up time (BUT), higher corneal staining scores, and looser corneal microvilli after orchiectomy in mice. Our findings were in agreement with the results of previous studies that showed low androgen level influences the function of lacrimal23 and meibomian glands,17 thereby causing low tear production and instability of the tear film. Except the above-reported results, shorter, flattened, and absent corneal epithelial microvilli were observed, and intercellular desmosomes were not tightly connected at the late stage after orchiectomy and caused
a nonintact corneal barrier. DHT supplement repaired the corneal epithelium ultrastructure in dry eye mice. As we know, corneal epithelial cell–ocular mucus–tear film interaction is a key mechanism of corneal defense,24,25 and flattened and shorter microvillus projections of corneal epithelium were found in dry eye animals.26,27 Mucin anchoring on the microvilli and decreased mucin expressions were correlated to the symptomatic severity of dry eye.28 Although the regulation of androgen on mucin of ocular surface has not been well illuminated, androgen-dependent regulation of human MUC1 mucin expression in breast and prostatic cell lines had been proved.29 Therefore, according to our present results and previous studies, we speculate that androgen may influence the corneal epithelial ultrastructure, cell–ocular mucus– tear film interaction, and tear film homeostasis. Moreover, the characteristics of the dry eye mouse model we used in this study are as follows: (1) our dry eye mouse model is very easy to set up without special agent or equipment; (2) it mimics extreme androgen deficiency (from baseline 43.4 ng/mL to nearly 0 ng/mL); (3) the mouse dry eye
338
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
| www.corneajrnl.com
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
Effects of Androgen on the Cornea and Tear Film
FIGURE 2. Corneal epithelial microvilli in different groups at several time points. Images from transmission electron microscopy (scale bar, 1 mm). Black arrows show corneal epithelial microvilli. A, Baseline. B, Orchiectomy 1 week. C, Orchiectomy 2 weeks. D, Orchiectomy 4 weeks. E, Orchiectomy 6 weeks. F, Orchiectomy 8 weeks. G–I, ORCH-DHTdrop 1 week, ORCH-DHTinj 1 week, ORCH-DMSOcontrol 1 week. J–L, ORCH-DHTdrop 3 weeks, ORCH-DHTinj 3 weeks, ORCH-DMSOcontrol 3 weeks. M–O, ORCH-DHTdrop 5 weeks, ORCH-DHTinj 5 weeks, ORCH-DMSOcontrol 5 weeks. P–R, ORCH-DHTdrop 9 weeks, ORCH-DHTinj 9 weeks, ORCH-DMSOcontrol 9 weeks.
model was set up without directly damaging ocular tissues and influencing the ocular surface homeostasis; (4) changes of tear film function in our mouse model persisted for at least 8 Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
weeks; (5) mouse model is more convenient for experimental research because there are more antibodies or agents available for mouse than for other species. However, some investigators www.corneajrnl.com |
339
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Li et al
Cornea Volume 34, Number 3, March 2015
FIGURE 3. Corneal epithelial intercellular desmosomes in different groups at several time points. Images from transmission electron microscopy (scale bar, 500 nm). White arrows show corneal epithelial intercellular desmosomes. ★ shows separation of desmosomes. A, Baseline. B, Orchiectomy 1 week. C, Orchiectomy 2 weeks. D, Orchiectomy 4 weeks; E. Orchiectomy 6 weeks; F. Orchiectomy 8 weeks. G–I, ORCH-DHTdrop 1 week, ORCH-DHTinj 1 week, ORCH-DMSOcontrol 1 week. J–L, ORCH-DHTdrop 3 weeks, ORCH-DHTinj 3 weeks, ORCH-DMSOcontrol 3 weeks. M–O, ORCH-DHTdrop 5 weeks, ORCH-DHTinj 5 weeks, ORCH-DMSOcontrol 5 weeks. P–R, ORCH-DHTdrop 9 weeks, ORCH-DHTinj 9 weeks, ORCH-DMSOcontrol 9 weeks.
have concluded that mice may be inappropriate models to understand human lacrimal and meibomian glands,30,31 because gene expressions in both tissues are very different
in humans and mice, and the activity is species specific. However, we consider that it is too early to deny mice as a dry eye animal model. For a dry eye study, researchers
340
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
| www.corneajrnl.com
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
can choose different animal models according to different research purposes (biological process or gene expression) or target tissues (eg, cornea, conjunctiva, and lacrimal or meibomian glands). To date, many studies used mice as their dry eye models and obtained meaningful results.32,33 Tear production, BUT, and corneal staining are recognized as important and typical indexes for dry eye diagnosis by the International Dry Eye Workshop.34 In our study, we found that tear functions were damaged after orchiectomy. After DHT supplement, regardless of eye drops or injection, tear production, BUT, and corneal staining improved to some extent. Especially in the DHT injection group, tear production and corneal fluorescein staining scores were close to baseline levels. We speculated that better outcomes of the DHT injection group were due to the higher serum androgen level after DHT injection than that after administration of eye drops. Results from our study support the fact that clinical application of methyltestosterone may alleviate dry eye symptoms in postmenopausal women.35 Therefore, androgen-based medicines or eye drops may repair corneal damage and maintain stability of the tear film. It could be a promising treatment for dry eye in the future. To conclude, a dry eye model could be established in a mouse through orchiectomy. All 3 layers of tear film were disrupted, which was evidenced by changes in tear production, BUT, and corneal epithelial ultrastructure. Although the specific mechanism of how androgen regulates tear film function and affects corneal epithelial ultrastructure and/or mucin is not clear, the dry eye mouse model helped us to reveal the relationship between dry eye and androgen, and will allow therapeutic development for dry eye by using androgen. Our future direction of the study will explore the signal pathway involving the regulation of tear film function through androgen. REFERENCES 1. McCarty CA, Bansal AK, Livingston PM, et al. The epidemiology of dry eye in Melbourne, Australia. Ophthalmology. 