Investigative Ophthalmology & Visual Science, Vol. 32, No. 2, February 1991
Copyright © Association for Research in Vision and Ophthalmology
Effect of Selenire on Epithelium of Cultured Rabbit Lens
K. R. Hightower and J. P. McCready
Selenite (Se) cataract in rabbit lenses was investigated in vitro to define target sites of Se that might be
involved in calcium elevation and lens opacification. Experiments in which the anterior or the posterior
surface of the lens was exposed to Se showed that anterior exposure led to ionic imbalances and
opacification in the whole lens. Posterior exposure to Se (1 mM, 2 hr) had no effect. Se treatment (0.1
mM) of epithelial homogenates led to a 56% loss of thiol (SH) groups, and treatment of lenses cultured
in Se led to a 22% loss. Experiments to assess the effects of Se on SH groups of Ca-ATPase showed
that the transport enzyme was not affected by the poison. To determine whether this negative finding
was due to the lack of accessibility of Se for SH sites in an ordered membrane, Ca-ATPase was also
assayed in homogenate preparations treated with Se; still no inhibition of Ca-ATPase activity was
observed. Therefore, an alternative explanation of calcium elevation was explored. The passive movement of labeled chloride(36Cl) was found to be twice as fast in Se-treated lenses as it was in control
lenses. Measurement of the lens voltage indicated an 18-mV depolarization in Se-treated lenses,
suggesting that Se increased membrane permeability. All cataractogenic changes that occurred after
Se treatment were irreversible—despite intervention with external application of reduced glutathione
or cysteine. This finding suggests that irreversible loss of SH groups in lens membranes is important
in maintaining ion homeostasis. Invest Ophthalmol Vis Sci 32:406-409,1991
meability in the lens,5 experiments were designed to
measure SH loss induced by Se and to assess the damage incurred by whole lenses when either the anterior
or posterior surface was exposed to Se.
Little is known of the mechanism by which Selenite (Se) produces cataract—beyond the fact that calcium elevation accompanies lens opacification in
vitro and in vivo and appears to activate proteolysis
via calpain.1 One hypothesis regarding the beginning
of cataract development suggests that Se could oxidize SH groups of cation transport enzymes that
maintain ion homeostasis. 2 Oxidation of membrane
SH groups, controlling passive fluxes of ions, could
also contribute to ion imbalances. Bergad's study, in
which Na/K-ATPase activity in lens homogenates
was inhibited by Se,3 suggests that SH groups might
be involved in cataracts caused by Se. Later studies
verified the finding that sodium transport in rabbit
lenses exposed to Se in vitro is inhibited by Se.4
This study investigated the hypothesis that Se can
modify membrane SH groups that are important in
ion transport and permeability. In particular, since
no explanation exists for calcium elevation in Se cataract, Ca-ATPase activity in lens epithelium and fiber
cell membranes was measured. Because a small loss
in membrane SH groups also affects membrane per-
Materials and Methods
In this study, four-week-old New Zealand white
rabbits were killed with T-61 Euthanasia solution
(Hoechst-Roussell Agri-Vet Company, Somerville,
NJ). These investigations conformed to the Arvo Resolution on the Use of Animals in Research. Lenses
were dissected and immediately put into tissue culture medium (TC-199, Gibco Laboratories, Grand
Island, NY). Lenses cultured for 48 and 72 hr were
dissected in a sterile hood and the medium was
changed every 24 hr to replenish the glucose supply.
Se was present only during the first 24 hr of culture.
Lenses were kept in tissue culture dishes containing 8
ml of medium and were maintained at 34°C. Aliquots of a 10-mM Se stock solution, Na2Se03 (Sigma
Chemical Co., St. Louis, MO), were added to the
medium to produce 0.1 mM Se. In experiments in
which only the anterior or posterior surface was exposed to Se, lenses were placed in tissue culture dishes
where the medium just reached the lens equator such
that one surface was exposed to moist air. The Se
concentration for these experiments was 1.0 mM and
the exposure time 2 hr.
From the Eye Research Institute, Oakland University, Rochester, Michigan.
Supported by grant No. RO1-EY-03681-10 from the National
Eye Institute, National Institutes of Health, Bethesda, Maryland.
Submitted for publication: March 26, 1990; accepted March 31,
1990.
