Original Paper
Ophthalmic Res 1992;24:265-273
T.C. Chua Charles H. Keith b Edward Yeea Keith Green a-c
Intracellular pH of Tissue-Cultured Bovine Corneal Endothelial Cells
Departments of Ophthalmology and Physiology & Endocrinology, Medical College of Georgia. Augusta. Ga.: Department of Zoology, University of Georgia, Athens. Ga.. USA
Key Words
Bovine corneal endothelium Tissue culture Intracellular pH Amiloride Ammonium chloride Video imaging
Intracellular pH (pH;) of bovine tissue-cultured corneal endo thelial cells has been measured under several experimental conditions. Determinations were made on individual cells using video-imaging techniques that allowed assessment of 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfiuorescein fluo rescence at 440 and 490 nm. Each experiment had a calibra tion performed on a cell monolayer: this was performed using a high K+-nigericin solution. Resting pH; was 7.25 ± 0.03 (n = 18) in bicarbonate solution at pH 7.4. Amiloride (1 mM) caused an acidification of approximately 0.2 U within 2 min: replacement with normal Ringer allowed a return to normal pH; after an alkali overshoot. Exposure to 20 m.\4 NH4C1 caused alkalinization that became acidic upon washout of NH4CI. In Na+-rich solution pH; returned to normal after acidification but pH; remained low in Na+-free solution until substituted by Na+-rich solution. Removal of HCO3- from the bathing solution caused a nonsignificant acidification of pH; by 0.1 U at 2 and 4 min, and 4,4'-diisothiocyanostilbene-2,2'disulfonic acid (DIDS: 1 mM) acidified pH; by 0.14 U at 2 min and 0.24 U at 4 min. Addition of DIDS (1 mM) in a HCC>3-free solution had no effect on pH;. Hydrogen peroxide acidified pH; by 0.3 U at 50 pM and 1 mM. These results indi cate that a Na+:H+ antiport exists that regulates pH; even at normal ambient pH in the presence of bicarbonate: this pro cess becomes highly activated after an acid load. There is a DIDS-sensitive HCO3 movement that is probably coupled to Na+ or Cl".
Received: June 28.1991 Accepted: January 17. 1992
Keith Green. PhD, DSc Department of Ophthalmology Medical College of Georgia Augusta. GA 30912-3400 (USA)
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Abstract
Initial measurements of intracellular pH (pHj) of corneal endothelial cells [1-3] used the intracellular to extracellular distribution ratio of l4C-labeled dimethyl oxazolidinedione (DMO) [4], Alternative methods now utilize fluorescent dyes trapped within cells [5], One such dye is 2',7'-bis(2-carboxyethyl)5(6)-carboxyfluorescein acetoxymethvlester (BCECF-AM) which freely enters cells. The AM group is cleaved trapping the fluorescent, pH-sensitive and relatively impermeant BCECF inside the cell. Studies of pHj have been made in several ocular tissues including cultured bovine reti nal pigment epithelium [6], cultured bovine pigmented ciliary epithelium [7] and cultured bovine corneal epithelial cells [8, 9] all using 5 (and 6)-carboxy-4',5'-dimethylfluorescein. Fresh rabbit corneal epithelium [ 10], frog reti nal pigment epithelium [11], toad lens epithe lium [12-14] and rabbit ciliary epithelium [15, 16] have all been studied using BCECF, while lens epithelium [ 17] and lens fiber and epithelial cell vesicles have been examined using other probes [18], The pHj of cultured bovine corneal endothelium with 5 (and 6)carboxydimethylfluorescein has been de scribed [19] as well as changes occurring in pHj as a function of external perturbations. Knowledge of the corneal endothelial cell pHi is important for the placement of the var ious transport systems, including bicarbon ate, at either the apical or basolateral surface of the cells. The net bicarbonate transport across the endothelium from the stroma to the aqueous-facing surface [20, 21] requires one of two systems to be operative. Bicarbonate entry occurs either actively across the stromal (basolateral) surface with cellular accumula tion and then subsequent diffusion to the aqueous, or passively into the cell and endoge nous production with active extrusion across
266
the apical (aqueous-facing) surface. pHj would be more alkaline in the first case due to bicarbonate accumulation and more acidic in the second case. The pHj of corneal endothelial cells has been reported as 7.1 in an ambient solution at pH 7.5 (A pH 0.4 U) using a tracer method [ 1-3], 7.15 in a bathing solution at pH 7.4 (A pH 0.25 U) [19] and 7.35 in a solution at 7.5 (A pH 0.15U) [22] using dye techniques. These differences require resolution. Na+:H+ exchange has been found to exist in almost all cell systems [6-19, 23], and we examined the endothelium for the presence of this exchange in cultured bovine corneal endothelial cells. The effects of a known oxidant, hydrogen per oxide, were examined on this tissue.
Materials and Methods Bovine eyes were routinely obtained from a local slaughterhouse and were stored on ice for not more than 4 h. They were washed with sterile saline to remove blood, and then the corneal epithelium and surrounding pigment were completely scraped off us ing a sterile scalpel blade. The globe was dipped in 70% ethanol for 10 s. The cornea was dissected out with aseptic instruments and placed, endothelial side up, in a sterile petri dish containing sterile normal Ringer’s solution: 110 mM NaCl, 4.5 mM K.C1. 1.0 mM NaH2PC>4, 1.2 mM MgCI2, 5.5 mM glucose. 1.8 mM CaCI2 and 25 mM NaHCO,. pH 7.4. The sterile dishes were taken under a laminar flow hood, and the corneas were washed through 4 dishes of culture me dium containing Dulbecco’s modified Eagle medium/ Ham’s FI2 mixture (1:1). 15 mM NaHCCb. 1 mM ascorbic acid, 100 U/ml penicillin. 100 gg/ml streptomycin, 5 gg/ml gentamicin and 2.5 gg/ml am photericin B. To each corneal cup was added 0.5 ml of 0.05% trypsin in 0.02% EGTA solution, and each was incubated at 37 °C in a humidifed incubator with 5% C 0 2 for 1 h. At the end of the incubation, the solution was aspirated from the center of each corneal cup. The endothelium was pelleted and washed several times with culture medium. The suspension was plated out with culture medium containing 20% fetal bovine serum and incubated at 37 °C in a humidified incuba tor with 5% C 0 2. The cells had ambient solution
Chu/Keith/Y ee/Green
Corneal Endothelial Intracellular pH
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Introduction
NaCI, that also contained 50 pM nigericin (made from a stock solution of 10 m M in dimethylformamide: 95% ethanol. 3:1. v/v). This solution allows equilibra tion of intracellular and extracellular pH, and a steadystate fluorescence ratio was determined after allowing 3-5 min for equilibration with a new external pH. Nigericin was present in each bathing solution used in the calibration series. Background fluorescence at the two wavelengths was obtained from cell-free areas in the chambers. Solution changes were achieved by rapid exchange through the use of pipettes: exchanges were usually complete (two rinses with new solution) within 3045 s. The basic Ringer solution was bicarbonatc-based [1-3, 14]. Additional normal Ringer solution was bi carbonate-rich and Hepcs-buffered (10 mM). NH4CI, low N a \ HCOj-free solutions were prepared by equi molar substitutions of NaCI by NH4C1, NaCI by cho line chloride, and HCO3 by CP. respectively. Amiloride (1 mM). 4.4'-diisothiocyano-stilbcne-2,2'-disulfonicacid(DIDS: 1 mM), hydrogen peroxide and oua bain (1 mM) were all obtained from Sigma (St. Louis. Mo., USA). The pH of normal Ringer was adjusted to the desired value using 1 N NaOH or HC1, and this solution, as well as all others, were used within 1 min of pH adjustment in experiments that lasted no longer than 6 or 8 min in the presence of one solution. The ambient pH in any experiment, therefore, was con stant.