1998;105:1114–1119. 2. Lin PY, Tsai SY, Cheng CY, et al. Prevalence of dry eye among an elderly Chinese population in Taiwan: the Shihpai Eye Study. Ophthalmology. 2003;110:1096–1101. 3. Lee AJ, Lee J, Saw SM, et al. Prevalence and risk factors associated with dry eye symptoms: a population based study in Indonesia. Br J Ophthalmol. 2002;86:1347–1351. 4. The epidemiology of dry eye disease: report of the epidemiology subcommittee of the international. Dry Eye WorkShop (2007). Ocul Surf. 2007;5:93–107. 5. Schaumberg DA, Sullivan DA, Buring JE, et al. Prevalence of dry eye syndrome among US women. Am J Ophthalmol. 2003;136:318–326. 6. Schaumberg DA, Dana R, Buring JE, et al. Prevalence of dry eye disease among US men: estimates from the physicians’ health studies. Arch Ophthalmol. 2009;127:763–768. 7. Mertzanis P, Abetz L, Rajagopalan K, et al. The relative burden of dry eye in patients’ lives: comparisons to a U.S. normative sample. Invest Ophthalmol Vis Sci. 2005;46:46–50. 8. Hashemi H, Khabazkhoob M, Kheirkhah A, et al. Prevalence of dry eye syndrome in an adult population. Clin Experiment Ophthalmol. 2014;42: 242–248. 9. Scuderi G, Contestabile MT, Gagliano C, et al. Effects of phytoestrogen in postmenopausal women with dry eye syndrome: a randomized clinical trial. Can J Ophthalmol. 2012;47:489–492. 10. Rocha EM, Mantelli F, Nominato LF, et al. Hormones and dry eye syndrome: an update on what we do and don’t know. Curr Opin Ophthalmol. 2013;24:348–355.
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
Effects of Androgen on the Cornea and Tear Film
11. Schaumbeg DA, Buring JE, Sullivan DA, et al. Hormone replacement therapy and dry eye syndrome. JAMA. 2001;286:2114–2119. 12. Sullivan BD, Evans JE, Dana MR, et al. Influence of aging on the polar and neutral lipid profiles in human meibomian gland secretions. Arch Ophthalmol. 2006;124:1286–1292. 13. Sullivan BD, Evans JE, Cermak JM, et al. Complete androgen insensitivity syndrome: effect on human meibomian gland secretion. Arch Ophthalmol. 2002;120:1686–1699. 14. Sweeney DF, Millar TJ, Raju SR. Tear film stability: a review. Exp Eye Res. 2013;117:28–38. 15. Gipson IK. Distribution of mucins at the ocular surface. Exp Eye Res. 2004;78:379–388. 16. Hori Y, Sugiyama H, Soma T, et al. Expression of membrane-associated mucins in cultivated human oral mucosal epithelial cells. Cornea. 2007; 26(9 suppl 1):S65–S69. 17. Schirra F, Suzuki T, Richards SM, et al. Androgen control of gene expression in the mouse meibomian gland. Invest Ophthalmol Vis Sci. 2005;46:3666–3675. 18. Krenzer KL, Dana MR, Ullman MD, et al. Effect of androgen deficiency on the human meibomian gland and ocular surface. J Clin Endocrinol Metab. 2000;85:4874–4882. 19. Yang CQ, Sun W, Gu YS. A clinical study of the efficacy of topical corticosteroids on dry eye. J Zhejiang Univ Sci B. 2006;7:675–678. 20. Peng QH, Yao XL, Wu QL, et al. Effects of extract of Buddleja officinalis eye drops on androgen receptors of lacrimal gland cells of castrated rats with dry eye. Int J Ophthalmol. 2010;3:43–48. 21. Yao XL, Peng QH, Peng J, et al. Effects of extract of Buddleja officinalis on partial inflammation of lacrimal gland in castrated rabbits with dry eye. Int J Ophthalmol. 2010;3:114–119. 22. Singh S, Moksha L, Sharma N, et al. Development and evaluation of animal models for sex steroid deficient dry eye. J Pharmacol Toxicol Methods. 2014;70:29–34. 23. Sullivan DA. Sex and sex steroid influence on dry eye syndromes. In: Pflugfelder S, Beuerman R, Stern ME, eds. Dry Eye and Ocular Surface Disease. New York City, NY: Marcel Dekker; 2004:165–190. 24. Ashutosh S. Energetics of corneal epithelial cell-ocular mucus–tear film interactions: some surface-chemical pathways of corneal defense. Biophys Chem. 1993;47:87–99. 25. Foster JB, Lee WB. The tear film. In: Edward JH, Mark JM, Barry WL, eds. Ocular Surface Disease: Cornea, Conjunctiva and Tear Film. New York City, NY: Elsevier Saunders; 2013:17–20. 26. Van Horn DL, Schutten WH, Hyndiuk RA, et al. Xerophthalmia in vitamin A-deficient rabbits. Clinical and ultrastructural alterations in the cornea. Invest Ophthalmol Vis Sci. 1980;19:1067–1079. 27. Kern TJ, Erb HN, Schaedler JM, et al. Scanning electron microscopy of experimental keratoconjunctivitis sicca in dogs: cornea and bulbar conjunctiva. Vet Pathol. 1988;25:468–474. 28. Zhang J, Yan X, Li H. Analysis of the correlations of mucins, inflammatory markers, and clinical tests in dry eye. Cornea. 2013; 32:928–932. 29. Mitchell S, Abel P, Madaan S, et al. Androgen-dependent regulation of human MUC1 mucin expression. Neoplasia. 2002;4:9–18. 30. Ozyildirim AM, Wistow GJ, Gao J, et al. The lacrimal gland transcriptome is an unusually rich source of rare and poorly characterized gene transcripts. Invest Ophthalmol Vis Sci. 2005;46:1572–1580. 31. Sullivan DA, RahimiDarabad R, Liu S, et al. Are mice relevant models for human meibomian gland research? ARVO Web site. Available at: https://arvo2014.abstractcentral.com. Accessed May 4, 2014. 32. Shim J, Park C, Lee HS, et al. Change in prostaglandin expression levels and synthesizing activities in dry eye disease. Ophthalmology. 2012;119: 2211–2219. 33. Xiao X, Luo P, Zhao H, et al. Amniotic membrane extract ameliorates benzalkonium chloride-induced dry eye in a murine model. Exp Eye Res. 2013;115:31–40. 34. Methodologies and monitor dry eye disease: report of the diagnostic and methodology subcommittee of the international. Dry Eye WorkShop (2007). Ocul Surf. 2007;5:108–152. 35. Scott G, Yiu SC, Wasilewski D, et al. Combined esterified estrogen and methyltestosterone treatment for dry eye syndrome in postmenopausal women. Am J Ophthalmol. 2005;139:1109–1110.
