Received 19 June 2009 | Accepted 5 August 2009 | Published 10 August 2009
1Schepens Eye Research Institute and Department of Ophthalmology, Harvard Medical School, Boston, MA; 2Department of Biological Sciences, Virginia Tech, Blacksburg, VA
Correspondence to: David A. Sullivan, Ph.D., Schepens Eye Research Institute, 20 Staniford Street, Boston, MA, 02114; Phone: (617) 912-0287; FAX: (617) 912-0101; email: email@example.com
Purpose: We hypothesize that sex steroids induce sex-specific and/or opposite effects in the lacrimal and meibomian glands and that these actions may influence the prevalence of dry eye syndrome. The objective of this study was to begin to test this hypothesis.
Methods: Lacrimal and meibomian glands were obtained from ovariectomized mice that had been treated with testosterone or control vehicle for 14 days. Samples were processed for the isolation of RNA, and analyzed for differentially expressed mRNAs using CodeLink Bioarrays and quantitative real-time PCR (qPCR) techniques. Data were compared to those obtained following testosterone treatment of orchiectomized mice, as well as after the administration of 17β-estradiol and/or progesterone to ovariectomized mice.
Results: Our findings demonstrate that testosterone regulates the expression of thousands of genes in the lacrimal and meibomian glands of ovariectomized mice. The magnitude and extent of these hormonal effects, which encompassed numerous biological, molecular, and cellular ontologies, was tissue-dependent. Particularly notable was the androgen stimulation of meibomian gland genes related to lipid metabolic pathways, and the suppression of genes associated with keratinization. Many of the genes regulated by testosterone in female tissues were identical to those controlled by androgens in male lacrimal and meibomian glands. However, some genes were modulated in a sex-specific manner. In addition, a number of the androgen-regulated genes in female glands were altered in the opposite direction by 17β-estradiol and/or progesterone.
Conclusions: Our results support our hypothesis that sex steroids may induce sex-specific and/or opposite effects in the lacrimal and meibomian glands. Whether these actions contribute to the prevalence of dry eye remains to be determined.
Sex steroid hormones, such as androgens and estrogens, exert significant effects on almost every cell, tissue, and organ system of the body . However, these hormone actions may not be the same in both males and females. Rather, sex steroid influence may be sex-specific. For example, sex steroids elicit sex-specific effects on many tissues (e.g. hippocampus, spinal cord, vasculature, muscle, and liver) and cells (e.g. preadipocytes, neutrophils, and antigen-presenting cells) [2-9]. Sex steroids may also induce opposite [10-14], and even antagonistic , influences, such as on sebaceous gland activity, cerebrovascular function, microtubule polymerization, chronic allograft nephropathy, autoimmunity, and the cystic fibrosis transmembrane conductance regulator (i.e. a chloride channel). These differential actions of sex steroids in males and females may contribute to a variety of conditions, including stress, atherosclerosis, coronary heart disease, fat distribution, inflammation, autoimmune disease, infection, pain syndromes, stroke, and lung development [2-5,8-10,14].
We hypothesize that sex-specific and/or opposite effects of sex steroids also occur in the lacrimal and meibomian glands and may influence the prevalence of dry eye disease. In support of this hypothesis, investigators have reported that sex steroid actions on the transcription of certain genes, the translation of the corresponding proteins, and the development of paradoxical inflammation in the lacrimal gland are sex-biased [15-21]. Moreover, we have discovered that sex and sex steroid hormones are critical factors in the pathogenesis of dry eye syndromes [22-31], which occur predominantly in women . We have also discovered that androgens may suppress, and estrogens may promote, aqueous-deficient and/or evaporative dry eye [22-26,31]. We hypothesize that these opposing sex steroid actions may involve antagonist effects between androgens and estrogens in the lacrimal and meibomian glands.
