| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Schepens Eye Research Institute (K.L.K., M.R.D., J.M.C., D.A.S.), Brigham and Women’s Hospital (M.R.D., J.M.C.), Department of Ophthalmology, Harvard Medical School (M.R.D., J.M.C., D.A.S.), and New England College of Optometry (K.L.K., D.B.T.), Boston, Massachusetts 02114; Edith Norse Rogers Veterans Memorial Hospital (M.D.U.), Bedford, Massachusetts 01730; and Eunice Kennedy Shriver Center for Mental Retardation (J.E.E.), Waltham, Massachusetts 02452
Address correspondence and requests for reprints to: David A. Sullivan, Ph.D., Schepens Eye Research Institute, 20 Staniford Street, Boston, Massachusetts 02114. E-mail: sullivan{at}vision.eri.harvard.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We hypothesize that androgens regulate meibomian gland function,
enhance the quality and/or quantity of lipids produced by this tissue,
and promote the formation of the tear film’s lipid layer. We also
hypothesize that androgen deficiency is a critical etiologic factor in
the pathogenesis of meibomian gland dysfunction and evaporative dry
eye. In support of these hypotheses, we and others have discovered
that: 1) the meibomian glands of rats, rabbits, and humans are androgen
target organs and contain androgen receptor messenger RNA (mRNA) and/or
androgen receptor protein within acinar epithelial cell nuclei
(8, 9); 2) human meibomian glands contain the mRNAs for
both Types 1 and 2 5
-reductase (9), an enzyme that
converts testosterone into the potent androgen 5-dihydrotestosterone
(DHT) (10); 3) application of dehydroepiandrosterone,
an androgen precursor (10), to the ocular surface of
rabbits, dogs, and/or a human stimulates the production and release of
meibomian gland lipids and prolongs the tear film breakup time
(11); 4) orchiectomy causes a significant alteration
in the lipid profile of rabbit meibomian glands, whereas the topical
administration of 19-nortestosterone for 2 weeks, but not placebo
compounds, begins to restore the lipid pattern to that found in intact
animals (12); and 5) the one common denominator in
menopause (13), aging in both sexes (10, 13),
and primary and secondary Sjögren’s syndrome
(14, 15, 16) seems to be androgen deficiency.
If our hypotheses are correct, we would predict that chronic androgen deficiency, such as induced by the extended use of antiandrogen medications, will lead to meibomian gland dysfunction, altered lipid profiles in meibomian gland secretions, decreased tear film stability, and evaporative dry eye. The purpose of the present investigation was to test this prediction.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Male subjects taking antiandrogen therapy for prostatic
indications were recruited from the Departments of Urology at Brigham
and Women’s Hospital and Boston University Medical Center (Boston,
MA). These patients (n = 15), whose average age was 70.9 ±
1.9 yr, had been treated with antiandrogen medications for periods
ranging from 3–96 months (median, 36 months). These medications
included leuprolide acetate (Lupron), goserelin acetate (Zoladex),
bicalutamide (Casodex), flutamide (Eulexin), and/or
finasteride (Proscar). Age-related male controls
(64.8 ± 1.0 yr old; n = 6), who were not receiving
antiandrogen treatment, as well as younger normal individuals (
30 yr
old; n = 3 males and 1 female) were recruited from the Boston
environs. The ages of the patients and their age-related controls were
not significantly different. These studies were approved by the Human
Studies Committee of the Schepens Eye Research Institute (Boston, MA)
and were conducted in accordance with guidelines established by the
Declaration of Helsinki.
