Measurement of the Volume Flow and Hydraulic Conductivity Across by iht11609


									      Measurement of the Volume Flow and Hydraulic
           Conductivity Across the Isolated Dog
               Retinal Pigment Epithelium
                                                           Shunji Tsuboi

            The isolated dog retinal pigment epithelium (RPE)-choroid was gently stretched on the inner surface
            of a spherical stainless mesh, retinal side upward, and clamped between half-chambers made of Kel-F.
            The volume flow across the tissue was monitored by the movement of water in capillary tubes
            connected to both chambers. With zero pressure difference across the RPE-choroid, retina-to-choroid
            fluid flow was determined to be 6.4 jul/hr/cm2 (absorption). Removal of HCO3~ from the solution did
            not affect the fluid flow. However, the flow was reduced 88% in Cl"-free medium, indicating a coupling
            between water and Cl~ absorption. The flow was also inhibited by ouabain ( 1 0 s M) and furosemide
            (10~4 M). Hydraulic conductivity (Lp) of the RPE-choroid was determined to be 0.0126 ^l/min/cm 2 /
            mm Hg which places the dog RPE-choroid in the category of a "leaky" epithelium. Invest Ophthalmol
            Vis Sci 28:1776-1782, 1987

   Evidence is accumulating in vivo that the retinal                        The present study was undertaken to determine the
pigment epithelium (RPE) absorbs fluid from the ret-                      volume flow across the isolated dog RPE-choroid.
inal to the choroidal surface.1"3 This volume flow                        The effect of HCO3~, Cl~, ouabain, and furosemide
may assist the sensory retina in attachment to the                        on the volume flow was then examined. Finally, the
RPE,1'4 and is one of the unconventional routes for                       hydraulic conductivity (Lp) was determined.
aqueous humor drainage in eyes with rhegmato-
genous retinal detachments (postiridial flow).2"5 Di-                                 Materials and Methods
rect measurement of the flow has also been made                           Tissue Preparation
using the isolated frog RPE-choroid, where the vol-
ume flow is 4.8-7.8 nl/hr/cm2, from retina to cho-                          The eyes were obtained from adult dogs of both
roid, without hydrostatic or osmotic pressure differ-                     sexes weighing 15-30 kg shortly after sacrifice. Prepa-
ence across the membrane, indicating an isotonic                          ration of the dog RPE-choroid was described in detail
electrolyte-linkedfluidflow.6-7                                           previously. 810 All procedures conformed to the
                                                                          ARVO Resolution on the Use of Animals in Re-
   Hughes, Miller, and Machen reported that HCO3~
is the major ion that drives water from retina to cho-
roid in the frog RPE-choroid, based on the observa-
                                                                          Bathing Solution
tion that removal of HCO3~ from the bathing solu-
tion both inhibited the fluid movement and de-                           The standard bathing solution was composed of (in
creased the short-circuit current (SCC).6 However, in                 mM) NaCl, 102; KC1, 0.8; NaHCO3, 24; KH2PO4,
the isolated RPE-choroid of dogs8 and chickens,9                      1.2; Na2SO4, 4; CaCl2, 2.5; MgSO4, 1.2; HEPES-Na,
SCC is not decreased by the elimination of ambient                    8.0; HEPES, 9.0; and glucose, 5.5. In addition, a so-
HCO3~. It is thus likely that a different mechanism is                lution where either HCO3" or Cl~ was replaced with
involved in the water transport of warm-blooded ani-                  equivalent SO42~ was prepared. The RPE-choroid
mals.                                                                 was always immersed in the identical solution on
                                                                      both sides. The pH of the solutions was 7.4 at 37°C.
  From the Department of Ophthalmology, University of Minne-          Osmolality was measured by a freezing point osmom-
sota, Minneapolis, Minnesota.                                         eter (micro osmette; Precision Systems, Inc., Natick,
  Supported by NIH grant EY-03277.                                    MA) and adjusted, if necessary, to 300 mosmol/kg
  Dr. Tsuboi is Visiting Research Fellow, Department of Ophthal-      using mannitol.
mology, Osaka University Medical School, Osaka, Japan.
  Submitted for publication: June 9, 1986.
                                                                         During initial volumetric experiments, it was
  Reprint requests: Shunji Tsuboi, MD, University of Minnesota,       readily noticed that expansion of the solution due to
Box 493 Mayo, Minneapolis, MN 55455.                                  gas bubbles resulted in a substantial error. The solu-

No. 11                                FLUID ADSORPTION IN DOG RPE / Tsuboi                                               1777

tions were thus slightly degassed using a vacuum
pump, so that air bubbles did not appear when the
temperature was raised to 37°C. This caused negligi-
ble change in the pH and pO2.