www.corneajrnl.com |
341
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Effects of Androgen on Ultrastructure of Corneal Epithelium and Function of the Tear Film in BALB/c Mice Li Li, PhD, MD, Qianyan Kang, PhD, MD, Shuangmei Wang, MD, and Xuan Zheng, MD
Purpose: The aim of this study was to investigate the effects of androgen on the ultrastructure of corneal epithelium and function of the tear film in a mouse model.
Methods: Healthy adult male BALB/c mice were randomly apportioned to normal control, sham-operated, or orchiectomy groups. In the orchiectomy group, 4 subgroups with treatment [blank group, dihydrotestosterone (DHT) eye drop group, DHT injection group, and dimethyl sulfoxide (DMSO) eye drop group] were further established at 3 weeks after orchiectomy. Tear production, tear break-up time, and corneal fluorescein staining were evaluated in all groups at multiple time points. Serum androgen concentrations were measured by competitive radioimmunoassay, and ultrastructure of corneal epithelial cells was examined by transmission electron microscopy. Results: Before orchiectomy, the mean serum concentration of androgen was 43.4 ng/mL, which decreased to approximately 0 ng/mL at 1 week after orchiectomy. Significantly less tear production, shorter tear break-up time, higher corneal fluorescein staining score, shorter and flattened corneal epithelial microvilli, and looser intercellular desmosomes were observed in mice after orchiectomy. DHT supplements increased serum androgen levels, and some of the tear film functions and morphological features of microvilli and desmosomes were gradually close to those at baseline in the DHT injection group. Conclusions: Persistent corneal staining, decreased tear production, short break-up time, and unhealthy corneal ultrastructure were observed in mice that received orchiectomy for at least 8 weeks. A mouse that received orchiectomy could be used as a dry eye model. Exogenous DHT supplement improved corneal epithelial ultrastructure and tear film function in this model. Key Words: dry eye, androgen, ultrastructure of corneal epithelium (Cornea 2015;34:334–341)
D
ry eye is the most commonly encountered disease in ophthalmology, which affects 5%1 to more than 35%2
MATERIALS AND METHODS
Received for publication July 9, 2014; revision received September 15, 2014; accepted September 16, 2014. Published online ahead of print December 19, 2014. From the Department of Ophthalmology of First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China. Supported by a grant from the National Science Foundation of China. The authors have no conflicts of interest to disclose. Reprints: Li Li, PhD, MD, Department of Ophthalmology of First Affiliated Hosptial of Xi’an Jiaotong University, 277 West YanTa Road, Xi’an, Shaanxi Province, China 710061 (e-mail: [email protected]). Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved.
334
of people at various ages worldwide. Asian people have higher prevalence of dry eye than whites.2,3 Prevalence of dry eye disease rises dramatically with increasing age, with a significantly higher incidence in postmenopausal women.4–7 Not only perimenopausal stages in women but also hormonal diseases that can seriously result in dryness on the ocular surface.8–10 However, currently, all treatments of dry eye cannot completely alleviate symptoms and stop the progress of the disease. Better understanding of the relationship between sex hormones and dry eye should be favorable. Postmenopausal estrogen therapy has been proved to be one of the risk factor for dry eye4 and was found to be related with increased prevalence of dry eye.11 Moreover, decreased androgen levels in men and women with aging12 or congenital androgen insufficiency13 was associated with dry eye. Therefore, some researchers have speculated that androgen, rather than estrogen, is one protective factor for dry eye. Alterations of tear film are the most common pathological changes in dry eye.14 The tear film comprises 3 layers: lipid layer, aqueous layer, and glycocalyx or mucin layer.15 Many studies have shown that transmembrane mucins and integrity of the corneal epithelium ultrastructure are of great significance in maintaining tear film stability and reducing the occurrence of dry eye.16 Previous studies have demonstrated that androgen can regulate and control differentiation and secretion functions of tear ducts and meibomian gland,17,18 but few studies have specifically addressed the effect of androgen on the mucin layer and corneal epithelial ultrastructure. We conjectured that the decreased androgen level may influence the function of the tear film and the ultrastructure of the corneal epithelium, which leads to dry eye. If so, an exogenous androgen supplement may be a viable treatment for dry eye. We investigated this hypothesis by using a mouse model of dry eye induced through orchiectomy, to determine a theoretical basis for the treatment of dry eye using androgen.
| www.corneajrnl.com
Animals The Committee for Ethical Animal Experiments of the First Affiliated Hospital of Xi’an Jiaotong University approved the experimental protocol. The research was performed in strict accordance with the Association of Research in Vision and Ophthalmology statement for the Use of Animals in Ophthalmic and Vision Research. Healthy male BALB/c mice weighing 20 6 2 g and 6 to 8 weeks old were obtained from the animal center at Medical Cornea Volume 34, Number 3, March 2015
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
School of the Xi’an Jiaotong University. Mice were reared under standard laboratory conditions.