Reprint requests: Kenneth R. Hightower, 420 Dodge Hall, Eye
Research Institute, Oakland University, Rochester, MI 483094401.
Concentrations of calcium and sodium were measured using atomic absorption spectroscopy (902
GBC Spectrophotometer, Applied Research Labora-
406
Downloaded From: http://iovs.arvojournals.org/ on 06/14/2017
No. 2
tories, Dearborn, MI). Lenses were homogenized in
10% trichloroacetic acid (TCA) with 0.2% lanthanum
added to prevent phosphate interference. Extracellular space in young rabbit lenses is less than 6% and is
limited primarily to the outer capsule; hence, no corrections are made in determining total ion levels.6
Ca-ATPase activity in the capsule epithelium was
measured by the colorimetric determination of inorganic phosphate present in incubation media. The
preparation of the tissue and specifics of the assay, as
reported earlier,7 consisted of a crude preparation of
membranes obtained after they were centrifuged at
37,000 X g for 30 min and washed twice. This preparation yields more than 95% of the Na/K-ATPase
activity observed in the whole epithelium.
Membrane permeability was assessed by measuring the lens voltage and the accumulation of 36C1, an
ion known to move into and out of the lens passively,
ie, according to Nernst electrochemical gradients
rather than active transport.8 Lenses were transferred
to 199m containing 9000 cpm/0.1 ml of 36C1 (NEN
Du Pont Co., Wilmington, DE). The rate of accumulation of 36C1 was measured over 16 hr, after which
time lenses were homogenized in 2 ml of 10% TCA
and 0.1-ml aliquots were counted (Beckman LS 5801
scintillation counter, Beckman Instruments, Inc.,
Fullerton, CA). The data are expressed as the activity
ratio measured in the lens and medium.
Results
Experiments were performed to define the anatomic site of Se action. Because the primary site of
active transport lay chiefly in the lens epithelium,
lenses were cultured so that only one surface of the
lens was exposed to Se (1 mM). Lenses were cultured
in a petri dish and half-immersed in medium, with
either the anterior or posterior surface exposed to a
moist-air environment. After culture at 34°C for 2 hr,
lenses were cultured in medium fully immersed in the
Table 1. Sodium and calcium concentrations after
selective treatment with selenite of anterior or
posterior surfaces of rabbit lens
Surface exposed
to medium
Anterior
No Se (4)
Se(6)
Posterior
No Se (5)
Se(6)
407
SELENITE BINDING TO MEMBRANE SH GROUPS / Highrower and McCready
Sodium cone
(mM)
Calcium cone
(mM)
14.5 ±3.1
25.3 ±2.7
0.18 ±0.009
0.50 ± 0.006
11.1 ±3.2
14.5 ±2.6
0.26 ± 0.02
0.30 ± 0.08
Values represent means ± I SE. Significant differences occur for sodium
and calcium values between control and Se experiments only for anterior
surface exposure, at the P = 0.01 level. The number in parentheses is the
number of lenses.
Downloaded From: http://iovs.arvojournals.org/ on 06/14/2017
Table 2. Membrane sulfhydryl groups
of rabbit lens epithelium
Experiment
Control
Epith. homogenates cultured in
0.1 mM Se for 15 min
Lens culture in 0.1 mM Se for
24 hr
nm SH
Loss
(4)
4.60 ±0.10
—
(7)
2.05 ± 0.05
56
(6)
3.60 ± 0.08
22
Values are means ± 1 SE. Differences are significant at the P = 0.01 level
Epith. homogenates refer to preparation of lens epithelium homogenized in
buffer containing 0.1 mM sodium selenite for no longer than 15 min. Number in parentheses is the number of lenses.
absence of Se for an additional 24 hr—sufficient time
to show ionic imbalances. Table 1 shows that control
lenses that were not exposed to Se survived this treatment well, as assessed by sodium and calcium levels.
Another indication that this treatment did not cause
significant, irreversible damage is that normal membrane potentials were observed in lenses immediately
on return to normal culture. Thus, the potential of
lenses partially immersed was —63 mV (n = 4) compared with -67 mV (n = 4, SE = ±2 mV) for normal
lenses.
