Download Inhibition of dexamethasone-induced cytoskeletal changes in

Inhibition of Dexamethasone-Induced Cytoskeletal
Changes in Cultured Human Trabecular Meshwork Cells
by Tetrahydrocortisol
Abbot F. Clark, Debbie Lane, Karen Wilson, Sharon T. Miggans,
and Mitchell D. McCartney
Purpose. To determine the cellular mechanism of action of the intraocular pressure (IOP)
lowering steroid tetrahydrocortisol (THF).
Methods. Tetrahydrocortisol was evaluated for glucocorticoid antagonist activity using in vitro
and in vivo assays. Systemically administered THF was evaluated for its ability to inhibit dexamethasone-induced body weight loss and systemic hypertension in rats. In vitro receptor
antagonism was tested using the supernatant fraction of IM9 cells as the source of soluble
glucocorticoid receptor in 3H-dexamethasone displacement binding assays. In addition, six
different primary human trabecular meshwork (TM) cell lines were cultured for 0 to 14 days
in the absence or presence of dexamethasone (10~7 M) and/or THF (10"b to 10~8 M). The
effects of these steroids on the TM cytoskeleton were determined by epifluorescent microscopy
and by transmission electron microscopy.
Results. Tetrahydrocortisol was unable to inhibit the dexamethasone (DEX)-induced systemic
hypertension and decrease in body mass in rats and was unable to displace 3H-DEX from the
soluble human glucocorticoid receptor. However, THF inhibited the DEX-induced formation
of cross-linked actin networks in cultured human TM cells in a progressive and dose-dependent
manner (IC50 = 5.7 X 10~7 M). Dexamethasone caused changes in the TM cell microtubules
that were reversed partially by concomitant treatment with THF. Tetrahydrocortisol alone
appeared to increase microfilament bundling in TM cells.
Conclusions. Tetrahydrocortisol was not a glucocorticoid antagonist at the level of die classical
glucocorticoid receptor and did not appear to antagonize systemically mediated glucocorticoid
activity in the rat. Tetrahydrocortisol inhibited DEX-induced changes in the TM microfilaments
and microtubules. These results may explain partially the IOP lowering activity of THF because
glucocorticoid-mediated changes in the TM cytoskeleton have been proposed to be involved in
the generation of ocular hypertension. Invest Ophthalmol Vis Sci. 1996;35:805-813.
X opical ocular or systemic administration of glucocorticoids can lead to the development of ocular hypertension in susceptible persons,1"4 and if glucocorticoid administration is continued, open angle glaucoma56 that in many ways mimics primary open angle
glaucoma will develop in many of these steroid responders.7 Glucocorticoid-induced ocular hypertension has been shown to be caused by increased resisFrom A Icon Laboratories, Inc., Fort Worth, Texas.
Presented in part at the 1993 and 1994 annual meetings of the Association for
Research in Vision and Ophthalmology.
Sulimitled for publication May I, 1995; revised December 5, 1995; accepted
Decembers, 1995.
Proprietary interest category: P, E.
Reprint requests: Abbot F. Clark, Glaucoma Research R2-41, Alcon Laboratories,
6201 South Freeway, Fort Worth, IX 76134.
tance to aqueous humor outflow1 4 and is associated
with biochemical89 and ultrastructural changes10"12 in
the trabecular meshwork (TM). Glucocorticoid-induced ocular hypertension also can be generated in
rabbits,13'14 cats,15'16 and monkeys.17 A number of studies have reported glucocorticoid-mediated changes in
cultured TM cells, including altered gene and protein
expression,1819 altered deposition of TM extracellular
matrix molecules,20'21 decreased extracellular proteinase activities,22'23 TM cell and nucleus enlargement,24"20 reorganization of TM cytoskeletal elements,24'25 inhibition of TM cell functions,18'25'27 and
activation of the endoplasmic reticulum and Golgi apparatus in TM cells.24
The ability of glucocorticoids to induce ocular hy-
Investigative Ophthalmology & Visual Science, April 1996, Vol. 37, No. 5
Copyright © Association for Research in Vision and Ophthalmology
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017
805
806
Investigative Ophthalmology & Visual Science, April 1996, Vol. 37, No. 5
pertension has inspired the search for steroids that
may have the opposite effect and, thus, lower IOP.28"31
One of these ocular hypotensive steroids is tetrahydrocortisol (THF), which has been reported to lower
the intraocular pressure (IOP) of dexamethasone
(DEX)-induced ocular hypertensive rabbits on topical
administration.32 Tetrahydrocortisol is a natural component of steroid metabolism and is the major metabolite of cortisol, the endogenous glucocorticoid, in
humans. In addition, a preliminary experiment suggested that THF was also an effective ocular hypotensive agent in patients with primary open angle glaucoma.