Results
The fluorescence image showed consistent labeling between cells with a differentiation between cytoplasm and nucleus. Background fluorescence, after dye labeling and subse quent rinsing, was usually near the zero level. Calibration of pH, was performed routinely after perfusion with K.“-Ringer containing ni gericin. External pH was varied between 6.8 and 8.0. Plots of the steady-state ratio, deter mined 3-5 min after each pH change, indi cate that the 440:490 ratio was high in more acidic solutions with a decreased ratio as am bient pH became more alkaline (fig. 1). Above pH 7.7, the change in fluorescence ratio per unit pH became reduced and the slope mark edly attenuated.
267
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
changed 3 times each week. Amphotericin B was no longer added to the culture medium containing 10% fetal bovine serum when cells reached conflucncy. For experiments, confluent cultures were trypsinized for 5 min at 37 °C. The cell suspension was centri fuged at 1,000 rpm for 5 min and washed twice with fresh culture medium. Cell count and viability tests were performed using a hemacytometer and by moni toring trypan blue exclusion, respectively. The cells were plated on plain coverslips in Bionique chambers (BIO-RAD. Cambridge. Mass., USA; approximately 1 million cells per coverslip) and allowed to settle for 48 h before use. BCECF-AM (Molecular Probes. Eugene. Oreg.. USA) w'as made in a stock solution of 10 mA/in dimethylsulfoxide and used as a 5 pMsolution in Ringer for loading cells with dye. After 45 min the cells were rinsed several times with normal Ringer prior to place ment on the stage of a Zeiss IM fluorescence micro scope that was maintained at 37 °C with an air curtain (Nicholson Precision Instruments. Gaithersberg, Md. USA). Cells were generally imaged with a X 63 objec tive. Epifluorescence excitation was attenuated with a 1.0- or 1,5-ND filter, passed through a 440- or 490-nm interference filter and reflected off a 505-nm dichroic mirror. Emitted signals were passed through the dich roic and a 520- to 560-nm bandpass filter before being imaged. The fluorescein emission upon a 440-nm exci tation is independent of pH. whereas the 490-nm exci tation is pH dependent [24]. Fluorescence was re corded using a DAGE SIT camera (DAGE-MTI. Mi chigan City, Ind., USA) [25,26], Images were digitized directly onto an Imaging Technology framestore (ImagingTechnology, Inc., Woburn, Mass, USA) in an image processor obtained from G.W. Harnaway and Associates (Boulder, Colo., USA). All images were cor rected for shading by dividing by a lowpass-filtered background image and were corrected for geometric distortion [25], Pairs of images were mathematically correlated and intensified, above local background, and were integrated blockwise using circles of a 5-pixel diameter. Several data points were obtained from the cytoplasmic regions of cells in the visual field and an average of 5 ratios (440:490 fluorescence) were ob tained from each corrected frame. Frequently, 6-9 cells were included in the measurement field. Cells were not exposed to light except during the brief peri ods of data collection, about 5 s per wavelength. Usually, only a short series of experiments was per formed with every monolayer, and each experiment was followed by a 5-point calibration using the nigericin-high K technique [27], The cells were washed with a K“-Ringer. in which KC1 had been substituted for
1mM Amiloride
Ringer
Fig. 1. Plot of calibration of intracellular pH as a function of signal ratio. Values are the mean ± SEM of at least 6 monolayer determinations at each point. SEM is maximally the size of the data point.
Fig. 2. Changes in pH, after amiloride. Values are the mean ± SEM of 4 monolayers. Where no SEM is shown, it is too small to be indicated.
A resting pH| of 7.25 ± 0.03 (mean ± SEM, n = 18 monolayers, each with 5-9 cells that were sampled, thus reflecting between 90 and 144 cells in total) was obtained. Each monolayer was initially bathed with Ringer at 7.4 and a base-line pHj value determined. There was no difference in resting pH; whether the cells were exposed to a HCO> rich or a HCOÿ-rich plus Hepes-buffered solu tion. Amiloride (1 mM) caused an acidifica tion of between 0.16 and 0.26 pH U at 2 min after drug addition. This value was main tained during an additional 2 min of amilo ride. Within 2 min after exposure to normal Ringer, pH; returned to near baseline values after a small overshoot. At 6 min after the return to Ringer solution, pHj was normal (fig. 2).
Changing the bathing solution from nor mal Ringer to a 20 mM NHgCl-containing solution caused an abrupt alkalinization of the cell that underwent a slight acidification over the next 4 min. Replacement of the am bient solution with normal or Na+-free Ringer caused a marked and immediate acidification of pHj within 2 min; a further 2 min allowed recovery towards normal pHj at a rate of about 0.2 pH U/min (fig. 3). Replacement of the NHgCl solution with a Na+-frce solution also caused acidification that was sustained. Replacement of the Na+-free solution with a Na+-containing solution initiated a rapid re turn towards normal pHj at the same rate as noted after immediate replacement with Na+rich solution, i.e.. 0.2 pH U/min or about half that noted in similar experiments previously in a bicarbonate-rich solution [ 19] (fig. 3). Re-
268
Chu/Keith/Yee/Green
Corneal Endothelial Intracellular pH
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Time in Minutes
pH,
20 mM NHaCI
covery of pHj after removal of NH4CI is, therefore, a sodium-dependent process. The different stages of this process are illustrated in figure 4 with photographs of the 490/440 images processed as described in Materials and Methods. Bathing in HCOTfree solution caused a nonsignificant acidification of pHj by 0 . 1 U at 2 and at 4 min (n = 5 experiments: 26 cells) as seen earlier [19], and no further change oc curred after DIDS (1 mM) was added. The addition of DIDS alone acidified pHj by 0.14 U at 2 min and 0.24 U after 4 min. Reex posure to NaCl Ringer caused a return of pHj to normal values. Ouabain (1 mM) caused a very small acidification of pHj ( < 0 .l U), and no change of pHj was induced by BaCB (2 mM) or high K+ Ringer (50 mM K+) within 10 min of exposure to these agents.
Na+ Ringer ( • )
Time in Minutes
Physiological concentrations of H2O2 (50 pM) caused an acidification of pHj by 0.3 U within 2 min, and the pHj recovered to resting state at 5 min. A 20-fold increase in concen tration (1 mM) also induced a decrease of pHj by 0.3 U within 3 min. Recovery from this acidification was also observed, and the pHj remained stable for at least 10 min after re covery.
Discussion
At steady-state in an ambient pH of 7.4, the pHj of bovine tissue-cultured corneal en dothelial cells is 7.25 ± 0.03 (n = 18; A pH 0.1 5 U). This value is closer to ambient solu tion pH than that of fresh rabbit endothelial cells [ 1-3] under identical bathing conditions.
