ABSTRACT

A COMPARATIVE STUDY OF WATER AND SODIUM PERMEABILITY
OF LAKE TROUT AND RABBIT CORNEAS

by Henry Francis Edelhauser

Experiments with corneas in vitro have shown that water
permeability of lake trout corneas is the same whether the
data are based on the transfer of tritiated water from the
epithelial to the "endothelial" bathing media or vice versa.
In these experiments there was present an osmotic gradient
of some 6.6 atmOSpheres which favored movement of water
through the cornea to the solution bathing the "endothelium."
Fish corneas were found to be impermeable to sodium ions in
both directions. Water permeability constants and sodium
flux for normal and denuded corneas indicate that the
"endothelium" offers some resistance to the transfer of water
and sodium into or across the fish cornea, but the major
resistance to this movement is provided by the epithelial
layer.

Correlations between water permeability, sodium flux and
histopathological conditions of diseased corneas are noted.
Corneas in the late stages of the disease are at maximal
hydration in situ, whereas, the normal and early disease stages

imbibe water in vitro from the bathing media. Corneal sodium

Henry Francis Edelhauser

ion concentration varied inversely with corneal hydration in
diseased lake trout corneas.

For fish corneas the permeability constant for movement
of sodium across the "endothelial" barrier in a cornea-
aqueous direction is 0.029mEq/cm2/hr some 16% lower than that
observed in rabbit corneas. HTO and Na22 flux in this fish
corneal perfusion study occurred independent of each other;
whereas, with rabbit corneas osmotic flow of water occurred.
Rabbit corneas had a 3 to 5 fold greater water exchange than
fish corneas and they were found to actively transport sodium
in the tear—aqueous direction; however, in the fish cornea
no sodium flux occurs. An adenosine triphosphatase was
identified in the basement layers of the corneal epithelium,

stroma and corneal "endithelium" of fish.

A COMPARATIVE STUDY OF WATER AND SODIUM PERMEABILITY

OF LAKE TROUT AND RABBIT CORNEAS

BY

Henry Francis Edelhauser

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Physiology

1966

ACKNOWLEDGMENT

The author wishes to express his appreciation to
Dr. P. O. Fromm, Department of Physiology, for his guidance
and assurance throughout this study. Sincere appreciation
is also extended to Dr. J. R. Hoffert, Department of
Physiology, for his assistance and consultation which con-
tributed greatly to this work.

He also wishes to thank Dr. B. V. Alfredson, chairman
of the Department of Physiology, for providing the opportun-
ity for graduate study in the department and providing the
facilities for such study. The opportunity for consultation
with the members of his guidance committee, Dr. W. L. Frantz,
Dr. W. D. Collings, and Dr. E. P. Reineke during his edu-
cational program and during the preparation of this disserta—
tion is greatly appreciated.

Thanks are also extended to Mr. K. Irish both for his
help in construction of the recirculation blocks and for his
interest in the project. The author is also indebted to his
fellow graduate students for their technical assistance and
helpful suggestions throughout this study.

In addition the writer is indebted to the U.S.P.H.S.,
Division of National Institute of General Medical Sciences,
for a Pre-doctoral Fellowship and funds in support of this
work. This research was also supported in part by grant
NB 04125 from the National Institute of Neurological Diseases
and Blindness.

**************

ii

Dedicated to my wife, Barbara, whose
sacrifices and encouragements made

the completion of this dissertation
a reality.

iii

TABLE OF CONTENTS

INTRODUCTION . . . . . . . . . . . . . . . . .
LITERATURE REVIEW. . . . . . . . . . . . . . .

Water Permeability . . . . . . . . . . .
Ionic Permeability . . . . . . . . . . .
Fish Cornea Water and Ionic Permeability

MATERIALS AND METHODS. . . . . . . . . . . . .

Experimental Animals . . . . . . . . . .
Apparatus. . . . . . . . . . . . . . . .
Experiments with Tritiated Water . . . .
Experiments with Sodiumfl . . . . . .

Dual Labeled Experiments Using HTO and Na22.

Corneal Adenosine Triphosphatase . . . .
Surface Area of Cornea . . . . . . . . .

RESULTS. . . . . . . . . . . . . . . . . . . .

I. Theoretical . . . . . . . . . . . . .
Experiments with HTO2 - Equations

Experiments with Na2 and Dual Labeling

Experiment Equations . . . . . .

Equation to Correct for Backdiffusion of

Sodium . . . . . . . . . . .

Equation to Calculate Kp' from T 1/2 .

Permeability Constants . . . . . .

II. Experiments with Tritiated Water. . .

III. Experiments with Na22 . . . . . . . .

IV. Dual Labeling Experiments for HTO and

V. Corneal ATPase - Histochemical. . . .
DISCUSSION . . . . . . . . . . . . . . . . . .
SUMMARY AND CONCLUSIONS. . . . . . . . . . . .
LITERATURE CITED . . . . . . . . . . . . . . .
APPENDIX . . . . . . . . . . . . . . . . . . .

Procedure for counting tritium and Na22
simultaneously . . . . . . . . . . .

Sample calculation for Na22 corrected in”

the tritium counts . . . . . . . . .

iv

Page

11
14
14
14
15
19
22
24
25
27

27
27

28
5O
52
52
55

45
47

51
65
68
72

75

74

LIST OF TABLES

TABLE Page

I. Permeability constants for water transfer and
corneal tritiated water content of normal and
denuded lake trout corneas in vitro. . . . . . . 54

 

II. Permeability constants, tritiated water content,
and hydration of normal and pathological lake
trout corneas. . . . . . . . . . . . . . . . . . 55

III. Average sodium fluxes over four hours; cornea
and media Specific activity; and corneal water
of normal and denuded lake trout corneas
in vitro . . . . . . . . . . . . . . . . . . . . 57

IV. Sodium fluxes; cornea and media specific activ-
ity; and hydration of normal and pathological
lake trout corneas . . . . . . . . . . . . . . . 40

V. Sodium and water content of diseased lake trout
corneas. . . . . . . . . . . . . . . . . . . . . 40

VI. Average permeability constants for water trans-
fer over four hours; cornea and media specific
activity; and water content of normal fish and
rabbit corneas in vitro . . . . . . . . . . . . 45

 

VII. Average sodium fluxes over four hours; cornea
and media Specific activity of normal fish and
rabbit corneas . . . . . . . . . . . . . . . . . 46

APPENDIX TABLES
A. Data from Mark I liquid scintillation system
used for standard curve for dual labeling

experiment . . . . . . . . . . . . . . . . . . . 75

B. Percent turnover of Na22 in fish and rabbit
corneas. . . . . . . . . . . . . . . . . . . . . 78

LIST OF FIGURES

FIGURE Page
1. Corneal Clamps (actual size) . . . . . . . . . . 16
2. Pictorial View of Fish Cornea Perfusion. . . . . 17
5. Pictorial View of Rabbit Cornea Perfusion. . . . 18
4. Schematic. . . . . . . . . . . . . . . . . . . . 28
5. Schematic. . . . . . . . . . . . . . . . . . . . 5O
6. Schematic. . . . . . . . . . . . . . . . . . . . 51

7. Reverse exchange--Equilibration of labeled
sodium between cornea and TC-199 . . . . . . . . 42
8. Corneal ATPase . . . . . . . . . . . . . . . . . 49

APPENDIX FIGURES

A. Standard curve I . . . . . . . . . . . . . . . . 76

B. Standard curve II. . . . . . . . . . . . . . . . 77

vi

INTRODUCTION

In the movement to restock waters of the Great Lakes
region with lake trout, which have been depleted by the sea
lamprey, both state and federal hatcheries have increased
their lake trout rearing programs. Concurrently with this
rearing program in the state of Michigan, it has been ob-
served by Allison (1965) and Hoffert and Fromm (1965) and
others that hatchery-reared lake trout are susceptible to
developing various types of eye abnormalities, such as severe
corneal and lenticular lesions.

If the literature on corneal physiology of the past
twenty years is examined, it will be found that corneas
adapted to an aerial environment have excited enormous inter-
est; in so doing they have caused a general disregard for
the phylogenetically more primitive corneas which develop in
an aquatic environment devoid of a tear film, and, in the
case of the fresh water fish, exposed to a hypoosmotic
medium. In such Species physiological regulation is accentu-
ated, for the tendency of water to enter the cornea and salt
to leave it is greater than in the case of mammals where
corneas are bathed with slightly hyperosmotic tears.

A comparative study of water and sodium permeability

of the isolated normal and pathological lake trout cornea

was undertaken. Also investigated were the roles of the
"endothelial" and epithelial layers of the cornea, as related
to water and sodium permeability. (The word "endothelium" is
enclosed in quotation marks because it is not really endo—
thelium, but mesothelium or mesenchymal epithelium.) Experi-
ments using rabbit corneas were also performed in order to
compare corneas of aerial environments to those from aquatic
environments.

The regulatory mechanisms by which the fish cornea re—
mains transparent are presently unknown. Physiological
processes involved in maintaining water balance in connective
tissues are fundamental; are of great importance in diseases
of connective tissue in general, and are eSpecially interest-
ing in a situation where there is present a potential osmotic
gradient of 6.6 atmOSpheres favoring the uptake of water by
the cornea across the epithelial layer. The ultimate ob-
jective of this investigation was to acquire some understand-
ing of how the transparency of the fish cornea is maintained.

Along with an experimental procedure whereby the

composition of fluid on the two Sides of a clamped in vitro

 

cornea can be controlled, analytical methods were applied
utilizing titrated water (HTO) and radiosodium (Naaa) in this
study of corneal permeability. Using these tracers, perme-
ability was quantitated in terms of microgram and microliter
quantities. With a more recent isotope technique "dual

labeling," one can investigate the simultaneous behavior of

two ions, a procedure of considerable importance in studies
of water balance. Using this technique HTO and Na22 were
studied concurrently to see: (1) if the water movement is
dependent upon the sodium movement (as is the case with

most other tissues), (2) if their fluxes are of prime im-
portance in understanding the maintenance of fish corneal
transparency, and (5) if the increased osmotic pressure of
the external environment had any effect. Data from this
study permitted comparisons of the permeability character-
istics of the aquatic cornea (normal and diseased lake trout)

to that of mammalian cornea (rabbit) to be made.

LITERATURE REVIEW

Studies of corneal permeability have been mainly
centered around mammalian Species. This research dates back
to 1855 and has been extensively reviewed by Maurice (1955)
who compared the reported rates of penetration of ionized
substances across the cornea and its layers with that of Na24.
In general, the evidence suggests that the substances in
solution within the corneal stroma exchange by passive dif—
fusion with those in the plasma at the limbus, and, within a
Size range of 10-25 A those substances in the aqueous humor
cross the "endothelium." Maurice postulated that there is a
small co-existent active transport of some substances out of
the cornea which accounts for the maintenance of its normal
thickness. In a more recent review (Chapter 6 in "The Eye"
edited by Davson), Maurice (1962) deals comprehensively with
such corneal problems as the physical basis of tranSparency,
hydration, permeability, nerve supply, and wound repair.
Dohlman (1965, 1964) composed the annual review on the cornea
and sclera for those respective years. The permeability of
the cornea as it relates to ion and water movement in teleosts
has been reviewed by Edelhauser (1964). Donn (1965, 1966)
wrote the corresponding annual review on the cornea and sclera
and in the most recent publication he described the physiology

of the posterior corneal membrane extensively.

