Exp. Eye Res. (1992) 54, 181-191

Regulatory Volume Decrease by Cultured Non-pigmented Ciliary Epithelial Cells MORTIMER

M. C I V A N a b L K I M P E T E R S O N - Y A N T O R N O L AND R O B E R T E. Y A N T O R N O

MIGUEL

COCA-PRADOS

'~

"c

Department of ~Physiology and bMedicine, University of Pennsylvania, PA, cDepartment of Electrical Engineering, Temple University, Philadelphia, PA, and dDepartment of Ophthalmology, Yale Medical School New Haven, CT, U.S.A. (Received Baltimore 11 November 1990 and accepted in revised form 19 February 1991) Cells (ODM C1-2/SV40) derived from human non-pigmented ciliary epithelial cells were studied by electronic cell sizing. The time course of the cell volume (vc) was monitored after suspending cells in paired experimental and control, isosmotic and hyposmotic solutions of identical ionic composition. Following anisosmotic cell swelling, the cells displayed the regulatory volume decrease (RVD) previously described. The RVD primarily reflects loss of cell KC1 since: (1) the K+-channel blockers quinidine and Ba 2+ both inhibit the RVD; and (2) replacement of external CI- with gluconate or addition of the CI- channel blocker NPPB also inhibits the RVD. Bicarbonate has previously been reported to speed the RVD. This action likely reflects pH dependence of the channels since: (1) increasing the external pH speeds the RVD, whether or not HCOa- is present; and (2) DIDS (a blocker of C1 channels and of C1 /HCOa- exchange) is an effective inhibitor of the RVD, even after blocking C1 /HCO 3 exchange by removing external HCQ . The RVD could also be inhibited by reducing the availability of Ca~+, either by omitting Ca~+ from the external medium or by blocking mobilization of intracellular Ca2+ with TMB-8. Furthermore, the RVD was slowed and incomplete in the presence of the calcium/calmodulin blocker trifluoperazine. We conclude that anisosmotic swelling triggers a series of events, mediated at least in part by calcium/calmodulin, leading to the extrusion of KC1 through parallel K+ and C1- channels. Key words: regulatory volume decrease; K+ channels ; Cl-,channels; coupled antiports ; quinidine; Ba~+; NPPB; calcium/calmodulin. 1. Introduction The application of anisosmotic changes in cell volume (vc) can be used to probe the mechanisms and regulation of aqueous formation. Swelling or shrinking cells elicits secondary regulatory responses in ion transport. The transduction linking the perturbation and response involves regulatory events of a general nature, such as changes in intracellular Ca 2+ activity and pH (Hoffmann, 1987). The same transduction mechanisms are likely triggered by other perturbations as well. Therefore, understanding volume-transport coupling should provide insights into the regulation of aqueous formation by the ciliary epithelium under a broad range of conditions. In principle, other experimental perturbations also provide useful probes for studying aqueous formation. We have been focusing on cell volume regulation (Yantorno et al., 1989; Civan et al., 1990) because of four considerations. (1) This approach has proved effective in probing the transport mechanisms and regulation of other cells (Grinstein et al., 1985b; Gilles, 1987; Hoffmann, 1987). For example, study of the volume-regulatory responses provided the first indication of a Na+/ * For correspondence at: Department of Physiology, Universityof Pennsylvania, Richards Building, Philadelphia, PA 19104-6085, U.S.A. 0014-4835/92/020181+ 11 $03.00/0

K+/2C1 - symport operating in avian erythocytes (Kregenow and Caryk, 1979). Another example is provided by the indication that the Na+/H + antiport in lymphocytes is volume-activated by changing the pH sensitivity of an internal modifier site (Grinstein et al., 1985a, b). (2) Preliminary data have already documented the value of applying this approach specifically to the study of ciliary epithelial cells (Farahbakhsh and Fain, 1988; Yantorno et al., 1989; Civan et al., 1990). (3) The basic mechanisms needed for cell volume regulation are also required for aqueous formation. For example, K+ channels are incorporated in the basal (aqueous) membrane of the non-pigmented ciliary epithelial cells (Yantorno et al., 1987; Civan et al., 1990). These channels appear to constitute the only mechanism by which K+ can be secreted by the non-pigmented cells into the aqueous, and are likely critical in their regulatory volume responses (Yantorno et al., 1989). (4) Cell volume regulation must be an intrinsically important aspect of normal aqueous secretion because of the rapid flow of solutes and water through the cells. Based on published data (Cole, 1966; Hogan, AIvarado and Weddell, 1971), we estimate that the turnover time for the intracellular fluid in the ciliary epithelium of h u m a n s and rabbits is approximately 3 min. Without volume regulation, even a small © 1992 Academic Press Limited

182

mismatch between the rates of fluid entry (from the underlying stroma) and extrusion (into the aqueous humor) could be catastrophic by producing either swelling and cell rupture, or shrinkage and distortion of the cytoskeleton and relationships among the intracellular organelles. Following anisomotic swelling of ciliary epithelial cells, the cell volume spontaneously returns towards its initial baseline value. This regulatory volume decrease (RVD) phenomenon has been observed both by optical measurements of intact rabbit ciliary processes (Farahbakhsh and Fain, 1988) and by electronic cell sorting of human non-pigmented ciliary epithelial cells in suspension (Yantorno et al., 1989). From our initial measurements of cell volume, we concluded that the RVD of non-pigmented ciliary epithelial cells probably reflects activation of the parallel K+ and C1 channels (Yantorno et al., 1989). However, we could not exclude a possible major contribution from parallel K+/H + and C1-/ HCQantiports. In the current study, we present further information supporting the concept that: (1) the RVD primarily reflects K+ and CI- movement through parallel channels; and (2) Ca2+ is one component in the coupling between cell swelling and the RVD.