269
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Fig. 3. Changes in pH; induced by exposure to a prepulse of NH4CI with subsequent perfusion with either Na+-rich or Na+-free solu tions. Note that cell acidification occurs after removal of NH4CI re gardless o f Na+-rich or Na+-free so lution. In Na+-rich solution, recov ery towards normal pH, occurs in stantly (•), but in Na*-free solution, cell remains acid until exposed to Na+-rich solution (o). Values are the mean ± SEM of at least 5 monolayer determinations at each point. SEM is too small to be indi cated at 0. 2, 4, 12 and 14 min. The values over the first 6 min were the same for both series of experiments since the solutions were identical until 4 min. The values here are the mean ± SEM of at least 10 deter minations. The 6-min value was the same in Na+-Ringer and Na+free Ringer (o). Na+-free Ringer was present from 4 to 10 min be fore replacement with Na* Ringer.
The pHj is, however, closer to values found for fresh rabbit corneal epithelial basal cells [ 10] and cultured bovine corneal epithelial cells [6-9], Tissue-cultured bovine corneal en dothelial cells in bicarbonate Ringer showed a pH; o f 7.15 at an external pH of 7.4 (A pH 0.25 U) [ 19] or a A pH of 0.15 in pH 7.5 [22], values very close to those found in the present studies. The difference between cultured and fresh endothelial cells may reflect the species difference and the fact that there may be a decrease in the ability of cultured cells to per form quantitatively the same as fresh cells. Recent results [22] comparing fresh and cul-
270
tured bovine endothelial cells also showed that fresh cells were more acidic relative to the ambient bathing solution. The experiments with prepulses of NH4CI were made to acid-load the cells. The influx of NH 3 increases pH|, and a slow acidification then occurs as NH4+ enters the cells down its electrochemical gradient. When the cell is then bathed in NH4Cl-free solution, either with or without Na+, the NH 3 is quickly lost and pH, falls far below the resting pH;. A recovery phase follows due to the pHj regula tory systems as the cell transports acid equiva lents. The data clearly indicate that these pro-
Chu/Keith/Yee/Grecn
Corneal Endothelial Intracellular pH
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Fig. 4. Series of photographs of cells that were used in the construction of figure 3. a At zero time, normal cells, b At 2 min, slightly darker cytoplasm, c At 4 min. little detectable change in the ratio, d At 6 min. after replacement of the NH4CI solution. The cells have become much lighter, reflecting the acidification caused by replacement of the bathing solu tion with (in this case) Na^-rich solution.
passive distribution of H+ by the Nernst po tential. taken here as -4 0 mV [Wigham CG. Green K, Hodson S. 1991, unpubl. data]. Changes in the external pH evoked altera tions in pHj. although the latter were rela tively small compared to the ambient pertur bations. We found that pHj increased by 0.2 U in face of an ambient change from pH 6.8 to 8.0. These data differ slightly from these found earlier [ 1, 2] where the A pH between pHj and the ambient solution re mained almost constant at 0.4 pH U as am bient pH increased from 7.0 to 8.0. The effect of ouabain on pHj is predictable based upon the known inhibition of the Na+:K~ pump by this inhibitor. Use of oua bain causes the intracellular Na+ to increase, thus decreasing the Na+ gradient into the cell. This, in turn, leads to a decreased sodium entry via the Na+:H+ exchanger and a second ary decrease in pHj to a more acidic environ ment. That this decrease is not related to the change in membrane voltage per sc is indi cated by the absence of an effect of either a 50 mM K+ Ringer solution, that normally depo larizes the membrane potential by 20-15 mV within 20-30 s [Wigham CG, Green K. Hodson S, 1991. unpubl. data], or BaCb. a K+channcl blocker, on pHj. There appears to be little effect of membrane potential, at least depolarization, on pHj although a depen dence of intracellular electrical potential dif ference on pHj has been strongly suggested previously [32]. In these experiments, acidifi cation of the cellular interior usually leads to a hyperpolarization, and alkalinization was as sociated with depolarization. These changes were usually transient in nature and lasted, under the given experimental conditions, less than 10 min [32], These data are not incom patible since pHj could drive membrane po tential through alterations of membrane charges, whereas alterations in membrane po tential induced by a different mechanism
271
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
cesses occur in endothelial cells and that re covery is blocked by Na+ removal from the bathing solution in the presence of bicarbon ate (fig. 3). supporting prior studies [19], The cellular acidification after amiloride under normal bathing conditions (fig. 2) is also strongly suggestive of the presence of a Na+:H+ exchanger that normally regulates pH| [19, 28]. Previous studies, however, noted lit tle or no effect of amiloride under resting con ditions [ 19], Similarly, experiments using ace tate (data not included) to induce cellular acidification also caused an overshoot alkalinization of pHj when the ambient solution was returned to normal Ringer. These data confirm those obtained in other studies using similar techniques [ 19], That pH| is lower than that of the ambient solution supports the notion that bicarbonate transport across this membrane [20. 21] is directed out of the cell. Inwardly directed active movement of bicarbonate across the basolateral membrane (with subsequent dif fusion out of the cell across the opposite mem brane) should lead to alkalinization of the cell, whereas outwardly directed bicarbonate transport should lead to acidification. Given the vectorial orientation of bicarbonate trans port from the stromal to the aqueous-facing side of the endothelium, the bicarbonate transport systems may be rightfully placed, in an outwardly facing direction, on the apical membrane of the cell [20, 21, 29-32], Other studies [19, 32] confirm that only a slight, nonsignificant decrease in pHj occurs after bicarbonate removal from the ambient solu tion. These data, plus those from the use of DIDS in the presence and absence of bicar bonate. suggest that a net uptake of bicarbon ate occurs that is DIDS inhibitable. The data also support the notion [19] that the Na+/ HCOî symport is oriented in a net inward direction. Furthermore, the pHj is more alka line than predicted if it was based upon the
would not necessarily induce a change in pHj. Preliminary' data [19], however, indicated that the use of a high K+ solution caused cellu lar alkalinization. It has been shown that an oxidant (H2O2) normally present in aqueous humor at 30-80 \iM [33, 34] induced a modest fall in pHj: of interest is that a high concentration of 1 mM also induced a small, reversible fall in pHj. The transient acidification of pHj by peroxide is possibly either directly affecting the Na"/H+ exchanger or damaging the Na+:K+ ATPase [35] or its metabolic supply [36] thereby indi rectly causing inactivation of exchanger. It is
noteworthy that high concentrations of hy drogen peroxide are needed in acute experi ments in order to induce physiological changes [37-39],
Acknowledgements Supported in part by NEI Research Grant EY04558 (KG), in part by an Unrestricted Depart mental Award from Research to Prevent Blindness, Inc., and in part by NS2510I (CHK). We thank Brenda Sheppard for her valuable secretarial assis tance and Lisa Cheeks and Tracey Slagle for their tech nical assistance.
1 Green K, Simon S. Kelly GM. Bow man KA: Effects of [Na+], [CL], car bonic anhydrase, and intracellular pH on corneal endothelial bicarbon ate transport. Invest Ophthalmol Vis Sei 1981;21:586-594. 2 Bowman KA. Elijah RD, Cheeks KE. Green K; Intracellular potential and pH of rabbit corneal endothelial cells. Curr Eye Res 1984:3:991— 1000.
3 Hull DS. Green K. Bowman K. Csukas S, Riley MV: Intracellular pH and glutathione levels in rabbit cor neal endothelium following storage in moist chamber and MK medium. Invest Ophthalmol Vis Sei 1983:24: 214-217. 4 Miller RB. Tyson I. Reiman AS: pH of isolated resting skeletal muscle and its relation to potassium con tent. Am J Physiol 1963:204:1048 1054. 5 Rink T. Tsien R. Pozzan T: Cyto plasmic pH and free Mg in lympho cytes. J Cell Biol 1982:96:189-196. 6 Keller SK. Jentsch TJ. Janicke 1. Wiederholt M: Regulation of intra cellular pH in cultured bovine reti nal pigment epithelial cells. Pflügers Arch 1988:411:47-52.