The data in Table 1 also show that little change
occurred in sodium or calcium levels in the six lenses
whose posterior surface was exposed to 1.0 mM Se for
2 hr and then cultured for 24 hr in normal medium.
The lenses remained transparent throughout both
culture periods. In contrast, lenses whose anterior
surface was exposed to Se gained significant amounts
of sodium and calcium, with levels of each ion approximately doubling after overnight culture. Moreover, lens appearance was notably more hazy after
anterior exposure. These findings suggest that the anterior surface of the lens might be the primary target
of Se toxicity that initiates cataract formation.
Because exposure of the anterior surface to Se appears to produce ionic imbalances and opacification,
detailed analysis of the lens epithelium was initiated.
To assess the vulnerability of membrane SH to Se, a
crude preparation of epithelial membranes was exposed to Se for 15 min (0.1 mM). Table 2 shows that
the total number of SH groups declined from 4.6
nmol in a freshly removed epithelium preparation to
2.1 nmol after Se treatment, a 56% loss.
To assess accessibility of SH groups in an intact
epithelium, lenses were cultured in medium containing 0.1 mM sodium Se for 24 hr. After incubation,
lenses were rinsed thoroughly. The epithelia were removed and homogenized, and the lenses were centrifuged to obtain a membrane pellet. SH groups of the
membrane pellet were again assayed. Table 2 shows
that the total number of SH groups decreased from
4.6 to 3.6 nmol, a 22% loss. Thus, regardless of the
Vol. 32
INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / Februory 1991
408
conditions under which the epithelium is exposed to
Se, a loss in SH groups occurs.
A loss of membrane SH groups in the lens epithelium could explain the inhibition of sodium export
from the lens and the corresponding loss of Na/KATPase activity observed by Bergad3 in lens homogenates. To determine whether calcium elevation might
be related to loss of SH groups in the Ca-ATPase
transporter, experiments were performed to measure
Ca-ATPase activity in lens epithelia of Se-treated
lenses. Epithelia were removed from lenses first exposed to Se for 24 hr, and then cultured an additional
48 hr, at which time calcium elevation had occurred.
Table 3 shows that Ca-ATPase in these lens epithelia
was not affected.
Se was added directly to homogenates of epithelia
membranes to determine whether lack of Ca-ATPase
inhibition by Se might be due to inaccessibility of
interior or buried SH groups. Thus, membranes obtained by centrifuging homogenates of epithelia were
resuspended in an ionic buffer in the presence of Se
during the 1-hr incubation to assay ATPase activity.
As seen in Table 3, Se appears to have had no
harmful effect on Ca-ATPase-mediated hydrolysis
of ATP.
Although cytotoxicity of Se may involve numerous
processes, a reasonable hypothesis must allow for the
potential chemical reaction of Se with membrane SH
groups. Externally applied glutathione (GSH) or cysteine to Se-treated lenses might prevent further ionic
imbalances or possibly reverse the cataractogenic
process initiated during thefirst24 hr in which Se was
present. When lenses were pretreated with Se, exposed to 5 mM GSH (in a complete medium), and
further cultured, a statistically significant ionic imbalance (Table 4) was seen, as measured by sodium
and calcium levels. Sodium levels increased whether
GSH was absent or present. Calcium increased from
0.24 mM to 0.32 mM in Se-treated lenses exposed to
GSH. The same experiment repeated with 5 mM
cysteine showed that it was also ineffective.
Although Se inhibition of sodium transport helps
Table 3. Ca-ATPase analysis of lens epithelium
after selenite treatment
Preparation
Cultured lens (72 hr)
Control
Se(0.1 mM)
Homogenate
Control
Se (0.1 mM)
Ca-A TPase
0.030 ± 0.006 (5)
0.029 ±0.004 (16)
0.030 ± 0.004 (25)
0.026 ± 0.005 (8)
Values are means ± 1 SE; units are micromoles inorganic phosphate liberated per hour per capsule epithelium from 4-week-old rabbit lens. No difference in means at 0.05 level. Number in parentheses is the number of capsule
epithelia.