Tissue Mounting
   The RPE-choroid was gently placed, retinal side
upward, on the inner surface of a spherical stainless
mesh (radius = 3.54 mm, depth = 3.0 mm, Fig. 1).            Fig. 1. Cross section of the tissue-holding ring. The RPE-choroid
The area of the RPE-choroid exposed to the solution       (1) is held between the silicone disk (2) and stainless mesh (3). The
was 0.67 cm2. The mesh with the tissue was mounted        mesh is spherical inside the ring, so that the area exposed to the
in a tissue-holding ring. An elastic doughnut-shaped      solution is enlarged (0.67 cm2).
silicone plate (Dow Corning 500-9; Midlands, MI)
was placed on the retinal side of the tissue. The tis-
sue-holding ring, mesh, and silicone plate were           tern was 30 nl. To prevent evaporation from the
coated with high-vacuum grease. These procedures          water surface, the ends of capillaries were covered
were done under an operating microscope.                  with wet gauze.
                                                             Water movement was always monitored at both
                                                          sides of the half-chambers, ie, retinal and choroidal
Chambers                                                  sides of the RPE-choroid. When the RPE-choroid
   Two pairs of Ussing-type chambers were used in         transports water from retinal to choroidal surface, the
each experiment. The shape of each half-chamber           volume of the bathing solution on the retinal side
was symmetrical, with a volume of 12.5 ml, and be-        should decrease, while that on the choroidal side
tween each half-chamber fit the tissue-holding ring       should increase over time. When other factors such as
(Fig. 2). The volume fluctuation originating from the     temperature change, gas bubbles appearing in the so-
system was occasionally tested by changing the orien-     lution, and evaporation are also involved, simulta-
tation of the tissue ring in the chamber. One of the      neous increase or decrease at both ends may occur.
chambers, constructed especially for the volume flow      Experiments where simultaneous increase or de-
measurement, was made of Kel-F, an extremely hy-          crease occurred throughout the experiment were dis-
drophobic material, instead of Lucite, which absorbs      carded.
a considerable amount of water.6 Moreover, the               Piston-like movement of the tissue may cause a
Kel-F chamber did not contain electrode holes in          substantial error, since a rigid support was placed
order to avoid a possible leak.                           only on the choroidal side. Due to water surface ten-
   The mounted RPE-choroid was first clamped be-          sion in the capillaries, a pressure was required to initi-
tween Lucite half-chambers, which were connected          ate meniscus movement. In the present system, the
to an automatic voltage clamp device through two          frictional pressure was around 1 cm H2O, the pres-
pairs of 3% agar-3 M KC1 bridges and calomel elec-        sure capable of dislocating RPE-choroid from the net.
trodes as described in detail previously.810 The trans-   Initial experiments revealed that transepithelial water
epithelial potential difference (Et) and SCC were then    movement sometimes resulted in movement of the
determined. The resistance (Rt) was calculated from       tissue rather than meniscus. Thus, the following pro-
Et and SCC by Ohm's law. Ten to 30 min later, when        cession was followed in each experiment.
Et became constant, the mounted RPE-choroid was
moved to the Kel-F chamber. After the measurement
of volumeflowas described below, electrical parame-
ters were again examined in the Lucite chamber.

Measurement of the Volume Flow
   Each half of the Kel-F chamber was connected to a
glass capillary tube, with inner diameter of 0.66 mm,
via polyethylene tubing. The capillaries were set hor-
izontally at the same height. Volume flow across the
                                                            Fig. 2. Kel-F chambers and tissue holding ring. 1. Water jacket.
tissue was determined from the meniscus movement          2. Stopper and hole for air release and drug administration. 3.
in the capillary." Since the movement of 0.1 mm was       Connector to the capillary tube through polyethylene tubing. 4.
detectable using a magnifier, the accuracy of the sys-    Gasket holding temperature probe. 5. Tissue-holding ring.
1778                        INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / November 1987                             Vol. 28