Water Permeability

Cogan and Kinsey (1942a) presented data on water trans-
fer through the excised cat cornea in the aqueous to tears
direction. This transfer results in establishment of an
osmotic gradient by an intact epithelium. Their results
Showed no net transfer of water in the anterior to posterior
direction because bullae developed and consequently epithelial
damage occurred. Nevertheless, they inferred that deuterium
oxide is transferred through the excised cornea by diffusion
in both directions, Since the fluid volume remained constant.
Therefore, every molecule of heavy water that was transferred
necessarily was replaced by a molecule of water coming from
the opposite side of the cornea. They (Cogan and Kinsey,
1942b) proposed a theory of fluid movement within the cornea,
where water is continuously abstracted from the anterior and
posterior surface by osmotic forces.

Anseth and Dohlman (1957) reported on the influence of
intraocular pressure on the water—binding capacity of the
corneal stroma in isolated bovine corneas. They noted that
"endothelial" damage causes considerable corneal swelling
whereas epithelial damage has a lesser effect and can be ex-
plained on a purely physical basis of pressure. They observed
that intraocular pressure counteracts the corneal inhibition
of tear fluid, but not that of aqueous humor; therefore, they
implied that the effect of hydrostatic pressure is probably

reSponSible for the difference in swelling when the epithelium

and "endothelium" are removed. Hedbys and Mishima (1962)

have measured the resistance of the corneal stroma to the flow
of water both along and across the collagen fibers of the
cornea as well as with various degrees of hydration. In these
investigations of flow along the stroma, a marked difference
in thickness was generated by allowing a part of the corneal
strip to swell and the adjacent part to dry. Water equilibrium
along the strip was established by the water movement from the
swollen to the dry part and the profile was evened out. By
repeated measurements of thickness along the strip, they
established the flow of water. Since the relationship between

thickness and swelling pressure was known, the pressure

,4

gradient along the cornea was determined. This flow of water
across the stroma was determined by pressing 0.9% NaCl or
water through a cornea tissue tightly sealed in a holder be—
tween two porous stainless-steel filters and collecting the
water in a capillary tube. They found the resistance to flow
in the two directions to be Similar at normal hydration, but
for dried cornea the resistance was higher across the tissue
than along it. The calculations Show that the collagen fibrils
alone offer only 1/50 of the measured resistance. They sug-
gested that the interfibrillar substance is reSponSible for
the high resistance to the flow of water in the corneal stroma°
Donn (1962) reviewed information on the movement of
solutes and water across the cornea and summarized the recent

work on water diffusion and "endothelial" pinocytosis of the

mammalian cornea. Direct measurements (with tritiated water)
of water movement across the in vitro rabbit cornea were made
by Donn §£_§l. (1965). They found the water flux across the
rabbit cornea was the same in either direction. This movement
of water across the cornea can be accounted for by diffusion
alone, without any bulk flow; the loss of water from the
aqueous into the tears under conditions of normal hydrostatic
pressure is negligible; and the resistance of the epithelium
and "endothelium" to the diffusion of water is comparable to
that of the stroma.

Hedbys §£_al, (1965) have shown that when the cornea of
a living rabbit is cannulated a negative pressure with respect
to atmospheric pressure is recorded. When the normal intra-
ocular pressure is accounted for, the resulting negative pres—
sure would then correspond to the swelling pressure of the
excised tissue. This proves that the stroma has a tendency
to swell in vivo which normally must be countered by an active
tranSport mechanism located in the limiting layers of the
cornea. Mishima (1965) studied the water permeability of the
rabbit cornea with reference to the superficial oily precorneal
tear film. He found this layer to retard the evaporation of
the tear film from the corneal surface; the "evaporation
tonicity" of the precorneal tear film increases on opening the
eye and the cornea becomes Slightly thinner and vice-versa.
The tonicity increase in open eyes results in an osmotic water

flow from aqueous through the cornea to the tear film.

Information on corneal hydration as related to corneal
metabolism, a possible water pump, and passive diffusion as
a result of active ionic transport, has been reviewed by

Edelhauser (1964).
Ionic Permeability

The effect of hydrostatic pressure on the transfer of
sodium chloride through the excised cat cornea has been studied
by Cogan and Kinsey (1942c). They found that the cornea is
impermeable to 1 percent sodium chloride solution in the out—
in direction with a hydrostatic pressure up to 400 mm Hg. but
permeable in the in-out direction under various hydrostatic
pressures. The permeability to sodium chloride is in one
direction only and is controlled by the epithelium. Maurice
(1951) investigated the permeability of the in vivo rabbit
cornea to sodium ions. First Na24 was placed in a conjunctival
sac and allowed to remain in contact with the corneal epithel-
ium for various times at the end of which the cornea and
aqueous humor were removed and their activities measured.

It was then possible from these results to derive most of the
permeability relationship between the cornea and its external
media. Secondly, Na24 was injected into the blood and the
concentration in the plasma, aqueous humor and cornea measured.
Thirdly, radioactive salt was introduced directly into the
anterior chamber and the changes in activity with time in the
cornea and aqueous humor were determined. With the data ob—

tained Maurice was able to calculate permeability constants

for the cornea "endothelium,” epithelium, entire cornea and
blood-aqueous barrier. He concluded that the corneal
epithelium has a finite but low permeability to inorganic

ions in both directions (0.00080-0.0024 cm.hr‘l); the corneal
"endothelium" has a much higher permeability to inorganic ions
in both directions (0.072 cm.hr‘l) than does the epithelium;
and this sodium extrusion resident in the "endothelium"
accounts for the maintenance of corneal dehydration.

Potts (1954) reviewed literature dealing with the perme—
ability of corneal layers as demonstrated by tracer experi-
ments and stated that only with isotopic tracers have analyti-
cal methods become available allowing analysis of sub-microgram
amounts of inorganic salts and water in ocular tissue.

Active transport of a substance has been proposed for
the mammalian cornea and, strictly defined, requires its
movement to be against an electrochemical potential gradient
from one side of a membrane to another. The proof of such
transport requires control of the composition of fluid and
electrical potentials on the two Sides of the membrane.

In the case of the cornea the necessary conditions have been
achieved by clamping a freshly excised tissue from a rabbit
between two chambers (a modified Ussing-Zerahn Chamber,

USSing 1949) so that the fluid could be circulated over each
surface, Donn 23 _l. (1959a, 1959b). The electrical potential
was measured by using non-polarizable calomel electrodes, and

it was found that a potential difference of 10 to 40 mv was

10

built up across the cornea with the outside of the cornea
negative with respect to the inside. By removal of differ—
ent layers it was found that the potential was generated by
the epithelium (Potts and Modrell, 1957; Donn 2: _l., 1959b).
These experiments demonstrated that sodium is actively trans-
ported across the epithelium into the corneal stroma and
that the net flux of sodium is nearly equivalent to the
neutralizing current.

Green (1965, 1966) measured ion fluxes which occurred
during the first hour after excision of the rabbit cornea.
He found that both sodium and chloride ions are actively
transported across the cornea from the epithelial to
"endothelial" surface. The location of this transport "pump"
suggested by Green was in the epithelium. His idea of a
chloride pump is consistent with that of Cogan and Kinsey
(1924a). Active chloride tranSport by the cornea is also
supported by the work of Zadunaisky (1966) who has Shown that
chloride ions are actively transported across frog corneas,
and that this represents the only source of current across
this "membrane." He measured the net rransport of chloride
from the aqueous to the tear side of the cornea. Davis and
Zadunaisky (1966) have proposed that potassium ions are
required for this chloride transport and that both of these

ions are involved in maintenance of corneal transparency.

11
Fish Cornea Water and Ionic Permeability

Aquatic corneal physiology has been studied very little
except from the anatomical point of view (Walls, 1942a;
Varbec, 1959). Varbec provided evidence for the existence
of a doubted fundamental corneal "endothelium” of the teleost
which can be clearly demonstrated in a flat preparation.

Fish eyes are of physiological interest because they have
tranSparent corneas adapted to media of extremely varied
osmotic pressure. The cornea of the fresh water fish is
constantly in contact with a hypoosmotic solution, whereas,
for most salt water fishes, sea water is hyperosmotic with
respect to their tissue fluids. In one group of marine fish,
the elasmobranchs, the internal osmotic pressure is so high
that their external environment is relatively hypoosmotic to
the internal one.

In 1954 Smelser and Chen did a comparative study of the
structure and hydration properties of corneas adapted to air
and aquatic environments. They have Shown that guinea pig
and carp corneas are Similar anatomically but have different
hydration properties. In his Procter Award Lecture, Smelser
(1962) reported on three types of aquatic corneas: carp-
fresh water,scup-marine,dogfish (elasmobranch)-marine. Ike
found corneas to be metachromatic indicating the presence of
mucopolysaccharides but markedly less than mammalian corneas.
Fish corneas were found to swell in a manner similar to

mammalian corneas when placed in various physiological salt

12

solutions; however, the degree of corneal swelling in distilled
water was greater in the scup, which is adapted to life in sea
water, and less in the carp which lives in a hypoosmotic
medium. The elasmobranch cornea differs from those of all
other Species so far studied in that they do not hydrate in
any solution ranging from hypoosmotic to hyperosmotic. These
results led to the conclusion that the mechanism which keeps
the teleost cornea dehydrated and transparent is diffusion

of water from the cornea. This mechanism probably depends
upon the colloid-osmotic pressure of aqueous-humor rather than
on an active pump, which in turn would be dependent upon the
metabolic activity of the epithelium or ”endothelium" as is
typical of mammals.

Cation concentrations and water content in the normal
rainbow trout cornea, aqueous humor, lens, and vitreous humor
were presented by Edelhauser gt__l. (1965). The in vitro
rainbow trout cornea showed a limited permeability to both
water and ions, and corneal ion concentration appeared to
vary inversely with corneal hydration. Also postulated was
an exchange of water across the trout cornea rather than a
net movement which may have been caused by the osmotic
gradient.

Allison (1965) reported that when control of the sea
lamprey appeared imminent, production of lake trout in
hatcheries was increased. Coincident with this increase,

certain diseases of lake trout became evident. One disease,

15

cataract, results in permanent damage to the eye. Another
eye disease which occurs in hatchery-reared lake trout from
the Harrietta, Michigan, hatchery is a corneal lesion.
Pathological progression of this lesion has been categorized

into stages and described by Hoffert and Fromm (1965).

MATERIALS AND METHODS

 

Experimental Animals

 

The two- to four—year-old (18-29 cm long) lake trout

(Salvelinus namycush) used in these experiments were provided

 

by the Michigan Conservation Department from their hatchery
at Harrietta, Michigan. This age group was selected because
their corneal diameter was one centimeter which was necessary
in order to clamp the cornea for in vitro studies. Lake trout
were transported from the hatchery in a galvanized metal tank
lined with non-toxic paint and fitted with an agitator for
aeration. In the laboratory they were kept in fiberglass
lined tanks with continuous inflow of dechlorinated water at
one end and an overflow Spout at the other end of the tank.
Conditions of 90C and 15 hours light and 9 hours darkness
were maintained each day.

Small rabbits of 0.5—1 kg. were used because their
corneas had a diameter of one cm. which was desirable for

the apparatus.