Downloaded From: http://iovs.arvojournals.org/ on 06/14/2017
Table 4. Effect of glutathione and cysteine on
sodium and calcium levels in selenite-treated
rabbit lenses
Sodium cone (mM)
Experiment
Control
Se
Se, GSH
Se, Cysteine
(5)
(3)
(9)
(3)
18.3 ±2.4
26.2 ± 1.7
26.5 ± 1.6
30.9 ± 3.6
Calcium cone (mM)
0.24
0.40
0.32
0.42
±0.01
± 0.05
± 0.02
± 0.06
All lenses were cultured a total of 72 hr. GSH or cysteine was added (5
mM) after the first 24 hr of Se treatment, after selenite was removed from
media. Values represent means ± SE. Experimental means for Na and Ca
levels are statistically greater than control means at the P = 0.01 level.
Number in parentheses is the number of lenses.
explain sodium elevation in Se cataracts,4 the lack of
inhibition of the calcium ATPase does little to explain calcium elevation. In the absence of calcium
pump inhibition, the alternative explanation for calcium elevation is increased membrane permeability
in the presence of Se. A reliable indication of altered
membrane permeability is membrane depolarization
and a corresponding increase in Cl influx.8 Labeled
Cl uptake was measured because Cl is passively distributed and movee independently of active transport
mechanisms; its rate is governed only by membrane
voltage and permeability. Lenses were exposed to Se
for 24 hr and cultured for extended periods in the
absence of Se. After these various culture times,36Cl
accumulation was measured over 16 hr. Uptake
values are given as the ratio of the labeled Cl accumulated in the lens to the amount of labeled Cl in the
medium (Table 5). In the first group of lenses (control), Cl uptake after culture for 48 hr was 0.19, compared with 0.16 in freshly excised lenses. Se-treated
lenses, however, showed a Cl uptake value of 0.23
± 0.02, a 21% increase (statistically significant at P
= 0.01 level). When control and Se-treated lenses
were cultured for 72 hr, with Se still present for 24 hr,
the Cl uptake differed markedly, ie, 0.25 compared
with 0.42 ± 0.04 for experimental lenses, a 68% increase. Measurement of membrane voltage in these
lenses showed that Se had little effect on the voltage,
ie, from - 6 5 mV (n = 4, SE ± 2 mV) to - 6 0 mV in
the lenses cultured for 48 hr (n = 4, SE ± 2 mV) and
Table 5. Effect of selenite on 36C1uptake into
cultured rabbit lenses
Time (hr)
Control
Experimental
0
48
72
0.16 ±0.01 (4)
0.19 ±0.02 (4)
0.25 ± 0.02 (4)
0.23 ± 0.02 (6)
0.42 ± 0.04 (6)
Values represent means ± SE. Experimental means are statistically greater
than control means at the 0.05 level. Data are expressed as ratios of counts
per minute in the lens to media after 16-hr uptake. Number in parentheses
represents number of lenses.
No. 2
SELENITE BINDING TO MEMBRANE SH GROUPS / Highrower ond McCreody
from -63 mV (n = 5, SE ± 3 mV) to -45 mV in the
lenses cultured for 72 hr (n = 5, SE ± 3 mV).
Discussion
As summarized recently by Shearer,2 Se cataract
may be a consequence of oxidation of protein SH
groups or nonenzymatic reaction of Se and protein
SH to form selenotrisulfides. Bunce and Hess,9 as
well as Shearer,10 provided evidence that Se initiates a
decline in reduced glutathione, stimulation of the
hexosemonophosphate shunt, and a decrease in /3Nicotinamide Adenine Dinucleotide Phosphate (reduced form NADPH). No evidence of an increase in
whole-lens oxidized glutathione (GSSG) has been reported and no increase in insoluble lens protein disulfide has been seen in Se cataract. Therefore, although
widespread oxidation is not occurring in Se cataract,
Se may be reacting with membrane SH groups to
form selenotrisulfide complexes.2 A small number of
critical membrane SH groups must be involved to
cause severe ionic imbalances, such as those observed
in epithelial cells of the rabbit lens treated by hydrogen peroxide or membrane sulfhydryl probes such as
p-chloromercuriphenyl sulfonic acid (PCMBS).'' In
these studies, a specific loss of membrane SH groups
in the lens epithelium was accompanied by a change
in ionic fluxes and ion imbalances. These data are
consistent with the hypothesis that membrane SH
groups may be targets of Se.