The current study was conducted to determine
the IOP-lowering mechanism of action of THF. Tetrahydrocortisol was tested in rats for its ability to inhibit
DEX-induced systemic hypertension and loss in body
weight. Tetrahydrocortisol also was evaluated for its
ability to interact with the classical human glucocorticoid receptor in an in vitro ligand-binding assay. In
addition, because it has been suggested that one
mechanism for DEX-induced ocular hypertension is
caused by the effects of DEX on the TM cell cytoskeleton,25 we also evaluated the effects of THF on DEXinduced cytoskeletal changes in cultured human trabecular meshwork cells.
MATERIALS AND METHODS
Effect of Steroids on Body Weight and Blood
Pressure
All animal experimentation was conducted in strict
compliance with the ARVO Statement for the Use of
Animals in Ophthalmic and Vision Resesarch. Four
groups (n = 5 per group) of normotensive SpragueDawley rats, each weighing 200 to 300 g, were treated
daily for 14 days with subcutaneous injections at 0.1
ml/100 g body weight of sesame oil (control), dexamethasone (0.1% in sesame oil), THF (1% in sesame
oil), or DEX + THF (0.1% and 1%, respectively, in
sesame oil). The rats received water and laboratory
chow ad libitum and were housed under a 12-hour
light-12-hour dark cycle. Body weight and blood pressure were measured three times per week. Indirect
blood pressure (tail cuff) was determined using a
Narco (Austin, TX) Bio-Systems Physiograph.
Human Glucocorticoid Receptor IigandBinding Assay
The human lymphoblast cell line IM9 (ATCC,
Bethesda, MD) was used as a source of the soluble
glucocorticoid receptor (GR).34 The cells were grown
to densities of 1 to 10 X 105 cells per milliliter in RPMI
1640 media (Gibco, Grand Island, NY) containing
10% fetal bovine serum (HyClone, Logan, UT), penicillin (100 U/ml), streptomycin (100 /xg/ml), and 2
mM L-glutamine (Gibco) at 37° and 7% CO2 in a
humidified incubator. The IM9 cells were harvested
from the media by centrifugation for 10 minutes at
1500g\ Cells were washed in 12 volumes of Dulbecco's
phosphate-buffered saline (PBS; Gibco) and repelleted. Washed cells were resuspended in five to six
volumes (per volume of packed cells) of homogenization buffer (10 mM TES, 10 mM sodium molybdate,
1 mM EDTA, pH 7.4, 20 mM 2-mercaptoethanol, and
10% glycerol), and the cells were broken by nitrogen
cavitation using 2 X 15 minutes at 600 to 750 psi
nitrogen in the N2 cavitator (Parr Instrument, Moline,
IL) at 0°C. Cell disruption was confirmed by Hoffman
contrast microscopy using a Nikon (Garden City, NY)
Diaphot. The broken cell preparation was then centrifuged at 27,000g for 15 minutes, and the resultant
supernatant was centrifuged at 103,000g for 60 minutes at 4°C. The amount of protein in the supernatant
fraction was determined using a BCA assay kit (Pierce
Chemical, Rockford, IL) with a bovine serum albumin
standard. Aliquots of the supernatant fraction were
snap frozen in a dry ice-acetone bath and stored at
— 70°C. Competitive binding assays were done in duplicate in homogenization buffer (total volume of 200
lA) by mixing 1 mg of IM9 cytosol, 0.05 /xCi (3 nM)
of 3H-dexamethasone (Amersham, Arlington,
Heights, IL), and unlabeled competitor steroids (10~5
to 10~" M) consisting of dexamethasone, prednisolone, cortisol, triamcinolone acetonide, progesterone, cortexolone (Sigma Chemical, St. Louis, MO),
and tetrahydrocortisol (THF; 5/?-pregnan-3a, 11/?,
17a, 21-tetrol-20-one) (Steraloids, Wilton, NH). After
incubation at 0°C for 16 to 18 hours, the assay was
stopped by the addition of 100 fi\ of a charcoal-dextran mixture (2% activated charcoal, 0.5% dextran in
10 mM Tris, 1 mM EDTA, pH 7.4). The assay mixture
was further incubated at 0°C for 10 minutes before
being centrifuged for 5 minutes at 8200g. A 100-^tl
sample of the supernatant (protein-bound steroid
fraction) was assayed for radioactivity by liquid scintillation spectrometry, and the IC50 values were determined graphically.