Clinical assessment
After providing informed consent, subjects were asked to
complete without supervision a questionnaire that assessed dry eye
symptoms (Ocular Surface Disease Index; Allergan, Inc.,
Irvine, CA), as well as to answer questions related to medical
histories and current medications (Table 1
). Individuals then underwent a complete
ocular surface and anterior segment examination of both eyes (Table 2) by corneal external disease
subspecialists. This examination, which was based on standard protocols
(17, 18, 19, 20, 21), included a slit lamp evaluation of the: 1) tear
film, for the presence of mucus and debris; 2) tear meniscus, by using
semiquantitative measurements. A normal meniscus appears as a clear and
continuous tear film over the ocular surface adjacent to the lower lid
margin, and is typically over 0.3 mm in height (20). An
abnormal meniscus is present when a clearly continuous tear film cannot
be identified over the ocular surface. An intermediate meniscus is
identified as a level between normal and abnormal; 3) lids, for the
identification of neovascularization, irregular posterior margins,
scurf, sleeve, and collarettes; 4) conjunctiva, for the appearance of
bulbar injection, tarsal injection and papillary hypertrophy; 5)
cornea, for the existence of punctate epithelial keratitis, adherent
mucus, filamentary keratitis, and neovascularization; 6) tear film
breakup time, which is a measure of tear film stability and a global
criterion for dry eye (17); 7) fluorescein staining of the
cornea (graded on a 0-3 scale for each of five areas)
(17); and 8) rose bengal staining (graded on a 0-3 scale)
of the conjunctiva (six nasal and temporal regions) and the entire
cornea (17). The tear film breakup time and fluorescein
and rose bengal staining procedures were performed according to
published methods (17). For the calculation of staining
results, grading scores for ocular surface regions from both eyes were
summed, thereby yielding a total score for each patient. Additional
parameters that were evaluated included: 1) assessment for metaplasia
of the meibomian gland orifices, a condition defined as an abnormal
growth and keratinization of duct epithelium (6); 2)
analysis of the quality of meibomian gland secretions, according to a
published classification system (21). In brief, the
grading scheme was ‘0‘ for clear excreta with small particles, ‘1‘
for opaque excreta with normal viscosity, ‘2‘ for opaque excreta with
increased viscosity, and ‘3‘ for secreta that retained shape after
digital expression; and 3) examination of the appearance of the
anterior chamber and iris (19).
|
|
Biochemical procedures
Meibomian gland secretions were analyzed for the relative content of cholesterol, cholesterol esters, wax esters, diglycerides, triglycerides, and specific molecular species in the diglyceride fraction by highperformance liquid chromatography (HPLC; Spectra-Physics Model 8700, Thermo Separation Products, San Jose, CA) and mass spectrometry (MS; Finnigan 4500, Finnigan, San Jose, CA). Samples were separated over a 10 cm x 2 mm Inertsil silica column (Keystone Scientific, Inc., Bellefonte, PA) with a complex, multistep gradient that combined mobile phases of isooctane-tetrahydrofuran (99:1,vol/vol), isopropanol-chloroform (4:1, vol/vol), and isopropanol-water (1:1, vol/vol) and had a linear flow velocity of 0.4 mL/min (23). The vaporizer temperature in the moving-belt interface of the Finnigan HPLC/mass spectrometer was 310 C. MS was conducted in positive ion, chemical ionization mode with ammonia reagent gas, and data were acquired with a Teknivent Vector/Two (Teknivent Corporation, Maryland Heights, MO). The predominant peaks in HPLC/MS elution plots were identified by the use of specific ions [i.e. cholesterol and cholesterol esters, mass/charge ratio (m/z) 369; wax esters, m/z 636, 650, 664, and 678; diglycerides and triglycerides, m/z 551, 577, 579, 603, and 605; and squalene, m/z 411], peak areas were determined, and the relative amounts of various lipid fractions were then calculated. Analysis of the m/z ratios of diglyceride fatty acids was facilitated through a time-based decomposition of HPLC/MS elution plots.
Statistical analyses
Statistical analyses of the data were performed by using the
unpaired, two-tailed Student’s t, Mann-Whitney
U, and 
2 tests.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To determine whether chronic androgen deficiency is associated with alterations of the meibomian gland and ocular surface, male subjects (n = 15) taking antiandrogen therapy (median, 3 yr) for prostatic indications, as well as their age-related controls (n = 6), were given thorough anterior segment examinations. In addition, subjects and controls completed questionnaires designed to assess dry eye symptoms.
Our results demonstrated that the use of antiandrogen medications is
associated with meibomian gland dysfunction, tear film instability, and
functional dry eye. Slit lamp examinations revealed that patients
taking antiandrogen pharmaceuticals, compared with controls, had a
significant increase in the frequency of appearance of tear film
debris, an abnormal tear film meniscus, irregular posterior lid
margins, lid sleeves and collarettes and conjunctival tarsal injection
(Table 3 and Fig. 1). In addition, patients had a
significant decrease in their tear film breakup time (Fig. 2) and a significant increase in the
degrees of corneal fluorescein and rose bengal staining (Fig. 3) and inferior bulbar conjunctival rose
bengal staining (control OD and OS, 0.29 ± 0.11; patient
OD and OS, 0.63 ± 0.10; P < 0.05, one-tail).
