Exp. Eye Res. (1980) 30, 53-58
Glucose Transport in Isolated Mammalian Pigment Epithelium G. J. PASCUZZO, J. E. JOHNSOK*
Department of Physiology and Biophysics, Department of Radiology and Radiation Biology, College of Veterinary Medicine and Biomedical Sciences, Colorndo &ate University, Port Collins, Colmdo 80523; U.X.A
(Received 20 October 1978 ad
in revised form 12 Jme
1979, Neuj York)
The isolated retinal pigment epithelium-choroid of sheep was studied in vitro as a single membrane mounted in a Ussing chamber. The average potential of 4-O tissue preparations was 6.3 mV. In all preparations the retina (apical) side was positive with respect to the choroid (basal) side. Changes in pl’a+, K’, or Ca?+ concentration on the choroid side did not affect the tissue potential, whereas a ten-fold increase in K+ concentratjion or Ca,2+ chelation by EGTA on the retina side decreased the potential. The potential was also decreased by 0.75 mlvr-phloridzin and 1.0 mlvr-dinitrophenol. Utilizing a glucose concentration gradient of 4 mlvr-choroid to 1 mlcl-retina side, the unidirectional glucose fluxes were found to be . hr retina to choroid. Both uni883 nmol/cm3 . hr choroid to retina and 222 nmol/cm” directional fluxes were decreased by 65% Na+ replacement with choline Cl. The ret’ina to choroid flux was also decreased by 0.75 m&I-phloridzin and I.0 mnr-dinitrophenol. The sodium dependence and inhibitor effects suggest that mechanisms of glucose transport other than simple diffusion are operative in sheep retinal pigment epithelium. Key lords: pigment epithelium; glucose; transport.
1. Introduction The electrical and transport-related properties of isolated amphibian retina’1pigment epithelium (RPE) h ave been extensively studied (Lasansky and DeFisch, 1966; Steinberg and Miller, 1973; Zadunaisky and Degnan, 1976; Miller and Steinberg, 1976, 1977a, 1~).Only recently has work on isolated mammalian RPE been reported, Steinberg, Miller and Stern (1978) have found the electrical responsesof cat RPE to be similar to those of frog RPE with the exceptions that the cat basal membrane seemsinsensitive to extracellular ionic changes and that the cat apical membrane has a larger Na+ and lower HCO, conductance. The present studieswere undertaken to characterize someRPE properties of a’nother mammal, i.e. sheep,and to investigate the movement of D-glLLCOSe through the isolated tissue. In addition, the effects of know inhibitors of metabolism and transport on RPE glucose translocation were determined. Initial experiments with equal glucose concentrations across the tissue resulted in fluxes too small to permit adequate statistical analysis of treatments. In order to have fluxes large enough for measurable differences and to approximate in situ conditions, a glucose concentration gradient of 4 m;Mchoroid (basal) side to 1 m&f retina (apical) side was used. Although absolute values of glucoseconcentration may, of course, vary in vivo, this gradient compared well with the data of Matschinsky, Passonneauy and Lowry (1968) which indicate a change in glucose concentration of approximately 3 111111 acrossthe RPE of rabbit. * Department of Radiology and Radiation I- Reprint requests to: Dr E. L. Pautler. OOi4-4835/SO/OlOO~3+06
Q 1980 Academic 53
Press Inc. (London)
G. J. PASCUZZO,
J. E. JOHNSOK
E. L. PAUTLER
Eyes were obtained immediately after the death of sheep at a local abattoir. The sclera was removed, eyes placed in oxygenated Ringer’s solution and transported to the laboratory (20-30 min travelling time). A l-cm2 non-tapetal section was cut from the eye, the vitreous and retina gently peeled away, the isolated RPE-choroid floated onto a [email protected]
gasket and then mounted as a membrane between halves of a Ussing chamber. B’orty ml bathing sohrtion were added to each side. By means of a [email protected]
pump and/or syringes, the solution on separate sides of the tissue could be changed completely within 3 min. Appropriate control recordings were made to determine the influence of solution pumping upon the tissue. In InM concentrations, the bath ionic composition was 1.08 Na,HPG,, 0.61 KHsPO,, 18.5 NaHCO,, 120 NaCl, 3.62 KCl, 0.69 MgSO, and 1.1’7 CaCl,. The fluid was circulated by a bubble lift using 95% O,-5% CO,, which maintained the pH at 7.4. Bathing solution temperature was maintained at 32-34°C by means of a heated reservoir. The tissue area exposed to solution was 0.20 cmz. Voltage-sensing electrodes were placed 1 mm from each side of the tissue via agarized cotton wick salt bridges. A Grass P16 micro electrode d.c. amplifier connected to either a Philips PM3200 oscilloscope or a Gould Brush 220 chart recorder provided transepithelial potential (TEP) measurements. Current-pulsing electrodes were placed 2.5 cm from and in line with the exposed sides of tissue via similar salt bridges. A current to voltage inverting amplifier connected to the chart recorder provided short-circuit current (SCC) measurements. Electrical measurements were adjusted for electrode tip potentials and solution