Apparatus

 

A technique for holding trout corneas in vitro between
two chambers in such a manner to allow the fluid bathing each
surface to be regulated has been previously described by

Edelhauser (1964, 1965). In this apparatus excised corneas

14

15

are held rigidly in place between two lucite blocks which
contain the aforementioned chambers. The corneal clamps

are of two types: number 1 with a diameter of 0.6 cm. and
number 2 with a diameter of 1.0 cm. These clamps are illus—
trated in full size in Figure 1. Figure 2 illustrates the
perfusion blocks for permeability studies on fish cornea
conducted at 150C. Figure 5 Shows the perfusion blocks for

similar studies conducted on the rabbit cornea at 550C.
Experiments with Tritiated Water

The experiments in this series were conducted with the
corneas clamped with holder number 1. In all experiments
chemically defined tissue culture medium (TC199) obtained
from Difco Laboratories, Detroit, Michigan, was used to bathe
the "endothelial" surface of the corneas and tap water was
used to bathe the epithelial surfaces. Depending upon which
direction water movement was to be investigated, 0.5 ml of
tritiated water (0.1 mc HTO/ml) was added to either the 5 ml
of TC199 or the 5 ml of tap water. Four hours after the
addition of tritiated water, 20 microliter samples of the two
media were obtained, using disposable micropipettes (Drummond
microcaps, Drummond Scientific Co., Broomall, Pa.), and
transferred to glass vials containing 15 ml of a liquid
scintillation counting solution. That part of each cornea
which was exposed to the two media was dissected free and
also placed in 15 ml of counting solution. The counting solu-

tion contained a primary scintillator (5 gm PPO; 2,5—

16

Corneal holder

Number 1. 0.6 mm diameter

 

 

 

 

 

 

 

 

 

 

 

 

 

6 mm
.1.
End
Front Side
Number 2. 10.0 mm diameter
Front Side

'Figure 1. Corneal Clamps (actual size).

17

 

Figure 2. Pictorial view of fish cornea perfusion.

18

 

Figure 5. Pictorial view of rabbit cornea perfusion.

19

diphenyloxazole), a secondary scintillator (50 mg a €NPO;
arnaphthylphenyloxazole), and a solvent system of 80 gm
naphthalene plus dioxane to make 1 liter. Since dioxane is
a good dehydrant, it removed the water and HTO from the
cornea, thus enabling the activity within the corneas to
be assayed. A Tricarb Liquid Scintillation Spectrometer
514A was used to assay the radioactivity in the samples.
Each sample was counted three times for 10 minutes at a
machine efficiency of 19 percent for tritium, and the average
number of counts per minute was corrected for background
activity.

In studies of the corneal layers, corneas were denuded
with a kerotone knife and histological sections were made to
verify that the epithelium and/or "endothelium" were com-

pletely removed by this procedure.
Experiments with Sodium22

A. Experiments in this series were conducted with corneas
clamped in the larger holder number 2. The fish were killed
by cutting off their heads, and the eye was protrused and
held rigidly in this position using a pair of forceps pressed
down on the outer edge of the ventral side of the eyeball.

The sclera was pierced approximately 5 mm lateral to the outer
edge of the cornea, and a neat cut was made around the sclera
at this distance from the cornea. The cornea was placed in a

petri dish in tap water and iris removed, then it was placed

20

in the clamp and held stationary by pressure from the inside
Sleeve of the clamp pressing on the sclera. All experiments
were conducted in a manner Similar to that outlined for
tritiated water except 0.2 ml of Na22 (1.0 pc/ml) was added
to an increased volume of 6 ml of TC199 or 6 ml of tap water.
After the addition of Na22, two 100 microliter samples were
taken hourly for 4 hours from the two media, using disposable
micropipettes. One sample for counting was transferred to a
five-dram plastic vial containing 10 ml of distilled water.
The sodium content of the other sample was determined using

a Coleman 21 flame photometer (Edelhauser, 1965). That part
of each cornea which was exposed to the two media was dis—
sected free, cut in half, and one half digested with 10 ml
concentrated nitric acid for counting. Tissue water content
of the other half was determined by drying and then, after
ashing, corneal sodium was determined (Edelhauser, 1965).

All Na22 samples from these experiments were counted using a
two-inch thallium-treated NaI well type scintillation de—
tector, Nuclear Measurements Corporation, Indianapolis,
Indiana, Model PHA-1CA pulse height analyzer, and Model DS—1A
decade Scaler used as a slave scaler. Counting was done at
the 5 percent level of error or less. Since counts per minute
of Na22 and total amount of sodium ion in both media and cor-

nea were known Specific activity was calculated.

B. In studies of the corneal layers, the procedure

described for tritiated water experiments was followed.

21

C. Corneal sodium and water concentrations: whole
corneas of normal and diseased Lake Trout were excised and
corneal sodium and water were determined as described by

Edelhauser (1964).

D. Back diffusion: corneas from a Single fish were
excised and mounted in the recirculation blocks as described,
and the same amount of Na22 was added to the TC199 which
bathed the "endothelium." After a two-hour “perfusion,"

22 Since the specific

the cornea was uniformally labeled with Na
activity of sodium in the cornea was approximately equal to
that of the "endothelial" solution. Two-hour samples were
taken of both media (199 and tap) then the media were removed
and one cornea was dissected free and cut in half: one half
for counting, the other half assayed for total corneal sodium
as previously described. This cornea was used as a control

22 in the cornea at two hours. With the

for the amount of Na
other paired cornea, the tap water and TC199 were removed

from the recirculation blocks and the blocks rinsed. The
cornea, which remained in the clamp, was rinsed with saline
and then placed back in the blocks, 5 ml tap water and 5 ml
TC199 without tracer were then added to the recirculating
system. The back diffusion of Na22 from the labeled cornea
was measured in separate experiments by determining the amount
of Na22 in the bathing fluids after 0.25, 0.5, 1.0 and 2.0

hours of perfusion. Corneal Na22 was assayed in the usual

manner .

22
Dual Labeling EXperiments Using HTO and Na22

,Experiments in this series were conducted with corneas
clamped in holder number 2 which allows more surface area
of the cornea to be exposed between the bathing solutions.
The fish cornea was excised and clamped as previously described.
The rabbit cornea procedure was as follows: Rabbits were
killed with an intravenous injection of sodium pentabarbital
(Nembutal) given via the marginal vein of an ear. An eye
was protrused and held rigidly in position using a pair of
forceps inserted under the eyeball and resting on the outer
edge of the orbit. The sclera was pierced and a neat cut was
made at a distance of approximately 5 mm lateral to the outer
edge of the cornea. The cornea and ring of sclera were im—
mediately transferred to a petri dish containing TC199 solu-
tion at 55°C and the iris removed along with any connective
tissue which remained attached to the sclera. The cornea
was rinsed in the TC199 and immediately mounted in the clamp
and placed in the recirculating blocks. TC199 was added to
both compartments and maintained at pH 7.2 by using a 95 per-
cent 02:5 percent C02 gas mixture in the siphons. Although
there is considerably more hydration of the rabbit cornea with
TC199 than with I'Kinsey" solution (Mishima, 1966), TC 199 was
selected in order to have experiments similar to those with
fish corneas. Temperature of the entire system was kept at
55°C. by conducting the experiments in a gravity convection

oven (Figure 5).

25

Depending upon which direction water and sodium perme-
ability was to be investigated 0.1 ml of HTO (100 uc/ml)
and 0.2 ml Na22 (1.0 pc/ml) were added either to the 6 ml of
TC199 in the "endothelial" (En-199) or epithelial (Ep-199)
compartment. Samples were taken as described in the Na22
experiments except that samples and halves of corneas for
counting were transferred to glass vials containing 15 ml of
the liquid scintillation counting solution described in the
experiments with tritiated water. The HTO and Na22 content
of these samples were counted simultaneously with a Nuclear
Chicago Mark I Liquid scintillation system (see Appendix A
for procedure). The counting efficiency in this procedure
was 50 percent for HTO and approximately 94 percent for Na22
and all counting was done at a 2 percent level of error or
less. Using a channels ratio technique (Nuclear Chicago Mark I)
the degree of quenching was essentially constant, therefore,
no correction for quenching was made. With this counting
scheme there is some overlapping of the Na22 Spectrum in the
channel peaked for tritium. The necessary corrections were
made according to the detailed procedure outlined in Appendix A.
Simultaneous counting for corneal HTO and Na22 was also
possible without digesting the cornea. Since dioxane is a
good dehydrant it removed the water and HTO from the cornea
thus enabling the activity within the cornea to be assayed.
Dioxane also removed some Na22 from the cornea but the majority

of the Na22 remained within the cornea. With the corneal Na22

24

distributed in this manner the Na22 activity was assayed and
corrected for quenching. In order to justify counting cornea
sodium in this manner for comparison with media sodium the
following experiments were run: (1) When corneas were di-
gested with Nuclear Chicago N.C.S.TM Solubilizer counted and
.corrected for quenching, the activity was Similar to that
received when corneas were not digested and located on the
bottom of the vial. (2) When the cornea was suspended in

the scintillation counting solution on a wire the counts were
the same as if the cornea was resting on the bottom of the
vial. (5) position of the cornea on the bottom of the vial
had no effect on determination of corneal Na22. (4) A Na22
solution on the bottom of a vial without fluors and Scintil-
lation counting solution and Na22 in dioxane without fluors
gave five or six counts above background.

After the perfusion experiment was completed the cornea
was removed from the clamp, and that part of the cornea ex-
posed to the two media excised from the sclera and cut in
half. One half placed in 15 ml of scintillation solution
the other half weighed, dried and analyzed in the usual manner.
Specific activities for Na22 and HTO in the tap water, TC199

and cornea were calculated.

Corneal Adenosine Triphosphatase

 

Lake trout corneas were excised fresh frozen, sectioned

immediately with a cryostat microtome at 16 n, sections

25

mounted on slides and incubated. This method, without fixa-
tion, for fish cornea differs from that used for the rabbit
cornea (Ehlers, 1965).

To demonstrate ATPase the method of Wachstein and
Meisel (1957) found in Pearse (1960) was used with Slight
modifications. The principle underlying this method is that
the phosphate ions, which the enzyme liberates from ATP, are
trapped as lead phOSphate. The lead phOSphate is subsequently
converted into black lead sulphide. Black or brown—black
deposits thus indicate the Site of enzyme activity. Differ-
ences between the method used and that of Wachstein and Meisel
are: (a) ATP (disodium salt) was dissolved in 0.8 M saline;
(b) the incubating medium was filtered before the sections
were placed in it; (c) the fresh sections were incubated for
20, 40, and 60 minutes at 250C and; (d) after incubation one
quick dip in 1 percent yellow ammonium sulphide was sufficient
to develop the deposits.

Control experiments were made on fish gut and cornea

incubated without ATP.
Surface Area of Cornea

By using cornea holder number 2 and clamping the sclera
rather than the cornea as hold number 1 does, the cornea of the
fish and rabbit holds its spherical shape. The superficial

area of the cornea was calculated using the following formula:

26

curved surface = 2vrh = (4h2 + C2)

Ma

where r = radius; h = height; and c chord length = 1.0 cm.

 

 

Rabbit Lake Trout
Radius 0.57 cm 0.69 cm
Height 0.50 cm 0.20 cm
Chord 1.00 cm 1.00 cm
Surface Area 1.06 cm2 0.88 cm2

One can observe from the calculated surface area that
the rabbit cornea is more Spherical than the fish cornea..
When placed in identical clamps the rabbit corneas have an
exposed surface area some 20 percent greater than for fish

corneas .