However, whether sex steroids exert sex-specific or antagonistic effects in ocular adnexal tissues is unknown. The purpose of this study was to determine whether sex steroids do elicit sex-specific and/or opposite effects in both the lacrimal and meibomian glands. Towards this end, we examined the impact of testosterone on glandular gene expression in ovariectomized mice, and compared these effects to those found in androgen-treated orchiectomized mice. We also compared the testosterone-induced gene alterations in the female lacrimal and meibomian glands to those elicited by 17β-estradiol, progesterone, or both hormones together in these tissues.
Age-matched and young adult BALB/c mice, that were ovariectomized at 8 weeks of age, were obtained from Taconic Laboratories (Germantown, NY). Animals were housed in constant temperature rooms with fixed light/dark intervals of 12 h duration. Ten days after surgery, pellets containing vehicle (cholesterol, methylcellulose, lactose) or testosterone (10 mg) were implanted subcutaneously in the ovariectomized mice. The pellets were purchased from Innovative Research of America (Sarasota, FL) and were designed for the constant release of placebo or physiological amounts of androgen (for a male [33-36]) for 3 weeks. After 14 days of treatment, mice (n=5-6 mice/ condition/ experiment) were sacrificed by CO2 inhalation and exorbital lacrimal and meibomian glands were removed. The meibomian glands were excised from the upper and lower lids under direct visualization with a biomicroscope. This surgical procedure involved making a small incision near the inner corner of the eyelid, separating skin and subcutaneous tissue from the inner to outer aspect of the lid, and then removing skin from the meibomian glands by cutting at the mucocutaneous junction. Following these steps, the palpebral conjunctiva was removed from the meibomian glands, and the glands were dissected from the remaining tissue by starting at the outer lid corner and carefully avoiding an adjacent vein. Tissues were pooled according to group (n=10-12 glands/ sample) and processed for RNA analysis. All studies with mice were approved by the Institutional Animal Care and Use Committee of The Schepens Eye Research Institute (Boston, MA) and adhered to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research.
To determine the influence of testosterone on lacrimal and meibomian gland gene expression, total RNA was extracted from tissues by using TRIzol reagent (Invitrogen Corp., Carlsbad, CA). Lacrimal tissue RNA was further purified with RNAqueous spin columns (Ambion, Austin, Tx). Glandular RNA samples were exposed to RNase-free DNase (Invitrogen), examined spectrophotometrically at 260 nm to determine concentration and analyzed with a RNA 6000 Nano LabChip and an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA) to verify RNA integrity.
Gene expression was assessed by using CodeLink Uniset Mouse I Bioarrays (~10,000 genes; Amersham Biosciences/GE Healthcare, Piscataway, NJ). The RNA samples were processed for CodeLink Bioarray hybridization, according to published procedures . In brief, cDNA was synthesized from RNA (2 µg) with a CodeLink Expression Assay Reagent Kit (Amersham, Piscataway, NJ) and purified with a QIAquick purification kit (Qiagen, Valencia, CA). Samples were then dried, and cRNA was produced with a CodeLink Expression Assay Reagent Kit (Amersham), recovered with an RNeasy kit (Qiagen) and quantified with an UV spectrophotometer. Fragmented, biotin-labeled cRNA was incubated and shaken (300 rpm shaker) on a CodeLink Bioarray at 37 °C for 18 h. After this time period, the Bioarray was washed, exposed to streptavidin-Alexa 647, and scanned by utilizing ScanArray Express software and a ScanArray Express HT scanner (Packard BioScience, Meriden, CT) with the laser set at 635 nm, laser power at 100%, and photomultiplier tube voltage at 60%. Scanned image files were appraised by utilizing CodeLink image and data analysis software (Amersham), which yielded both raw and normalized hybridization signal intensities for each array spot. The spot intensities (~10,000 ) on the microarray image were standardized to a median of 1. Normalized data, with signal intensities exceeding 0.50 for the meibomian gland and 0.75 for the lacrimal gland, were evaluated with GeneSifter.net software (Geospiza, Seattle, WA). This comprehensive program also generated gene ontology and z-score reports. These ontologies encompassed biological processes, molecular functions and cellular components and were organized according to the guidelines of the Gene Ontology Consortium . Gene expression data were analyzed without and with log transformation and statistical evaluation of these data was performed with Student’s t-test (two-tailed, unpaired) with GeneSifter.net software. Genes that were expressed in similar or opposite directions in different experiments were identified by using the GeneSifter.net intersector program (Geospiza). The data from the individual Bioarrays (n=6) are accessible for download through the National Center for Biotechnology Information’s Gene Expression Omnibus (GEO) via series accession number GSE3995.