2. Materials and Methods Cells

The cell line studied (ODM C1-2/SV40) was derived from a primary culture of human non-pigmented ciliary epithelium (Martin-Vasallo, Ghosh and CocaPrados, 1989). Cells were grown in Dulbecco's modified Eagle's medium (DMEM, 320-1965, Gibco Laboratories, Life Technologies, Inc., Grand Island, NY) with 10% fetal bovine serum (FBS, A-1115-6, HyClone Laboratories, Inc., Logan, UT; and F4884, Sigma Chemical Co., St Louis, MO) and 50/zg m1-1 gentamycin sulphate (garamycin, Schering Corp., Kenilworth, NJ; and Hazelton Biologics, Lenexa, KS), at 37°C in 5 % CQ. The osmolality of the medium was 328 mosmol. Cells were passaged every 6-7 days, and studied in suspension after reaching confluency, within 6-10 days of passage. Cells were harvested for passage and study, following the commonly-used trypsinization procedure described previously (Yantorno et al., 1989). A 0.5-ml aliquot of the cell suspension in DMEM (or in C1-- or Ca~÷-free medium, as appropriate) was added to 20 ml of each test solution, and volume measurements performed with a Coulter Counter (model ZBI-Channelyzer II), using a lO0-/zm aperture. The calibration ,of the instrument was verified by measuring the diameter of polystyrene microspheres of known dimensions (Epics Div. of Coulter Corp., Hialeah, FL). Measurements of the volume of spherical cells in suspension lead to vo values in agreement with those estimated from other techniques (Lee et al., 1988),

M . M . C I V A N ET AL.

The validity of the application of electronic cell sizing to the ODM C1-2/SV40 cells is further supported by the qualitatively similar results obtained by our analysis of these cultured cells in suspension (Yantorno et al., 1989) and by optical analysis of intact rabbit ciliary processes (Farahbakhsh and Fain, 1988). Both approaches have shown that the non-pigmented ciliary epithelial cells display a regulatory volume decrease over a period of minutes, which is partially blocked by: Ba2+; increased external [K÷]; and pretreatment with cardiotonic steroids. Solutions and Chemicals

The standard Ringer's solutions contained (mM): 55-0 NaC1, 0"6 MgC12, 2"4 KC1, 1"2 CaC12, 0"6 KH2PO4, 15"0 NaHCQ, 7-5 N-hydroxyethylpiperazine-N'-2ethanesulfonic acid (Hepes), and 10 mM glucose at pH 7-5. The osmolality of the test solutions was either 150-160mosmol (in the absence of sucrose) or 300-315 mosmol (after the addition of sucrose). Isosmotic and hyposmotic C1 -free Ringer's solutions were also prepared by substituting gluconate for C1-. Ceils were also studied in the nominal absence of external HCO3- or Ca2÷. Bicarbonate was replaced by CI-, and Ca2+ by Na ÷ in isoequivalent fashion. All chemicals were reagent grade. Quinidine, trifluoperazine (TFP) and acetazoleamide were obtained from Sigma Chemical Co., 4,4'-diisothiocyano-2,2'disulfonic acid (DIDS) from Pierce (Rockford, IL) and Calbiochem (San Diego, CA), and 8-(N,N-diethylamino)octyl 3,4,5-trimethoxy-benzoate (TMB-8) from Calibiochem (San Diego, CA). Gadolinium chloride hexahydrate was obtained from the Aldrich Chemical Co. (Milwaukee, WI). Statistics

Unless otherwise stated, all values are presented as the m e a n + l S.E. The number of experiments is indicated by the symbol N. The probability of the null hypothesis has been calculated using Student's t-test. 3. Results Baseline Measurements

The phenomenon of the regulatory volume decrease (RVD) is illustrated by the baseline measurements in Fig. 1 (0). Upon transferring the cells from DMEM to isosmotic test medium ('Iso'), the mean baseline cell volume (vc) was 2-52 4-0'05 pL Thereafter, ve declined by about 10-15% over the first lOmin, and was subsequently steady. In contrast, suspending cells in the 50% hyposmotic solution ('Hypo') produced maximum swelling within 2-4 min, followed by a spontaneous fall in vo to a value close to the initial volume. We have quantified the RVD response by: measuring the elapsed time before the peak swelling is

REGULATORY

VOLUME

D E C R E A S E BY C I L I A R Y

EPITHELIAL

CELLS

183

150

130

I10

¢ o.zx--~X

o - ~ ~ ~ ~

90

8~xT

70 -I0

T

T

1

I

I

I

I

I

I

0

IO

20

30 Time (min)

40

50

60

70

FIG. 1. Effect of 5 mM Ba 2+ on the response of cell volume (vo) to suspension in isotonic and 50% hypotonic solution. Further details concerning this and the other figures are provided in the text and in Table I. ( A ) Hypotonic + Ba; (Q) hypotonic; ( A ) isotonic + Ba; (O) isotonic.