RESULTS

I. Theoretical

It is convenient at this stage to formulate in a simple
manner some of the diffusion relationships between the cornea
and the fluids bathing the eye. This will help to establish
terminology, to make evident some of the assumptions that have
to be justified, and to emphasize numerical constants which

must be evaluated.

Experiments with HTO - Equations

The Equation (5) used for the calculation of the perme-
ability constant of the cornea to water, will indicate the
time course of change in radioactivity in the labeled compart-
ment when the opposite chamber has a finite volume rather than
an infinite volume (Mishima, 1966). It does not take specific
activities into account; therefore, the radioactivity measure-
ments are converted directly into pl of water per hour that
cross the cornea. It is assumed that there is perfect mixing
in each chamber. Equation (5) does take into account any HTO
that appears in chamber two and exchanges in the reverse
direction from which it has come. Permeability constants were
calculated over 4 hours since the diffusion rate is linear

with time over this period.

27

28

In Figure 4, symbols are the following:

 

 

 

V = volume of chamber 1 =
chamber 2 = 5 ml
CH 1 1 CH 2 A = area 0.28 cm2
HTO i Kp = permeability constant
. c1 = conc. of HTO in chamber
”27‘ 1 at given time
Cornea c2 = conc. of HTO in chamber
2 at given time
Figure 4 co = conc. of HTO in chamber
1 at time 0
Then %%1 = fig¥ = 3555- (C1-C2) (1)

Neglecting the HTO in the cornea (small volume)

C1 + C2 = Cb = constant

Therefore C2 = C0 — C1

Put this in equation 2:

dc -K A
dt v (2C1 Co) (2)

Transformation would go as follows:

fcl dC] = "KEA Id:

C1

[ lIl(2C1-CQ)]C = :E‘EI‘A t
O

4..

2

1n ZCJ—CQ = _2 KEA t
cO v

-2 'A
Q;,=.% [ 1 + e t ] (5)

Experiments with Na22

Experiment Equations

and Dual Labeling

The equations describing the exchange of radioactive

ions between two compartments have been outlined by Ussing

29

(1949), Green (1965, 1966) and others. If MNa(I) is the
total amount of Na+ in the inside solution, ANa22(O) the
amount of Na22 (CPM) in the total outside solution at end

of time "t", Na(I)22 the amount of Na22 (CPM) in the total
inside solution and A is the exposed surface area (cmg)
across which flux is occurring, then the flux (F) in time "t“
is described by:

(Mu. " (mm)
F = (D) r.

(4)

F is equal to the "endothelial“ to epithelial rate of entry
or flux of Na22 given as qu/cmZ/hr. The epithelial to

"endothelial" flux of Na22 may be readily calculated by sub—
stituting the appropriate values in the equation. The flux
calculated from equation (4) is total flux whereas equation

22 across the

(5) was used to describe the net flux of Na
fish and rabbit cornea. The theory behind equation (5) can
be found in Sheppard's (1962) book, where the solution yields
an exponential relation and p (permeability constant) is
constant.

In Figure 5 and equation (5) the symbols are the

following:

50

CH 1 CH 2 S = total amount of traced sub-

 

 

 

 

T stance in the system
Na22 I measured in moles, grams,
//”1 etc.
Coriea 81, $2 = amounts of S in chamber

1 and 2 reSpectively.
Figure 5 p = rate of exchange between
compartments measured in
microequivalents per hour.

a1, a2 = specific activities in
chambers 1 and 2.

t = time in hours.

a(0) = specific activity of zero

time in chamber 1.
A = area (fish 0.88 cma; rabbit =
1.06 cma)

Then log (al-ag) = log a(0) - p (5% + 5&9 t (5)

Solve for p = nEq/hr
Then §-= qu/Cmg/hr

Equation to Correct for Backdiffusion of Sodium

The mechanism of exchange of corneal Na+ with aqueous
humor is very Similar to the kinetics which have been widely
studied by Sheppard and Martin (1950), and summarized by
Comar (1955), for the exchange of potassium between plasma
and erythrocytes. With fish corneas, under the experimental
conditions described, the set—up may be considered as a
closed two-compartment system since the corneal epithelium
is impermeable to sodium ions. The process is represented

schematically in Figure (6).

51

 

 

 

 

 

 

 

 

Phase 1 Aqueous A1 and A2 = concentration of
humor sodium in aqueous humor
A1* , A1 and cornea respectively.
A1* and A2* = concentration of
"Endothelium" labeled sodium in aqueous
y humor and cornea re-
Phase 2 Cornea spectively at time "t".
A2* A2 A0* = A1* + A2* = concentration
of labeled sodium in

 

 

 

aqueous humor (TC199) at

Figure 6 zero t1me.

*
Al—-= S; and %?-= 82 are
2

specific activities in
aqueous humor and cornea
respectively, at time "t".

p = rate of movement of Na22
from phase 1 to phase 2
and vice versa.

Equation (6) enables one to calculate the net flux of sodium
into the cornea taking into account the exponential rate of
loss of sodium from the cornea to the solution bathing the
"endothelium."

22

The rate of movement of labeled Na out of phase 1

*
will be -p%l- and the rate of movement into phase 1 will be
1
*
p %:—-. Therefore, the net movement of labeled Na22 may be

represented as follows (Comar, 1955):

B511. = _p £1:.+ éai
dt A1 A2

converting to Specific activities

dS

substituting for 82 in terms of $1 and integrating where

A*
SOP-fi—

52

-K*t[<l)+<i)1
EL: A1+Age p A1 A2

s. (e)

A1+A2

Solve for Kp* = qu/hr

*
then §§—-= qu/cmZ/hr

where A = 0.78cm2 which equals the surface area.
Equation to Calculate Kp' from T1/2

A permeability constant for the corneal epithelium can
be calculated if the Specific activity and the rate of loss

(Tl/2) of corneal sodium are known (equation 7).

Kp' = -——-— (7)

Permeability constants

The permeability constants previously described are

listed with their reSpective units.

constant unit isotope equation
.. = {:Tiégz‘iéir m
F = qu/cme/hr Na22 (4)
p = qu/cma/hr Naaa (5)
Kp* = qu/Cma/hr Na22 (6)
Kp' = qu/cm2/hr Na22 (7)

55
II. Experiments with Tritiated Water

In all experiments the solution which bathed the
"endothelial“ surfaces of the corneas was very nearly equal
osmotically (about 290 mOS/1) to trout plasma, whereas, that
bathing the epithelial surface was tap water. Thus, in every
case there existed across the cornea an osmotic gradient of
some 6.6 atmospheres favoring an osmotic movement of water
through the cornea to the solution bathing the endothelial
surface. Permeability constants (Kp) were calculated accord—
ing to equation (5).

Data presented in Table I Show that the movement of
tritiated water in either direction across the healthy lake
trout cornea is equal irrespective of the existing osmotic
gradient. There was a slightly greater increase in the
permeability when the epithelia were removed than when the
"endothelia" were removed. When both layers were removed the
permeabilities were significantly increased above the value
for intact corneas but were not Significantly different from
the permeabilities of corneas with the epithelia alone re-
moved.

The triated water content of the corneas was determined
and the concentrations of HTO in the corneas are expressed
as CPM/mg of cornea. Data for normal intact corneas in the
two experimental Situations, A and B as given in Table I,
indicate that the epithelium offers a greater resistance than

the "endothelium" to entry of water into the cornea. With the

54

Table I. Permeability constants for water transfer and corneal
tritiated water content of normal and denuded lake
trout corneas in vitro.

Kp(5)**
Permeability Corneal HTO
Number Constant Content
Condition of Cornea of data (cm3/cm2/hr) (CPM/mg)

 

(A. With tritiated water in epithelial compartment at start of
experiment.)

Normal 6 0.16.1 0.02* 555
Epithelium removed 6 0.52 i 0.06 605
"Endothelium" removed 6 0.57 i_0.02 651
Both layers removed 8 0.55 i 0.05 827

(B. With tritiated water in "endothelial" compartment at start
of experiment.)

Normal 5 0.16.1 0.01 916
Epithelium removed 6 0.47 i.0.05 586
"Endothelium" removed 4 0.54.1 0.06 570
Both layers removed 6 0.55.1 0.05 410

 

*Standard error of mean
**Equation used for calculation of permeability constant

epithelia removed, the A-condition corneas showed expected
high concentrations of HTO. The B-condition corneas had less
HTO than normal controls presumably because of a greater loss
of HTO into the epithelial compartment. We are unable to ex-
plain the high HTO content of A-condition corneas with the
"endothelium" removed. AS expected, B-condition corneas with

the "endothelium" removed had higher concentrations of HTO

55

than those with the epithelium removed. With both layers re—
moved the B-condition corneas had less HTO than A-condition
corneas again presumably due to a greater loss of HTO into
the epithelial compartment by the former. With a single ex-
ception these data are consistent with the idea that corneal
water is much more freely exchangeable with water in the
"endothelial" than with water in the epithelial compartment.

Table II. Permeability constants, tritiated water content
and hydration of normal and pathological lake trout

 

 

corneas.
Kp(5)**
Permeability Corneal HTO Wet weight
Stage of No. of Constant content of cornea
disease corneas (ems/cma/hr) (CPM/mg) (mg)
Normal 6 0.16 i 0.02* 555 27.2 i 0.9*
1 8 0.20 i 0.02 577 27.1 i 0.6
2 9 0.25 i 0.05 725 29.5 i 0.6
5 6 0.17 i 0.02 676 58.8 i 5.1
4 5 1.46 i 0.02 758 45.1 i 8.4
1-a 8 0.22 i 0.02 978 25.6 i 1.8

 

*Standard error of mean
**Equation used for calculation of permeability constant

With the exception of the values for Stage 5, those given
in Table II indicate that as the corneal lesion progresses
from Stage 1 through Stage 4, there is a tendency for the
permeability of the cornea to increase. Viewed in the light

of previous histological work,these results were not at all

56

surprising. In all of these experiments tritiated water was
placed in the epithelial or tap water compartment and perme-
ability constants were calculated on the basis of appearance,

after 4 hours, of HTO in the originally unlabeled compartment.
III. Experiments with Na22

In all experiments corneas were subjected to the osmotic
gradient as described under II above. Sodium fluxes were
calculated with Eq. (4) and net permeability constants or
fluxes calculated with Eq. (5).

A. Data presented in Table III Show that there is no
sodium flux in either direction across the normal intact
fish cornea and that the specific activity of the cornea in
A-condition is almost equal to that of the TC-199. Under
B-conditions the Specific activity of the cornea is extremely
small compared to that of the tap water. This small Specific
activity could be due to surface absorption and does not
necessarily indicate true uptake of Na22 by the corneas.

There was a tremendous increase in the total Na flux when the
epithelia were removed in condition A whereas in condition B
the flux was very slight; however, the net flux in both con-
ditions was very similar. This demonstrates the impermeable
nature of the epithelium to Na+. With this layer removed
more sodium moves with than against the osmotic gradient.