272
7 Helbig H. Korbmacher C, Stumpff F, Coca-Prados M. Wiederholt M: Na’/H" exchange regulates intracel lular pH in a cell clone derived from bovine pigmented ciliary epithe lium. J Cell Physiol 1988:137:384389. 8 Korbmacher C. Helbig H. Förster C. Wiederholt M: Evidence for Na/1L exchange and pH sensitive mem brane voltage in cultured bovine corneal epithelial cells. Cun Eye Res 1988;7:619-626. 9 Korbmacher C. Helbig H, Förster C. Wiederholt M: Characterization of Na'/IL exchange in a rabbit corneal epithelial cell line (SIRC). Biochim BiophvsActa 1988:943:405-410. 10 Bonnano JA. Machen TE: Intracel lular pH regulation in basal corneal epithelial cells measured in corneal explants. Characterization of Na/H exchange. Exp Eye Res 1989:49: 129-142. 11 Lin H, Miller SS: [K]0-induced changes in apical membrane voltage (VA) modulate pH, in frog retinal pigment epithelium (RPE). Invest Ophthalmol Vis Sei I989:30(suppl): 168. 12 Wolosin JM. Alvarez IJ. Candia OA: HCO j transport in the load lens epithelium is mediated by an electrogenic Na‘-dependent symport. Am J Physiol 1990;258:C855C86I.
13 Wolosin JM. Alvarez U . Candia OA: Stimulation of toad lens epithe lial NV/H* exchange activity by hy pertonicity. Exp Eye Res 1989:48: 855-862. 14 Wolosin JM. Alvarez LJ, Candia OA: Cellular pH and Na'-H‘ ex change activity in lens epithelium of Bulb marlnus toad. Am J Physiol 1988:255:C595-C602. 15 Wolosin JM. Bonnano JA, Machen TE: Na -dependcnt HCOi" trans port and CL/HCO3' exchange in cil iary epithelium. Invest Ophthalmol VisSci I989;30(suppl):169. 16 Wolosin JM: Stoichiometry of the sodium bicarbonate cotransport of the ciliary body epithelium. Invest Ophthalmol VisSci 1990:31 (suppl): 442. 17 Bassnett S. Reinisch L, Beebe DC: Intracellular pH measurement using triple exitation-dual emission fluo rescence ratios. Am J Physiol 1990: 258:C171—Cl 78. 18 Ye J. Zadunaisky JA: Na/H exchan ger and stimulatory effect of EGF and fLO: on lens fiber and epithe lial membrane vesicles studied with fluorescent dyes. Invest Ophthalmol VisSci 1990:31(supplf:442.
Chu/Keith/Yee/Green
Corneal Endothelial Intracellular pH
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
References
27 Thomas JA. Buchsbaum RN, Zimniak A. Racker E: Intracellular pH measurements in Erlich ascites tu mor cells utilizing spectroscopic probes generated in situ. Biochemis try 1979:8:2210-2218. 28 Grinstein S. Rothstein A: Mecha nisms of regulation of the Na'/H' exchanger. J Membr Biol 1986:90:
1- 12. 29 FischbargJ. Lim JJ: Role of cations, anions and carbonic anhvdrase in fluid transport across rabbit corneal endothelium. J Physiol 1974:241: 647-675. 30 Green K, Cheeks L. Berdecia R. Kramer K. Hull DS: Effect of PCb and metabolic inhibitors or. ionic fluxes across the isolated rabbit cor neal endothelium. Lens Eye Tox Res 1990:7:103-119. 31 Jentsch TJ. Keller SK. Koch M. Wiederholt M: Evidence for coupled transport of bicarbonate and so dium in cultured bovine corneal en dothelial cells. J Membr Biol 1984: 81:189-204. 32 Jentsch TJ. Koch M, Bleckman H. Wiederholt M: Effect of bicarbon ate. pH. methazolamide and stilbenes on the intracellular potential of cultured bovine corneal endothe lial cells. J Membr Biol 1984:78: 103-117. 33 Spector A, Garner WH: Hydrogen peroxide and human cataract. Exp Eye Res 1981:33:673-681.
34 Costarides A. Recasens JF. Riley MV. Green K: The effects of ascor bate. 3-aminotriazole. and 1.3 bis(2chloroethyl)-1-nitrosourea on hy drogen peroxide levels in the rabbit aqueous humor. Lens Eye Tox Res 1989:6:167-173. 35 Garner WH. Garner MH, Spector A: HiO:-induced uncoupling of bo vine Na~/K* ATPase. Proc Natl Acad Sci USA 1983:80:2044-2048. 36 Hyslop PA, Hinshaw DB, Halsey Jr WA, Schraufstatter IU. Sauerhcber RD. Spragg RG. Jackson JH. Coch rane CG: Mechanisms of oxidantmediated cell injury: The glycolytic and mitochondrial pathways of ADP phosphorylation are major in tracellular targets inactivated by hy drogen peroxide. J Biol Chem 1988: 263:1665-1675. 37 Polansky JR. Wood IS. Maglio MT. Alvarado JA: Trabecular meshwork cell culture in glaucoma research: Evaluation of biological activity and structural properties of human tra becular cells in vitro. Ophthalmo logy 1984:91:580-595. 38 Polansky JR. Fauss DJ. Hydom T, Bloom E: Cellular injury from sus tained vs. acute hydrogen peroxide exposure in cultured human corneal endothelium and human lens endo thelium. CLAO J 1990:16(suppl): S23-S28. 39 Csukas S. Costarides A. Riley MV and Green K: Hydrogen peroxide in the rabbit anterior chamber: Effects on glutathione, and catalase effects on peroxide kinetics. Curr Eye Res 1987:6:1395-1402.
273
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
19 Jentsch TJ. Korbmacher C. Janicke J. Fischer DG, Stahl F. Helbig H. Hollwede H, Cragoe EJ. Keller SK, Wiederholt M: Regulation of cyto plasmic pH of cultured bovine cor neal endothelial cells in the absence and presence of bicarbonate. J MembrBiol 1988:103:29-40. 20 Huff JW. Green K: Characteristics of bicarbonate, sodium and chloride fluxes in the rabbit corneal endothe lium. Exp Eye Res 1983:36:607— 615. 21 Hodson S. Miller F: The bicarbon ate pump in the endothelium which regulates the hydration of rabbit cor nea. J Physiol 1976:263:563-577. 22 Bonnano JA. Giasson C: Bicarbon ate transport and pH, regulation in fresh and cultured bovine corneal endothelium. Invest Ophthalmol VisSci 1991 ;32(suppl): 1078. 23 Aronson P: Kinetic properties of the plasma membrane Na-H exchanger. Annu Rev Physiol 1985:47:545560. 24 Matsushima Y, Yoshitomi K, Koseki C. Kawamura M. Akabanc S. Imanishi M. Imai M: Mechanics of intracellular pH regulation in the hamster inner medullary collecting duct perfused in vitro. Pfliigers Arch 1990:416:715-721. 25 Keith CH. Blane K: Sites of tubulin polymerization in PC 12 cells. J Neurochem 1990:54:1258-1268. 26 Keith CH: Quantitative fluores cence techniques for the determina tion of local microtubule polymeri zation equilibria in cultured neu rons. J Neurosci Methods 1991:39: 141-152.