The differential expression of selected genes was confirmed by using quantitative real-time PCR (qPCR), as previously described [33,39]. In brief, sense and anti-sense primers were designed by utilizing Primer Express Software, version 1.5a (Applied Biosystems, Inc., Foster City, CA). The qPCR reactions were performed with Applied Biosystems’ SYBR Green PCR Master Mix, MicroAmp Optical 96-Well Reaction Plates, ABI PRISM Optical Adhesive Covers and the GeneAmp 7900 HT Sequence Detection System. Gene expression was calculated by using either the Relative Standard Curve Method or the Comparative CT Method, and standardizing levels to that of glyceraldehyde-3-phosphate dehydrogenase or tubulin δ1 mRNA. Dissociation curves were examined to ensure the absence of secondary PCR products.
To determine the influence of testosterone on lacrimal and meibomian gland gene expression, tissues were obtained from placebo- and androgen-treated, ovariectomized mice (n=5-6/group/ experiment) and processed for analysis by utilizing CodeLink Uniset Mouse I Bioarrays. Examination of non- and log-transformed data from 3 separate experiments demonstrated that testosterone significantly altered the expression of thousands of genes in female ocular tissues. Androgen action significantly upregulated 768 genes (e.g. transforming growth factor β1 and β3), and significantly downregulated 1,350 genes (e.g. pancreatic lipase related protein 1; Table 1 and Table 2), in the lacrimal gland. Moreover, testosterone administration increased the expression of 726 genes (e.g. glutathione peroxidase 3) and decreased the activity of 283 genes (e.g. small proline-rich protein 2A) in the meibomian gland (Table 1 and Table 3). The magnitude of these hormonal effects was tissue-dependent. Testosterone treatment induced a ≥10 fold change in the expression of over 45 lacrimal gland genes, and elicited a ≥2 fold response in the activity of 23 meibomian gland genes.
The influence of androgen exposure was particularly notable in its significant stimulation of such ontologies as cell cycle, transferase activity and chromosomes in the lacrimal gland (Table 4), and protein transport, oxidoreductase activity and mitochondria in the meibomian gland (Table 5). Testosterone treatment was also associated with a significant suppression of pathways linked to translation, structural molecule activity and ribosomes in lacrimal tissue, and the immune response, receptor binding and plasma membranes in the meibomian gland. Of particular interest was the significant androgen impact on the expression of genes linked to multiple lipid metabolic processes (e.g. biosynthesis, transport, oxidation) in meibomian tissue (Table 6). Also striking was the testosterone suppression of genes related to keratinization (z score=10.2) and cell fate commitment (z score=3.2) in the meibomian gland (Table 7).
To verify in part the CodeLink Bioarray results, selected genes were analyzed by qPCR. This experimental approach confirmed the effect of testosterone on all tested genes (Table 8).
Surprisingly, the effects of testosterone on the lacrimal and meibomian gland were not entirely tissue-specific. As shown in Table 9, androgen treatment caused analogous changes in the expression of 127 genes in both tissues, and opposite responses in 165 genes. Genes regulated in the same manner included vascular endothelial growth factor A, hyaluronan mediated motility receptor, and lipocalin 3 (Table 10). Those controlled in a different way included chemokine binding protein 2, cholecystokinin, and matrix metalloproteinase 3 (Table 10).
We have previously discovered that testosterone treatment significantly influences the expression of over 2,000 and 1,500 genes in the lacrimal and meibomian glands, respectively, of orchiectomized mice . To examine whether these androgen actions are analogous to those found in ovariectomized mice, we compared the data from the earlier studies (n=10-14 glands/ sample/ experiment; n=3 experiments/ study) with orchiectomized mice to those in the present investigation. All gene expression results were generated with CodeLink Uniset Mouse I Bioarrays and analyzed with GeneSifter software.