                7.3±0.4(mV) _                      -^ 3.6 ±0.7       bathing solution was preheated at 37°C and added to
                378±47 (O cm2)                        330126
     0.15                                                            the solutions on both sides of the tissue chamber
                                 Sham                                using a micropipette (final concentrations, 10~5 M
                                                                     ouabain and 10~4 M furosemide). Symmetrical ad-
     0.10       H-h-                                                 ministration of the drug precluded any change in os-
                                                                     motic pressure difference across the membrane. The
                                                                     solutions were stirred for 1 to 5 min using magnetic
E                                                                    stirring bars. A preliminary experiment showed that
^    0.05                                                            this mixing procedure was enough to cause a typical
                                                                     decrease in Et and SCC by these drugs in the isolated
                                                                     dog RPE-choroid. Control experiments with the
        0                                                            same procedure without drugs were also performed.

                  30         60         90              120          Hydraulic Conductivity
                               Time (min)
                                                                       After the volume flow measurement, where only
   Fig. 3. Time course of the volumeflow(Jv) in the control exper-   compensatory and transient pressure was applied,
iment. Each point represents means ± SE of seven experiments. At     volume flow was also determined with a continuous
t = 60 min, "sham" procedure is performed. An almost linear
decline is noted regardless of the procedure. Et and Rt before and
                                                                     hydrostatic pressure on the retinal side. Three or 8 cm
after volume flow measurement are also shown.                        H2O was applied.

   First, 3 cm H2O static pressure was applied on the                                       Results
retinal side for 10 min by lowering the capillary tube
                                                                     Control Experiments
connected to the choroidal side. This produced a
rapid movement of the meniscus, due to retina-to-                       Figure 3 summarizes seven control experiments.
choroid tissue movement, but in seconds the menis-                   After the tissue-holding ring was placed in the Kel-F
cus movement stopped, indicating that the tissue was                 chamber, less than 30 min was required for the tem-
entirely attached to the stainless steel net. Experi-                perature to become stabilized. Thereafter, fluid trans-
ments with continuous meniscus movement result-                      port was observed from retina to choroid (absorption)
ing from a leak were discarded. Capillaries were then                for more than 2 hr. Mean Et before the volume flow
set at the same height. Every 10 min, 3 cm H2O static                measurement, 7.3 mV, retinal side positive, de-
pressure was placed on the retinal side for 15 sec, and              creased 50% after the measurement. Mean Rt, 378
readings were made. This transient pressure compen-                  ohm-cm2, decreased 9%. An almost linear decline in
sated tissue dislocation. The 3 cm H2O pressure was                  Jv was noted regardless of the sham procedure. The
shown to have negligible effect on the volume flow                   decay of Jv was, however, small enough to examine
determination (see Results). Hydrostatic pressure was                drug effects within the 2 hr time period.
calibrated using a pressure transducer (Hewlett Pack-
ard, 7754 B; Waltham, MA).
                                                                     HCO3~- and CP-Free Solutions
Temperature Control                                                     It was virtually impossible in the present study to
                                                                     determine the volume flow of the RPE-choroid in
   The Kel-F chamber including tissue was placed in-                 two consecutive solutions, because of the spontane-
side an oven whose temperature was kept around                       ous decay of Jv and relatively large time loss until the
32°C. In addition, the temperature of the circulating                temperature of the substituted solution became stabi-
water around the chamber (Fig. 2) was controlled by                  lized. Therefore, each preparation was used for only
a proportional controller (Yellow Springs Instrument                 one of the bathing solutions.
Co., Yellow Springs, OH; Model 72), so that the
                                                                        Table 1 summarizes Jv, Et, Rt, and SCC of the dog
bathing solution, monitored by a tele-thermometer
                                                                     RPE-choroid determined in the standard, HCO3~-
(Yellow Springs Model 43 TA) and temperature
                                                                     free, and Cr-free solutions. Jv values in this Table
probes (Yellow Springs Model 423), was kept con-
                                                                     were taken one hour after the beginning of volume
stant at 36-37 ± 0.05°C.                                             flow measurement. Et, Rt, and SCC were measured
                                                                     in the Lucite chamber just prior to (standard and
Effect of Ouabain and Furosemide                                     HCO3"-free) or after (Cr-free) the volume flow mea-
  After the volume flow became constant, 125 /A of                   surement. Retina-to-choroid Jv in the HCO3"-free
10~3 M ouabain or 10~2 M furosemide in standard                      HEPES was 0.093 /il/min/cm 2 (5.6                     2
No. 11                                            FLUID ADSORPTION IN DOG RPE / Tsuboi                                                            1779