TABLE I

Time constant (~) of the regulatory volume decrease ~ (min)

Vo,o (%)

Atpeak (min)

Atpl~t (rain)

(N)

Atm (min)

P

Control* Ba 2+ (5 mM)

7"6 _+0"4 18.1 _+0.2

148 _+ 2 158'3 _+0.3

4 2

2 6

4 4

4-12 8-20

-< 0'001

Control Quinidine (0"5 raM)

11'7_+0'7 26'5_+1'6

165 _+4 160_+2

2 2

2 2

6 6

4-20 4-20

-< 0"001

Control NPPB (100 #M)

15"0 ± 0'9 41.2_+2"2

160 _+ 3 159+1

4 4

---

6 6

4-20 4-30

-< 0"001

14"62 _+0"02 8"8_+1'2

153"2 _+0-04 156_+6

4 2

2 --

7 7

8-20 2-12

-< 0"001

Conditions

pH 7I" pH 8I" Control* DIDS (0"5 mM)

11" 1 _+O"8 32"2_+1"1

158 _+ 3 171+1

2 4

2 6

6 6

4-12 8-30

-< 0'001

Control 0 Ca ~+

19'0 _+ 1'9 31'9+4'9

149 _+ 2 162_+2

4 4

---

6 6

4-12 4-12

-< 0-05

Controls Gd 3+ (20 ffM)

8'5 ± 1"1 11"1_+0'3

146 _+4 160_+1

2 2

2 2

4 4

4-12 4-12

-> 0"05

Control TMB-8 (70 #M)

17"6 _+ 1"6 23"9_+3"0

159 ± 4 180±5

8 4

2 6

6 6

8-20 8-20

->0-05

Control TFP§ (20-25 #M)

12'7 _+ 1'2 30"1 -+2"1

164 +_ 5 184_+4

4 12

-6

4 4

4-20 12-30

-< 0'001

Control TFP (10/*M)

7"9 -+0-6 27-4_+3"1

178 -+ 10 141_+2

2 2

-6

4 4

8-20 4-30

-< 0"01

* HCOa- not included in test solutions. "~ HCO3 omitted from four of the seven suspensions at both pH 7 and 8. 1: HCQ- and phosphate not included in test solutions. § TFP symbolizes trifluoperazine, Following suspension in 50% hypotonic medium, the cells displayed peak values of cell volume at the times entered under 'Atpe~k'. The durations of the measured plateaux at the maximum values are listed under ' Atpi~t'. The control suspensions displayed plateau durations of 0-2 rain. The time courses of the cell volumes have been fitted with single exponentials (r) over the time intervals following suspension entered under 'Atrit'. ~o,ois the cell volume extrapolated to the time of suspension (t = 0), a t t a i n e d (Atpeak) a n d t h e d u r a t i o n of t h e p l a t e a u p e r i o d (if p r e s e n t , Atp~at) at t h e m a x i m u m swelling, a n d fitting t h e r e g u l a t o r y r e s p o n s e to t h e m o n o - e x p o n e n t i a l e x p r e s s i o n ( Y a n t o r n o et al., 1 9 8 9 ) : vc = (~c,o-V~)e-'/~ + v~

(1)

w h e r e fe,o is t h e e x t r a p o l a t e d v a l u e of v~ at t = O, vo~ is t h e s t e a d y - s t a t e v o l u m e ( t a k e n to be 1 1 0 % of t h e

initial v o l u m e in i s o t o n i c solution), a n d r is t h e t i m e c o n s t a n t of t h e RVD. T h e t i m e p e r i o d (Atf~t) o v e r w h i c h t h e fit w a s c o n d u c t e d a n d t h e d e r i v e d p a r a m e t e r s (Atpeak, Atpl~t, vo,o a n d z) u n d e r b a s e l i n e a n d exp e r i m e n t a l c o n d i t i o n s are e n t e r e d in T a b l e I. W e e m p h a s i z e t h a t Eqn (1) c o n s t i t u t e s a n e m p i r i c a l e x p r e s s i o n w h i c h w e h a v e f o u n d useful for d a t a reduction. Additional exponential contributions with

184

M.M.

ClVAN

ET AL.

170

I-I

,so °E~ ,30

~ ,~\~

"~{ :

~

-~

I,

>,

±

o

70 -I0

~

' I0

0

' 20

' 30

~ 40

' 50

60

70

Time (rain)

FIG. 2. Effectof quinidine on the regulatory volume decrease (RVD). (A) Hypotonic + quinidine ; ( • ) hypotonic; (O) isotonic; (~) isotonic + quinidine. 200

T T T -~ I I

180

04

T l

[ t

T

m

T ±

160

IT'--T T 140

::3

¢ 120 -

± T

I00 80 -I0

T

I

I

I

I

I

I

I

0

I0

20

30

40

50

60

70

Time (rain)

FIG. 3. Effect of external C1 replacement with gluconate on the response to anisosmotic swelling. The total calcium concentration was 1.25 mM. (A) Hypotonic (0 CI); ( • ) hypotonic; (O) isotonic; (A) isotonic (0 Cl). time constants of tens of seconds would not have been detected with our current approach. Such contributions are probably small (Yantorno et al., 1989), but possibly significant. This possibility is given some support by the observation that certain experimental perturbations do increase Vo,o (e.g. Figs 8 and 10). Given the limitations in resolving r (imposed by the time required for measurement and our wish to avoid cellular damage by increasing the stirring rat), we have addressed changes in ~c.o only qualitatively. We have found that the baseline values of 7 can vary significantly (by a factor of approximately 2). For this reason, all of the current measurements were conducted with parallel controls obtained from the same cell harvesting. In order to obviate the effects of time-dependent change, the chronological order of sequence of measurements was reversed in half of each series of experiments. Importance of K ÷ in the RVD

One of the central findings of our previous study of the RVD of the ODM C1-2/SV40 cells was that the K+channel blocker Ba > prolonged r by about 80% (P < 0-05) at a concentration of 2 raM. Because of the

importance we have ascribed to this observation, the experiment has been repeated with 5 mMBa 2+ (Fig. 1, Table I). At the higher concentration, Ba > produced greater effects, increasing r by about 140% (P < 0-001) and prolonging Atp~t. K+ channels have also been reported to be blocked by quinidine and quinine in a variety of preparations (Germann, Ernst and Dawson, 1986; Germann et al., 1986; Hoffmann, Lambert and Simonsen, 1986; Lee et al., 1988; Tang, Peterson-Yantorno and Civan, 1989). Our recent patch-clamp data have suggested that quinidine also blocks baseline K÷ channels in ODM C1-2/SV40 cells (Civan et al., 1990). Therefore, 0.5 mM quinidine was applied to cells suspended in isosmotic and 50%-hyposmotic media (Fig. 2, Table I). In comparison to the control preparation, quinidine prolonged r by approximately 125 % (P < 0.001). We conclude from the experiments with K+-channel blockers that loss of cell K÷ is likely to play a major role in the regulatory volume decrease. Importance of C1- in the R VD