Our analysis of log transformed data demonstrated that testosterone administration significantly altered the expression of many of the same genes in both ovariectomized and orchiectomized mice. This hormonal control included 1,072 similar genes in lacrimal glands and 285 similar genes in meibomian glands. The nature (i.e. up- or down-regulation) of the androgen effects was identical in 95.5% of the lacrimal gland genes, and 77.6% of the meibomian gland genes.
To further compare testosterone influence on glandular gene expression in female and male mice, we focused on genes in female tissues with relatively high array intensities, expression ratios (i.e. compared to the placebo-treated group), and significant differences (i.e. versus placebo). These more stringent criteria (lacrimal=p <0.001; intensity >2.0; ratio >1.8; meibomian=p <0.01; intensity >1.0; ratio >1.35) identified 173 (8.7% of the total) and 80 (8.4% of the total) genes in female lacrimal and meibomian glands, respectively. Comparative analyses showed that, with the exception of 4 genes, all lacrimal gland genes were increased or decreased in the same direction in males. The 4 exceptions were genes that were significantly reduced (ratios between 2.5 to 3.6 fold) by testosterone in female mice, but were not lower following androgen exposure in male lacrimal tissues (Table 11). In the meibomian gland the expression of 2 genes, that were significantly decreased (ratios=1.4 fold) in female tissues, were significantly increased (ratios=1.3 fold) in males (Table 11). Another 13 genes in the female meibomian gland, that showed significant expression differences (ratios between 1.4 to 1.8 fold) after testosterone exposure, were not influenced in the same direction in males (Table 11). Overall, these comparisons did not reveal any gene that was unique to female lacrimal or meibomian glands.
Additional comparisons demonstrated that many of the androgen effects on biological process, molecular function and cellular component ontologies in lacrimal and meibomian glands were the same in both females and males (data not shown).
We have found that 17β-estradiol and/or progesterone administration for 2 weeks significantly alters the expression of hundreds of genes in the lacrimal and meibomian glands of ovariectomized mice [41,42] (Table 12). To determine whether these sex steroid actions are unique, or the same as, or opposite to, those elicited by testosterone exposure, we compared the data from our previous investigations (n=14 glands/ sample/ experiment; n=3 experiments/ study) [41,42] to those in the current study. The ages of the BALB/c mice used in all experiments were similar. In addition, all gene expression results were obtained with CodeLink Uniset Mouse I Bioarrays and analyzed with GeneSifter software.
As shown in Table 12 and Table 13, the majority of genes regulated by testosterone, 17β-estradiol and/or progesterone in the lacrimal and meibomian glands were unique (i.e. hormone-specific). Treatment with 17β-estradiol, progesterone, or both sex steroids in combination significantly influenced less than 6.4% of genes controlled by androgens (Table 13). Conversely, testosterone significantly influenced less than 23.3% of estrogen- and progestin-regulated genes (Table 13).
However, although the total number of common genes was limited, the nature of the hormone response to testosterone, as compared to 17β-estradiol, progesterone or both hormones together, was typically different. Between 62.3 to 68.8% of the sex steroid effects (i.e. testosterone versus other hormones) in the lacrimal gland were in the opposite direction. Moreover, between 45.2 to 54.0% of the hormone responses in the meibomian gland were in the opposite direction. Examples of estrogen- and progestin-regulated genes that were modulated in a similar or opposite manner relative to testosterone are shown in Table 14 and Table 15.
Our findings demonstrate that testosterone regulates the expression of thousands of genes in the lacrimal and meibomian glands of ovariectomized mice. The magnitude and extent of these hormonal effects, which encompassed numerous biological, molecular and cellular ontologies, was tissue-dependent. Particularly notable was the androgen stimulation of meibomian gland genes related to lipid metabolic pathways, and the suppression of genes associated with keratinization. Many of the genes regulated by testosterone in female tissues were identical to those controlled by androgens in male lacrimal and meibomian glands. However, some genes were modulated in a sex-specific manner. In addition, a number of the androgen-regulated genes in female glands were altered in the opposite direction by 17β-estradiol and/or progesterone. Overall, our results support our hypothesis that sex steroids may induce sex-specific and opposite effects in the lacrimal and meibomian glands.