Table 1. Jv (retina-to-choroid), Et, Rt, and SCC
                                           Jv                              Et                              Rt                       SCC
          Solution                     (nl/min/cm2)                      (mV)                          (ohm-cm2)                (nEq/hr/cm2)

         Standard                     0.106 ±0.009                      7.0 ± 1.1                        365 ± 19                0.71 ±0.11
           (n)                             (10)                            (14)                            (14)                      (14)
         HCOj-free                    0.093 ±0.010                      6.1 ± 1.0                        365 ±31                 0.60 ± 0.06
           (n)                             (10)                            (10)                            (10)                      (10)
         Cr-free                      0.012 ±0.008                      1.7 ±0.3                         289 ± 12                0.21 ±0.04
             (n)                            (6)                            (5)                              (5)                       (5)
  Values are mean ± SE.

which was significantly larger than zero (P < 0.001)                             bain was added into both half-chambers 1 hr after the
but not significantly different from Jv in the HCO3~-                            beginning of volume flow measurement. HCO3"-free
rich standard solution. Et, Rt, and SCC of these two                             HEPES solution was used in these experiments. After
sets of experiments were not significantly different.                            the ouabain application, Jv decreased markedly. The
   In the Cl~-free medium Jv reduced markedly, and                               difference from control was obvious 20 min after the
in 60 min it became not statistically different from                             application.
zero (Fig. 4). Jv at t = 60 min, 0.012 ± 0.008 (SE)                                 Reduction rates of Jv, Et, and Rt are summarized
/il/min/cm2, was markedly lower than the control ex-                             in Table 2. Jv values used in this Table were taken
periment superimposed in the Figure (0.099 ± 0.010,                              right before and 60 min after the drug administration
P < 0.001). Et reduced 77% from the value measured                               (or sham) procedure. Et and Rt were taken right be-
in the standard HEPES prior to the volume flow mea-                              fore and after the volume flow measurement. The
surement (Fig. 4). The reduction rate was signifi-                               reduction of Jv by ouabain application was 83%, sig-
cantly higher than control (25%/hr, P < 0.001, see                               nificantly larger than control (P < 0.002). At the end
Table 2). Change in Rt was not significant, however.                             of the experiment Et was reduced 91%, whereas Rt
                                                                                 decreased insignificantly.
10~5 M Ouabain
                                                                                 10~4 M Furosemide
  Addition of preheated 125 n\ aliquots including
ouabain or furosemide into the 12.5 ml half-                                        The effect of 10~4 M furosemide applied to both
chambers had little effect on the temperature, so that                           half-chambers is summarized in Figure 6, which con-
volume flow measurement was disturbed only for                                   sists of four experiments with HCO3~-free HEPES
several minutes from the drug administration. Figure                             solution and three with HCO3~-rich solution. Sixty
5 summarizes five experiments where 10~5 M oua-
                                                                                                  6.6 + 0.9 (mV) .                        0.6 10.2
                                                                                                  328116 (Ocm2)                           296 ± 15
                 7.5±1.9 (mV) _                         1.7±0.3
                 360 ±21 (O cm2)                        289±12
                                                                                      0.15    r               n-5
                                                                                                             10"° M Ouabain
      0.15   r

                                                                                 ST 0.10
                                                                                   E                                                                         o
  E                                                                              |   0.05
 |    0.05

                                                                                                    30              60        90            120
                           30          60          90             120                                                Time (min)
                                   Time (min)
                                                                                    Fig. 5. Effect of 10"5 M ouabain on the volume flow (Jv, solid
   Fig. 4. Volume flow is not maintained in the Cl"-free medium                  line). Each point represents mean ± SE of five experiments. After
(solid line). Each point repcesents mean ± SE of six experiments.                ouabain application Jv decreases markedly, and in 60 min it be-
Note the difference from the control experiment (dotted line). Et                comes statistically not different from zero. Note the difference from
reduces 77% in the Cl~-free medium.                                              the control experiment (dotted line). Et decreases 91%.
1780                                INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / November 1987                                     Vol. 28

Table 2. Reduction rate(%)ofJv, Et, and Rt                                                  O.b
  Experiment             (n)             Jv             Et           Rt
Sham (control)           (7)          39 ± 8       50   ±4         9± 6
Ouabain                  (5)          83 ± 8*      91   ±5f       14± 12   CVJ
Furosemide               (7)          84 ± 11$     76   ±2*       18± 4       E 0.3