In our initial study of volume regulation by the ODM C1-2/SV40 cells (Yantorno et al., 1989), glu-

REGULATORY

VOLUME

D E C R E A S E BY C I L I A R Y 210

T TI~T

190

EPITHELIAL

CELLS

185

T

/

1

3_

170 15o

1

t

I\T

~ 130 -0

J_

llO

-

.....

7O -I0

T 1 _a {

I

I

I

I

I

I

I

0

I0

20

30 Time (min)

40

50

6O

70

FI6. 4. Effect of external C1 replacement with gluonate on the response to anisosmotic swelling. The total calcium concentration of the gluconate-containing solution was increased fivefold to 6.25 mM. (A) Hypotonic (0 C1); O) hypotonic; (A) isotonic (0 C1); (O) isotonic. 160 150

i -q\ I

140

o~ 130 ~: 120 > II0

I00

Q

90 80

-I0

_|

0'

' I0

2'0

'o

5 40' Time (rain)

5i0

±

610

70

FiG. 5. Effect of external pH on the RVD. (O) Isotonic, pH 7; (0) hypotonic, pH 7; (A) isotonic, pH 8; (A) hypotonic, pH 8. conate substitution for external C1- blocked the regulatory volume response. However, we were unable to monitor the initial swelling phase following suspension in hypotonic, C1--free solution. We have reexamined the role of C1- in the current study. The experimental and 'control' aliquots of harvested cells were both suspended initially in C1--free medium before resuspension 3 9 - 5 8 m i n later in the test solutions (with or without C1-). The time course of the swelling phase has been more clearly resolved with the present protocol (Fig. 3). The effects of C1replacement by an impermeant anion have been reproduced. In the absence of external CI-, the cells were much more swollen relative to the isosmotic suspensions. Furthermore, in contrast to the 'control' suspensions, no regulatory volume decrease was detected. The effects of the chloride replacement with gluconate could have reflected not only the removal of CI-, but also the reduced Ca 2+ activity resulting from chelation by the anion (v.i.). In past studies we have tripled the total calcium concentration in gluconatecontaining solutions (Duffey et al., 1986; Leibowich, DeLong and Civan, 1988; Yantorno et al., 1989) to maintain the Ca 2+ activity constant [within about

10-15 % (Duffey et al., 1986)]. To examine this point, we conducted an additional series of four experiments in which the total calcium concentration was increased fivefold in the gluconate media. As demonstrated by Fig. 4, the results were very little changed. Independent of external Ca ~+ activity, replacement of external C1- with an impermeable anion markedly reduced the regulatory response to cell swelling. The potential importance of C1- has also been explored by observing the effect of 5-nitro-2-(3phenylpropylamino)-benzoate (NPPB, Wangemann et al., 1986). At concentrations of ~ 100 FM, NPPB blocks C1- channels of renal tubular cells. Like almost all other inhibitors, NPPB can inhibit more than a single site, particularly at concentrations > 100/ZM (Wangemann et al., 1986). Whole-cell patch-clamp analysis has verified that 1 0 - 1 0 0 FM NPPB blocks C1channels in ODM C1-2/8V40 cells (R. E. Yantorno, D. A. Carrd, M. Coca-Prados, T. Krupin and M.M. Civan, in press). At a concentration of 100/ZM, NPPB prolonged ~ by about 175% (P < 0.001, Table I). We conclude from the experiments with C1 substitution and the C1--channel blocker that loss of cell C1- is also likely to play a major role in the regulatory volume decrease.

186

M.M. CIVAN

ET A L .

150

130

IIO 2

90

70' -10

I

I

0

I0

I 20

I 30 Time (rain)

I 40

I 50

I 60

70

Fro. 6. Effectof O.5 mM acetazoleamide on the RVD of cells suspended in solutions without added bicarbonate. ( • ) Hypotonic; (A) hypotonic + acetazoleamide (C)) isotonic ; (/k) isotonic + acetazoleamide. 160

140

o J20 "5 I00

80

0

I I0

I 20

I 30

I 40

I 50

1I 60

70

Time (rain)

FIG. 7. Effect of DIDS on the RVD, in the absence of added bicarbonate. ( • ) Hypotonic, D1DS; (©) hypotonic; ([]) isotonic, DIDS; (11) isotonic.

Role of HCO3- in the RVD Bicarbonate has been reported to speed the regulatory volume decrease (Yantorno et al., 1989). The bicarbonate stimulation could arise from a direct action of bicarbonate, by participating in the operation of coupled C1-/HCQ- and K÷/H + antiports (Hoffmann, 1987). Alternatively, the bicarbonate could interact indirectly. For example, by altering pH, bicarbonate could affect pH-sensitive K+ and/or CI- channels (Steels and Boulpaep, 1976; Biagi et al., 1981; BelloReuss, 1982; Kubota, Biagi and Giebisch, 1983; Helbig, Korbmacher and Wiederholt, 1987 ; Helbig et al., 1989). We have approached the problem in two ways. First, cells were suspended in test solutions at two different values of pH (7'0 and 8"0). In this series of experiments, three sets of preparations were bathed with bicarbonate-containing solutions, and four were suspended in solutions nominally free of HCQ-. At the lower pH, Atpe~k was prolonged, a plateau was observed at maximum swelling, and r was lengthened by approximately 165 % (Fig. 5, Table I). The results of