Experimental Models We divided healthy male BALB/c mice into 3 groups: normal control group, sham-operated group, and orchiectomy group (ORCH). According to treatment options, ORCH group was randomly separated into 4 subgroups as follows: blank ORCH, dihydrotestosterone (DHT) eye drop group (ORCHDHTdrop), DHT injection group (ORCH-DHTinj), and dimethyl sulfoxide (DMSO) eye drop control group (ORCH-DMSOcontrol). DHT or DMSO treatment was started at 3 weeks after orchiectomy. In mice receiving orchiectomy, the animals were anesthetized by intraperitoneal injection of 10% chloral hydrate (0.4 mL/kg). The skin of the mice was cut, bilateral testes were resected, and incisions were sewn up. After surgery, 0.1 mL penicillin G was injected once a day for 3 days to prevent infection. In the sham-operated group, mice received the same procedure as in the ORCH group, except for resection of the bilateral testes. In the ORCH-DHTdrop group, 5 mL of 0.03% DHT eye drops (DHT dry powder dissolved in 0.1% DMSO) was used for each eye 4 times per day. In the ORCH-DHTinj group, 50 mL of DHT solution (0.04%, DHT dry powder dissolved in 0.1% DMSO) was injected subcutaneously once daily. In the ORCH-DMSOcontrol group, 5 mL of 0.1% DMSO eye drops was used 4 times daily for each eye as vehicle control.
Serum Androgen Concentration After death, eyes of mice were removed and more than 500 mL of blood was collected. The serum was transferred to a 1-mL centrifuge tube and stored in a 220°C refrigerator. Testosterone levels were detected by competition radioimmunoassay (Testo, Tianjin, China).
Effects of Androgen on the Cornea and Tear Film
fluorescein. Corneal fluorescein staining was classified using a grading system that was modified by Yang et al,19 based on the size of corneal staining. The cornea was divided into 4 quadrants, the extent of staining in each quadrant was classified as: grade 0, no punctuate staining; grade 1, ,5 staining points; grade 2, staining points more than grade 1 but less than grade 3; and grade 3, massive staining. The total grade for corneal fluorescein staining for 1 eye was the sum of all 4 quadrants and could be potentially any number from 0 to 12.
Transmission Electron Microscopy
Corneas were fixed in 4% glutaraldehyde solution for 2 to 4 hours, washed with 1/15 M phosphate buffer, postfixed in osmium tetroxide for 1 to 2 hours, washed again, and dehydrated in an acetone series. Specimens of the cornea were embedded, in accordance with the standard method. The embedding blocks were sliced to 50 nm. After baking and dyeing, the ultrastructure of corneal epithelial cells was observed under an electron microscope (Hitachi H7650; Hitachi, Tokyo, Japan).
Statistical Analysis Statistical analyses were performed using SPSS for Windows version 13.0. Measurements were presented as mean 6 SD. One-way analysis of variance was used for androgen level, tear production, TBUT, and corneal scores analysis, followed by either Bonferroni or Dunnett t test to compare data at several time points within the same group. P , 0.05 was considered statistically significant.
RESULTS Serum Androgen Concentration
Tear break-up time (TBUT) was evaluated under cobalt blue light after application of 5 mL of 1% sodium fluorescein (Sigma-Aldrich, St Louis, MO) by a micropipette into both eyes of the nonanesthetized mouse. The average of 3 observations of the time required for dry spots to appear on the corneal surface after blinking was recorded.
Serum androgen concentrations (ng/mL) in the normal control group, sham-operated group, and ORCH group (baseline, before orchiectomy) were 43.0 6 6.7, 41.1 6 6.1, and 43.4 6 5.1, respectively. There were no significant differences among them (P . 0.05). There were also no significant differences between androgen concentrations of normal control and sham-operated groups at each time point. However, 1 week after orchiectomy in the ORCH group, mean serum androgen concentration decreased nearly to 0 ng/mL, a statistically significant change compared with baseline (43.4 ng/mL; P , 0.05). After 1 week of DHT treatments of castrated mice, by either eye drops (ORCH-DHTdrop) or injection (ORCHDHTinj), serum androgen concentrations were much higher compared with those of untreated ORCH mice, and those of ORCH-DHTinj group were higher than baseline values (Table 1). Androgen levels of ORCH-DHTdrop group remained lower than those at baseline. Serum androgen concentrations in the ORCH-DMSOcontrol group were only slightly higher (0–3 ng/ mL) and significantly lower than those of the ORCH-DHTtreated groups at each time point (P , 0.05).
Corneal Fluorescein Staining
Tear Function
Tear Production Tear production was measured with phenol redimpregnated cotton thread (no systemic or topical anesthesia). The thread (PCOT test; Testo) was held with jeweler forceps and applied to the ocular surface in the lateral canthus for 60 seconds. The wet length of the thread was measured in millimeters using the scale on the cotton thread. In this study, we observed the comprehensive and integrated effect on tear production, but we did not measure porphyrin levels in the tear samples.