                                                                             Jv (jul/min/
   Values are mean ± SE. */>< 0.002, -fP< 0.001, tP < 0.01 with respect
to "control" values.                                                                        0.2

minutes after the furosemide application, mean Jv
decreased 84%. This rate of decline was significantly                                                                  -8cm H o 0-
larger than the control experiments (P < 0.01, Table
2). Et decreased 76% at the end of the experiment.                                                30       60             90             120
                                                                                                             Time (min)
Hydraulic Conductivity
                                                                              Fig. 7. A representative experiment for Lp measurement. After
   Application of continuous 3 cm H2O hydrostatic
                                                                           the volume flow (Jv) measurement with zero pressure (only tran-
pressure on the retinal side did not significantly in-                     sient 3 cm H 2 O pressure, see Materials and Methods), 8 cm H 2 O
crease the Jv, affirming the negligible effect of the                      pressure is applied continuously on the retinal side. Note that the
transient 3 cm H2O pressure on the volume flow                             leveled Lp value is slightly higher during the 8 cm H 2 O pressure.
measurement (n = 9). However, by 8 cm H2O pres-
sure, Jv increased markedly, followed by a gradual
decrease, and in about 20 min leveled off at a value                       roid, confirms that the dog RPE-choroid transports
slightly larger than the initial value (Fig. 7). Using the                 fluid from retinal to choroidal surface (absorption),
constant post-peak values, the Lp of six tissues was                       since fluid absorption occurs without any hydrostatic
determined to be 0.0126 ± 0.0018 (SE) /il/min/cm2/                         or osmotic pressure difference across the membrane.
mm Hg in the HCO3~-free HEPES solution at 37°C.                            The amount of fluid transport, 6.36 Atl/hr/cm2 in the
The mean (±SE) Et and Rt of these tissues measured                         HCO3"-rich HEPES, is comparable to the frog RPE-
before Lp determination were 7.7 ± 0.6 mV and 329                          choroid,6'7 rabbit corneal endothelium,12 and mam-
± 15 ohm-cm2, respectively.                                                malian proximal tubules.13 Negi and Marmor re-
                                                                           ported that in the rabbit, RPE metabolic water trans-
                                Discussion                                 port accounts for 70% of the in vivo resorption rate of
                                                                           the subretinal fluid, 8.4 /ul/hr/cm2,14 which is close to
  The present study, the first direct determination of                     the present in vitro value.
the volume flow across the mammalian RPE-cho-                                 In a wide variety of epithelia, fluid transport is
                                                                           linked to electrolyte transport.15 Assuming isotonic
                 6.9 + 0.9 (mV) _                             1.7 + 0.3    volumeflowacross the dog RPE-choroid, Js/Jv = 300
                 334 + 21 (Ocm2)                              276+25       mosmol/kg, where Js is the sum of net ion fluxes in
     0.15    r                 -4                                          the open-circuit condition. Using the Jv value of ex-
                          10        M Furosemide
                                                                           periments with HCO3"-rich HEPES, 6.36 /il/hr/cm2,
                                                                           Js is calculated to be 1.91 /u,osmol/hr/cm2. In contrast
~ 0.10                                                                     to the frog RPE,6 HCO3~ is not the primary driving
o                                                                          force for Jv, since Jv is not inhibited by the elimina-
c                                                                          tion of ambient HCO3~. However, Cl" contributes
^ 0.05                                                                     substantially to the Js, since Jv is inhibited 88% in the
                                                                           Cl~-free medium. In the short-circuited dog RPE, net
                                                                           Cl~ flux is determined to be 0.67 /ueq/hr/cm2 from
             L                                                             retina to choroid,8 about one-third as large as the Js.
                                                                           Therefore, participation of the ions other than Cl~
                    30               60        90              120         and HCO3~ is also suggested.
                                      Time (min)                              Since K+ is transported from retina to choroid in
                                                                           the frog RPE-choroid,16 K+ is another likely constitu-
   Fig. 6. Effect of 10 4 M furosemide on the volume flow (Jv, solid
line). Each point represents mean ± SE of seven experiments. After
                                                                           ent. However, the net K+ flux is only a small fraction
furosemide application Jv decreases significantly. Note the differ-        of Js.6 Frambach and Misfeldt reported that in the
ence from the control experiment (dotted line). Et decreases 76%.          chicken RPE-choroid-sclera, net Na+ movement is
No. 11                                      FLUID ADSORPTION IN DOG RPE / Tsuboi                                           1781