When the endothelia are removed there is a slight Na+ flux,
but no net flux which indicates that the corneal "endothelium"

+
does to some extent control corneal Na ; however, the

57

mmxdam mo coaumHSUHmo wow wow: SOHumsqm**
some map mo Houum pumpcmum*

 

 

 

 

Um>ofimm
m.aflm.mm m.m mm «ow oa m.fiflw.am 6.0Ha.m a mummmq :aom
pm>o§mm
0.fiH0.mm 0.0 n >>m¢ a 0.0 0.0 w :HMHHmQDOUcm=
Um>OEmm
N.Hflm.mm m.m mma mmm aa $.0Hm.mm w.0wm.m 0H amaamnuflmm
m.NHm.0> v.0 m hawm 0 0.0 0.0 m HmEHOZ
ucmfiuummaoo HMflHmnuHmm SH mmmz .m
Um>ofimm
m.fiflk.mm m.m mm mm as N.mwm.mm m.mflm.¢¢ a mumsmq auom
Um>OEmm
m.0Hm.mm 0.0 00 um 00 0.0 m.mflm.> m zamaamsuopcmz
Um>08mm
m.0na.mm a.m was 45 mm *fi.mum.om *m.mua.mm Ha Hmflflmnuflmm
*m.aHm.Hm N.m mu 0 >0 0.0 0.0 m HmEuoz
ucwEuummfioo =HMHHm£uopcm= SH mmmz .<
Omm R mE\Emo H£\NEU\UmQ_ H£\Nfin\vm1
mmcuoo mmquou mmcuoo Hmum3 mos mdeUB xsam umz stm S mmauou mo SOHDHUSOU
amaxsmu sufl>wuu< oamaomam **Am0m **A¢0m
.OH#H> CH .mmmGHOU “DOM“ OMMH COUSGOG USN HMEHOG MO kum3 HMOGHOU
tam “>5fl>fluom oHMHoomm mflooe 0cm mocnoo “musog usom um>o mwaHm asapom wmmum>4 .HHH manna

58

epithelium acts as the main barrier to preserve corneal Na.
When both layers are removed the Na+ fluxes were significantly
greater than the fluxes across corneas with the epithelial
alone removed.

The Na22 content of the cornea was determined and
Specific activities were calculated. With the epithelia re—
moved both A and B condition corneas showed similar Specific
activities presumably because the Specific activity of the
tap water (condition B) was 4-fold higher than the TC199 in
condition A, thus the probability of labeled Na+ (condition B)
penetrating the cornea was much greater. When the "endothelia"
are removed in condition A the corneal Specific activity is
equal to the Specific activity of the TC-199, but in the B
condition the corneal Specific activity is extremely small
compared to the Specific activity of the tap water.

These corneal Specific activities are as expected if we
assume that the corneal epithelia controls the sodium flux.
With both layers removed both condition A and B corneas had
Similar specific activities. Since condition B tap water
Specific activity was 4 times greater than the Specific
activity of TC-199 in condition A; these results are in-
terpreted as indicating that there is a greater tendency for
Na22 to move with the gradient rather than against it. With-
out a Single exception these data are consistent with the
idea that corneal sodium is more freely exchangeable with

sodium in the "endothelial" than with sodium in the epithelial

59

compartment. Hydration was greatest during the four hour
perfusion of corneas with the "endothelium removed," less in
those with the epithelium removed and least in those with
both layers removed. Thus, hydration of the stroma is re—
stricted and indicates that possibly corneal hydration in the
fish is controlled by the epithelium and "endothelium" as in

mammalian cornea.

B. In Table IV are data for sodium flux across Lake
Trout corneas representative of the various stages in patho-
logical progression of the corneal lesion described by
Hoffert and Fromm (1965). Only in stages 5 and 4 is there a
sodium flux across the cornea. This is consistent with
previous histological work where one observes in the early
stage 5 a definite erosion, thinning and general disorgani—
zation of the corneal epithelium. Once the corneal epithelium
is disrupted (stage 5) an outward sodium flux will occur.
Stage 4 corneas have a complete hole; therefore, there is
nothing to prevent an outward flux of sodium. In all cases
the cornea CPM/mg in the diseased eyes are lower than the
normal indicating that these diseased corneas take up less
radiosodium. This may be related to hydration of the cornea
which will be considered in connection with Table V. We are
unable to explain the high and variable Specific activities
of the diseased corneas unless these corneas have the ability
to differentiate between Na22 and Na23 which is highly unlikely.

In calculating a corneal Specific activity there may be an

40

Table IV. Sodium fluxes; cornea and media_Specific activity; and
hydration of normal and pathological lake trout

 

corneas.
F(4)** Specific Activity
Stage of Flux cpm/qu, Cornea Cornea

 

Disease n qu/cma/hr Tc-199 Tap Water Cornea cpm/mg % H2O

 

Normal 9 0.0 87 0 7815* 9.2 81.511.5*
1 5 0.0 78 0 120112 7.4 81.810.6
2 5 0.0 79 0 152149 8.2 85.511.4
5 5 5.811.0 79 67 109117 8.5 85.410.4
4 5 96.510.9 68 61 151155 5.1 86.111.2
1-a 5 0.0 78 0 185165 8.1 85.411.0

 

*Standard error of mean
**Equation used for calculation of flux

Table V. Sodium and Water content of diseased lake trout corneas.

- i
l -

 

Stage of Na

Disease n % H2O mEq/l

Normal 9 75.911.5* 12518.7*
1 5 79.012.5 12616.9
2 5 85.812.5 10518.9
5 2 85.611.6 9115.7
4 9 85.511.2 9016.8
1-a 6 80.411.2 11516.8

‘

*Standard error of mean

41

over-correction for hydration which would decrease values

. +
for the concentratIOn of corneal Na .

C. Data in Table V indicate that as the corneal lesion
progresses in the lake trout, increased corneal hydration
occurs and simultaneously, corneal sodium concentration
decreases. During a four—hour perfusion the normal, stages
1 and 1-a corneas hydrate 5.6 percent, 2.8 percent, and 5.0
percent, respectively, whereas stage 2 corneas hydrate
slightly and stage 5 and 4 corneas do not hydrate at all.
This indicates that stage 2, 5 and 4 corneas are at maximal
hydration in situ, whereas, the normal, stage 1 and 1-a
corneas are not hydrated maximally in situ and they do imbibe

water from the bathing media in vitro.

D. Figure 7 represents the reverse exchange—equilibration
of labeled sodium between cornea and TC-199 (see also Figure
6). This semi-log plot of Specific activity of the cornea vs.
time was fitted for a straight line (2 = 1.7145 — 0.7816 x)
and T1/2 was calculated. In the fish 1/2 of the corneal
sodium exchanges with the TC-199 every 22.8 min. and the per-
cent turnover per minute is 5.0 percent (see appendix for
calculations). For rabbits, the exchange T 1/2 of the corneal
stroma sodium is 14.0 min. (Maurice, 1951) and from this a
percent turnover' of 4.8% per minute can be calculated. Thus
by comparison the fish cornea is able to retain its corneal

sodium longer than the rabbit. The permeability constant Kp*

Corneal Specific Activity

42

 

 

 

60"
50 Figure 7. Reverse Exchange--Equilibration
. of labeled sodium between
.0 cornea and TC-199.
40r-
O
50-— 3
20 .
A
10 Y = 1.7454 — 0.7816 x
‘ two—tailed 99% confidence limits
for population of Slope B
Pr (-0.1408 <B> 1.4224) = 0.99
T 1/2 = 22.8 min.
a
O
o
J . I g . l
0.25 0.50 1 2

Hours

 

45

for the fish corneal "endothelium" is 0.02910.005 mEq/cmZ/hr
calculated using equation (6) and corrected for backdiffusion.
The Kp' calculated from the T 1/2 using equation 7, is 0.050
mEq/cmg/hr. Since Kp* = Kp', it appears that Simple diffusion
and not active transport is involved in this exchange. No
data were presented by Maurice (1951) and, a Kp* for the rabbit
cornea "endothelium" using formula (6) could not be calculated;
however, a Kp' was calculated from the T 1/2 yielding a value
of 0.048 mEq/cma/hr. From the permeability constants (Kp*

and Kp') it appears that the rabbit cornea "endothelium" has

a greater permeability for sodium than the fish corneal

"endothelium."
IV. Dual Labeling Experiments for HTO and Na22

The results of an experiment with Na22 and HTO are given
in Tables VI and VII and these tables Should be viewed in
light of each other. HTO and Na22 fluxes were calculated with
equations (5) and (4) and net sodium flux calculated with
equation (5). In the fish cornea there is no apparent move-
ment of water drawn along osmotically with the movement of
sodium ions; thus HTO flux and Na22 flux in this corneal per-
fusion study occur independent of each other. However, with
rabbit corneas there is evidence of an osmotic flow of water.
In Table VI condition B and C the flux of water through the
rabbit cornea is significantly increased over condition A.
Data presented under condition B and C in Table VII also Show

a flux of sodium across the rabbit cornea. These results are

44

consistent with the previously proven active transport of
Na+ across the rabbit cornea epithelium in the tears to
aqueous direction. It is possible that with active transport
of Na+ across the corneal epithelium one would expect an
increased osmotic water flux in the tears-aqueous direction
and maximally hydrated corneas as indicated by data for
conditions B and C in Tables VI. In all cases the rabbit
corneas had a 5 to 5—fold greater water exchange than fish
corneas. Values for the water flux across the fish corneas
(Table VI) are Slightly lower than the values given in

Table I. This could possibly be due to differences between
the corneal clamps used. Corneal clamp 1 does damage the
outer edge of the corneal disc, whereas, clamp 2 holds the
cornea in place by clamping sclera tissue and does not touch
the cornea per se.

The water flux across the fish and rabbit cornea is
mostly an exchange with only a small net water flux found in
condition B and C rabbit corneas. In condition A the Specific
activity of sodium in the fish cornea and cornea CPM/mg are
much higher than comparable values for rabbit corneas, perhaps
due to the greater exchange of water by the latter. Values
for corneal Specific activity and corneal CPM/mg are reversed
in condition B, indicating the fish corneal epithelium main-
tains a greater impermeability to water than that of the rabbit.
When one compares all of the data in condition A and B, it

becomes apparent that water flux is impeded to a much greater

45

magnoe asuamnuoocm I am Nmmauoa ssflamnunmm u mm+
ucmumcoo hyaaflnmenmm mo cowumHSUHmu How poms SOHumsqm**
came may mo Honnm pumpcmum*

 

¢.HHm.mm 00m Hmm I «mm 50H mmwdmw w uHQQmm

unmum um ucmfiuummfioo HMHHmsuHQm SH mmmz 0cm Gem .0

>.0H¢.mm mmm mmm I mmm Nmfi 0NH00¢ ¢ uflflflmm

0.0H>.mm mud mam N00 I mm mfimw d nmflm

unmum um ucmauummEoo HMHHm£UHmm SH 083 ucmEuHMQEOU amaamnuoocmifimmmmz .m

m.mHN.Hm hmm mmm I 00d ,fimm 0mH0¢m m UHQQMM

*0.fiHH.mm >m¢ 000 0¢ I N05 *m.mHNOH fl Smflm

unmum um ucmEuHmmEoo HMHHmnuOUcm SH mmmz 0cm 08m .¢

 

 

0mm R mE\EmU mos +mmalmm +mmdlcm H£\NEU\HQ SOHuHUGOU
mmcuoo mmcuoo mmcuoo Hn\EQo . usmumcoo : HmucmEHummxm
mufl>fluo< UHMHommm huflaflflmmfiumm
**AWVQM

 

ouufl> SH mmmcuou
UHQQMH cam swam HmEHoc mo ucmucou HmuNB paw “mufl>fluom Uflwfiummm MHUmE 0cm
mmsuoo “undo: Hsom um>o nmmmcmuu Hmum3 Mom mucmumcou huHHHQmmEHmm mmmum>¢ .H> magma

46

mmauoa ssflamnuoccm I am immanoa ssflamnugmm u Sm+
mmxSHm mo SOHumHSUHmo MOM 0mm: mSOHuMSUm

SmmE may mo Honum UHMSSMum**
.x.