Ophthalmic Res 1992;24:265-273
T.C. Chua Charles H. Keith b Edward Yeea Keith Green a-c
Intracellular pH of Tissue-Cultured Bovine Corneal Endothelial Cells
Departments of Ophthalmology and Physiology & Endocrinology, Medical College of Georgia. Augusta. Ga.: Department of Zoology, University of Georgia, Athens. Ga.. USA
Key Words
Bovine corneal endothelium Tissue culture Intracellular pH Amiloride Ammonium chloride Video imaging
Intracellular pH (pH;) of bovine tissue-cultured corneal endo thelial cells has been measured under several experimental conditions. Determinations were made on individual cells using video-imaging techniques that allowed assessment of 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfiuorescein fluo rescence at 440 and 490 nm. Each experiment had a calibra tion performed on a cell monolayer: this was performed using a high K+-nigericin solution. Resting pH; was 7.25 ± 0.03 (n = 18) in bicarbonate solution at pH 7.4. Amiloride (1 mM) caused an acidification of approximately 0.2 U within 2 min: replacement with normal Ringer allowed a return to normal pH; after an alkali overshoot. Exposure to 20 m.\4 NH4C1 caused alkalinization that became acidic upon washout of NH4CI. In Na+-rich solution pH; returned to normal after acidification but pH; remained low in Na+-free solution until substituted by Na+-rich solution. Removal of HCO3- from the bathing solution caused a nonsignificant acidification of pH; by 0.1 U at 2 and 4 min, and 4,4'-diisothiocyanostilbene-2,2'disulfonic acid (DIDS: 1 mM) acidified pH; by 0.14 U at 2 min and 0.24 U at 4 min. Addition of DIDS (1 mM) in a HCC>3-free solution had no effect on pH;. Hydrogen peroxide acidified pH; by 0.3 U at 50 pM and 1 mM. These results indi cate that a Na+:H+ antiport exists that regulates pH; even at normal ambient pH in the presence of bicarbonate: this pro cess becomes highly activated after an acid load. There is a DIDS-sensitive HCO3 movement that is probably coupled to Na+ or Cl".
Received: June 28.1991 Accepted: January 17. 1992
Keith Green. PhD, DSc Department of Ophthalmology Medical College of Georgia Augusta. GA 30912-3400 (USA)
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Abstract
Initial measurements of intracellular pH (pHj) of corneal endothelial cells [1-3] used the intracellular to extracellular distribution ratio of l4C-labeled dimethyl oxazolidinedione (DMO) [4], Alternative methods now utilize fluorescent dyes trapped within cells [5], One such dye is 2',7'-bis(2-carboxyethyl)5(6)-carboxyfluorescein acetoxymethvlester (BCECF-AM) which freely enters cells. The AM group is cleaved trapping the fluorescent, pH-sensitive and relatively impermeant BCECF inside the cell. Studies of pHj have been made in several ocular tissues including cultured bovine reti nal pigment epithelium [6], cultured bovine pigmented ciliary epithelium [7] and cultured bovine corneal epithelial cells [8, 9] all using 5 (and 6)-carboxy-4',5'-dimethylfluorescein. Fresh rabbit corneal epithelium [ 10], frog reti nal pigment epithelium [11], toad lens epithe lium [12-14] and rabbit ciliary epithelium [15, 16] have all been studied using BCECF, while lens epithelium [ 17] and lens fiber and epithelial cell vesicles have been examined using other probes [18], The pHj of cultured bovine corneal endothelium with 5 (and 6)carboxydimethylfluorescein has been de scribed [19] as well as changes occurring in pHj as a function of external perturbations. Knowledge of the corneal endothelial cell pHi is important for the placement of the var ious transport systems, including bicarbon ate, at either the apical or basolateral surface of the cells. The net bicarbonate transport across the endothelium from the stroma to the aqueous-facing surface [20, 21] requires one of two systems to be operative. Bicarbonate entry occurs either actively across the stromal (basolateral) surface with cellular accumula tion and then subsequent diffusion to the aqueous, or passively into the cell and endoge nous production with active extrusion across
266
the apical (aqueous-facing) surface. pHj would be more alkaline in the first case due to bicarbonate accumulation and more acidic in the second case. The pHj of corneal endothelial cells has been reported as 7.1 in an ambient solution at pH 7.5 (A pH 0.4 U) using a tracer method [ 1-3], 7.15 in a bathing solution at pH 7.4 (A pH 0.25 U) [19] and 7.35 in a solution at 7.5 (A pH 0.15U) [22] using dye techniques. These differences require resolution. Na+:H+ exchange has been found to exist in almost all cell systems [6-19, 23], and we examined the endothelium for the presence of this exchange in cultured bovine corneal endothelial cells. The effects of a known oxidant, hydrogen per oxide, were examined on this tissue.
Materials and Methods Bovine eyes were routinely obtained from a local slaughterhouse and were stored on ice for not more than 4 h. They were washed with sterile saline to remove blood, and then the corneal epithelium and surrounding pigment were completely scraped off us ing a sterile scalpel blade. The globe was dipped in 70% ethanol for 10 s. The cornea was dissected out with aseptic instruments and placed, endothelial side up, in a sterile petri dish containing sterile normal Ringer’s solution: 110 mM NaCl, 4.5 mM K.C1. 1.0 mM NaH2PC>4, 1.2 mM MgCI2, 5.5 mM glucose. 1.8 mM CaCI2 and 25 mM NaHCO,. pH 7.4. The sterile dishes were taken under a laminar flow hood, and the corneas were washed through 4 dishes of culture me dium containing Dulbecco’s modified Eagle medium/ Ham’s FI2 mixture (1:1). 15 mM NaHCCb. 1 mM ascorbic acid, 100 U/ml penicillin. 100 gg/ml streptomycin, 5 gg/ml gentamicin and 2.5 gg/ml am photericin B. To each corneal cup was added 0.5 ml of 0.05% trypsin in 0.02% EGTA solution, and each was incubated at 37 °C in a humidifed incubator with 5% C 0 2 for 1 h. At the end of the incubation, the solution was aspirated from the center of each corneal cup. The endothelium was pelleted and washed several times with culture medium. The suspension was plated out with culture medium containing 20% fetal bovine serum and incubated at 37 °C in a humidified incuba tor with 5% C 0 2. The cells had ambient solution
Chu/Keith/Y ee/Green
Corneal Endothelial Intracellular pH
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Introduction
NaCI, that also contained 50 pM nigericin (made from a stock solution of 10 m M in dimethylformamide: 95% ethanol. 3:1. v/v). This solution allows equilibra tion of intracellular and extracellular pH, and a steadystate fluorescence ratio was determined after allowing 3-5 min for equilibration with a new external pH. Nigericin was present in each bathing solution used in the calibration series. Background fluorescence at the two wavelengths was obtained from cell-free areas in the chambers. Solution changes were achieved by rapid exchange through the use of pipettes: exchanges were usually complete (two rinses with new solution) within 3045 s. The basic Ringer solution was bicarbonatc-based [1-3, 14]. Additional normal Ringer solution was bi carbonate-rich and Hepcs-buffered (10 mM). NH4CI, low N a \ HCOj-free solutions were prepared by equi molar substitutions of NaCI by NH4C1, NaCI by cho line chloride, and HCO3 by CP. respectively. Amiloride (1 mM). 4.4'-diisothiocyano-stilbcne-2,2'-disulfonicacid(DIDS: 1 mM), hydrogen peroxide and oua bain (1 mM) were all obtained from Sigma (St. Louis. Mo., USA). The pH of normal Ringer was adjusted to the desired value using 1 N NaOH or HC1, and this solution, as well as all others, were used within 1 min of pH adjustment in experiments that lasted no longer than 6 or 8 min in the presence of one solution. The ambient pH in any experiment, therefore, was con stant.