   Fig. 8. A hypothetical scheme of the             Retina                        RPE                            Choroid
electrolyte-linked water transport                 (apical)                                                      (Basal)
across the dog retinal pigment epithe-
lium. Ouabain-sensitive Na-K-ATP-                   Et(+)                                                          Et(-)
ase is distributed in the apical cell
membrane, and produces the electro-         Water
chemical potential difference of Na+
across the membrane. Backflux of Na+
down the electro-chemical potential
difference is linked with Cl~ via a furo-                                                                             Water
semide-sensitive neutral carrier. Cl"
diffuses across the basolateral mem-
brane into the paracellular space. Dif-          Furosemide
fusional movement of Na+ is absorp-
tive because of the retina-positive Et.           Ouabain
Absorption of NaCl allows water to be
absorbed isotonically.

 from retina to choroid under the open-circuit condi-           This is in good agreement with a previous report
 tion.9 They also noted that this movement of Na + is a         where the diffusion of carboxyfluorescein (MW
 passive diffusion resulting from the potential differ-         = 376) across the isolated dog RPE-choroid was
 ence, retinal side positive, across the RPE. Therefore,        shown to occur through a paracellular pathway.22 Al-
 passive Na+ absorption may account for a part of Js,           though it is debatable whether water also moves
 even if in the dog, net Na+ flux is from choroid to            through the "tight" junctions of the "leaky" epithe-
 retina in the short-circuited condition.8 The ions con-        lium,23 there is some evidence that in the cynomolgus
 stituting Js are, of course, only concluded by flux            monkey, inward (choroid-to-retina) diffusion of car-
 studies under the open-circuit condition.                      boxyfluorescein interacts with outward fluid flow
    Ouabain is an inhibitor of Na-K ATPase located in           across the RPE. 24 Metabolic (electrolyte-linked)
the RPE apical membrane.17 In a previous study                  water transport across the dog RPE is summarized in
using dogs, it was shown that 10~5 M ouabain inhibits           Figure 8.
active transport of both Na+ and Cl".8 In the present
study, 10~5 M ouabain inhibits Jv in the HCO3~-free             Key words: volume flow, hydraulic conductivity, retinal
HEPES solution, suggesting that ouabain-induced Jv              pigment epithelium, ouabain, furosemide, chloride trans-
                                                                port, dog
reduction is independent of ambient HCO3~. Thus,
ouabain-induced Jv reduction may result from re-
duced Cl~ transport. In addition, abolition of Et by
ouabain would cause a reduced passive diffusion of               I thank Jonathan E. Pederson, MD, for his encourage-
                                                                ment and helpful suggestions.
Na+ in the retina-to-choroid direction. Ouabain-sen-
sitive Jv is also encountered in the isolated frog RPE.6                                 References
    Furosemide at 10~4 M is an inhibitor of Cl~ trans-           1. Marmor MF, Abdul-Rahim AS, and Cohen DS: The effect of
port, but not Na+ transport, in the dog RPE.9 There-                metabolic inhibitors on retinal adhesion and subretinal fluid
fore, inhibition of Jv by the 10~4 M furosemide, inde-              resorption. Invest Ophthalmol Vis Sci 19:893, 1980.
pendent of the ambient HCO3~, further supports the               2. Pederson JE and Cantrill HL: Experimental retinal detach-
coupling between water and Cl~ transport. Furose-                   ment: V. Fluid movement through the retinal hole. Arch Oph-
                                                                    thalmo! 102:136, 1984.
mide also inhibits Cl~ transport in the isolated RPE             3. Tsuboi S, Taki-Noie J, Emi K, and Manabe R: Fluid dynamics
of frog18 and chicken.9 Furthermore, in the cyno-                   in eyes with rhegmatogenous retinal detachments. Am J Oph-
molgus monkey eye with retinal detachment, retina-                  thalmol 99:673, 1985.
to-choroid RPE permeability to fluorescein is de-                4. Tsuboi S and Pederson JE: Permeability of the blood-retinal
creased by vitreous 10~4 M furosemide, suggesting                   barrier to carboxyfluorescein in eyes with rhegmatogenous ret-
                                                                    inal detachment. Invest Ophthalmol Vis Sci 28:96, 1987.
furosemide-sensitive fluid transport in vivo.19                  5. Brubaker RF and Pederson JE: Ciliochoroidal detachment.
   In the present study Lp of the dog RPE-choroid                   Surv Ophthalmol 27:281, 1983.
was determined to be 0.0126 /il/min/cm2/mm Hg.                   6. Hughes BA, Miller SS, and Machen TE: Effects of cyclic AMP
Although this vdlue may possibly be an overestimate,                on fluid absorption and ion transport across frog retinal pig-
                                                                    ment epithelium. J Gen Physiol 83:875, 1984.
because of the edge damage, it is about one-third as
                                                                 7. Frambach DA, Weiter JJ, and Adler AJ: A photogrammetric
large as the bullfrog RPE-choroid,20 placing the RPE-               method to measure fluid movement across isolated frog retinal
choroid in the category of a "leaky" epithelium.21                  pigment epithelium. Biophys J 47:547, 1985.
1782                         INVESTIGATIVE OPHTHALMOLOGY G VISUAL SCIENCE / November 1987                                          Vol. 28