resistance. The criterion for judging a tissue preparation as “viable” was a TEP of at least 3 mV after the SCC had stabilized. Although certain preparations were studied for 8-10 hr in the chamber, the viability of most tissues was limited to 5-6 hr. Data reported here was obtained only within the first 4 hr of the preparations. The ionic dependence of the TEP was studied by replacing the bathing solution on one side of the tissue with a bathing solution altered in one of the following fashions: 65% Na+ substitution with either choline Cl or LiCl; a IO-fold increase in Kf concentration balanced by corresponding Na+ decrease; chelation of C2+ by 15 m&r-EGTA. Each of the altered solutions was bubled with 95% O,-5% CO, and adjusted to pH 7.4 before use. Flux experiments utilizing the 4 m&r choroid to 1 1nM retina side glucose concentration gradient consisted of a continuous TEP recording with periodic passage of de current through the tissue for SCC and resistance measurements. When the SCC had stabilized, 8-10 @IX of [14C]D-glucose was added to one side. Twenty 0.225 ml samples were taken from each side in simultaneous pairs over a 4-hr period. Sample d/min were determined by a Beckman LX9000 counter. Certain samples were analyzed via paper chromatography to determine the content of [14C]D-glucose; D-glucose accounted for greater than 95% of the total sample activity on the appearance side. Unidirectional fluxes from choroid to retina side (C&R) and retina to choroid side (R+C) were determined on separate tissues. Two methods were used to calculate the unidirectional fluxes. (1) In a manner similar to that described by Zadunaisky and Degnan (1976) and Miller and Steinberg (1976), the slope of the linear (constant flux) portion of the [‘“Clglucose appearance curve was taken. The linear portion usually occurred from 2-4 hr after the addition of the labeled glucose. (2) Using a standard technique of compartmental analysis (Shipley and Clark, 1972), a function of the form Sa(t) = SA(t--co [l -exp( --kt)] was fit to the specific activity (SA) appearance data. The derivative of this function with respect to time evaluated at t = 3 hr provided the unidirectional flux measurement. Methods 1 and 2 yielded comparable flux magnitudes; however, method 2 tended to yield more decisive statistical comparisons of treatment groups. For example, the effect
of phloridzin on the CL-R flux was rather tentative (P-0.05) using method 1, hut the effect was found to be insignificant (P > 0.2) using method 2. The inhibitors used in the glucose flux experiments were O-75 rnsf phloridzin, a known inhibitor of glucose transport, and I.0 mM dinitrophenol (DSP), an “uncoupler” of oxidative phosphorylation. Sodium dependence of glucose movement was studied using the 65% Naf replacement with choline Cl altered bathing solution. Solutions were pre-
bubbled and pH adjusted beforeuse.Flux experimentswereperformed by adding solution with a singleinhibitor or choline Cl to both sidesof the tissue, measuringthe SCCuntil stable, then proceedingas describedabove. 3. Results
In all viable RPE preparations the retina side was positive with respect to the choroid side. When the chamber was completely filled with solution and the tissue SCC stabihzed; continued pumping of the bathing solution had no effect upon the TEP recordings. The average electrical parameters of sheep RPE are shown in Table I. These values are comparable to those reported in other species,although resistancesabove 320 emawere not seeneven for TEP’s as high as 24 mV. 5mV
L IO min
FIG. 1. Transepithelial potential responses to RPE. The agents were tenfold increase in Kf 65% Ma+ replacement using choline Cl or LiCl. changes in TEP for as long as 15 min after t.he the retina side were not consistent. The EGTA biphasic response as shown.
various agents when added to one side of isolated sheep concentration. Ca2+ chelation with 1.5 mix-EGTA, and Kate that on the choroid side there were no significant addition of an agent. The choline Cl and LiCl effects on always decreased the TEP but occasionally elicited a
As illustrated in Fig. 1; none of the altered ionic solutions produced any sizeable changesin the TEP when they were placed on the choroid side. A tenfold increase in K+ concentration on the retina side did result in an abrupt decreasein TEP. These observations were also reported for cat RPE (Steinberg, Miller and Stern, 1978). The addition of EGTA on the retina sidealways causeda reduction of TEP but occasionally an initial transient increase was noted. The TEP was not affected by 65% Na+ substitution with choline Cl on either side during the period of observations. Using LiCl, 65% Na+ substitution on the retina side produced somedecreasesin TEP, but, these decreasesdid not appear consistently. Plux
In terms of electrical parameters there were no differences in mean initial conditions among control or treatment groups. In particular, there were no statistical differences among the tissue resistanceswithin the groups. The unidirectional glucoseflux data