Evaluation of Effects of Steroids on Trabecular
Meshwork Cell Cytoskeleton
Human TM cells were cultured and characterized as
described previously.202425 Briefly, TM cells were
grown from explants dissected from human donor
eyes (obtained from a variety of regional eye banks)
placed in Ham's F10 media containing 10% fetal bovine serum (HyClone), penicillin, streptomycin, and
2 mM L-glutamine (Gibco). The TM cells were propagated by serial passage using Cytodex 3 (Sigma) microcarrier beads. Human TM cells were grown to confluence on glass coverslips for light microscopic analysis
or on Formvar-coated nickel grids for wholemount
transmission electron microscopy (TEM) analysis. The
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017
Tetrahydrocortisol Effects on the Trabecular Meshwork Cytoskeleton
807
100
O
•
V
T
80
60
40
Control
DEX
THF
THF+DEX
20
0
-20
-40
-60
-80
-100
-120
0
2
4
6
8
10
12
14
16
0
2
TIME (days)
4
6
8
10
12
14
TIME (days)
FIGURE l. Effects of dexamethasone (DEX) and tetrahydrocortisol (THF) on rat body weight
and blood pressure. Four groups of rats (n = 5 per group) were given daily subcutaneous
injections of sesame oil, dexamethasone (0.1 mg/100 g body weight), tetrahydrocortisol (1
mg/100 g body weight), or a combination of both DEX and THF for 14 days. (A) Mean
change in initial body weight ± SEM. The DEX-treated group was statistically different from
the control group on days 4, 7, 9, 11, and 14 (P < 0.05). The DEX + THF group was
statistically different from the control group on days 7, 9, 11, and 14 (P < 0.02). (B) Mean
change in systemic blood pressure. The DEX group was statistically different the predosing
blood pressure (day 0) on days 5, 8, 10, and 12 (P < 0.05). The DEX + THF group was
statistically different from the predose blood pressure (day 0) on days 5, 10, and 12 (P <
0.05).
TM cells were treated with dexamethasone (10
6
7
8
7
M)
and/or THF (KT , 10" , or 10" M) for 0 to 14 days.
Stock solutions of the steroids were prepared by dissolving DEX (1(T4 M) and THF (1(TH to 1(T5 M) in
absolute ethanol. Stock solutions were diluted in the
media (1 fi\ stock solution per milliliter of media)
immediately before use. Control cells received equivalent volumes of absolute ethanol (0.1% final concentration). Trabecular meshwork cell microfilaments
were examined by epifluorescence after fixing the
cells in 1% glutaraldehyde (Sigma), 0.5% Triton X100 (Sigma), 50 mM phosphate buffer (pH 7.2) and
staining with rhodamine-phalloidin (Molecular
Probes, Eugene, OR) as previously described.25 The
percentage of TM cells with cross-linked actin networks (CLANs) was determined by examining approximately 200 cells on each of two coverslips per experimental condition. Each experiment was performed
two or more times. The definition and characterization of CLANs have been described in detail.24'25 Trabecular meshwork cell microtubules were visualized by
fixing the cells in methanol at — 20°C for 10 minutes,
rinsing with PBS and incubating with a 1:25 dilution
of an anti-tubulin primary antibody (Boehringer
Mannheim, Indianapolis, IN) in PBS per 1% bovine
serum albumin (Sigma) for 1 hour. Cells were then
rinsed with PBS, incubated with a 1:20 dilution of
fluorescein isothiocyanate-labeled rabbit anti-mouse
secondary antibody (Boehringer Mannheim), and examined by fluorescent microscopy using a Nikon Optiphot Photomicroscope (Nikon).
Trabecular meshwork cells grown on Formvarcoated (Electron Microscopy Supplies, Fort Washington, PA) nickel grids were prepared for wholemount
transmission electron microscopy as described24 to examine the effects of steroids on microfilament and
microtubule organization. Cells were fixed for 30 minutes in 0.25% glutaraldehyde and blocked for 20 minutes in 4% nonfat dry milk. They were then incubated
overnight with monoclonal anti-actin or anti-tubulin
primary antibodies (Amersham) using 1:100 dilutions.
After rinsing five times for 2 minutes each in Trisbuffered saline, the cells were incubated for 2 hours
with 1:5 dilutions of immunogold-conjugated goat
anti-mouse secondary antibodies (Amersham). Cells
subsequendy were fixed for 30 minutes in buffered
1% glutaraldehyde, rinsed, osmicated (2% osmium)
for 2 minutes, dehydrated, critical point dried, and
examined in a Zeiss (Thornwood, NY) CEM-902 transmission electron microscope.
RESULTS
Effect of Dexamethasone and
Tetrahydrocortisol on Body Weight and Blood
Pressure
Results of daily steroid treatment for 2 weeks on rat
body weight are shown in Figure 1A. Vehicle (sesame
oil)-treated as well as THF-treated rats continued to
gain weight during the 2 weeks of treatment, with an
average weight gain of 26 to 29 g per week. In contrast,
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017
808
Investigative Ophthalmology & Visual Science, April 1996, Vol. 37, No. 5
120
•
A
A
DEX
Cortisol
Prednisolone
D
Triamcinolone Acetonide
100806040200-
,-11
10-10
1Q-9
1Q-8
1Q-7
1Q-6
1Q-5
1Q-4
Concentration of Compound (M)
10 11
1Q-10
1Q-9
10
-8
1Q-7
1Q-6
1Q-5
1Q-4
Concentration of Compound (M)
FIGURE 2. Binding affinity of different steroids for the human glucocorticoid receptor. Various concentrations of unlabeled steroids were allowed to compete for 3H-DEX (3 nM)
binding to the human glucocorticoid receptor (cultured IM9 cell supernatant). (A) Displacement binding for glucocorticoid agonists (triamcinolone acetonide, dexamethasone, prednisolone, cortisol). (B) Displacement binding for glucocorticoid antagonists (progesterone,
cortexolone) and tetrahydrocortisol (THF).
the DEX-treated group progressively lost weight after
the first 2 days of treatment, with an average loss of
approximately 46 g per week. The weight loss of the
group treated with the combination of THF plus DEX
was identical to the group treated with DEX alone.