 8. Tsuboi S, Manabe R, and Iizuka S: Aspects of electrolyte trans-    16. Miller SS and Steinberg RH: Active transport of ions across
    port across isolated dog retinal pigment epithelium. Am J               frog retinal pigment epithelium. Exp Eye Res 25:235, 1977.
    Physiol 250 (Renal Fluid Electrolyte Physiol 19):F781, 1986.       17. Ostwald TJ and Steinberg RH: Localization of frog retinal
 9. Frambach DA and Misfeldt DS: Furosemide-sensitive Cl                    pigment epithelium Na+-K+ ATPase. Exp Eye Res 31:351,
    transport in embryonic chicken retinal pigment epithelium.              1980.
    Am J Physiol 244 (Renal Fluid Electrolyte Physiol 13):F679,        18. DiMattio JK, Degnan J, and Zadunaisky JA: A model for
     1983.                                                                 transepithelial ion transport across the isolated retinal pigment
10. Tsuboi S, Fujimoto T, Uchihori Y, Emi K, Iizuka S, Kishida             epithelium of the frog. Exp Eye Res 37:409, 1983.
    K, and Manabe R: Measurement of retinal permeability to            19. Tsuboi S and Pederson JE: Experimental retinal detachment:
    sodium fluorescein in vitro. Invest Ophthalmol Vis Sci                  XI. Furosemide-inhibitable fluid absorption across the retinal
    25:1146, 1984.                                                         pigment epithelium in vivo. Arch Ophthalmol 104:602, 1986.
11. Reid EW: Transport of fluid by certain epithelia. J Physiol       20. Brown JA and Zadunaisky JA: Ion coupled fluid movements
    26:436, 1900-1901.                                                     across the bullfrog retinal pigment epithelium. ARVO Ab-
12. Fischbarg J, Lim JJ, and Bourguet J: Adenosine stimulation of          stracts. Invest Ophthalmol Vis Sci 22(Suppl):102, 1982.
    fluid transport across rabbit corneal endothelium. J Membr        21. Fromter E and Diamond JM: Route of passive ion permeation
    Biol 35:95, 1977.                                                      in epithelia. Nature (London) 235:9, 1972.
13. Schafer JA, Troutman SL, and Andreoli TE: Volume reab-            22. Tsuboi S and Pederson JE: Permeability of the isolated dog
    sorption, transepithelial potential differences, and ionic per-        retinal pigment epithelium to carboxyfluorescein. Invest Oph-
    meability properties in mammalian proximal straight tubules.           thalmol Vis Sci 27:1767, 1986.
    JGen Physiol 64:582, 1974.                                        23. Levitt DG: Routes of membrane water transport. Comparative
14. Negi A and Marmor MF: Quantitative estimation of metabolic             Physiology. In Water Transport Across Epithelia, Ussing HH,
    transport of subretinal fluid. Invest Ophthalmol Vis Sci               Bindslev N, Lassen NA, and Sen-Knudsen O, editors. Copen-
    27:1564, 1986.                                                         hagen, Munksgaard, 1981, pp. 248-257.
15. Diamond JM and Bossert WH: Standing-gradient osmotic              24. Tsuboi S and Pederson JE: Acetazolamide effect on the inward
    flow: A mechanism for coupling of water and solute transport           permeability of the blood-retinal barrier to carboxyfluorescein.
    in epithelia. J Gen Physiol 50:2061, 1967.                             Invest Ophthalmol Vis Sci 28:92, 1987.

To top