this experiment establish the plausibility, but do not prove, that HCO 3- acts indirectly on r by affecting pHsensitive channels. We have also approached the problem more directly by using the following reasoning. If the RVD largely reflects KC1 loss through coupled C1-/HCQ- and K+/H ÷ antiports, we should markedly inhibit the process by removing external HCO3-. Even without adding HCQ-, a small amount of bicarbonate will arise from CO2 generation and hydration by the tissue. This endogenous level of HCQ- must be low. Inhibition of carbonic anhydrase activity by 0"5 mM acetazoleamide had no significant effect on the RVD of cells suspended in HCQ--free medium (Fig. 6). In the absence of external HCQ-, the further addition of 0.5 mM DIDS to block the CI-/HCQ- exchange should have little or no effect. On the other hand, DIDS not only blocks CL /HCQ- exchange, but also inhibits C1channels in many cells (GSgelein, 19 8 8). Therefore, if KC1 loss proceeds largely through pH-sensitive K÷ and CI- channels, and if the inhibitor also blocks C1channels in these cells, administration of DIDS would further slow the RVD. As illustrated by Fig. 7, the

REGULATORY

VOLUME

DECREASE

BY CILIARY

EPITHELIAL

CELLS

187

160 150 140 150 120 '5 >

I10 I00 90 8O -10

~ I 0

±

± I I0

j.

±

•

I 20

I 30 Time (min)

I 40

I 50

I 60

70

FIG. 8. Effect of external Ca2+ withdrawal on the RVD. (1) Hypotonic (0 Ca); (0) hypotonic; (/k) isotonic (0 Ca); (O) isotonic.

150

T~

130

0

90

~

~±

o

6--__

÷ 1

-I0

0

I0

20

30 Time (min)

40

50

6

70

FI6. 9. Effect of 20/AM Gda+ on the response of ve to suspension in hypo- and isotonic medium. ( i ) Hypotonic+Gd; (O) hypotonic ; (ZX) isotonic + Gd ; (O) isotonic. addition of 0.5 mM DIDS to nominally bicarbonatefree solutions increased the degree of swelling and prolonged Atpe~k, Atp~t and z (Table I). The time constant was lengthened by nearly 2 0 0 % ( P < 0-001). We conclude from the experiments presented in Figs 1-7 that the RVD primarily reflects loss of KC1 through parallel K+ and Cl- channels, rather than through coupled C1-/HCO 3- and K*/H* antiports.

Role of Ca2+ in Signal Transduction The mechanisms coupling cell swelling and the regulatory volume decrease are incompletely understood. However, Ca 2+ has been reported to play a role in the RVD of some cells (Hoffmann, 1987). We have begun to examine the possible importance of Ca 2+ in triggering the RVD of the non-pigmented ciliary epithelial ceils. As illustrated by Fig. 8, the nominal removal of external Ca 2+ does affect the RVD without altering the time course of vc for cells suspended in isosmotic solution. The maximum swelling was enhanced by about 25 % (P < 0-005), increasing from an extrapo-

fated swelling of 49 _+2 to 6 2 _ 2 % (Table I). ~" was also prolonged by about 65% (P < 0.05). However, the steady-state volumes were little changed by omitting Ca 2+ from the test suspension medium. Thus, external Ca 2+ appears to participate in the RVD phenomenon, but is not crucial in its expression, either because of the availability of intracellular Ca 2+ stores or because of additional coupling mechanisms. In some tissues, Ca ~+ entry is increased by stretch activation of cation-non-selective channels (Christensen, 1987; Millet and Pickard, 1988 ; Yang and Sachs, 1988). These channels have been reported to be blocked by 1 0 - 2 5 0 # M Gd ~+ (Millet and Pickard, 1988; Yang and Sachs, 1988). To examine this possibility, cells were suspended in the iso- and hypotonic test solutions with and without 20 FM Gd 3+. The inhibitor slightly increased the degree of swelling, but otherwise had little effect (Fig. 9, Table I). We have concluded that the stretch-activated cation-nonselective channels are unlikely to play a major role in triggering the RVD of the ODM C1-2/SV40 cells. We have also examined the effect of reducing Ca 2+ release from the endoplasmic reticulum (Streb et al., 1984). TMB-8 has been used to block mobilization of

188

M.M.

160

T

T

&-A

.L

ET AL

3-

T

±

~

±

80 -I0

CIVAN

I 0

I I0

.1.

I 20

T

I 30

[ 40

I 50

I 60

70

Time (rain)

F i e . 1 0 . Effect o f 7 0 # M T M B - 8 o n t h e R V D . ( • )

Hypotonic

+ TMB- 8 ; (O)

hypotonic

; (&)

isotonic + TMB- 8 ; (O)

isotonic.

170

150

o~ v

130

IlO --

L

90

I

70

I

I

I

I

I

6O B

E

v o

[

50 4O 3O

5'

20

o. >.,

I0 -

-I0

I 0

I

I

I

I

I

I

I0

20

30

40

50

60

70

Time (rain)