Tear Break-up Time
Corneal fluorescein staining was evaluated under cobalt blue light and imaged after application of 5 mL of 1% sodium
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
Significant differences in tear production and TBUT were not detected among normal control group, sham-operated www.corneajrnl.com |
335
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
Li et al
TABLE 1. Serum Androgen Concentrations in the ORCH group ORCH-DHTdrop
Baseline Postorchiectomy Treatment duration
ORCH-DHTinj
Weeks
Serum Androgen Concentration, ng/mL
P*
— 1 2 1 3 5 9
43.4 6 5.1 0 0 28.5 6 5.0*† 32.1 6 4.3*† 31.1 6 6.3*† 31.1 6 5.4*†
— — — 0.020 0.014 0.016 0.014
P†
Serum Androgen Concentration, ng/mL
P*
P†
— — — ,0.001 ,0.001 ,0.001 ,0.001
43.7 6 5.1 0 0 53.1 6 9.6† 79.9 6 4.2*† 57.6 6 7.1*† 56.9 6 3.8*†
— — — 0.061 ,0.001 0.038 0.042
— — — ,0.001 ,0.001 ,0.001 ,0.001
*Compared with baseline. †Compared with 2 weeks after castration.
group, and baseline of orchiectomy group. No corneal staining was observed in these groups (Table 2). Two weeks after orchiectomy, tear production and TBUT in the ORCH group were significantly lower than those at baseline (Table 2). Corneal fluorescein staining scores were significantly higher after 2 weeks in these mice and continued to increase in severity until at 8 weeks when massive fluorescein staining was observed in each quadrant of the cornea (P , 0.05; Fig. 1). To determine the effect of exogenous DHT on tear function, we used different treatments for 9 weeks starting 3 weeks after orchiectomy. In the ORCH-DHTdrop group, phenol red line length at 3, 5, and 9 weeks was almost that of the baseline value (Table 3). TBUT and corneal fluorescein scores (Table 3 and Fig. 1) recovered only slightly and remained significantly worse than at baseline. In the ORCH-DHTinj group, treatment was associated with near recovery of baseline tear production at 3, 5, and 9 weeks and corneal staining scores at 5 and 9 weeks, but TBUT remained significantly lower throughout the experimental period. Tear functions in the ORCH-DMSOcontrol group did not improve (Table 3 and Fig. 1).
(Fig. 2) were long and thick; the intercellular desmosomes anchored tightly (Fig. 3). After orchiectomy, the epithelial microvilli gradually became shorter, flattened, and fewer. Eight weeks after orchiectomy, some fissure zones were found among corneal epithelial layers. Until 6 weeks after orchiectomy, quantity and morphological characteristics of intercellular desmosomes remained normal. However, at 8 weeks after orchiectomy, separation of desmosomes was observed in the superficial layer of corneal epithelial cells (Fig. 3). After 1 week of treatment, corneal epithelial microvilli in mice of ORCH-DHTdrop and ORCH-DMSO groups were not significantly different compared with the untreated ORCH group 2 weeks after orchiectomy (Fig. 2). In mice of ORCHDHTinj group, after 1 week of treatment, quantity of corneal epithelial microvilli increased slightly compared with that of nontreated ORCH group. With continued treatment, corneal epithelial microvilli increased gradually in the ORCH-DHTinj and ORCH-DHTdrop groups, and near recovered to baseline levels in the ORCH-DHTinj group at 9 weeks of treatment. The epithelial microvilli in mice of the ORCH-DMSOcontrol group remained shorter and looser, and fissure zones were observed (Fig. 2). Corneal epithelial intercellular desmosomes remained in regular arrangement and normal density in mice of ORCHDHTinj and ORCH-DHTdrop groups. However, separation of some desmosomes at 9 weeks after treatment was obvious in the superficial layer of corneal epithelial cells in the ORCHDMSOcontrol group (Fig. 3).
Transmission Electron Microscopy In the normal control group, sham-operated group, and baseline of orchiectomy group, corneal epithelial microvilli TABLE 2. Tear Functions in Different Groups
Phenol Red Line, mm P
Weeks Normal control Sham operated Baseline of orchiectomy group Postorchiectomy without treatment
— — — 1 2 4 6 8
TBUT, s
5.0 4.9 5.3 4.6 4.3 3.4 2.6 2.1
6 6 6 6 6 6 6 6
0.6 0.4 1.0 0.8 0.6* 0.6* 0.3* 0.4*
Corneal Staining, Score P
0.928 0.963 — 0.157 0.034 0.001 ,0.001 ,0.001
68.2 62.5 68.3 58.7 47.2 39.5 32.7 31.0
6 6 6 6 6 6 6 6
2.4 3.8 12.9 10.0 7.6* 3.4* 3.9* 2.4*
0.389 0.682 — 0.087 0.041 0.002 0.001 ,0.001
P
0.7 2.2 2.7 4.7 7.8
0 0 0 6 0.8 6 1.5* 6 0.5* 6 0.8* 6 0.8*
— — — 0.354 0.004 ,0.001 ,0.001 ,0.001
*Compared with baseline.
336
| www.corneajrnl.com
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
Effects of Androgen on the Cornea and Tear Film
FIGURE 1. Corneal fluorescein staining in different groups at several time points. The green color on the cornea shows fluorescein staining. A, Baseline. B, Orchiectomy 1 week. C, Orchiectomy 2 weeks. D, Orchiectomy 4 weeks. E, Orchiectomy 6 weeks. F, Orchiectomy 8 weeks. G–I, ORCHDHTdrop 1 week, ORCH-DHTinj 1 week, ORCH-DMSOcontrol 1 week. J–L, ORCHDHTdrop 3 weeks, ORCH-DHTinj 3 weeks, ORCH-DMSOcontrol 3 weeks. M–O, ORCHDHTdrop 5 weeks, ORCH-DHTinj 5 weeks, ORCH-DMSOcontrol 5 weeks. P–R, ORCHDHTdrop 9 weeks, ORCH-DHTinj 9 weeks, ORCH-DMSOcontrol 9 weeks.