Results
The fluorescence image showed consistent labeling between cells with a differentiation between cytoplasm and nucleus. Background fluorescence, after dye labeling and subse quent rinsing, was usually near the zero level. Calibration of pH, was performed routinely after perfusion with K.“-Ringer containing ni gericin. External pH was varied between 6.8 and 8.0. Plots of the steady-state ratio, deter mined 3-5 min after each pH change, indi cate that the 440:490 ratio was high in more acidic solutions with a decreased ratio as am bient pH became more alkaline (fig. 1). Above pH 7.7, the change in fluorescence ratio per unit pH became reduced and the slope mark edly attenuated.
267
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
changed 3 times each week. Amphotericin B was no longer added to the culture medium containing 10% fetal bovine serum when cells reached conflucncy. For experiments, confluent cultures were trypsinized for 5 min at 37 °C. The cell suspension was centri fuged at 1,000 rpm for 5 min and washed twice with fresh culture medium. Cell count and viability tests were performed using a hemacytometer and by moni toring trypan blue exclusion, respectively. The cells were plated on plain coverslips in Bionique chambers (BIO-RAD. Cambridge. Mass., USA; approximately 1 million cells per coverslip) and allowed to settle for 48 h before use. BCECF-AM (Molecular Probes. Eugene. Oreg.. USA) w'as made in a stock solution of 10 mA/in dimethylsulfoxide and used as a 5 pMsolution in Ringer for loading cells with dye. After 45 min the cells were rinsed several times with normal Ringer prior to place ment on the stage of a Zeiss IM fluorescence micro scope that was maintained at 37 °C with an air curtain (Nicholson Precision Instruments. Gaithersberg, Md. USA). Cells were generally imaged with a X 63 objec tive. Epifluorescence excitation was attenuated with a 1.0- or 1,5-ND filter, passed through a 440- or 490-nm interference filter and reflected off a 505-nm dichroic mirror. Emitted signals were passed through the dich roic and a 520- to 560-nm bandpass filter before being imaged. The fluorescein emission upon a 440-nm exci tation is independent of pH. whereas the 490-nm exci tation is pH dependent [24]. Fluorescence was re corded using a DAGE SIT camera (DAGE-MTI. Mi chigan City, Ind., USA) [25,26], Images were digitized directly onto an Imaging Technology framestore (ImagingTechnology, Inc., Woburn, Mass, USA) in an image processor obtained from G.W. Harnaway and Associates (Boulder, Colo., USA). All images were cor rected for shading by dividing by a lowpass-filtered background image and were corrected for geometric distortion [25], Pairs of images were mathematically correlated and intensified, above local background, and were integrated blockwise using circles of a 5-pixel diameter. Several data points were obtained from the cytoplasmic regions of cells in the visual field and an average of 5 ratios (440:490 fluorescence) were ob tained from each corrected frame. Frequently, 6-9 cells were included in the measurement field. Cells were not exposed to light except during the brief peri ods of data collection, about 5 s per wavelength. Usually, only a short series of experiments was per formed with every monolayer, and each experiment was followed by a 5-point calibration using the nigericin-high K technique [27], The cells were washed with a K“-Ringer. in which KC1 had been substituted for
1mM Amiloride
Ringer
Fig. 1. Plot of calibration of intracellular pH as a function of signal ratio. Values are the mean ± SEM of at least 6 monolayer determinations at each point. SEM is maximally the size of the data point.
Fig. 2. Changes in pH, after amiloride. Values are the mean ± SEM of 4 monolayers. Where no SEM is shown, it is too small to be indicated.
A resting pH| of 7.25 ± 0.03 (mean ± SEM, n = 18 monolayers, each with 5-9 cells that were sampled, thus reflecting between 90 and 144 cells in total) was obtained. Each monolayer was initially bathed with Ringer at 7.4 and a base-line pHj value determined. There was no difference in resting pH; whether the cells were exposed to a HCO> rich or a HCOÿ-rich plus Hepes-buffered solu tion. Amiloride (1 mM) caused an acidifica tion of between 0.16 and 0.26 pH U at 2 min after drug addition. This value was main tained during an additional 2 min of amilo ride. Within 2 min after exposure to normal Ringer, pH; returned to near baseline values after a small overshoot. At 6 min after the return to Ringer solution, pHj was normal (fig. 2).
Changing the bathing solution from nor mal Ringer to a 20 mM NHgCl-containing solution caused an abrupt alkalinization of the cell that underwent a slight acidification over the next 4 min. Replacement of the am bient solution with normal or Na+-free Ringer caused a marked and immediate acidification of pHj within 2 min; a further 2 min allowed recovery towards normal pHj at a rate of about 0.2 pH U/min (fig. 3). Replacement of the NHgCl solution with a Na+-frce solution also caused acidification that was sustained. Replacement of the Na+-free solution with a Na+-containing solution initiated a rapid re turn towards normal pHj at the same rate as noted after immediate replacement with Na+rich solution, i.e.. 0.2 pH U/min or about half that noted in similar experiments previously in a bicarbonate-rich solution [ 19] (fig. 3). Re-
268
Chu/Keith/Yee/Green
Corneal Endothelial Intracellular pH
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Time in Minutes
pH,
20 mM NHaCI
covery of pHj after removal of NH4CI is, therefore, a sodium-dependent process. The different stages of this process are illustrated in figure 4 with photographs of the 490/440 images processed as described in Materials and Methods. Bathing in HCOTfree solution caused a nonsignificant acidification of pHj by 0 . 1 U at 2 and at 4 min (n = 5 experiments: 26 cells) as seen earlier [19], and no further change oc curred after DIDS (1 mM) was added. The addition of DIDS alone acidified pHj by 0.14 U at 2 min and 0.24 U after 4 min. Reex posure to NaCl Ringer caused a return of pHj to normal values. Ouabain (1 mM) caused a very small acidification of pHj ( < 0 .l U), and no change of pHj was induced by BaCB (2 mM) or high K+ Ringer (50 mM K+) within 10 min of exposure to these agents.
Na+ Ringer ( • )
Time in Minutes
Physiological concentrations of H2O2 (50 pM) caused an acidification of pHj by 0.3 U within 2 min, and the pHj recovered to resting state at 5 min. A 20-fold increase in concen tration (1 mM) also induced a decrease of pHj by 0.3 U within 3 min. Recovery from this acidification was also observed, and the pHj remained stable for at least 10 min after re covery.
Discussion
At steady-state in an ambient pH of 7.4, the pHj of bovine tissue-cultured corneal en dothelial cells is 7.25 ± 0.03 (n = 18; A pH 0.1 5 U). This value is closer to ambient solu tion pH than that of fresh rabbit endothelial cells [ 1-3] under identical bathing conditions.
269
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Fig. 3. Changes in pH; induced by exposure to a prepulse of NH4CI with subsequent perfusion with either Na+-rich or Na+-free solu tions. Note that cell acidification occurs after removal of NH4CI re gardless o f Na+-rich or Na+-free so lution. In Na+-rich solution, recov ery towards normal pH, occurs in stantly (•), but in Na*-free solution, cell remains acid until exposed to Na+-rich solution (o). Values are the mean ± SEM of at least 5 monolayer determinations at each point. SEM is too small to be indi cated at 0. 2, 4, 12 and 14 min. The values over the first 6 min were the same for both series of experiments since the solutions were identical until 4 min. The values here are the mean ± SEM of at least 10 deter minations. The 6-min value was the same in Na+-Ringer and Na+free Ringer (o). Na+-free Ringer was present from 4 to 10 min be fore replacement with Na* Ringer.