Therefore, THF treatment alone did not have any adverse effect on body weight, and THF did not block
the catabolic effect of DEX. Results of 2 weeks of steroid treatment on rat blood pressure are shown in
Figure IB. Blood pressures of control and THF-treated
groups were not significantly changed during the 2
weeks of treatment. In contrast, DEX treatment
caused a progressive and significant increase in systolic
blood pressure, with a pressure elevation of 8 to 9 mm
Hg after 10 to 12 days of DEX administration. The
group treated with the combination of THF plus DEX
had progressive and significant rises in systolic blood
pressure that were identical to the group treated with
DEX alone. Therefore, THF did not appear to block
the DEX-induced systemic hypertension.
Glucocorticoid Receptor Binding
Glucocorticoid receptor agonists bound to the soluble
receptor with affinities that closely matched their antiinflammatory potencies (Fig. 2A): triamcinolone acetonide ^ dexamethasone > prednisolone > cortisol,
with IC5Os of 1.1 X 1(T8 M, 1.2 X lfr 8 M, 2.2 X 10"8
M, and 5.6 X 10~8 M, respectively. The glucocorticoid
antagonists progesterone and cortexolone also bound
to the glucocorticoid receptor and displaced radiolabeled DEX (Fig. 2B) with IC50s of 1.9 X 10~7 M and 5.8
X 10~7 M, respectively. In contrast, THF was unable to
bind competitively to the glucocorticoid receptor and
to displace DEX even at high concentrations (10~'1
M). In addition, there was no binding of radiolabeled
THF to the human glucocorticoid receptor in direct
binding assays (data not shown).
Effect of Dexamethasone and
Tetrahydrocortisol on Trabecular Meshwork
Cytoskeleton
Dexamethasone treatment caused a time-dependent
reorganization of cultured human TM cell microfilaments to form cross-linked actin networks (Fig. 3A)
as seen by epifluorescent microscopy of rhodaminephalloidin-stained cells.24'25 Concomitant treatment of
the TM cells with DEX (1(T7 M) and THF (lfr 6 M)
caused a progressive inhibition of DEX-induced CLAN
formation (Fig. 3A). There was no difference in TM
CLAN formation at days 2 and 4 between DEX and
DEX + THF treatment. However, beginning at day 6,
THF caused an inhibition and reversal of DEX-induced CLAN formation. In 16 different experiments
using seven different TM cell lines (each cell line was
examined two to three times), 10~h M THF treatment
for 14 days inhibited and reversed the DEX-induced
CLAN formation almost completely (Fig. 3B). Treatment with THF alone was comparable to the untreated
control cells. In dose-response studies, THF (10~'\
1(T7, and 10~8 M) was added to TM cell cultures with
DEX (10~7 M) for 14 days. Tetrahydrocortisol inhibited DEX-induced CLAN formation in a dose-dependent manner with an IC50 of 5.7 X 10~7 M (Fig. 3C).
The THF dose-response study (Fig. 3C) was performed
in a TM cell line that was a very high responder to
DEX treatment (i.e., almost all the DEX-treated TM
cells developed CLANs in the absence of THF). The
heterogeneity in responsiveness between TM cell lines
has been reported.25'35
We examined as well the morphologic effects of
DEX and THF on the TM cytoskeleton by light and
electron microscopic analysis of microfilaments and
microtubules. Data shown are representative of the
results from hundreds of photomicrographs taken
from all the TM cell lines examined. Normal cultured
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017
Tetrahydrocortisol Effects on the Trabecular Meshwork Cytoskeleton
809
60
w
50
O
40
O
#
V
T
4
6
8
Control
DEX
THF
THF+DEX
10
12
14
16
CON
TIME (days)
FIGURE 3. Effect of dexamethasone (DEX) and tetrahydrocortisol (THF) on cross-linked
actin network (CLAN) formation in cultured human trabecular meshwork cells. (A) Percent
of TM cells that develop CLANs when incubated without (control) or with DEX (10~7 M)
and/or THF (10~b M) for 0 to 14 days (average of two independent experiments). Control
(O), DEX (•), THF (V), DEX + THF (T). (B) Mean (± SEM) CLAN response (n = 16
assays) of seven different TM cell lines cultured with DEX (1(T7 M), THF (10~6 M), or DEX
+ THF for 14 days. The response of the DEX-treated group is significandy different from the
other three groups (P< 0.001). (C) Effect of THF dose on DEX-induced CLAN formation in
cultured human TM cells. Trabecular meshwork cells were incubated with DEX (10~7 M)
in the presence or absence of 10~6, 10~7, or 10~H M THF for 14 days (n = 4). The IC50 for
THF inhibition is 5.7 X 10~7 M, and r = 0.98.