Because exposure of the anterior surface of the lens
to Se results in whole-lens damage, the critical SH
groups would appear to reside with either the peripheral anteriorfibersor the epithelium. The likely target
site is the lens epithelium, since epithelia preparations show a loss of SH groups. Moreover, there is no
evidence that anterior fibers are different from posteriorfibers,which are insensitive to Se but contain the
same number of membrane SH groups.5
Since inhibition of the calcium transport enzyme
by Se is not evident in Se cataract, alternative explanations for calcium elevation include impaired
Na/Ca exchange processes or increased membrane
permeability. In the rabbit lens, Na/Ca exchange has
not been reported. Data obtained in this study suggest
that altered membrane permeability is a likely mechanism by which Se causes calcium to accumulate in
the lens. This conclusion is based on the fact that
exposure of lenses to Se caused both a voltage depolarization and an increased rate of Cl uptake—an
indication that membrane integrity has been compromised. In the absence of evidence for calcium
channels in the lens, we assume that calcium entry
occurs by way of the many nonselective ion channels.12
The identification of the particular membrane SH
groups is not possible from the results in this study,
Downloaded From: http://iovs.arvojournals.org/ on 06/14/2017
409
but there is some evidence to suggest that external
(exofacial) membrane SH groups are probably not
the critical sites. This conclusion is indicated by the
finding that treatment of Se-exposed lenses to GSH
or cysteine had little effect on restoring ion homeostasis. If internal, ie, cytosol-exposed, membrane SH
groups were target sites of Se, would not intracellular
GSH at a level of 9 mM be sufficient to reduce the
affected SH groups? It is more likely that the SH
groups affected are those located or buried in the interior of the membrane and not so easily accessed by
intracellular GSH or externally applied GSH.
In summary, these findings indicate that in addition to the previously known target of Se toxicity, ie,
the sodium pump,3'4 membrane SH groups of the
anterior aspect of the lens, probably the epithelium,
are also susceptible. The presumed function of these
SH groups may be associated with maintaining normal membrane permeability, that is, nonselective ion
channels described by Rae,12 which might also allow
calcium entry. It is noteworthy that calcium ATPase
is considerably less sensitive to Se than Na/K-ATPase. Nevertheless, elevation of calcium levels can be
initiated in Se cataract, despite uninhibited Ca-ATPase activity.
Key words: lens, Se, Ca-ATPase, SH groups, epithelium
References
1. Hightower KR, David LL, and Shearer TR: Regional distribution of free calcium in Selenite cataract: relation to Calpain II.
Invest Ophthalmol Vis Sci 28:1702, 1987.
2. Shearer TR, David LL, and Anderson RS: Selenite cataract: a
review. Curr Eye Res 6:289, 1987.
3. Bergad PL and Rathbun WB: Inhibition of Na/K-ATPase by
sodium Selenite and reversal by glutathione. Curr Eye Res
5:919, 1986.
4. Hightower KR and McCready JP: Effects of selenium on ion
homeostasis and transparency in cultured lenses. Invest
Ophthalmol Vis Sci 30:171, 1989.
5. Hightower KR and Harrison SE: Valinomycin cataract: the
relative role of calcium and sodium accumulation. Exp Eye
Res 34:941, 1982.
6. Hightower KR, Leverence V, and Reddy VN: Calcium transport in the lens. Invest Ophthalmol Vis Sci 19:1059, 1980.
7. Hightower KR: The importance of membrane sulfhydryl
groups to calcium homeostasis in the lens. Curr Eye Res 4:857,
1985.
8. Kinsey VE and Hightower KR: Studies on the Crystalline Lens
XXII: characterization of Chloride Movement Based on the
Pump-Leak Model. Exp Eye Res 23:425, 1976.
9. Bunce GE and Hess JL: Biochemical changes associated with
Selenite-induced cataract in the rat. Exp Eye Res 33:505, 1981.
10. Shearer TR, Anderson RS, and Britton JL: Uptake and distribution of radioactive selenium in cataractous rat. Curr Eye Res
2:561, 1982.
11. Hightower KR, Reddan JR, and Dziedzic DC: Susceptibility
of lens epithelial membrane SH groups to hydrogen peroxide.
Invest Ophthalmol Vis Sci 30:569, 1989.
12. Rae JL: The application of patch clamp methods to ocular
epithelia. Curr Eye Res 4:409, 1985.