Our observation that testosterone influences the expression of multiple genes in female ocular adnexal tissues was anticipated. Androgens are known to exert a tremendous impact on the structure and function of the lacrimal gland, including such aspects as the cellular morphology, nuclear architecture, protein synthesis, enzyme activity, receptor expression and fluid and protein secretion . Similarly, the meibomian gland is an androgen target organ, and androgens appear to regulate this tissue’s function (e.g. lipid production) [22-25,43-45]. Androgen deficiency, in turn, has been linked to lacrimal and meibomian gland dysfunction, and a corresponding aqueous tear deficiency and evaporative dry eye [22-26,28-30]. A number of these ocular effects, due to the presence or absence of androgens, may be associated with glandular alterations in gene activity [39,40].
Of the many androgen actions on gene expression in the lacrimal and meibomian glands, two that stand out are the stimulation of meibomian gland genes associated with lipid metabolic pathways, and the suppression of genes related to keratinization. First, the upregulation of genes involved with lipid biosynthesis, transport, and metabolism is reminiscent of testosterone’s similar influence on the male meibomian gland [39,46,47]. Androgens stimulate numerous lipid pathway genes in this tissue, including those related to lipogenesis, steroidogenesis, and cholesterogenesis [39,46,47]. Second, the testosterone downregulation of meibomian gland genes associated with keratinization may explain on a molecular level how androgens inhibit this process. Keratinization is believed to be a primary cause of meibomian gland dysfunction [48,49], which leads to tear film hyperosmolarity and instability and evaporative dry eye [50-53]. Androgens appear to prevent this keratinization, because androgen insufficiency (e.g. during anti-androgen treatment, complete androgen insensitivity syndrome and/or aging) is associated with keratinization of the meibomian gland ductal epithelium (i.e. orifice metaplasia) and the lid [22,25,54]. It may be that these combined androgen actions, promoting lipogenesis and suppressing keratinization, are the reason why topical androgens reportedly enhance the synthesis and secretion of meibomian gland lipids, prolong the tear film breakup time, and alleviate dry eye syndrome [44,45].
Testosterone's effects on the female lacrimal and meibomian gland most likely involve an association with saturable, high-affinity and androgen-specific receptors in acinar epithelial cell nuclei. Androgen receptors are members of the steroid/thyroid hormone/retinoic acid family of ligand-activated transcription factors and mediate the classical actions of androgens throughout the body [55,56]. After androgen binding to the receptor, the monomeric, activated hormone-receptor complex associates with an androgen response element in the regulatory region of specific target genes, typically dimerizes with another androgen-bound complex and, in combination with appropriate coactivators and promoter elements, modulates gene transcription [55,56]. In support of this hypothesis, androgen receptors have been shown to exist in lacrimal and meibomian gland epithelial cells [57-63] and androgen activity in these cells may be compromised by androgen receptor mutations or antagonists [22-25,64-69]. Another mechanism of androgen action may involve binding to glandular membrane receptors, stimulation of signal transduction cascades and consequent alteration of gene transcription [56,70]. Testosterone may also act indirectly, by regulating the release of anterior pituitary hormones that may influence the lacrimal and meibomian glands.