TM cells have abundant microfilaments that are in
linear arrays and are bundled into stress fibers (Figs.
4A, 5A). Dexamethasone treatment caused the TM
microfilaments to reorganize into cross-linked, geodesic, dome-like structures (Figs. 4B, 5B) composed of
actin filaments but not as heavily labeled with immunogold particles compared to the untreated control
TM cells. Treating the TM cells with THF alone caused
an increase in micron" lament bundling (Figs. 4C, 5C),
and the microfilaments appeared to be arranged in a
more linear fashion along the cell axis. The concomitant treatment of cultured TM cells with DEX and
THF for 14 days resulted in a mixture of microfilament
bundles and remnants of CLANs (Figs. 4D, 5D). Actin
immunogold staining appeared to return to control
levels in these cells.
In addition to the steroid-induced changes in TM
cell microfilament organization, DEX and THF altered the organization of TM cell microtubules. Microtubules of untreated TM cells are organized in astral
arrays extending from the microtubule organizing
center at the periphery of the nucleus to the edges of
the cell (Fig. 4E). Linear stretches of microtubules are
decorated readily with the anti-tubulin gold complex
(Fig. 5E). Treatment with THF alone does not dramatically alter TM cell microtubule organization (Fig.
5G). However, DEX-treatment caused two major
changes in the TM cell microtubule structure. In many
of the DEX-treated cells, the microtubule organizing
center appeared to migrate to a position above the
nucleus (Fig. 4F). Immuno-ultrastructural analysis revealed abundant microtubule tangles throughout the
TM cell cytoplasm (Fig. 5F). The addition of THF to
the DEX-treated cells appeared to normalize partially
the TM microtubule organization (Fig. 5H).
DISCUSSION
The administration of glucocorticoids by a variety of
routes can lead to the development of ocular hypertension and glaucoma in susceptible persons. 1 " 7 In
addition, it is possible to generate ocular hypertension
in animals by ocular administration of a potent glucocorticoid.13"17 Elevated IOP associated with glucocorticoid administration is caused by increased aqueous
humor outflow resistance 1 " 4 and is associated with biochemical and ultrastructural changes in the trabecular
meshwork. 8 " 12 Accordingly, numerous investigators18"27 have studied the effects of glucocorticoids on
cultured TM cells to discover the molecular mechanism (s) responsible for glaucomatous IOP elevation.
We propose the following hypothesis for glucocorticoid-induced ocular hypertension. Trabecular meshwork cells contain classical GR36'37 and are, therefore,
targets for glucocorticoid action. Binding of the glucocorticoid with the TM cell GR alters TM cell gene
expression,18 leading to the differential expression of
a subset of proteins. 1819 There is increased deposition
of extracellular matrix molecules,20'21 decreased expression of extracellular proteinases, 22 ' 23 and a reorganization of the TM cytoskeleton.24'25 The TM nucleus
and cell enlarge, 24 " 26 and various important TM cell
functions are inhibited. 182527 The combination of
these glucocorticoid-mediated effects on the TM leads
to progressive increased resistance in aqueous humor
outflow and to the development of ocular hyperten-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017
810
Investigative Ophthalmology & Visual Science, April 1996, Vol. 37, No. 5
FIGURE 4. Effect of dexamethasone (DEX) and tetrahydrocortisol (THF) on cultured trabecular meshwork (TM) cell microfilament and microtubule cytoskeletal elements. Confluent
TM cells were treated without (control) or with DEX (1(T7 M), THF (10~6 M), or DEX +
THF for 14 days. Actin filaments were visualized by rhodamine-phalloidin staining (A to
D). Microtubules were visualized using anti-tubulin immunofluorescence (E,F). Control
(A,E), DEX (B,F), THF (C), and DEX + THF (D). Microtubule organizing centers are
shown by arrows. Magnification bar = 50 fxm.