 

m Hm I 0mm 0 *5N1Qfld>.0 *H00.Qflm¢.m fl uHQQmm

pumum um uSmEuummEou HMHHmSuHmm SH mmmz USM 08m .0

w¢ wmm I N.N mum 0.0 0H.H w uHQQmm
mm mum 0 I fifim 0.0 0.0 fl anm

unmum um uSmEuummEoo HMHHmSuHmm SH 09m uSmEunmmEoo HMHHmSuopSm SH mmmz .m

0m own I 0 NHH 0.0 0.0 w uHQQmm

mm sum 0 I Nmm 0.0 0.0 w LmHm

unnum um uSmEuHmmEoo HMHHmSDOUSm SH Nmmz USm 093 .<

 

 

mE\Emo awe 1.mmHImm +mmHlSm HS\NEo\qm1. H£\NEU\Um1 SOHuHUSOU
mmSuou mmSuoo qmaVEQU xSHm umz xSHm S HmuSwEHHmmxm
Sun>nuo¢ UHGHomSm **im0m **i«0m

 

.mmmSnoo uHQQmH cSm Sme HMEHOS mo
>SH>HuUm UHMHummm MHUmE USm mmSuoo “muson Hsom Hm>o mmxvam ESHUOm wmmum>4 .HH> wHQMB

47

extent by the epithelium in the fish cornea than in corneas
of rabbits.

Data presented in Table VII shows that there is no move-
ment of sodium in either the fish or rabbit in the aqueous-
tears direction. In the reverse direction there is a sodium
flux in the rabbit (active transport) giving rise to a corneal
Specific activity of 21. Under comparable conditions the
specific activity of the fish cornea is only 0.7 (Table III)
indicating no active transport of sodium.

In all experiments the permeabilities of the corneal
endothelia of the fish and rabbit are much greater for water

and sodium than the corneal epithelia.

V. Corneal ATPase - Histochemical

Localization of the enzyme activity: the basement layer
of the corneal epithelium shows the highest ATPase activity
(Figure 5, Plates A and B). There was also some diffuse
enzyme activity within the stroma and the corneal "endothelium"
of the fish shows some ATPase activity also. A very limited
amount of activity was found in the corneal epithelium which
is very different from mammals where most of the ATPase activity
is located in the epithelium. Ehlers (1965) reported no
ATPase activity in the stroma of rabbit corneas.

Controls were run (Figure 5, Plate C) where the corneas
were incubated without the substrate ATP and gave a negative

reaction. B—glycerophosphate in the same concentration as ATP

48

.me¢ usozun3

.mumnmmosmoumohHmlm SSHB
.0 mumHm UmquSUSH SOHuomm HouuSou .0 oumHm

UmquSUSH SOHuomm HouuSOU

.mSmuQEmE Hmmmn
.mEOSum nmsonnu map um SOHuommn mSonumIImuoz

mmSmmHU >DH>HuUw mmwmam .mmSHOU SmHm SH mmmmem
0N mmSHoo mo SOHmmH HmQEHA .m mpmHm mx mmSuou mo HmpSoO .4 mumHm

m wusmHm

49

 

 

fx

‘3.
I
//
I-

Figure 8

50

was used as the second substrate for localization of any non-

Specific alkaline phosphatases which might be present. A very

slight reaction was seen (Plate D) but there was little dif-

ficulty in differentiating this reaction from the reaction of

ATPase.

DISCUSSION

 

Data presented in Table I indicate that the permeability
of normal lake trout corneas to water is the same irrespective
of whether the values are based on the transfer of tritiated
water from the epithelial to the endothelial bathing media or
vice-versa. These results are Similar to what has been ob—
served with corneas of fresh-water rainbow trout. Permeability
constants for both Species of fish are of the same order of
magnitude and are much lower than published values for corneas
of terrestrial animals. Once again we are forced to believe
that what we have measured represents an exchange of water
across the cornea and not a net flow of water from one side
to the other.

The data for mammalian corneas presented by Donn gt _l,
(1965), are probably most comparable to those given here for
trout, although conditions were different in the two experi—
ments. Their experiments with rabbit corneas were conducted
at 20-250 C., and the solutions (identical except for the
presence of HTO in one) bathing both surfaces of the corneas
were equal osmotically and were renewed by perfusion. In our
procedure a small recirculated volume of fluid bathed each
surface of the cornea. They applied a slight hydrostatic

pressure to the endothelial surface, whereas, no hydrostatic

51

52

pressure difference was maintained in our experiments. In
the calculations of permeability constants, neither research
group took into account the specific activity of corneal
water. The assumption was made in both cases that the amount
of HTO which appeared in the originally unlabelled solution
represented the transfer of a certain volume of water as
determined by the initial radioactivity of the labelled solu-
tion.

Donn gt _l. (1965), reported that the movement of water
across the rabbit cornea was equal in either direction. The
permeability constants for intact rabbit corneas were some
500 percent higher (0.47 vs. 0.16 cma/cmz/hr) than values
presented herein for lake trout corneas. After removal of
the epithelium, endothelium or both layers of the fish cornea
the rate of transfer of water across the cornea increased.
These denuded corneas had permeabilities comparable to those
for intact rabbit corneas. Based on the permeability
constants given in Table I, it has been concluded that some
resistance to water transfer across the fish cornea is offered
by the ”endothelium," but that the major resistance resides with
the epithelium. With the exception of a single datum
(A-condition, “endothelium" removed) the values for the tri-
tiated water content of the corneas after removal of different
layers are consistent with the above conclusion. Donn g£_al,

(1965), on the other hand, showed that when the epithelium or

the "endothelium" of the rabbit cornea is removed, the flux of

55

water does not change greatly, indicating that these layers
do not offer a large resistance to the passage of water.

The water permeability of lake trout corneas is the re-
sult of the interplay of many factors such as: presence of
physical (anatomical) barriers, metabolism of the cornea as
related to maintenance of the barriers, ion movement, effect
of environmental (chemical) agents on physical barriers,
osmotic pressure, intraocular pressure, and others. Some of
these factors are implicated in the corneal lesion of lake
trout which has been observed.

Stage 1 of the lake trout corneal lesion as described
by Hoffert and Fromm (1965) is characterized by keratectasia,
an increased anterior chamber and aqueous humor volume, and
a possible thinning of the central area of the cornea. The
movement of water across the cornea at this stage is not
significantly different from that across normal corneas
(Table II). Stage 2 corneas are characterized by a marked
keratoconuS-keratoglobus with pronounced changes in optical
tranSparency. There are some degenerative changes in the
epithelium including erosion of the superficial epithelium.
It is at this stage when the effect of osmotic pressure ap-
parently comes into play resulting in a great influx of water
into the cornea causing it to become opaque. Various degrees
of ulcerative keratitis characterize Stage 5. In this stage
one also observes a prominent erosion and thinning of the

corneal epithelium, an irregular thickening of Bowman's

54

membrane, and a localized hyperplasia of the epithelium.

The cornea shows a significant hydration and an increase in
over-all thickness. Values in Table I indicate that more
water gets into the cornea at this stage but less gets through
into the "endothelial" compartment, hence the permeability
constant is lower than that for the preceding stage. Perme—
ability constants for Stage 4 corneas have little meaning Since
these corneas are characterized as having total perforations
in their central areas. Solutions in the immediate vicinity
of both surfaces of Stage 4 corneas contained considerable
quantities of HTO and this probably accounts for the high HTO
content of these corneas. Stage 1-a may be a continuation of
Stage 1 or it may represent a second form of the onset of the
lesion. In this stage the anterior chamber depth is normal
and the cornea has a normal radius of curvature, but it takes
on a milk cloudiness which may be diffuse or discrete. The
normal cuboidally arranged basement epithelial cells are dis-
organized, and the epithelium does not remain attached to its
base membrane. It is assumed that the increased permeability
and hydration observed in Stage 1-a corneas is associated with
this disruption of the basement layer of the epithelium.

It appears that as long as the epithelium and Bowman's
membrane remain intact, there is only a limited movement of
water into or across the lake trout cornea. Some of the changes
noted in the permeabilities of pathologic lake trout corneas

can be explained on the basis of destruction of the anatomical

55

barriers to water transfer which are normally present in intact,
healthy tissue. It is most probable that there are alterations
in the metabolism of the pathologic corneas which may also
affect their water permeability. Smelser and Chen (1954)

report that the corneal epithelium of carp is rich in phospha-
tase and this enzyme(s) may be important in the maintenance of
normal corneal hydration. We have positive histochemical
evidence of the presence of various enzyme systems indicating
the epithelium of lake trout corneas is quite active meta-
bolically.

Data presented in Table III Show that the normal lake
trout cornea is impermeable to sodium in the "endothelial"—
epithelial direction and vice-versa. This is of interest Since
the lake trout lives in a hypoosmotic environment conducive to
loss of sodium to the surrounding water. Apparently there is
a "barrier" within the cornea which maintains it impermeable
and prevents a loss of sodium from the aqueous humor to the
exterior. Histologically, the corneal epithelium consists of
approximately five cell layers packed closely together
(Edelhauser, 1964) and does not Show any tendency of becoming
hydrated, as would be expected of epithelium bathed with a
hypoosmotic medium. Also, the corneal epithelium (basement
membrane) cannot be regarded merely as a passive protective
membrane in the fish, for it is rich in enzymes, including
ATPase (Figure 5, Plate A) which suggests that it is an active

metabolic tissue. Whether this barrier, which has an

56

impermeable nature, contains an active pump mechanism as found
in mammalian corneas (Donn et al., 1965; Green, 1966; and
others) remains to be answered. However, when the epithelium
is removed (Table III), there is a unidirectional net flux of
sodium across the cornea, indicating that the impermeable barrier
within the cornea is located either in Bowman's membrane or the
corneal epithelium or both. After removal of the epithelium,
"endothelium," or both layers of the fish cornea, the flux of
sodium across the cornea increased. When the "endothelium"
was removed, there was a slight sodium flux from the cornea,
which is consistent with the water flux observed (Table I) and
it can be concluded that the "endothelium" does offer some
resistance to sodium and water movement. Possibly the resist-
ance of the "endothelium" could be due to a mucinous substance
layered on the "endothelium" similar to that Baramy (1957)
observed in the owl. Thus upon removal of the ”endothelium,"
this mucinous layer, high in hyaluronic acid, is also removed,
and there results an increased sodium flux and water exchange
across the fish cornea. Based on the permeability constants
given in Table III, it can be concluded that some resistance
to sodium flux across the fish cornea is maintained by the
"endothelium" (in the TC-199 tap direction), but the major re-
sistance to movement in both directions resides with the
epithelium. Sodium flux across the fish cornea is more likely
to follow the diffusion gradient (condition A) than to oppose

the gradient (condition B). When a Na22 gradient was

57

established by putting the Na22 on the epithelial surface
compartment, only a small fraction of the labelled Na pene-
trated the cornea.