|
|
|
|
) was significantly increased and that the quality
of secretions was significantly reduced (i.e. higher
viscosity; Fig. 4), in patients taking
antiandrogen therapy, relative to controls. Moreover, clinical
impressions were that the meibomian glands of ‘antiandrogen‘ patients
had a much higher frequency of altered appearance (e.g.
inspissation, cysts) and severe disease. These observations were
particularly notable, given that three control individuals had rosacea
and that 83% of controls had some evidence of scurf, blocked, or
dysfunctional meibomian glands.
|
).
|
We also compared the relative frequency of the signs and symptoms of
dry eye between patients taking finasteride (n = 4;
age, 67.3 ± 2.8 yr; diagnosis, prostatic hypertrophy; average
duration of treatment, 3 yr) vs. other antiandrogen
medications (i.e. predominantly leuprolide acetate, as well
as bicalutamide, goserelin acetate, and flutamide) (n = 11; age,
72.2 ± 2.3 yr; diagnosis, prostatic cancer; average duration of
treatment, 3 yr). The rationale for this comparison was that
finasteride, but not the other antiandrogen compounds,
might act to inhibit the local conversion of testosterone to DHT
(24). If so, finasteride actions might
reflect the importance of local steroidogenesis per se in
providing potent androgens to ocular surface tissues. These comparisons
showed that the left and right eyes of patients taking
finasteride had a significantly higher frequency of
appearance of conjunctival bulbar injection (finasteride
group, 100%; other treatment group, 50%; P < 0.05),
lid collarettes (finasteride group, 66.7%; other
treatment group, 15%; P < 0.05), metaplasia of
meibomian gland orifices (finasteride group, 100%; other
treatment group, 54.6%; P < 0.01), and corneal
fluorescein staining (finasteride group, 87.5%; other
treatment group, 31.8%; P < 0.01).
Finasteride-treated patients also had a significantly
greater sensitivity to wind (finasteride group, 75%
positive responses; other treatment group, 0% positive responses;
P < 0.005). In contrast, patients receiving other
antiandrogen therapies had a significantly higher frequency of
appearance of conjunctival papillary hypertrophy
(finasteride group, 0%; other treatment group, 57.1%;
P < 0.01%) in their left and right eyes.
Effect of antiandrogen treatment on the neutral lipid profile in meibomian gland secretions
To determine whether chronic androgen deficiency is associated with altered neutral lipid profiles in meibomian gland secretions, secretion samples (n = 2/individual) were obtained from the right and left eyes of patients (n = 15) taking antiandrogen medications and their age-related controls (n = 6), as well as younger individuals (n = 4), and analyzed for various lipid fractions by HPLC/MS.
As shown in Fig. 6, the use of
antiandrogen pharmaceuticals was associated with significant changes in
the relative amounts of lipids in meibomian gland secretions. Patients
had a significant attenuation in the levels of cholesterol esters, wax
esters, and diglycerides plus triglycerides, relative to those of
cholesterol, as well as a significant increase in the percentage of
cholesterol. In addition, patients taking antiandrogen therapy had a
decreased expression of specific molecular species (e.g. m/z
620) in the diglyceride fraction of meibomian gland secretions,
compared with that of controls (Fig. 7).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mechanism by which antiandrogen therapy interferes with meibomian
gland function and alters the lipid profile in meibomian gland
secretions is undoubtedly due to androgen deficiency. The medications
used by the patients in this study included: 1) leuprolide acetate and
goserelin acetate, which are analogs of LH-releasing hormone that
decrease testicular steroidogenesis and cause a dramatic reduction in
the serum levels of testosterone (24); 2) bicalutamide and
flutamide, which are nonsteroidal antiandrogens that inhibit androgen
uptake and/or binding to nuclear androgen receptors (24);
and 3) finasteride, an inhibitor of Type 2 5-reductase
that prevents conversion of testosterone to DHT (24).
These treatments would effectively reduce exposure of the meibomian
gland to active androgens. Consequently, given that this tissue is an
androgen target organ (26), contains both androgen
receptor protein and 5-reductase mRNA (9), and responds
to androgens with an enhanced lipid synthesis, production and release
(11, 12), it would seem that antiandrogen therapy and the
resulting androgen deficiency would lead to meibomian gland
dysfunction. In support of this interpretation are two observations.