sion. The glucocorticoid-induced elevated IOP can
lead to optic nerve head damage and glaucomatous
optic neuropathy. Details on the role of glucocorticoids in the generation of ocular hypertension can be
found in a recent review.38
Given this hypothesis, it may be possible to intervene in the glucocorticoid-induced damage to the TM
at various steps. For example, concomitant treatment
with a potent glucocorticoid antagonist, such as RU486, should prevent glucocorticoid-induced ocular hypertension at the TM cell by binding to the glucocorticoid receptor at the beginning of this cascade. There
have been several reports of topically administered
RU-486 partially blocking glucocorticoid-induced ocular hypertension in rabbits.3'140 It was suggested that
THF may be a glucocorticoid antagonist because it has
been reported to lower IOP in DEX-induced ocular
hypertensive rabbits.32 In addition, glucocorticoid antagonists have been shown to inhibit DEX-induced
changes in the cytoskeleton of cultured human TM
cells.25 The administration of glucocorticoids also can
cause systemic hypertension'11'42 and catabolic loss in
body weight.43M Tetrahydrocortisol was administered
systemically to determine whether it could block these
glucocorticoid-mediated systemic effects. Our results
show that THF is not a glucocorticoid antagonist in
the classical sense because it was unable to block the
effects of systemically administered DEX on body
weight and on systemic blood pressure, as has been
previously demonstrated with a more conventional
glucocorticoid antagonist such as RU-486.44 We have
shown that THF is not a GR antagonist because it does
not bind to the GR.
Glucocorticoids have been shown to generate
many changes in cultured TM cells.18"27 One of the
more dramatic changes is a major reorganization of
the TM cytoskeleton, which involves the actin microfilaments24'25 as well as the microtubules.45 This glucocorticoid-mediated change in the TM cytoskeleton is
associated with altered TM cell function25 and may be
responsible for corticosteroid-induced ocular hypertension. Although THF is not an antagonist at the
level of the glucocorticoid receptor, it was nonetheless
able to inhibit and reverse DEX-induced microfilament reorganization (CLAN formation) in TM cells.
There appear to be subtle effects of THF alone on TM
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017
Tetrahydrocortisol Effects on the Trabecular Meshwork Cytoskeleton
811
FIGURES. Effect of dexamethasone (DEX) and tetrahydrocortisol (THF) on cultured trabecular meshwork
(TM) cell micron lament and
microtubule cytoskeletal elements. Confluent TM cells
were treated without (control) or with DEX (1(T7 M),
THF (10"(i M), or DEX +
THF for 14 days. As described
in the Materials and Methods
section, TM cells were prepared for immunogold transmission electron microscopy
for actin (A to D) and tubulin
(E to H). Control (A.E), DEX
(B,F), THF (C,G), and DEX
+ THF (D,H). Magnification
bar = 1 //m.
cell actin microfilaments, causing increased bundling
and density of the stress fibers. Topical ocular administration of THF alone did not alter the IOP of rabbits,32
suggesting that this effect of THF on the TM cell cytoskeleton does not cause a significant change in the
outflow facility. However, the THF-mediated bundling
effect on actin appears to compete with the DEX-induced formation of cross-linked actin networks and
may thereby inhibit the IOP-elevating activity of DEX.
Although these steroid-induced cytoskeletal effects
have been shown only in cultured TM cells, it is possible that similar cytoskeletal changes occur in organ-
cultured eyes in situ as well as steroid-treated eyes in
vivo. We are currendy testing this hypodiesis by determining whether DEX-induced CLANs are generated in TM tissue.
Although the specific molecules responsible for
these steroid-induced cytoskeletal changes are as yet
unknown, it is possible that the IOP activity of both
THF and DEX may be mediated by actin-binding or
actin-associated proteins that regulate the organizational structure of actin microfilaments. Dexamethasone treatment also induced changes in the microtubule organization of TM cells. This may be the result
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017
812
Investigative Ophthalmology 8c Visual Science, April 1996, Vol. 37, No. 5
of a direct effect of DEX on the expression of microtubule organizing proteins, or it may indirectly be the
result of the DEX-induced reorganization of the TM
microfilaments or other indirect DEX effects. We have
recently shown that THF and DEX can independently
regulate the expression of specific proteins in cultured
TM cells,46 and some of these proteins may be important candidates in steroidal regulation of the TM
cytoskeleton.
Tetrahydrocortisol is a natural metabolite of cortisol and, as such, may be relatively free of ocular or
systemic side effects. Data from the current study show
that, although it is not a GR antagonist, THF does
modify the cytoskeleton of cultured human trabecular
meshwork cells and is capable of reversing DEX-induced changes in the cytoskeleton. Preliminary data
from other studies suggest that THF may also have an
IOP lowering activity in patients with primary open
angle glaucoma 33 as well as angiostatic activity.47
Whether these findings can be correlated direcdy is
unknown and will require additional study.
13.
14.
15.
16.
17.
18.
19.
Key Words
cytoskeleton, dexamethasone, microfilaments,
drocortisol, trabecular meshwork
tetrahy-
References
1. Becker B, Mills DW. Corticosteroids and intraocular
pressure. Arch Ophthalmol. 1963;70:500-507.
2. Armaly MF. Effect of corticosteroids on intra ocular
and fluid dynamics: I: The effect of dexamethasone
in the normal eye. Arch Ophthalmol. 1963;70:482-491.