FIG. 11. Effect of trifluoroperazine (TFP) on the response of vo to suspension in hypotonic medium. A, In the presence of 20-25 #M, a transient undershoot in v0 was noted in the isotonic suspension. B, The undershoot was abolished by suspending the cells in 10 ~M TFP, although the vo of the phenothiazine cells was about 5 % lower than that of the control cells in isotonic medium. The data are presented as the point-for-point differences between suspension in hypotonic versus isotonic solutions, with and without the TFP. A: ( • ) hypotonic + 20-25 #M TFP: ( 0 ) hypotonic; (©) isotonic', (/k) isotonic + 20-25/~M TFP. B : (O) hypotonic- isotonic ( + l 0 #M TFP) ; (C)) hypotonic- isotonic. intracellular Ca 2+ in other preparations (Chiou and Malagodi, 1 9 7 5 ; Mix, Dinerstein and Villereal, 1 9 8 4 ; Sawamura, 1 9 8 5 ; Yada, Rose and Loewenstein, 1 9 8 5 ; Rose, Yada and Loewenstein, 1986). In these experiments, we have followed the approach of Rose, Yada and Loewenstein (1986), applying only a 70-#M concentration. Substantially higher concentrations of TMB-8 m a y exert a direct inhibitory effect on PKC

(Kojima, Kojima and Rasmussen, 1 9 8 5 ; Sawamura, 1985). At 70 #M, TMB-8 had no significant effect on cells suspended in isosmotic solution (Fig. 10). However, in the hypotonic suspensions, TMB-8 increased the degree of swelling (P < 0.01) and prolonged &tplat (Table I). The prolongation of ~ was not significant at the 0-05 probability level. In the presence of the inhibitor, v c remained significantly higher than in the

R E G U L A T O R Y V O L U M E DECREASE BY C I L I A R Y E P I T H E L I A L CELLS

control suspensions, even after 60 min of analysis (Fig. 10). These results suggest that mobilization of Ca 2+ from intracellular stores plays a significant role in triggering the regulatory volume decrease. Calcium's role as a second messenger is frequently mediated by Ca2+-binding proteins. For example, binding of Ca 2+ to calmodulin is thought to produce a conformational change in the protein, leading to activation of a wide range of cellular functions (Klee and Newton, 1985). A number of pharmacological agents, such as the phenothiazines, have been used to block calcium/calmodulin stimulation of target molecules. Like most inhibitors, the phenothiazines may act at additional sites (Pollard et al., 1982; Ben-Gigi, Polacheck and Eilam, 1988). Application of 20-25 ~M trifluoperazine (TFP) produced marked changes in the time course of the RVD [Fig. 11 (A), Table I]. The drug increased ~.... and prolonged Atpe~k. Atpl~t and 7 (P < 0"001). However, the changes induced by TFP appeared more complex than those produced by the other experimental manipulations of this study. In particular, 20-25 #M TFP caused a transient fall in the volume of cells suspended in isosmotic solution. Furthermore, the volume measured at the final time point (t = 60 rain) was no different from that of the control suspension. These observations suggested that TFP might be exerting more than one effect at concentrations of 20-25 #M. For this reason, the experiment was repeated at a lower concentration (10 #M). The initial transient response of the isotonic suspension was thereby eliminated, and the TFP still produced an increase in the duration of the plateau and a prolongation of 7 by about 250% (P < 0-01, Table I). Even at a concentration of 10 #M, TFP produced a consistently lower vo in the isosmotic suspension than that of the paired isosmotic control suspensions. For this reason, the data have been plotted in Fig. 11 (B) as the point-for-point differences between the relative volumes measured in hypotonic and in isotonic media. With this alternative presentation, the marked inhibition exerted by TFP is clearer. After 60 min, the llVD of the TFP-treated cells was still incomplete.

4. Discussion

Ionic Basis Underlying the RVD Phenomenon

Biological cells in suspension uniformly display a regulatory volume decrease (RVD) following anisosmotic swelling in suspension (Siebens, 1985; Gilles, 1987; Hoffmann, 1987). The RVD of other cells has been thought to reflect the operation of at least five different mechanisms: (1) release of non-electrolytes (such as amino acids and taurine); (2) increased activity of the Na,K-activated pump; (3) loss of KCI from the cell through an electroneutral symport; (4) loss of KCI through coupled C1-/HCQ and K+/H ÷ antiports; and (5) loss of KC1 through parallel K÷ and

189

C1- channels. As discussed below, the current data, taken together with published information (Yantomo et al., 1989) exclude all but the last possibility. (1) Non-electrolyte release can scarcely be contributing significantly to the RVD of the non-pigmented ciliary epithelial cells in view of the central importance of K+ and C1- in the expression of this phenomenon. In particular, the regulatory response is slowed by: reducing the chemical gradient favoring K÷ loss (Farahbakhsh and Fain, 1988; Yantorno et al., 1989), and by inhibiting K÷ channels with either Ba 2÷ (Fig. 1, Table I; Farahbakhsh and Fain, 1988; Yantorno et al., 1989) or quinidine (Fig. 2, Table I). Similarly, external C1- replacement with the impermeable anion gluconate completely blocks the RVD (Fig. 3, Table I: Yantorno et al., 1989) and application of the C1--channel inhibitor NPPB markedly prolongs the time constant (r) of the response (Table I). (2) The Na,K-activated ATPase cannot be playing a direct significant role, since inclusion of an inhibitor (strophanthidin) in the suspending medium does not (acutely) affect the RVD (Yantorno et al., 1989). Cardiotonic steroids do affect the RVD indirectly by reducing the intracellular content of K+ during periods of preincubation (Farahbakhsh and Fain, 1988; Yantorno et al., 1989). (3) K÷/CI- symport exit is unlikely to be playing a role. Ascribing a major contribution to this mechanism is not readily reconciled with the observed responses of the RVD to: K+-channel blockers (Ba 2÷ and quinidine), the C1--channel blocker NPPB, and changes in the HCQ concentration (Yantorno et al,,~ 1989). (4) On the basis of our previous studies (Yantorno et al., 1989), we had been unable to exclude the possibility that substantial amounts of KC1 were lost through the operation of coupled CI-/HCQ- and K+/H+ antiports activated by cell swelling. The data in Figs 5 and 7 and Table I render this possibility highly unlikely. The speeding of the RVD produced by external bicarbonate is likely to reflect the demonstrated pH dependence of 7 (Fig. 5). This phenomenon probably arises from the pH dependence of K÷ channels. Electrophysiological evidence for this concept has been reported for a number of preparations (Steels and Boulpaep, 1976; Biagi et al., 1981; BelloReuss, 1982; Kubota, Biagi and Giebisch, 1983; Helbig, Korbmacher and Wiederholt, 1987), including ODM C1-2/SV40 cells (Helbig et al., 1989). Futhermore, the data in Fig. 7 cannot be ascribed to the operation of coupled antiports. With the omission of bicarbonate from the suspending medium, C1-/HCQmust have been proceeding at a much reduced rate. Under such circumstances, the further inhibition of the putative antiport by DIDS should have had very little effect on the RVD. The marked inhibition actually observed (Fig. 7, Table I) strongly suggests that an additional DIDS-sensitive mechanism was inhibited, namely the C1- channels (GSgelein, 1988). It will be appreciated that additional, more complex models of