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.corneajrnl.com |
337
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
Li et al
TABLE 3. Tear Functions in ORCH Group With DHT or DMSO Treatments Postorchiectomy, wk Phenol red line length, mm/min ORCH-DHTdrop P* P† ORCH-DHTinj P* P† DMSO control P* P† TBUT, s ORCH-DHTdrop P* P† ORCH-DHTinj P* P† DMSO control P* P† Corneal staining, score ORCH-DHTdrop P* P† ORCH-DHTinj P* P† DMSO control P* P†
Treatment Time, wk
Baseline
1
2
1
3
5
9
5.3 6 1.0 — — 5.3 6 1.0 — — 5.3 6 1.0 — —
4.6 6 0.9 0.185 0.468 4.6 6 0.8 0.185 0.468 4.6 6 0.8 0.185 0.468
4.3 6 0.6* 0.042 1.000 4.3 6 0.6* 0.042 1.000 4.3 6 0.6* 0.042 1.000
4.2 6 1.0* 0.025 0.826 4.3 6 0.9* 0.024 0.897 3.6 6 0.7*† ,0.001 0.030
4.4 6 1.1 0.081 0.765 4.6 6 0.8 0.122 0.390 3.2 6 0.7*† ,0.001 ,0.001
4.4 6 1.8 0.073 0.803 5.2 6 1.2† 0.867 0.027 3.1 6 0.6*† ,0.001 ,0.001
4.6 6 1.4 0.177 0.483 5.4 6 1.3† 0.819 0.010 2.3 6 0.4*† ,0.001 ,0.001
68.3 6 12.9 — — 68.3 6 12.9 — — 68.3 6 12.9 — —
58.7 6 10.0 0.086 0.089 58.7 6 10.0 0.086 0.089 58.7 6 10.0 0.086 0.089
47.2 6 7.6* 0.044 1.000 47.2 6 7.6* 0.044 1.000 47.2 6 7.6* 0.044 1.000
32.6 6 8.7*† ,0.001 ,0.001 36.8 6 10.9*† ,0.001 0.012 31.6 6 6.5*† ,0.001 ,0.001
31.1 6 5.6*† ,0.001 ,0.001 33.3 6 5.5*† ,0.001 0.001 27.7 6 7.8*† ,0.001 ,0.001
34.4 6 4.4*† ,0.001 0.001 41.3 6 6.8* ,0.001 0.151 27.6 6 6.1*† ,0.001 ,0.001
35.8 6 9.7*† ,0.001 0.002 53.9 6 12.8* 0.001 0.098 21.5 6 5.4*† ,0.001 ,0.001
0 — — 0 — — 0 — —
0.7 6 0.8† 0.335 0.032 0.7 6 0.8† 0.335 0.032 0.7 6 0.8† 0.335 0.032
2.2 6 1.5* 0.002 1.000 2.2 6 1.5* 0.002 1.000 2.2 6 1.5* 0.002 1.000
2.9 6 2.2* ,0.001 0.278 2.7 6 2.3* ,0.001 0.379 3.2 6 1.9* ,0.001 0.179
3.6 6 2.5* ,0.001 0.056 1.8 6 1.7* 0.003 0.464 3.2 6 1.9* ,0.001 0.179
2.5 6 1.8* ,0.001 0.629 0.9 6 0.9† 0.109 0.030 3.9 6 2.1*† ,0.001 0.020
1.7 6 1.7* 0.018 0.469 0.8 6 1.3† 0.145 0.021 4.6 6 2.8*† ,0.001 0.002
*Compared with baseline. †Compared with 2 weeks after castration.
DISCUSSION An ideal animal model is very helpful for understanding the mechanism of a disease and making progress toward diagnosis or treatment. In this study, we performed orchiectomy for BALB/c mice to mimic human androgen deficiency and successfully set up a dry eye mouse model. Similar dry eye animal models through orchiectomy were established in rat20 and rabbit.21 Recently, 5a-reductase inhibitor was reported to be used to mimic androgen deficiency and to establish a dry eye rat model.22 In this study, we observed a significant reduction in tear production, shorter break-up time (BUT), higher corneal staining scores, and looser corneal microvilli after orchiectomy in mice. Our findings were in agreement with the results of previous studies that showed low androgen level influences the function of lacrimal23 and meibomian glands,17 thereby causing low tear production and instability of the tear film. Except the above-reported results, shorter, flattened, and absent corneal epithelial microvilli were observed, and intercellular desmosomes were not tightly connected at the late stage after orchiectomy and caused
a nonintact corneal barrier. DHT supplement repaired the corneal epithelium ultrastructure in dry eye mice. As we know, corneal epithelial cell–ocular mucus–tear film interaction is a key mechanism of corneal defense,24,25 and flattened and shorter microvillus projections of corneal epithelium were found in dry eye animals.26,27 Mucin anchoring on the microvilli and decreased mucin expressions were correlated to the symptomatic severity of dry eye.28 Although the regulation of androgen on mucin of ocular surface has not been well illuminated, androgen-dependent regulation of human MUC1 mucin expression in breast and prostatic cell lines had been proved.29 Therefore, according to our present results and previous studies, we speculate that androgen may influence the corneal epithelial ultrastructure, cell–ocular mucus– tear film interaction, and tear film homeostasis. Moreover, the characteristics of the dry eye mouse model we used in this study are as follows: (1) our dry eye mouse model is very easy to set up without special agent or equipment; (2) it mimics extreme androgen deficiency (from baseline 43.4 ng/mL to nearly 0 ng/mL); (3) the mouse dry eye
338
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
| www.corneajrnl.com
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
Effects of Androgen on the Cornea and Tear Film
FIGURE 2. Corneal epithelial microvilli in different groups at several time points. Images from transmission electron microscopy (scale bar, 1 mm). Black arrows show corneal epithelial microvilli. A, Baseline. B, Orchiectomy 1 week. C, Orchiectomy 2 weeks. D, Orchiectomy 4 weeks. E, Orchiectomy 6 weeks. F, Orchiectomy 8 weeks. G–I, ORCH-DHTdrop 1 week, ORCH-DHTinj 1 week, ORCH-DMSOcontrol 1 week. J–L, ORCH-DHTdrop 3 weeks, ORCH-DHTinj 3 weeks, ORCH-DMSOcontrol 3 weeks. M–O, ORCH-DHTdrop 5 weeks, ORCH-DHTinj 5 weeks, ORCH-DMSOcontrol 5 weeks. P–R, ORCH-DHTdrop 9 weeks, ORCH-DHTinj 9 weeks, ORCH-DMSOcontrol 9 weeks.