First, the meibomian gland is a large sebaceous gland, and androgens
are known to control the development, differentiation, and lipid
elaboration of sebaceous glands in nonocular sites (27, 28). Antiandrogen treatment and the related androgen
insufficiency, in turn, lead to a marked decline in sebaceous gland
activity and lipid output (27, 29). Second, our recent
clinical studies on women with complete androgen insensitivity
syndrome, which is characterized by the absence of functional androgen
receptors (30), have demonstrated that affected
individuals have both meibomian gland disease (31) and
altered lipid patterns in their meibomian gland secretions
(32).
Our finding that antiandrogen treatment changed the neutral lipid profile of meibomian gland secretions might have been anticipated. Androgens have been shown to exert a significant influence on lipid metabolic pathways throughout the body (33, 34, 35, 36, 37, 38, 39). This hormone action includes the regulation of genes involved in fatty acid and cholesterol synthesis, the activity of lipogenic enzymes, the incorporation of fatty acids into neutral lipids, the content of cholesterol and other neutral lipids, and the secretion rate of wax esters (33, 34, 35, 36, 37, 38, 39). In addition, patients with Sjögren’s syndrome, a disease associated with androgen deficiency (14, 15, 16), have heightened levels of cholesterol in their tear film (40). In the current study, the antiandrogen patients had a relative decrease in the amount of wax esters, cholesterol esters, and diglycerides and triglycerides in their meibomian gland secretions, and a comparative rise in the quantity of cholesterol. This enhanced cholesterol content would promote tear film instability (2). Moreover, augmented cholesterol increases the melting point and viscosity of meibomian gland secretions, thereby leading to a stagnation and plugging of meibomian glands (2). In fact, these associations may explain the increased prevalence of obstructive meibomitis in these patients, as evidenced by an apparently higher frequency of meibomian gland inspissation and cysts found in individuals taking antiandrogen medications.
The impact of antiandrogen therapy on the conjunctiva, cornea, lid, and ocular surface symptomatology may have been due, in part, to decreased meibomian gland function. Meibomian gland dysfunction typically leads to an increase in the signs and symptoms of evaporative dry eye (1, 2, 3, 17). Indeed, this condition has been estimated to be a contributing factor in over 60% of all dry eye patients (4). Of interest, the symptoms associated with meibomian gland dysfunction may or may not be prominent and the ocular surface manifestations are often not correlated with the degree of meibomian gland disease (2). These observations may account for the relatively low score of antiandrogen patients on the Ocular Surface Disease Index questionnaire.
Another consideration in the response of the conjunctiva and cornea to
antiandrogen therapy is that these tissues express Types 1 and 2
5-reductase mRNA and/or androgen receptor mRNA and protein (8, 9). Furthermore, androgens have been shown to influence the
functional activity of both the conjunctiva and cornea
(41, 42, 43, 44, 45, 46). Therefore, the greater tarsal injection and
vital dye staining observed in the ocular tissues of antiandrogen
patients may have been the consequence not only of meibomian gland
dysfunction and evaporative dry eye, but also of androgen deficiency
per se. Theoretically, it is also possible that the
conjunctival and corneal effects might be partially attributed to a
decreased tear output from the lacrimal gland. Androgens regulate
multiple aspects of lacrimal gland function (47), and
investigators have speculated that a loss of androgens may result in
an ‘aqueous-deficient‘ dry eye (48). However, this
possibility is unlikely, given that we have recently found that
androgen insufficiency by itself does not cause aqueous tear deficiency
in nonautoimmune humans (22).
The ocular impact of antiandrogen treatment seemed to be influenced by
the nature of the medication. Thus, finasteride
administration, compared with the analogs of LH-releasing hormone or
the nonsteroidal antiandrogens, seemed to be associated with a greater
frequency of conjunctival bulbar injection, lid collarettes, metaplasia
of meibomian gland orifices, corneal fluorescein staining, and wind
sensitivity. In contrast, patients receiving the other antiandrogen
therapies had more frequent conjunctival papillary hypertrophy. The
reason for these different response patterns remains to be elucidated.
At present, it is known that Types 1 and 2 5-reductase mRNAs occur
in ocular surface tissues (9), but whether, as in other
sites, these mRNAs are translated, display tissue-specific degrees of
enzymatic activity, or show differential responsiveness to
finasteride has yet to be clarified. Similarly, the
relative importance of local steroidogenesis (10)
vs. systemic delivery for the accumulation of androgens in
ocular tissues, and the relative activity of classical (i.e.
nuclear receptor) vs. nonclassical mechanisms
(49) in mediating androgen action in the eye, are not
known. Thus, to explain the differential effects of the various
antiandrogen medications (i.e. suppressors of testicular
androgen synthesis, receptor antagonists, reductase inhibitors), it
would seem necessary to first determine the origin, form of metabolism,
and mode of action of androgens in ocular tissues. It is of particular
interest, though, that our findings extend those of a previous report,
which stated that leuprolide acetate administration was associated with
ophthalmic problems and blurred vision in some patients
(24).