3. Kass MA, Johnson T. Corticosteroid-induced glaucoma. In: Ritch R, Shields MB, Krupin T, eds. The
Glaucomas. St. Louis: CV Mosby; 1989; 1161-1168.
4. Bernstein NH, Schwartz B. Effects of long-term systemic steroids on ocular pressure and tonographic
values. Arch Ophthalmol. 1962;68:742-753.
5. Francois J. Cortisone et tension oculaire. Ann Ocul.
1954; 187:805-816. In French.
6. Goldmann H. Cortisone glaucoma. Arch Ophthalmol.
1962; 68:621-626.
7. Shields MB. Textbook of Glaucoma. 3rd ed. Baltimore:
Williams & Wilkins; 1992:374-380.
8. Johnson DH, Bradley JMB, Acott TS. The effect of
dexamethasone on glycosaminoglycans of the human
trabecular meshwork in perfusion organ culture. Invest Ophthalmol Vis Sci. 1990;31:2568-257l.
9. Johnson DH, Bradley JM, Scott TS, Fauss DJ, Polansky
JR. The effect of steroids on human trabecular meshwork in perfusion organ culture. ARVO Abstracts. Invest Ophthalmol Vis Sci. 1989; 30:223.
10. Rohen JW, Linner E, Witmer R. Electron microscopic
studies on the trabecular meshwork in two cases of
corticosteroid-glaucoma. Exp Eye Res. 1973; 17:19-31.
11. Roll R, Benedikt O. Electronmicroscopic investigation
of the trabecular meshwork in cortisonglaucoma. Klin
Mbl Augenheilk. 1979; 174:421-428.
12. Toriyama K. An electron microscopic study on the
20.
trabecular meshwork in corticosteroid-glaucoma. Folia
Ophlhalmoljpn. 1979;30:1583-1589.
Lorenzetti OJ. Effects of corticosteroids on ocular dynamics in rabbits. JPharmacol Exp Ther. 1970; 175:763772.
Knepper PA, Breen M, Weinstein HG, Blacik LJ. Intraocular pressure and glycosaminoglycan distribution in
the rabbit eye: Effect of age and dexamethasone. Exp
Eye Res. 1978; 27:567-575.
Zhan G-L, Miranda OC, Bito LZ. Steroid glaucoma:
Corticosteroid-induced ocular hypertension in cats.
Exp Eye Res. 1992;54:211-218.
Palkama A, Alaranta S, Oksala O, Krootila K, Virtanen
I, Uusitalo H. Glucocorticoid induced glaucoma in
the cat eye. ARVO Abstracts. Invest Ophthalmol Vis Sci.
1994; 35:2774.
DeSantis L, Garthwaite C, Knepper PA. Dexamethasone-induction of ocular hypertension in the primate.
ARVO Abstracts. Invest Ophthalmol Vis Sci. 1990; 31:99.
Polansky JR, Kurtz RM, Fauss DJ, Kim RY, Bloom E. In
vitro correlates of glucocorticoid effects on intraocular
pressure. In: Krieglstein GK, ed. Glaucoma Update IV.
Berlin: Springer-Verlag; 1991:20-29.
Tripathi BJ, Millard CB, Tripathi RC. Corticosteroids
induce a sialated glycoprotein (Cort-GP) in trabecular
cells in vitro. Exp Eye Res. 1990;51:735-737.
Steely HT, Browder SL, Julian MB, Miggans ST, Wilson KL, Clark AF. The effects of dexamethasone on
fibronectin expression in cultured human trabecular
meshwork cells. Invest Ophthalmol Vis Sci. 1992; 33:
2242-2250.
21. Yun AJ, Murphy CG, Polansky JR, Newsome DA, Alvarado JA. Proteins secreted by human trabecular cells:
Glucocorticoid and other effects. Invest Ophthalmol Vis
Sci. 1989;30:2012-2022.
22. Snyder RW, Stamer WD, Kramer TR, Seftor REB. Corticosteroid treatment and trabecular meshwork proteases in cell and organ culture supernatants. Exp Eye
Res. 1993; 57:461-468.
23. Samples JR, Alexander JP, Fisk A, Acott TS. Regulation
of the levels of human trabecular matrix metalloproteinases and inhibitor by Interleukin-1 and dexamethasone. Invest Ophthalmol Vis Sci. 1993; 34:3386-3395.
24. Wilson K, McCartney MD, Miggans ST, Clark AF. Dexamethasone induced ultrastructural changes in cultured human trabecular meshwork cells. Curr Eye Res.
1993; 12:783-793.
25. Clark AF, Wilson K, McCartney MD, Miggans ST, Kunkle M, Howe W. Glucocorticoid-induced cross-linked
actin networks in cultured human trabecular meshwork cells. Invest Ophthalmol Vis Sci. 1994;35:281-293.