The pHj is, however, closer to values found for fresh rabbit corneal epithelial basal cells [ 10] and cultured bovine corneal epithelial cells [6-9], Tissue-cultured bovine corneal en dothelial cells in bicarbonate Ringer showed a pH; o f 7.15 at an external pH of 7.4 (A pH 0.25 U) [ 19] or a A pH of 0.15 in pH 7.5 [22], values very close to those found in the present studies. The difference between cultured and fresh endothelial cells may reflect the species difference and the fact that there may be a decrease in the ability of cultured cells to per form quantitatively the same as fresh cells. Recent results [22] comparing fresh and cul-
270
tured bovine endothelial cells also showed that fresh cells were more acidic relative to the ambient bathing solution. The experiments with prepulses of NH4CI were made to acid-load the cells. The influx of NH 3 increases pH|, and a slow acidification then occurs as NH4+ enters the cells down its electrochemical gradient. When the cell is then bathed in NH4Cl-free solution, either with or without Na+, the NH 3 is quickly lost and pH, falls far below the resting pH;. A recovery phase follows due to the pHj regula tory systems as the cell transports acid equiva lents. The data clearly indicate that these pro-
Chu/Keith/Yee/Grecn
Corneal Endothelial Intracellular pH
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
Fig. 4. Series of photographs of cells that were used in the construction of figure 3. a At zero time, normal cells, b At 2 min, slightly darker cytoplasm, c At 4 min. little detectable change in the ratio, d At 6 min. after replacement of the NH4CI solution. The cells have become much lighter, reflecting the acidification caused by replacement of the bathing solu tion with (in this case) Na^-rich solution.
passive distribution of H+ by the Nernst po tential. taken here as -4 0 mV [Wigham CG. Green K, Hodson S. 1991, unpubl. data]. Changes in the external pH evoked altera tions in pHj. although the latter were rela tively small compared to the ambient pertur bations. We found that pHj increased by 0.2 U in face of an ambient change from pH 6.8 to 8.0. These data differ slightly from these found earlier [ 1, 2] where the A pH between pHj and the ambient solution re mained almost constant at 0.4 pH U as am bient pH increased from 7.0 to 8.0. The effect of ouabain on pHj is predictable based upon the known inhibition of the Na+:K~ pump by this inhibitor. Use of oua bain causes the intracellular Na+ to increase, thus decreasing the Na+ gradient into the cell. This, in turn, leads to a decreased sodium entry via the Na+:H+ exchanger and a second ary decrease in pHj to a more acidic environ ment. That this decrease is not related to the change in membrane voltage per sc is indi cated by the absence of an effect of either a 50 mM K+ Ringer solution, that normally depo larizes the membrane potential by 20-15 mV within 20-30 s [Wigham CG, Green K. Hodson S, 1991. unpubl. data], or BaCb. a K+channcl blocker, on pHj. There appears to be little effect of membrane potential, at least depolarization, on pHj although a depen dence of intracellular electrical potential dif ference on pHj has been strongly suggested previously [32]. In these experiments, acidifi cation of the cellular interior usually leads to a hyperpolarization, and alkalinization was as sociated with depolarization. These changes were usually transient in nature and lasted, under the given experimental conditions, less than 10 min [32], These data are not incom patible since pHj could drive membrane po tential through alterations of membrane charges, whereas alterations in membrane po tential induced by a different mechanism
271
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
cesses occur in endothelial cells and that re covery is blocked by Na+ removal from the bathing solution in the presence of bicarbon ate (fig. 3). supporting prior studies [19], The cellular acidification after amiloride under normal bathing conditions (fig. 2) is also strongly suggestive of the presence of a Na+:H+ exchanger that normally regulates pH| [19, 28]. Previous studies, however, noted lit tle or no effect of amiloride under resting con ditions [ 19], Similarly, experiments using ace tate (data not included) to induce cellular acidification also caused an overshoot alkalinization of pHj when the ambient solution was returned to normal Ringer. These data confirm those obtained in other studies using similar techniques [ 19], That pH| is lower than that of the ambient solution supports the notion that bicarbonate transport across this membrane [20. 21] is directed out of the cell. Inwardly directed active movement of bicarbonate across the basolateral membrane (with subsequent dif fusion out of the cell across the opposite mem brane) should lead to alkalinization of the cell, whereas outwardly directed bicarbonate transport should lead to acidification. Given the vectorial orientation of bicarbonate trans port from the stromal to the aqueous-facing side of the endothelium, the bicarbonate transport systems may be rightfully placed, in an outwardly facing direction, on the apical membrane of the cell [20, 21, 29-32], Other studies [19, 32] confirm that only a slight, nonsignificant decrease in pHj occurs after bicarbonate removal from the ambient solu tion. These data, plus those from the use of DIDS in the presence and absence of bicar bonate. suggest that a net uptake of bicarbon ate occurs that is DIDS inhibitable. The data also support the notion [19] that the Na+/ HCOî symport is oriented in a net inward direction. Furthermore, the pHj is more alka line than predicted if it was based upon the
would not necessarily induce a change in pHj. Preliminary' data [19], however, indicated that the use of a high K+ solution caused cellu lar alkalinization. It has been shown that an oxidant (H2O2) normally present in aqueous humor at 30-80 \iM [33, 34] induced a modest fall in pHj: of interest is that a high concentration of 1 mM also induced a small, reversible fall in pHj. The transient acidification of pHj by peroxide is possibly either directly affecting the Na"/H+ exchanger or damaging the Na+:K+ ATPase [35] or its metabolic supply [36] thereby indi rectly causing inactivation of exchanger. It is
noteworthy that high concentrations of hy drogen peroxide are needed in acute experi ments in order to induce physiological changes [37-39],
Acknowledgements Supported in part by NEI Research Grant EY04558 (KG), in part by an Unrestricted Depart mental Award from Research to Prevent Blindness, Inc., and in part by NS2510I (CHK). We thank Brenda Sheppard for her valuable secretarial assis tance and Lisa Cheeks and Tracey Slagle for their tech nical assistance.
1 Green K, Simon S. Kelly GM. Bow man KA: Effects of [Na+], [CL], car bonic anhydrase, and intracellular pH on corneal endothelial bicarbon ate transport. Invest Ophthalmol Vis Sei 1981;21:586-594. 2 Bowman KA. Elijah RD, Cheeks KE. Green K; Intracellular potential and pH of rabbit corneal endothelial cells. Curr Eye Res 1984:3:991— 1000.
3 Hull DS. Green K. Bowman K. Csukas S, Riley MV: Intracellular pH and glutathione levels in rabbit cor neal endothelium following storage in moist chamber and MK medium. Invest Ophthalmol Vis Sei 1983:24: 214-217. 4 Miller RB. Tyson I. Reiman AS: pH of isolated resting skeletal muscle and its relation to potassium con tent. Am J Physiol 1963:204:1048 1054. 5 Rink T. Tsien R. Pozzan T: Cyto plasmic pH and free Mg in lympho cytes. J Cell Biol 1982:96:189-196. 6 Keller SK. Jentsch TJ. Janicke 1. Wiederholt M: Regulation of intra cellular pH in cultured bovine reti nal pigment epithelial cells. Pflügers Arch 1988:411:47-52.