model was set up without directly damaging ocular tissues and influencing the ocular surface homeostasis; (4) changes of tear film function in our mouse model persisted for at least 8 Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
weeks; (5) mouse model is more convenient for experimental research because there are more antibodies or agents available for mouse than for other species. However, some investigators www.corneajrnl.com |
339
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Li et al
Cornea Volume 34, Number 3, March 2015
FIGURE 3. Corneal epithelial intercellular desmosomes in different groups at several time points. Images from transmission electron microscopy (scale bar, 500 nm). White arrows show corneal epithelial intercellular desmosomes. ★ shows separation of desmosomes. A, Baseline. B, Orchiectomy 1 week. C, Orchiectomy 2 weeks. D, Orchiectomy 4 weeks; E. Orchiectomy 6 weeks; F. Orchiectomy 8 weeks. G–I, ORCH-DHTdrop 1 week, ORCH-DHTinj 1 week, ORCH-DMSOcontrol 1 week. J–L, ORCH-DHTdrop 3 weeks, ORCH-DHTinj 3 weeks, ORCH-DMSOcontrol 3 weeks. M–O, ORCH-DHTdrop 5 weeks, ORCH-DHTinj 5 weeks, ORCH-DMSOcontrol 5 weeks. P–R, ORCH-DHTdrop 9 weeks, ORCH-DHTinj 9 weeks, ORCH-DMSOcontrol 9 weeks.
have concluded that mice may be inappropriate models to understand human lacrimal and meibomian glands,30,31 because gene expressions in both tissues are very different
in humans and mice, and the activity is species specific. However, we consider that it is too early to deny mice as a dry eye animal model. For a dry eye study, researchers
340
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
| www.corneajrnl.com
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Cornea Volume 34, Number 3, March 2015
can choose different animal models according to different research purposes (biological process or gene expression) or target tissues (eg, cornea, conjunctiva, and lacrimal or meibomian glands). To date, many studies used mice as their dry eye models and obtained meaningful results.32,33 Tear production, BUT, and corneal staining are recognized as important and typical indexes for dry eye diagnosis by the International Dry Eye Workshop.34 In our study, we found that tear functions were damaged after orchiectomy. After DHT supplement, regardless of eye drops or injection, tear production, BUT, and corneal staining improved to some extent. Especially in the DHT injection group, tear production and corneal fluorescein staining scores were close to baseline levels. We speculated that better outcomes of the DHT injection group were due to the higher serum androgen level after DHT injection than that after administration of eye drops. Results from our study support the fact that clinical application of methyltestosterone may alleviate dry eye symptoms in postmenopausal women.35 Therefore, androgen-based medicines or eye drops may repair corneal damage and maintain stability of the tear film. It could be a promising treatment for dry eye in the future. To conclude, a dry eye model could be established in a mouse through orchiectomy. All 3 layers of tear film were disrupted, which was evidenced by changes in tear production, BUT, and corneal epithelial ultrastructure. Although the specific mechanism of how androgen regulates tear film function and affects corneal epithelial ultrastructure and/or mucin is not clear, the dry eye mouse model helped us to reveal the relationship between dry eye and androgen, and will allow therapeutic development for dry eye by using androgen. Our future direction of the study will explore the signal pathway involving the regulation of tear film function through androgen. REFERENCES 1. McCarty CA, Bansal AK, Livingston PM, et al. The epidemiology of dry eye in Melbourne, Australia. Ophthalmology. 1998;105:1114–1119. 2. Lin PY, Tsai SY, Cheng CY, et al. Prevalence of dry eye among an elderly Chinese population in Taiwan: the Shihpai Eye Study. Ophthalmology. 2003;110:1096–1101. 3. Lee AJ, Lee J, Saw SM, et al. Prevalence and risk factors associated with dry eye symptoms: a population based study in Indonesia. Br J Ophthalmol. 2002;86:1347–1351. 4. The epidemiology of dry eye disease: report of the epidemiology subcommittee of the international. Dry Eye WorkShop (2007). Ocul Surf. 2007;5:93–107. 5. Schaumberg DA, Sullivan DA, Buring JE, et al. Prevalence of dry eye syndrome among US women. Am J Ophthalmol. 2003;136:318–326. 6. Schaumberg DA, Dana R, Buring JE, et al. Prevalence of dry eye disease among US men: estimates from the physicians’ health studies. Arch Ophthalmol. 2009;127:763–768. 7. Mertzanis P, Abetz L, Rajagopalan K, et al. The relative burden of dry eye in patients’ lives: comparisons to a U.S. normative sample. Invest Ophthalmol Vis Sci. 2005;46:46–50. 8. Hashemi H, Khabazkhoob M, Kheirkhah A, et al. Prevalence of dry eye syndrome in an adult population. Clin Experiment Ophthalmol. 2014;42: 242–248. 9. Scuderi G, Contestabile MT, Gagliano C, et al. Effects of phytoestrogen in postmenopausal women with dry eye syndrome: a randomized clinical trial. Can J Ophthalmol. 2012;47:489–492. 10. Rocha EM, Mantelli F, Nominato LF, et al. Hormones and dry eye syndrome: an update on what we do and don’t know. Curr Opin Ophthalmol. 2013;24:348–355.