Overall, these observations in patients taking antiandrogen therapy are consistent with our hypothesis that androgen deficiency is a critical etiological factor in the pathogenesis of meibomian gland dysfunction and evaporative dry eye. In further support of our hypothesis are the findings that: 1) reduced serum levels of testosterone are more prevalent in women with dry eye and correlate with the subjective severity of ocular symptoms (50); and 2) serum levels of total androgens decline during menopause (13) and aging in both sexes (10, 13), and these time periods coincide with an increased appearance of meibomian gland dysfunction and dry eye (51, 52, 53). As an additional consideration, this apparent interrelationship between androgen deficiency, meibomian gland dysfunction, and dry eye might help to explain why systemic androgen administration has been reported to alleviate the signs and symptoms of dry eye (54, 55, 56, 57, 58). Given these results, it is possible that efforts directed at alleviating this endocrine imbalance (e.g. topical application of androgens) may prove beneficial as a treatment for meibomian gland dysfunction and the associated evaporative dry eye, in androgen-deficient individuals.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received May 26, 2000.
Revised August 14, 2000.
Accepted September 6, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-reductase mRNA in
human ocular tissues. Br J Ophthalmol. 84:76–84.
sa a
könnymirigyre. Orv Hetil. 95:580–581.[Medline]
This article has been cited by other articles:
 |
|
 |
|
  I. A. Butovich, E. Uchiyama, and J. P. McCulley Lipids of human meibum: mass-spectrometric analysis and structural elucidation J. Lipid Res., October 1, 2007; 48(10): 2220 - 2235. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. D. Sullivan, J. E. Evans, M. R. Dana, and D. A. Sullivan Influence of aging on the polar and neutral lipid profiles in human meibomian gland secretions. Arch Ophthalmol, September 1, 2006; 124(9): 1286 - 1292. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P S Tsai, J E Evans, K M Green, R M Sullivan, D A Schaumberg, S M Richards, M R Dana, and D A Sullivan Proteomic analysis of human meibomian gland secretions. Br. J. Ophthalmol., March 1, 2006; 90(3): 372 - 377. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Schirra, T. Suzuki, S. M. Richards, R. V. Jensen, M. Liu, M. J. Lombardi, P. Rowley, N. S. Treister, and D. A. Sullivan Androgen Control of Gene Expression in the Mouse Meibomian Gland Invest. Ophthalmol. Vis. Sci., October 1, 2005; 46(10): 3666 - 3675. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Smith, S. Vitale, G. F. Reed, S. A. Grieshaber, L. A. Goodman, V. H. Vanderhoof, K. A. Calis, and L. M. Nelson Dry Eye Signs and Symptoms in Women With Premature Ovarian Failure Arch Ophthalmol, February 1, 2004; 122(2): 151 - 156. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. C. Pflugfelder Hormonal Deficiencies and Dry Eye Arch Ophthalmol, February 1, 2004; 122(2): 273 - 274. [Full Text] [PDF]
|
|
|
|
|
|
B. D. Sullivan, J. E. Evans, J. M. Cermak, K. L. Krenzer, M. R. Dana, and D. A. Sullivan Complete Androgen Insensitivity Syndrome: Effect on Human Meibomian Gland Secretions Arch Ophthalmol, December 1, 2002; 120(12): 1689 - 1699. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. SULLIVAN, B. D. SULLIVAN, J. E. EVANS, F. SCHIRRA, H. YAMAGAMI, M. LIU, S. M. RICHARDS, T. SUZUKI, D. A. SCHAUMBERG, R. M. SULLIVAN, et al. Androgen Deficiency, Meibomian Gland Dysfunction, and Evaporative Dry Eye Ann. N.Y. Acad. Sci., June 1, 2002; 966(1): 211 - 222. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. LeDoux, Q. Zhou, R. B. Murphy, M. L. Greene, and P. Ryan Parasympathetic Innervation of the Meibomian Glands in Rats Invest. Ophthalmol. Vis. Sci., October 1, 2001; 42(11): 2434 - 2441. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
,
P < 0.005) different between control and patient
groups (