The values of corneal Specific activity, corneal CPM/mg,
and percent corneal water are consistent with the previous
section for water movement and with the above conclusions.
Cogan and Kinsey (1942a) have Shown Similar results for the
cat cornea when its layers were denuded. Maurice (1951),

Green (1966), and data given in Table VII Show that the normal 1

 

rabbit cornea possesses a limited permeability to Na in both I
directions. The fish cornea is impermeable to sodium in both
directions. The low rate of water exchange across the cornea

(Table I) and impermeable nature of the fish cornea to sodium

explains why Smelser and Chen (1954) found the excised fish

cornea to swell less than mammalian corneas when placed in

isotonic and hypertonic sodium chloride solutions.

The lake trout corneal lesions have been briefly dis-
cussed in the section on water permeability. Values in Table
IV indicate that sodium diffuses across the cornea only in
stages 5 and 4. A cornea denuded of its epithelium permits
an outward diffusion of sodium. Thus it is feasible to con-
clude that the stage 5 cornea does have epithelial damage.

This supports previous histological work done by Hoffert and
Fromm (1965). Corneal hydration appears to be a very important
phenomena in these diseased lake trout corneas. Table V indi-

cates that the normal, stage 1 and 1-a corneas are able to

58

maintain an equilibrium condition with relation to corneal
hydration and sodium ion concentration. It appears that in
stage 2, 5, and 4 corneas the mechanism which maintains the
cornea dehydrated has been lost for they are at maximal or
near-maximal hydration in situ. When the data for corneal
hydration (after a four-hour perfusion) given in Tables IV and
V are compared, it is obvious that the normal, stages 1 and 1-a
corneas do hydrate, but do not reach the maximum hydration
observed in those of stages 2, 5, and 4. This indicates the
presence of a hydration control mechanism in the normal, 1
andjsiicorneas. According to Smelser and Chen (1954), the
resistance to hydration in fish cornea is due to the organi—
zation of mucoids (polymerization), electrolyte control and
thickness of the epithelium all of which act together to
inhibit the penetration of water. A theory has been suggested
by Donn (1966), Smelser (1962), and Hedbys (1961) where a
control of stroma sodium ion is needed to maintain a physio-
logical pH and to neutralize the negative charges furnished
by acid stroma polysaccharides. When for some reason the
homeostasis of the cornea is altered as in the various stages
of the lake trout corneal lesion, and the corneal sodium ion
decreases, the negative charges on the mucopolysaccharide are
not neutralized. The negative charges repel each other and
cause the nucopolysaccharides to uncoil, allowing the cornea
to take up more fluid. Even though this is still a theory,

it may be the case in this lesion. In stages 2, 5, and 4

59

corneas, where hydration is maximal, the critical point could
have been reached between ionic balance and mucopolysaccharide
uncoiling; therefore, no control of corneal hydration can be
maintained.

In the pathological progression of this corneal lesion
(Hoffert and Fromm, 1965), stage 1-a has been questioned as to
its place in the chronologigal progression. Data presented in
Tables IV and V generally support the classification; however,
sodium flux data, cornea CPM/mg, corneal water, and corneal
sodium ion concentrations all indicate that stage 1-a is a
continuation of stage 1 rather than a second form of the onset
of the lesion.

According to Donn (1966) the mammalian "endothelium"
interferes with corneal swelling in two ways: it provides
some sort of barrier decreasing the access of aqueous humor
to the corneal stroma, and its metabolic activity, aside from
maintaining the barrier property, acts on the stroma in some
poorly understood way to keep it dehydrated. Our data also
indicates the fish corneal "endothelium" acts similarly for
when the "endothelium" was removed, the water transfer, sodium
flux, and corneal hydration all increased.

The barrier function of the "endothelium" does not stem
from any unusual impermeability of this membrane. There is no
great resistance to the diffusion of inorganic ions which have
been studied in the mammal, and labk of resistance appears to

be similar in fish. In fish, one half of the sodium in the

 

60

stroma exchanges across the ”endothelium” with that of the
aqueous humor every 22.8 minutes, whereas in the rabbit T2
is 14 minutes. It can be concluded that the fish corneal
"endothelium" is of the order of 16 percent less permeable
to sodium than the rabbit. In View of such a large sodium
movement across the "endothelium," it would appear that most
of the movement of these ions is by way of a system of rela-
tively large, uncharged pores as suggested by Maurice (1961).
Since all of the corneal permeability studies to date
in other laboratories have been on mammalian corneas, rabbit
corneas were used in order to compare fish and rabbit corneal
permeability under nearly identical experimental conditions.
From the results of experiments using two-labeled com-
pounds, it is evident that there are differences in perme-
ability between mammalian and aquatic corneas. Corneas of
all Species which have been studied, except the elasmobranch
reported by Smelser (1962), imbibe water from the surrounding
media after damage to the limiting cellular membranes or after
excision. With this in mind one would expect the fish cornea
to exhibit a greater permeability to water than the rabbit
cornea under the conditions of the experiments reported here,
but this is not the case. AS illustrated by data in Table I
and VI the exchange of water is 5 to 5 fold higher in the
rabbit. These data for water exchange across rabbit corneas
are Similar to what Donn e; 21. (1965), reported. Apparently

the epithelial "barriers" in fish do not house the same

 

61

mechanism as they do in mammals. A metabolic "pump" for
active transport of sodium into the cornea was found in the
fish, but it was very evident in the rabbit. Nevertheless,
both epithelia and "endothelia" are of great importance in
maintaining corneal deturgescene in fish because if they are
removed, the exchange of water across the cornea is increased,
and it imbibes water. Possibly these layers, by virtue of
their being relatively impermeable to water, may passively

limit entrance of water into the corneal stroma from the

 

aqueous humor or the external fresh-water. I

It is known that damage to the corneal epithelium
drastically increases corneal permeability to water and ions
(Maurice, 1955, 1962) and utmost care was taken to avoid any
damage to the corneas in the present experiments. Values for
the unidirectional fluxes and net fluxes of sodium across the
rabbit cornea (Ep. to Endo. 5.45.1 0.001 and net 0.70.1 0.27
qu/cma/hr) are higher than those reported by Donn 33 _l.
(1959), Ep. to Endo. 0.28 and net 0.17 qu/cmZ/hr; and Green
(1966), Ep. to Endo. 0.87 .t 0.06 and net 0.5? .t 0.005
qu/cmZ/hr.

Green stated the most plausible explanation for the dis-
crepancy between his and Donn‘s data was that his solutions
bathing either side of the corneas were well stirred, whereas,
in Donn's investigation they were not. Proper stirring is
important to minimize the diffusion barrier to solutes offered

by the presence of the so-called "unstirred layers."

62

When investigating membrane permeability with isotopes, these
barriers may offer a great resistance to solute entering and
leaving the tissue (Green, 1965). There is no significant
difference between the net flux that Green reported and that
given in Table VII, but the unidirectional flux recorded in
Table VII is much higher than what Green reported. Since the
net fluxes are the same, it is believed that circulation in
the system used here was sufficient; however, there are dif-
ferences between the experimental set-ups which may account
for this unidirectional flux difference: (1) TC-199 was used
instead of Ringer's solution, (2) Green recorded a Short-
circuit current every ten minutes, and we did not. It is
generally accepted that when a membrane is Short circuited,
the membrane potential and permeability characteristics of
the membrane are altered for a given period of time after
short-circuiting. Since this was not the case in the present
study on the rabbit cornea (Table VII), the difference between
the two reported unidirectional fluxes may be due to this
short circuiting technique. Apparently short-circuiting had
no effect on the net flux of sodium across the cornea.

The method employed (Wachstein and Meisel, 1957) appears
to be the one best suited for histochemical localization of
ATPases. However, it reveals only a small part of the total
enzyme activity (Pearse, 1960), and it is not known if some
ATPases (e.g., Na—K—activated) are more completely inhibited

than others. With sodium B—glycerophosphate as a substrate

65

in the incubating medium no reaction was observed. The enzyme
demonstrated with ATP as a substrate cannot, therefore, be a
non-Specific alkaline phOSphatase. An active transport of
sodium across the mammalian corneal epithelium is known to
occur (Maurice, 1962), and this transport was found to be
inhibited by ouabain (Lambert and Donn, 1964). Ehlers (1965)
was unable to demonstrate ouabain inhibition of the corneal
ATPase, and he Stated the method of Wachstein and Meisel is
probably unsuitable for quantitative analysis of this enzyme.
The histochemical technique as used in the rabbit and fish
studies was for localization and not quantitation of ATPase.

A most interesting comparison pertaining to corneal
sodium permeability in those animals which have been studied
has been made. In the rabbit cornea there is a net sodium
flux in the direction of outside-in. This net flux has been
attributed to the active tranSport across the epithelium (Donn
.EE.§l-I 1959, and Green, 1966). Data in Tables III and VI
show that there is no flux or net flux of sodium across the
freshwater fish cornea; the cornea is impermeable enabling
adaptation to its aquatic environment. Recently Zadunaisky
(1966) worked on the bull frog cornea which has a nicitating
membrane to prevent tear loss while in an aerial environment
or submerged in a hypoosmotic medium. He found in the frog
cornea that sodium fluxes measured with Na22 indicated that no
net transport of this cation occurs; fluxes in both directions

were Similar. However, these corneas were found to actively

64

transport chloride in an in-out direction while sodium moved
passively.

Corneas of these three animals each handle sodium ions
in a different manner in order to adapt to their environment,
and normally each of these Species maintains a transparent
cornea. In the absence of more conclusive data, it is unwise
to attempt to formulate a concept, that phylogenetically, fish

and amphibia have more primitive corneas than mammals.

 

SUMMARY AND CONCLUSIONS

The investigation of the water and sodium permeability

of the isolated normal and pathological lake trout cornea

has Shown the following:

1.

4.

The water permeability of lake trout corneas, in vitro,

 

was found to be the same whether the data are based on
the transfer of tritiated water from the epithelial to
the "endothelial" bathing media or vice versa. This
occurred despite the presence of an osmotic gradient
which favored movement of water through the cornea to

the solution bathing the "endothelium."

The fish cornea was found to be impermeable to sodium
ions in both directions deSpite the presence of a
gradient which favored a sodium flux through the

cornea to the solution bathing the epithelium.

Permeability constants and sodium flux for normal
and denuded corneas indicate that the "endothelium"
offers some resistance to the transfer of water and
sodium into or across the fish cornea but the major
resistance to this movement is provided by the

epithelial layer.

Correlations between water permeability, sodium flux,
and histopathological condition of diseased corneas

are noted.

65

66

Corneal hydration in fish, as in mammals, is con-
trolled by the epithelium and "endothelium" Since
removal of these layers resulted in stromal hydra-
tion. Stage 2, 5 and 4 corneas are at maximal
hydration in situ, whereas, the normal, stage 1 and
1-a corneas do imbibe water in vitro from the bath-

ing media.

Corneal sodium ion concentration varied inversely
with corneal hydration in these diseased lake trout

corneas .

The permeability constant of sodium for the
"endothelial" barrier from cornea to aqueous in fish
is 0.029 mEq/cma/hr which is 16% lower than that

observed in rabbit corneas.

In the fish cornea there is no apparent movement of
water drawn along osmotically with the movement of
sodium ions. HTO movement and Na22 flux in this
corneal perfusion study occurred independent of each
other; however, with rabbit corneas osmotic flow of

water does occur.

Rabbit corneas had a 5 to 5 fold greater water ex-
change than fish and they actively transported sodium
in the tear-aqueous direction, whereas, in the fish

cornea no sodium flux occurred.