190

coupled antiport transport are not necessarily excluded. For example, the data are still compatible with the operation of DIDS-sensitive CI- exchange either with OH- or with an unidentified organic anion (such as formate) recirculating after leaving the cells in the acid form. (5) The possibility that the RVD primarily reflects the loss of KC1 t h r o u g h parallel K ÷ and C1- channels is consistent with the data thus far reported for nonpigmented ciliary epithelial cells in suspension (Yantorno et al., 1 9 8 9 ; present study) and in intact rabbit ciliary processes (Farahbakhsh and Fain, 1988). This interpretation is strongly supported by the observation that the RVD can be inhibited by blocking K ÷ channels with Ba 2+ (Fig. 1, Table I; F a r a h b a k h s h and Fain, 1 9 8 8 ; Yantorno et al., 1989) and quinidine (Fig. 2, Table I), and C1- channels with NPPB (Table I). This conclusion is further supported by recent patch-clamp analyses of the response of whole-cell currents to lowering the external osmolality (Civan et al., 1 9 9 0 ; R. E. Yantorno, D.A. CarrY, M. Coca-Prados, T. Krupin and M. M. Civan, in press). Transduction Mechanisms Triggering the R VD Ca 2+, calmodulin and arachidonic acid metabolism m a y mediate the triggering of the RVD by anisosmotic swelling of other cells (Hoffmann, 1 9 8 7 ; Christensen et al., 1988). The data of the present study suggest that Ca ~+ m a y be one second messenger of this transduction in ODM C1-2/SV40 cells as well. Omitting Ca ~÷ from the external m e d i u m (Fig. 8, Table I) delays the RVD to a limited extent. Blocking mobilization of intracellular stores of Ca ~+ exerts a more p r o n o u n c e d inhibition (Fig. 10, Table I). Furthermore, in the presence of the calcium/calmodulin inhibitor trifluoperazine, the RVD was slowed and incomplete (Fig. 11, Table I). The phenothiazines have also been reported to inhibit the RVD in other preparations. W h e t h e r Ca 2+ is the sole mediator, or whether other factors such as intracellular pH and arachidonic acid metabolites play a role, is yet to be resolved.

Acknowledgements We are grateful to Professor Carol Deutsch for graciously permitting us to use her Coulter counter for the current studies, and to Professor Rainer Greger for his kindness in providing us with a generous sample of NPPB. This work was supported in part by grant EYO8343 from the National Institutes of Health. References Bello-Reuss, E. (1982). Electrical properties of the basolateral membrane of the straight portion of the rabbit proximal renal tubule. ]. Physiol. (Lond.) 326, 49-63. Ben-Gigi, G., Polacheck, I. and Eilam, Y. (1988). In vitro synergistic activity of ketoconazole with trifluoperazine and with chlorpromazine against medically important yeasts. Chemotherapy 34, 96-100.

M . M . C I V A N ET AL.