Copyright Ó 2014 Wolters Kluwer Health, Inc. All rights reserved.
Effects of Androgen on the Cornea and Tear Film
11. Schaumbeg DA, Buring JE, Sullivan DA, et al. Hormone replacement therapy and dry eye syndrome. JAMA. 2001;286:2114–2119. 12. Sullivan BD, Evans JE, Dana MR, et al. Influence of aging on the polar and neutral lipid profiles in human meibomian gland secretions. Arch Ophthalmol. 2006;124:1286–1292. 13. Sullivan BD, Evans JE, Cermak JM, et al. Complete androgen insensitivity syndrome: effect on human meibomian gland secretion. Arch Ophthalmol. 2002;120:1686–1699. 14. Sweeney DF, Millar TJ, Raju SR. Tear film stability: a review. Exp Eye Res. 2013;117:28–38. 15. Gipson IK. Distribution of mucins at the ocular surface. Exp Eye Res. 2004;78:379–388. 16. Hori Y, Sugiyama H, Soma T, et al. Expression of membrane-associated mucins in cultivated human oral mucosal epithelial cells. Cornea. 2007; 26(9 suppl 1):S65–S69. 17. Schirra F, Suzuki T, Richards SM, et al. Androgen control of gene expression in the mouse meibomian gland. Invest Ophthalmol Vis Sci. 2005;46:3666–3675. 18. Krenzer KL, Dana MR, Ullman MD, et al. Effect of androgen deficiency on the human meibomian gland and ocular surface. J Clin Endocrinol Metab. 2000;85:4874–4882. 19. Yang CQ, Sun W, Gu YS. A clinical study of the efficacy of topical corticosteroids on dry eye. J Zhejiang Univ Sci B. 2006;7:675–678. 20. Peng QH, Yao XL, Wu QL, et al. Effects of extract of Buddleja officinalis eye drops on androgen receptors of lacrimal gland cells of castrated rats with dry eye. Int J Ophthalmol. 2010;3:43–48. 21. Yao XL, Peng QH, Peng J, et al. Effects of extract of Buddleja officinalis on partial inflammation of lacrimal gland in castrated rabbits with dry eye. Int J Ophthalmol. 2010;3:114–119. 22. Singh S, Moksha L, Sharma N, et al. Development and evaluation of animal models for sex steroid deficient dry eye. J Pharmacol Toxicol Methods. 2014;70:29–34. 23. Sullivan DA. Sex and sex steroid influence on dry eye syndromes. In: Pflugfelder S, Beuerman R, Stern ME, eds. Dry Eye and Ocular Surface Disease. New York City, NY: Marcel Dekker; 2004:165–190. 24. Ashutosh S. Energetics of corneal epithelial cell-ocular mucus–tear film interactions: some surface-chemical pathways of corneal defense. Biophys Chem. 1993;47:87–99. 25. Foster JB, Lee WB. The tear film. In: Edward JH, Mark JM, Barry WL, eds. Ocular Surface Disease: Cornea, Conjunctiva and Tear Film. New York City, NY: Elsevier Saunders; 2013:17–20. 26. Van Horn DL, Schutten WH, Hyndiuk RA, et al. Xerophthalmia in vitamin A-deficient rabbits. Clinical and ultrastructural alterations in the cornea. Invest Ophthalmol Vis Sci. 1980;19:1067–1079. 27. Kern TJ, Erb HN, Schaedler JM, et al. Scanning electron microscopy of experimental keratoconjunctivitis sicca in dogs: cornea and bulbar conjunctiva. Vet Pathol. 1988;25:468–474. 28. Zhang J, Yan X, Li H. Analysis of the correlations of mucins, inflammatory markers, and clinical tests in dry eye. Cornea. 2013; 32:928–932. 29. Mitchell S, Abel P, Madaan S, et al. Androgen-dependent regulation of human MUC1 mucin expression. Neoplasia. 2002;4:9–18. 30. Ozyildirim AM, Wistow GJ, Gao J, et al. The lacrimal gland transcriptome is an unusually rich source of rare and poorly characterized gene transcripts. Invest Ophthalmol Vis Sci. 2005;46:1572–1580. 31. Sullivan DA, RahimiDarabad R, Liu S, et al. Are mice relevant models for human meibomian gland research? ARVO Web site. Available at: https://arvo2014.abstractcentral.com. Accessed May 4, 2014. 32. Shim J, Park C, Lee HS, et al. Change in prostaglandin expression levels and synthesizing activities in dry eye disease. Ophthalmology. 2012;119: 2211–2219. 33. Xiao X, Luo P, Zhao H, et al. Amniotic membrane extract ameliorates benzalkonium chloride-induced dry eye in a murine model. Exp Eye Res. 2013;115:31–40. 34. Methodologies and monitor dry eye disease: report of the diagnostic and methodology subcommittee of the international. Dry Eye WorkShop (2007). Ocul Surf. 2007;5:108–152. 35. Scott G, Yiu SC, Wasilewski D, et al. Combined esterified estrogen and methyltestosterone treatment for dry eye syndrome in postmenopausal women. Am J Ophthalmol. 2005;139:1109–1110.
www.corneajrnl.com |
341
Copyright © 201 4 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