67

10. An adenosine triphosphatase was identified in the
basement layer of the corneal epithelium, stroma,

and corneal "endothelium."

LITERATURE CITED

Allison, L. N., 1965. Cataract in Harchery Lake Trout.
Trans. Am. Fish. Soc., 92: 54-58.

Anseth, T., and C. H. Dohlman, 1957. Influence of the Intra-
ocular Pressure on Hydration of the Corneal Stroma.
Acta Ophth., 55: 85-90.

Barany, E., L. Berggren, and F. Vrabec, 1957. The Mucinous
Layer Covering the Corneal Endothelium in the Owl
Strix aquco. Brit. J. Ophth., 41: 25-50.

Cogan, D. G., and V. E. Kinsey, 1942a. The Cornea I: Transfer
of Water and Sodium Chloride by Diffusion Through the
Excised Cornea. AMA Arch. Ophth., 27: 466-476.

Cogan, D. G., and V. E. Kinsey, 1942b. The Cornea V: Physio-
logical Aspects. AMA Arch. Ophth., 28: 661-669.

Cogan, D. G., and V. E. Kinsey, 1942c. The Cornea II: Transfer
of Water and Sodium Chloride by Hydrostatic Pressure
Through the Excised Cornea. AMA Arch. Ophth., 27: 696-
704.

Comar, C. L., 1955. Radioisotopes in Biology and Agriculture.
McGraw-Hill, Inc., New York, 18-58.

Davis, T. L., and J. A. Zadunaisky, 1966. Potassium Perme-
ability of the Frog Cornea. Fed. Proc. 25: 652.
(Abstract)

Dohlman, C. H., 1965. Annual Review, Cornea and Sclera.
AMA Arch. Ophth., 69: 257-278.

Dohlman, C. H., 1964. Annual Review, Cornea and Sclera.
AMA Arch. Ophth., 71: 249-265.

Donn, A., D. M. Maurice, and N. L. Mills, 1959a. The Active
Transport of Sodium Across the Corneal Epithelium in
the Rabbit. Trans. N. Y. Acad. of Sci., 21: 578-581.

Donn, A., D. M. Maurice, and N. L. Mills, 1959a. Studies on
the Living Cornea I§_yitrg. I. Method and Physio-
logical Measurements; II. The Active Transport of
Sodium Across the Epithelium. AMA Arch. Ophth., 62:
741-747; 748-757.

68

69

Donn, A., 1962. The Movement of the Ions and Water Across the
Cornea. Invest. Ophth., 1: 170-177.

Donn, A., S. L. Miller, and N. M. Mallett, 1965. Water Perme-
ability of the Living Cornea. AMA Arch. Ophth., 70:
515—521.

Donn, A., 1965. Annual Review, Cornea and Sclera. AMA Arch.
Ophth., 75: 278-294.

Donn, A., 1966. Annual Review, Cornea and Sclera. AMA Arch.
Ophth., 75: 261-288.

Edelhauser, H. F., 1964. In_Vitro Corneal Ion and Water Move-
ment in the Rainbow Trout, Salmo Gairdneri. M. S. thesis,
Michigan State University.

 

Edelhauser, H. F., J. R. Hoffert, and P. O. Fromm. 1965.
In Vitro Ion and Water Movement in Corneas of Rainbow
Trout. Invest. Ophth., 4: 290-296.

Ehlers, N., 1965. Corneal Adenosine Triphosphate: A Histo-
chemical Investigation. Exp. Eye Res., 4: 48-55.»

Green, K., 1965. Ion TranSport in Isolated Cornea of the
Rabbit. Am. J. Physiol., 209: 1511-1516.

Green, K., 1966. Ion TranSport Across the Isolated Rabbit
Cornea. Exp. Eye Res. 5: 106-110.

Hedbys, B., 1961. The Roll of Polysaccharides in Corneal
Swelling, Exp. Eye Res., 1: 81-91.

Hedbys, B., and S. Mishima, 1962. Flow of Water in the Corneal
Stroma. 'Exp. Eye Res. 1: 262-275.

Hedbys, B., S. Mishima, and D. M. Maurice, 1965. The Imbibition
_ Pressure of the Corneal Stroma. Exp. Eye Res. 2: 99-111.

Hoffert, J. R., and P. O. Fromm, 1965. Biomicroscopic, Gross,
and Microscopic Observations of Corneal Lesions in Lake
Trout, Salvelinus namaycush. J. Fish. Res. Bd., Canada,
22:761-766.

Lambert, B., and A. Donn, 1964. Effect of Ouabain on Active
Transport of Sodium in the Cornea. AMA Arch. Ophth.,
72: 525-528.

Maurice, D. M., 1951. The Permeability to Sodium Ions of the
Living Rabbit's Cornea. J. Physiol., 122: 567-591.

70

Maurice, D. M., 1955. The Permeability of the Cornea. Ophth.
Lit. Lond., 7:5-26.

Maurice, D. M., 1961. The Use of Permeability Studies in the
Investigation of Sub-microscopic Structure, in Structure
of the Eye, G. Smelser (ed.), Academic Press, New York.

Maurice, D. M., 1962. The Cornea and Sclera in the Eye, edited
by H. Davson, The Academic Press, New York, 289—568.

Mishima, S., 1965. Some Physiological ASpects of the Pre-
corneal Tear Film. AMA Arch. Ophth., 75:255-241.

Mishima, S., 1966. Personal communication.

Pearse, A. G. E., 1960. Histochemistry, Theoretical and
Applied. J. & A. Churchill, London, 876-877.

Potts, A. M., 1954. Permeability of Corneal Layers as
Demonstrated by Tracer Experiments. ACTA XVII Conc.
Ophth., 1: 506-511.

Potts, A. M., and Modrell, R. W., 1957. The Transcorneal
Potential. Am. J. Ophth., 55: 405-419.

Sheppard, C. W., and W. R. Martin, 1950. Cation Exchange Be-
tween Cells and Plasma of Mammalian Blood. I. Methods
and Application to Potassium Exchange in Human Blood.
J. Gen. Physiol., 55: 705-722.

Sheppard, C. W., 1962. Basic Principles of the Tracer Method.
John Wiley and Son, New York, 1-14.

Smelser, G. K., and D. K. Chen, 1954. A Comparative Study of
the Structure and Hydration Properties of Corneas
Adapted to Air and Aquatic Environments. ACTA XVII Conc.
Ophth., 1: 490-499.

Smelser, G. K., 1962. Corneal Hydration: Comparative Physi-
ology of the Fish and Mammals. Invest. Ophth., 1: 11-52.

Ussing, H. H., 1949. Active Ion Transport Through the Iso-
lated Frog Skin in the Light of Tracer Studies. ACTA
Physiol. Scand., 17: 1-19.

Vrabec, F., 1959. Studies of the Corneal and Trabecular Endo-
thelium. III. Corneal Endothelium in Teleostei,
Vestnik Ceskoslovenski Zoologicki Spolecnosti, ACTA Soc.
2001. Bohemosloven, XXII: 161-165.

71

Wachstein, M., and E. Meisel, 1957. Histochemistry of Hepatic
Phosphatase at A Physiological pH. Am. J. Clin. Path.,
27: 15-25.

Walls, G. L., 1942. The Vertebrate Eye and Its Adaptive
Radiations. Bloomfield Hills, Mich., Cranbrook Institute
of Science, 576-588.

Zadunaisky, J. A., 1966. Active TranSport of Chloride Across
the Cornea, Nature, 209: 1156-1157.

APPEND IX

72

75

Procedure for countingytritium and sodium22
simultaneously.

The Mark 1 Liquid Scintillation system was peaked for

HTO and Na22 with the following settings:

 

 

 

 

 

upper lower differential
discrimi- discrimi- integral
isotope channel attenuator nator nator switch
HTO A A-507 9.9 0.0 L-U
Na22 B C-000 9.9 9.2 L-oo
30% c A-926 1.9 0.0 L—U

channel A

With channel C set at 50% channel A one is able to have a
calculated quench correction made by the Mark 1.

To account for the amount of Na22 found in the HTO channel
the following procedure was used. A standard curve (see curves
I & II) was made by diluting Na22 through the range of
0.00002uc to 0.2uc. These various solutions were then counted

(see table A) and the counts per minute plotted on the ordinate

22 22

against a channels ratio (channel B Na cpm/channel A Na cpm
in tritium channel) which are plotted on the abscissa. Total
counts of Na22 in channel A (tritium channel) are calculated

by dividing the total counts of Na22 in channel B by the ratio
B/A obtained from the standard curve of cpm channel B vs.

ratio B/A (curves I & II). Channel A total counts minus calcu-

lated Na22 counts in channel A gives HTO counts in channel A

corrected for Na22 overlap (see sample calculation).

74

Sample calculation for sodium22 correction in
the tritium counts.

_ channel B Na22 cpm

(1) E—
A channel A Na22 cpm

=K

(2) Take sample Na22 counts in channel B read them off
standard curve to get B/A.
Example: HTO counts channel A = 51284

)

Naaa counts channel B = 197207

(a) find ratio of 196207 Naaa cpm from standard
curve = 44.1 = B/A

(b) calculation

B _ 196207 =
K --—-ZZTI 4440 cpm

(c) subtract Na22 counts from HTO counts
51284
- 4440
46844

46844 cpm = HTO counts in channel A

corrected for Na?2

75

 

 

rTable A: Data from Mark 1 Liquid Scintillation system used
for standard curve for dual labeling experiment.
Bottle Na22 Channel A Channel B Ratio
number go HTO cpm Na22 cpm B/A
Curve I
A 0.2 8555 585725 46.04
B 0.1 4425 196679 44.22
C 0.09 5960 176787 44.65
D 0.08 5271 155882 45.65
E 0.07 5125 156150 45.57
F 0.06 2740 116807 42.65
G 0.05 2285 97048 42.47
H 0.04 1801 77402 42.16
I 0.05 1444 58275 40.56
J 0.02 975 57909 58.95
K 0.01 551 19759 57.20
Curve II
M 0.002 127 5695 29.02
N 0.00155 108 2859 26.26
0 0.0010 85 1892 22.77
P 0.0005 61 1019 16.64
Q 0.0002 47 476 10.02
R 0.0001 42 275 6.57
S 0.00008 59 219 5.57
T 0.00004 58 147 5.82
U 0.00002 55 128 5.92
V 0.00000 55 77 2.24

IIBII

CPM Channel

76

 

580,000b

r

540,000L-

500,000"

260,000-

220,000”

180,000

I

140 I 000 '-

100,000-

60,000-

20,000"

 

 

Figure A.

Standard Curve I

40 42
Channels Ratio B/A

44

 

46

 

IIBII

CPM Channel

5800

5400

5000

2600

2200

1800

1400‘-

1000

600

200

77

 

P

 

 

Figure B. Standard Curve II.
I -l 1 n l J 11 1 l 1 l L 1 J k
2.0 6.0 10.0 14.0 18.0 22.0 26.0 50.0

Channels Ratio B/A

 

Table B:

Fish

Rabbit

78

Percent turnover of Na22

in fish and rabbit corneas.

 

_ 0.695 -Kt -Kt
T1/2 Kp - T—172' e 1 - e
22.9 min. 0.05059 0.97005 0.050 = 5.0%
14.0 min. 0.0495 0.95170 0.0490 = 4.8%

 

   

WIITIII‘I‘IINIISIWIIEI IIIIIIIIIIIIIIII
3 1293 030711687