Many of the genes modulated by testosterone in female tissues are identical to those regulated by androgens in male lacrimal and meibomian glands. However, some genes are controlled in a sex-specific manner. There were 4 genes downregulated by testosterone in the female, but not male, lacrimal gland. These include lymphocyte antigen 78, which is involved in innate immunity , a potassium voltage-gated channel member, and homer protein homolog 3, that helps couple surface receptors to intracellular calcium release . In the meibomian gland, several genes were upregulated in the female, but not male, including prostaglandin (PG) E synthase 2 (catalyzes the conversion of PG H2 to PG E2 ), estrogen related receptor α (may modulate the estrogen signaling pathway ), and tropomodulin 4 (blocks the elongation and depolymerization of actin filaments ). In addition, a series of meibomian gland genes were downregulated by testosterone solely in the female, such as glial fibrillary acidic protein (an intermediate filament ), chemokine (C-C motif) ligand 7 (attracts monocytes and eosinophils ) and interleukin 1β. Our analyses did not identify any gene that was uniquely and significantly expressed in female, as compared to male, glands.
We were especially interested in comparing the effects of the different sex steroids on gene expression in the lacrimal and meibomian glands. Androgens, estrogens and progestins play essential roles in the health and well-being of both men and women [72,73]. However, these hormones may also have antagonistic effects. An example is the influence of androgens and estrogens on the sebaceous gland. Androgens regulate the development, differentiation and lipid production of sebaceous glands throughout the body [74-78], and many of these actions appear to involve androgen receptors and the control of gene transcription in acinar epithelial cells [78-82]. Conversely, estrogens reduce the size, activity and lipid output of sebaceous glands [74,76,82-86] and for years were used clinically to decrease sebaceous gland function and secretion [74,75,84,85,87,88]. One mechanism proposed for estrogen action is that this hormone induces the release of lysosomal enzymes within sebocytes, resulting in premature cellular destruction and attenuated sebum elaboration [86,89]. Additional mechanisms, though, are that estrogens decrease testosterone uptake, interfere with testosterone's conversion to dihydrotestosterone, and antagonize androgen action in the sebaceous gland [15,83,86]. Indeed, estrogens have been described as the mainstay of treatment to reduce androgen effects on the sebaceous gland . These 'anti-androgen' actions of estrogens are dose-dependent, and may be overridden by exposure to physiological levels of androgens [74,84].
Androgen treatment also causes a significant decrease in the number of estradiol binding sites [89,90], and both hormones antagonize each other's regulation of their own receptor , in sebaceous glands. Indeed, some androgen effects are thought to be dependent upon low levels of estrogen . This steroid antagonism is not limited to androgens and estrogens. Androgens and progestins, for instance, may also show opposite effects on the same processes .
In the lacrimal and meibomian glands, a number of identical genes were influenced by testosterone, and 17β-estradiol and/or progesterone. The nature of the gene response to these hormones was sometimes similar, but often opposite. One lacrimal gland gene upregulated by 17β-estradiol and downregulated by testosterone is asialoglycoprotein receptor 1, which has been linked to the development of exocrine gland inflammation and dry eye [93-95]. These differential actions could contribute to the estrogen pro-inflammatory , and the androgen anti-inflammatory [33-36], effects in lacrimal tissue in Sjögren's syndrome. Genes stimulated by testosterone, but suppressed by 17β-estradiol, in the meibomian gland include secreted acidic cysteine rich glycoprotein (regulates cell growth ), vascular endothelial growth factor A (promotes cell migration, among many other actions ), matrix metalloproteinase 3 (degrades fibronectin, laminin, gelatins and collagens ) and cathepsin K (degrades extracellular matrices ). These genes could theoretically be involved in cell maturation, migration and holocrine secretion in the meibomian gland. If so, this would be consistent with a pro-sebaceous effect of androgens and an anti-sebaceous activity of estrogens. Such opposite effects of androgens and estrogens could also involve post-transcriptional  and non-genomic  pathways.
In summary, our data demonstrate that testosterone, 17β-estradiol and progesterone exert multiple effects on the lacrimal and meibomian glands, and that certain of these sex steroid actions are sex-specific and/or opposite. It is quite possible that these opposing sex steroid effects may play a role in the pathogenesis of dry eye syndrome.
The authors would like to express their appreciation to Mr. Michael J. Lombardi and Ms. Patricia Rowley for their technical assistance. The first author, David A. Sullivan, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analyses. This research was supported by a grant from NIH (EY05612).