272
7 Helbig H. Korbmacher C, Stumpff F, Coca-Prados M. Wiederholt M: Na’/H" exchange regulates intracel lular pH in a cell clone derived from bovine pigmented ciliary epithe lium. J Cell Physiol 1988:137:384389. 8 Korbmacher C. Helbig H. Förster C. Wiederholt M: Evidence for Na/1L exchange and pH sensitive mem brane voltage in cultured bovine corneal epithelial cells. Cun Eye Res 1988;7:619-626. 9 Korbmacher C. Helbig H, Förster C. Wiederholt M: Characterization of Na'/IL exchange in a rabbit corneal epithelial cell line (SIRC). Biochim BiophvsActa 1988:943:405-410. 10 Bonnano JA. Machen TE: Intracel lular pH regulation in basal corneal epithelial cells measured in corneal explants. Characterization of Na/H exchange. Exp Eye Res 1989:49: 129-142. 11 Lin H, Miller SS: [K]0-induced changes in apical membrane voltage (VA) modulate pH, in frog retinal pigment epithelium (RPE). Invest Ophthalmol Vis Sei I989:30(suppl): 168. 12 Wolosin JM. Alvarez IJ. Candia OA: HCO j transport in the load lens epithelium is mediated by an electrogenic Na‘-dependent symport. Am J Physiol 1990;258:C855C86I.
13 Wolosin JM. Alvarez U . Candia OA: Stimulation of toad lens epithe lial NV/H* exchange activity by hy pertonicity. Exp Eye Res 1989:48: 855-862. 14 Wolosin JM. Alvarez LJ, Candia OA: Cellular pH and Na'-H‘ ex change activity in lens epithelium of Bulb marlnus toad. Am J Physiol 1988:255:C595-C602. 15 Wolosin JM. Bonnano JA, Machen TE: Na -dependcnt HCOi" trans port and CL/HCO3' exchange in cil iary epithelium. Invest Ophthalmol VisSci I989;30(suppl):169. 16 Wolosin JM: Stoichiometry of the sodium bicarbonate cotransport of the ciliary body epithelium. Invest Ophthalmol VisSci 1990:31 (suppl): 442. 17 Bassnett S. Reinisch L, Beebe DC: Intracellular pH measurement using triple exitation-dual emission fluo rescence ratios. Am J Physiol 1990: 258:C171—Cl 78. 18 Ye J. Zadunaisky JA: Na/H exchan ger and stimulatory effect of EGF and fLO: on lens fiber and epithe lial membrane vesicles studied with fluorescent dyes. Invest Ophthalmol VisSci 1990:31(supplf:442.
Chu/Keith/Yee/Green
Corneal Endothelial Intracellular pH
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
References
27 Thomas JA. Buchsbaum RN, Zimniak A. Racker E: Intracellular pH measurements in Erlich ascites tu mor cells utilizing spectroscopic probes generated in situ. Biochemis try 1979:8:2210-2218. 28 Grinstein S. Rothstein A: Mecha nisms of regulation of the Na'/H' exchanger. J Membr Biol 1986:90:
1- 12. 29 FischbargJ. Lim JJ: Role of cations, anions and carbonic anhvdrase in fluid transport across rabbit corneal endothelium. J Physiol 1974:241: 647-675. 30 Green K, Cheeks L. Berdecia R. Kramer K. Hull DS: Effect of PCb and metabolic inhibitors or. ionic fluxes across the isolated rabbit cor neal endothelium. Lens Eye Tox Res 1990:7:103-119. 31 Jentsch TJ. Keller SK. Koch M. Wiederholt M: Evidence for coupled transport of bicarbonate and so dium in cultured bovine corneal en dothelial cells. J Membr Biol 1984: 81:189-204. 32 Jentsch TJ. Koch M, Bleckman H. Wiederholt M: Effect of bicarbon ate. pH. methazolamide and stilbenes on the intracellular potential of cultured bovine corneal endothe lial cells. J Membr Biol 1984:78: 103-117. 33 Spector A, Garner WH: Hydrogen peroxide and human cataract. Exp Eye Res 1981:33:673-681.
34 Costarides A. Recasens JF. Riley MV. Green K: The effects of ascor bate. 3-aminotriazole. and 1.3 bis(2chloroethyl)-1-nitrosourea on hy drogen peroxide levels in the rabbit aqueous humor. Lens Eye Tox Res 1989:6:167-173. 35 Garner WH. Garner MH, Spector A: HiO:-induced uncoupling of bo vine Na~/K* ATPase. Proc Natl Acad Sci USA 1983:80:2044-2048. 36 Hyslop PA, Hinshaw DB, Halsey Jr WA, Schraufstatter IU. Sauerhcber RD. Spragg RG. Jackson JH. Coch rane CG: Mechanisms of oxidantmediated cell injury: The glycolytic and mitochondrial pathways of ADP phosphorylation are major in tracellular targets inactivated by hy drogen peroxide. J Biol Chem 1988: 263:1665-1675. 37 Polansky JR. Wood IS. Maglio MT. Alvarado JA: Trabecular meshwork cell culture in glaucoma research: Evaluation of biological activity and structural properties of human tra becular cells in vitro. Ophthalmo logy 1984:91:580-595. 38 Polansky JR. Fauss DJ. Hydom T, Bloom E: Cellular injury from sus tained vs. acute hydrogen peroxide exposure in cultured human corneal endothelium and human lens endo thelium. CLAO J 1990:16(suppl): S23-S28. 39 Csukas S. Costarides A. Riley MV and Green K: Hydrogen peroxide in the rabbit anterior chamber: Effects on glutathione, and catalase effects on peroxide kinetics. Curr Eye Res 1987:6:1395-1402.
273
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/4/2018 1:33:16 PM
19 Jentsch TJ. Korbmacher C. Janicke J. Fischer DG, Stahl F. Helbig H. Hollwede H, Cragoe EJ. Keller SK, Wiederholt M: Regulation of cyto plasmic pH of cultured bovine cor neal endothelial cells in the absence and presence of bicarbonate. J MembrBiol 1988:103:29-40. 20 Huff JW. Green K: Characteristics of bicarbonate, sodium and chloride fluxes in the rabbit corneal endothe lium. Exp Eye Res 1983:36:607— 615. 21 Hodson S. Miller F: The bicarbon ate pump in the endothelium which regulates the hydration of rabbit cor nea. J Physiol 1976:263:563-577. 22 Bonnano JA. Giasson C: Bicarbon ate transport and pH, regulation in fresh and cultured bovine corneal endothelium. Invest Ophthalmol VisSci 1991 ;32(suppl): 1078. 23 Aronson P: Kinetic properties of the plasma membrane Na-H exchanger. Annu Rev Physiol 1985:47:545560. 24 Matsushima Y, Yoshitomi K, Koseki C. Kawamura M. Akabanc S. Imanishi M. Imai M: Mechanics of intracellular pH regulation in the hamster inner medullary collecting duct perfused in vitro. Pfliigers Arch 1990:416:715-721. 25 Keith CH. Blane K: Sites of tubulin polymerization in PC 12 cells. J Neurochem 1990:54:1258-1268. 26 Keith CH: Quantitative fluores cence techniques for the determina tion of local microtubule polymeri zation equilibria in cultured neu rons. J Neurosci Methods 1991:39: 141-152.