G. J. PASCUZZO.
J. E. JOHXSOS
E. L. PAUTLER
is summarized in Table II. The R + C flux was significantly reduced by phloridzin, DNP and Ka+ substitution. However; the C + R flux was significantly reduced by only the Naf substitution. The variability of data makes it difficult to evaluate any physiologically significant effects on net fluxes. Addition of either phloridzin or DXP decreased the TEP immediately, as shown in Fig. 2. The TEP again appeared unaffected by 65% Y L a+ replacement with choline Cl. TABLE
of sheep RPE
COIltXOl C-tR R+C l’hloridzin C-R R-tC DKP C+R R-tC Choline Cl C-R RX
:ci.2*5.1 [email protected]
x83+ 110 635*150 (r~O.2) 756&114 (P>O.4)
Resistance ( cw
A [glu] 0.75 m&r-phloridzin A [glu] I.0 miwD?\iP A [glu] 65% Xa+ replacement with choline Cl
n = 5 for each unidirectional A [glu] indicates 4 mv-glucose
(n = 40)
SU& 9(P<0.005) O-018)
group except for CAR control where 71 = 6. on choroid (C) side and 1 m&l on retina (R) side.
FIG. 2. Transepithelial potential responses to various agents when added to both sides of isolated sheep RPE. The agents were 0.75 mlir-phloridzin, I.0 mnf-DNP. and G5% Na+ replacement with choline Cl. Glucose flux experiments were performed in the presence of each of these agents.
In general, the electrical properties of isoiated sheep RPE reported here are conparable to those of other species; though somewhat smaller in magnitudes. The apical membrane is sensitive to changes in Kf concentration. However, the sheep apical membrane does not respond uniformly to 65% Na+ substitution with either choline The effective reduction of calcium Cl or LiCl, yet it is sensitive to Ca”+ chelation. concentration may alter the TEP by reducing the net calcium current flow and/or by a regulatory action on other ionic permeabilities. The occasional biphasic response observed after application of EGTA suggests a complex mode of action. on t,he The TEP was not affected by altering Na, K+, or Ca2+ concentrations chorcjd side. Either the RPE basal membrane is illsensitive to these ionic changes or else there exists a significant choroidal diffusion barrier to small ions. The barrier would have to be sufficiently restrictive to prevent the non.-tapetal RPE for responcling for up to 15 min after the cha,nge in bathing solution.
The ratio of unidirectional glucose fluxes in the control group is very close to the ratio of the glucose concentrations across the tissue. According to the ITssing flux ratio test (Kotyk and Janacek, 1975) this implies that the observed fluxes are consistent with; though not limited to, a simple diffusion model. In order to completely verify this conclusion, a variety of different glucose gradients would have to be employed. However, the effects of inhibitors suggest that mechanisms other than simple diffusion are operative in glucose movement across sheep RPE. Phloridzin, an established inhibitor of active glucose t#ra,nsport, affected only the R + C flux. However, at a 0.75 MM concentration, phloridzin could also affect facilitated diffusion systems (Stein, 1967) so that the phloridzin effect on the R + C flux i-s not necessaril: prcof of active glucose transport. The metabolic inhibitor DNP also affected only the R + C flux. Fluxes in bot,h directions were decreased by 65% Na+ replacement. There&e, glucose movement across sheep RPE may include mechanisms that are carrier-mediated, directly or indirectly dependent upon metabolic energy, and are sodium dependent. ACHNOWLEDGNESTS
This research was supported
A. and Janacek, K. (1975). Cell Nentbmne Tranqort. Plenum Press, Xew York and London. Lasansky, A. and DeFisch, F. (1966). Potential, current, and ionic fluxes across the isolated retinal pigment epithelium and choroid. J. Gen. PiLysiol. 49, 913-24. Matschinsky, F. N., Passonneau, J. V. and Lowry, 0. H. (1968). Quantitative histochemical analysis of glycolytic intermediates and cofactors with an oil well technique. J. Histochem. Cytochem. 16, 29-39. Miller, S. S. and Steinberg, R. H. (1976). Transport of taurine, L-methionine and 3.O-methyl-nglucose across frog retinal pigment epithelium. Exp. Eye Res. 23, 177-89. Miller, S. S. and Steinberg, R. H. (1977a). Active transport of ions across frog retinal pigment epithelium. Exp. Eye Res. 25, 23548.
G. J. PASCUZZO,
J. E. JOHNSON
E. L. PAUTLER
S. S. and Steinberg, R. H. (1977b). Passive ionic properties of frog retinal pigment epithelium. J. Membrane Biol. 36,337-72. Shipley, R. A. and Clark, R. E. (1972). Tracer Methods for in vivo Kinetics. Academic Press, New York and London. Stein, W. D. (1967). The Movement of Molecules Across Cell ikfembranes. Academic Press, New York and London. Steinberg, R. H. and Miller, S. S. (1973). Aspects of electrolyte transport in frog pigment epithelium. Exp. Eye Res. 16,365-72. Steinberg, R. H., Miller, S. 8. and Stern, W. H. (1978). Initial observation on the isolated retinal pigment epithelium-choroid of the cat. Invest. Ophthalmol. Visual Xci. 17, 675-8. Zadunaisky, J. A. and Degnan, K. J. (1976). Passage of sugars and urea across the isolated retina pigment epithelium of the frog. Exp. Eye Res. 23, 191-B.