26. Tripathi BJ, Tripathi RC, Swift HH. Hydrocortisoneinduced DNA endoreplication in human trabecular
cells in vitro. Exp Eye Res. 1989;49:259-270.
27. Shirato S, Bloom E, Polansky J, Alvarado J, Stilwell L.
Phagocytic properties of confluent cultured human
trabecular meshwork cells. ARVO Abstracts. Invest
Ophthalmol Vis Sci. 1988;29:125.
28. Treister G, Mannor S. Intraocular pressure and outflow facility: effect of estrogen and combined estrogen-progestin treatment in normal human eyes. Arch
Ophthalmol. 1970; 83:311-318.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017
Tetrahydrocortisol Effects on the Trabecular Meshwork Cytoskeleton
29. Green K, Phillips CI, Gore SM, Elijah RD, Bowman
KA, Cullen P. Ocular fluid dynamics response to topical RU486, a steroid blocker. CurrEyeRes. 1985; 4:605612.
30. Witzmann R. Effect of spironolactone on intraocular
pressure in glaucoma patients. Klin Monatsbl Augenheilkd. 1980; 177:445-446.
31. LambleJW, Lamble AP. Some effects of progestogens,
oestrogens and androgens on the ocular tension of
rabbits and owl monkeys. Exp Eye Res. 1978; 26:599610.
32. Southren AL, l'Hommedieu D, Gordon GG,
Weinstein BI. Intraocular hypotensive effect of a topically applied cortisol metabolite: 3a,5/?-tetrahydrocortisol. Invest Ophthalmol Vis Sd. 1987; 28:901-903.
33. Southren AL, Wandel T, Gordon GG, Weinstein BI.
Treatment of glaucoma with 3a,5/?-tetrahydrocortisol:
a new therapeutic modality. / Ocul Pharmacol.
1994; 10:38-391.
34. Eliard PH, Rousseau GG. Thermodynamics of steroid
binding to the human glucocorticoid receptor. Biochemj. 1984; 218:395-404.
35. Clark AF, Miggans ST, Wilson K, Browder S, McCartney MD. Cytoskeletal changes in cultured human glaucoma trabecular meshwork cells. / Glaucoma.
1995;4:183-188.
36. Weinreb RN, Bloom E, Baxter JD, et al. Detection of
glucocorticoid receptors in cultured human trabecular cells. Invest Ophthalmol Vis Sd. 1981; 21:403-407.
37. Hernandez MR, Wenk EJ, Weinstein BI, et al. Glucocorticoid target cells in human outflow pathway: autopsy and surgical specimens. Invest Ophthalmol Vis Sd.
1983;24:1612-1616.
38. Clark AF. Steroids, ocular hypertension, and glaucoma. / Glaucoma. 1995;4:354-369.
39. Green K, Cheeks L, Slagle T, Phillips CI. Interaction
813
between progesterone and mifepristone on intraocular pressure in rabbits. CurrEyeRes. 1989;8:317-32O.
40. Denis P, Elena PP, Nordmann JP, Sarrieau A, Saraux
H, Rostene W. Mifepristone (RU 486): In vitro binding to glucocorticoid receptors in iris-ciliary body and
in vivo effects on intraocular pressure in rabbits.
ARVO Abstracts. Invest Ophthalmol Vis Sd. 1993;
34:1115.
41. Mall G, Reinhard H, Stopp D, Rossner JA. Morphometric
observations on the rat heart after high dose treatment
with cortisol. Virchmus Arch. 1980;385:169-180.
42. Nasletti A, Erman A, Cagen LA, Baer PG. Plasma concentrations, renal excretion, and tissue release of prostaglandins in the rat with dexamethasone-induced hypertension. Endocrinology. 1984; 114:1033-1040.
43. Clark AF, DeMartino GN, Wildenthal K. Effects of
glucocorticoid treatment on cardiac protein synthesis
and degradation. AmJ Physiol. 1986;250:C821-C827.
44. Konagaya M, Bernard PA, Max SR. Blockade of glucocorticoid receptor binding and inhibition of dexamethasone-induced muscle atrophy in the rat by
RU38486, a potent glucocorticoid antagonist. Endocrinology. 1986; 119:375-380.
45. McCartney MD, Wilson K, Miggans ST, Clark AF. Effect of dexamethasone and tetrahydrocortisol on cultured human trabecular meshwork cell microfilament
and microtubule structure. ARVO Abstracts. Invest
Ophthalmol Vis Sd. 1994; 35:2735.
46. Kawase K, Steely HT, Clark AF. The effects of dexamethasone and tetrahydrocortisol on protein expression in cultured human trabecular meshwork cells.
ARVO Abstracts. Invest Ophthalmol Vis Sd. 1994;
35:2734.
47. Clark AF, McNatt L, Knepper PA. Angiostatic steroids
as a new class of IOP lowering compounds. ARVO
Abstracts. Invest Ophthalmol Vis Sd. 1994; 35:1057.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933416/ on 05/08/2017