Biagi, B., Kubota, T., Sohtell, M. and Giebisch, G. (1981). Intracellular potentials in rabbit proximal tubules perfused in vitro. Am. J. Physiol. 240, F200-F210. Chiou, C.Y. and Malagodi, M.H. (1975). Studies on the mechanism of a new Ca 2+ antagonist, 8-(N,N-diethylamino)octyl 3,4, 5-trimethoxybenzoate hydrochloride in smooth and skeletal muscle. Br. ]. Pharmacol. 53,279-85. Christensen, O. (1987). Mediation of cell volume regulation by Ca 2+ influx through stretch-activated channels. Nature (Lond.) 330, 66-8. Christensen, S., Hoffmann, E.K., Saermark, T.S. and 8imonsen, L. O. (1988). Inositol triphosphate may be a second messenger in regulatory volume decrease in Ehrlich mouse ascites-tumour cells. J. Physiol. (Lond.) 403, 109P. Civan, M. M., Coca-Prados, M., Krupin, T., and Yantorno, R. E. (1990). Whole-cell patch clamp analysis of ciliary epithelial cells during anisosmotic cell swelling. FASEB J. 4, 549 (Abst.). Cole, D.F. (1966). Aqueous humor formation. Doc. Ophthalmol. 21, 116-238. Duffey, M.E., Kelepouris, E., Peterson-Yantorno, K. and Civan, M.M. (1986). Micro-electrode study of intracellular pH in frog skin: dependence on serosal C1-. Am. J. Physiol. 251, F468-F474. Farahbakhsh, N.A. and Fain, G. L. (1988). Volume regulation of non-pigmented cells from ciliary epithelium. Invest. Ophthalmol. Vis. Sci. 28, 934-44. Germann, W.J., Ernst, S.A. and Dawson, D.C. (1986). Resting and osmotically induced basolateral K conductances in turtle colon. ]. Gen. Physiol. 8 8 , 2 5 3 - 7 4 . Germann, W. J., Lowy, M. E., Ernst, S. A. and Dawson, D. C. (1986). Differentiation of two distinct K conductances in the basolateral membrane of turtle colon. ]. Gen. Physiol. 88, 237-51. Gilles, R. (1987). Volume regulation in cells of euryhyaline invertebrates. Curr. Top. Membr. Transp. 30, 205-47. G6gelein, H. (1988). Chloride channels in epithelia. Biochim. Biophys. Acta 947, 521-47. Grinstein, S., Cohen, S., Goetz, J. D., Rothstein, A., Mellors. A. and Gelfand, E.W. (1985b). Activation of the Na+-H + antiport by changes in cell volume and by phorbol esters; possible role of protein kinase. Curr. Top. Membr. Transp. 26, 115-34. Grinstein, S., Goetz, J. D., Cohen, S., Furuya, W., Rothstein, A. and Gelfand, E.W. (1985a). Mechanism of regulatory volume increase in osmotically shrunken lymphocytes. Mol. Physiol. 8, 185-98. Helbig, H., Korbmacher, C. and Wiederholt, M. (1987). K÷conductance and electrogenic Na+/K + transport of cultured bovine pigmented ciliary epithelium. ]. Membr. Biol. 99, 173-86. Helbig, H., Korbmacher, C., Wohlfarth, J., Coca-Prados, M. and Wiederholt, M. (1989). Electrical membrane properties of a cell clone derived from human nonpigmented ciliary epithelium. Invest. Ophthalmol. Vis. Sci. 30, 882-9. Hoffmann, E. K. (1987). Volume regulation in cultured cells. Curr. Top. Membr. Transp. 30, 125-80. Hoffmann, E. K., Lambert, I. H. and Simonsen, L. O. (1986). Separate, Ca2+-activated K+ and C1- transport pathways in Erlich ascites tumor cells. ]. Membr. Biol. 91, 227-44. Hogan, M.J., Alvarado, J.A. and Weddell, J.E. (1971). Histology of the Human Eye: An Atlas and Textbook. W. B. Saunders: Philadelphia. Klee, C.B. and Newton, D.L. (1985). Calmodulin: an overview. In: Control and Manipulation of Calcium Movement (Ed. Parratt, ]. R.). Pp. 131-46. Raven Press: New York. Kojima, I., Kojima, K. and Rasmussen, H. (1985). Mechanism of inhibitory action of TMB-8 [8-(N,N-diethyl-

R E G U L A T O R Y V O L U M E DECREASE BY C I L I A R Y E P I T H E L I A L CELLS

amino) octyl- 3,4,5 -trimethoxybenzoate] on aldosterone secretion in adrenal glomerulosa cells. Biochem. ]. 232, 87-92. Kregenow, F.M. and Caryk, T. (1979). Co-transport of cations and CI during the volume regulatory responses of duck erythrocytes. Physiologist 22, 73. Kubota, T., Biagi, B. A. and Giebisch, G. (1983). Effects of acid-base disturbances on basolateral membrane potential and intracellular potassium activity in the proximal tubule of Necturus. ]. Membr. Biol. 73, 61-8. Lee, S.C., Price, M., Prystowsky, M.B. and Deutsch, C. (1988). volume response of quiescent and interleukin 2-stimulated T-lymphocytes to hypotonicity. Am. ]. Physiol. 254, C286-C296. Leibowich, S., DeLong, J. and Civan, M. M. (1988). Apical Na + permeability of frog skin during serosal C1replacement. ]. Membrane Biol. 102, 121-30. Martin-Vasallo, P., Ghosh, S. and Coca-Prados, M. (1989). Expression of Na,K-ATPase alpha subunits isoforms in the human ciliary body and cultured ciliary epithelial cells. J. Cell. Physiol. 1 4 1 , 2 4 3 - 5 2 . Millet, B. and Pickard, B. G. (1988). Gadolinium ion is an inhibitor suitable for testing the putative role of stretchactivated ion channels in geotropism and thigmotropism. Biophys. J. 53, 155a. Mix, L.L., Dinerstein, R.J. and Villereal, M.L. (1984). Mitogens and melittin stimulate an increase in intracellular free calcium concentration in human fibroblasts. Biochem. Biophys. Res. Commun. 119, 69-75. Pollard, H. B., Creutz, C. E., Fowler, V., Scott, J. and Pazoles, C.J. (1982). Calcium-dependent regulation of chromaffin granule movement, membrane contact, and fusion during exocytosis. Cold Spring Harbor Syrup. Ouant. Biol. 46, 819-34. Rose, B., Yada T. and Loewenstein, W.R. (1986). Down regulation of cell-to-cell communication by the viral src gene is blocked by TMB-8 and recovery of communication is blocked by vanadate. J. Membr. Biol. 94, 129-42. Sawamura, M. (1985). Inhibition of protein kinasc C activation by 8-N,N-(diethylamino) octyl-3,4,5-tri-

191

methoxy-benzoate (TMB-8), an intracellular Ca ~+ antagonist. Kobe J. Med. ScL 31, 221-32. Siebens, A. W, (1985). Cellular volume control. In The Kidney: Physiology and Pathophysiology (Eds Seldin, D. W. and Giebisch, G.). Pp. 91-115. Raven Press: New York. Steels, P. S. and Boulpaep, E. L. (1976). Effect ofpH on ionic conductance of the proximal tubule epithelium of Necturus and the role of buffer permeability. Fed. Proc. 35,465. Streb, H., BayerdOrffer, E., Haase, W., Irvine, R.F. and Schulz, I. (1984). Effect of inositoI-1,4,5-triphosphate on isolated subcellular fractions of rat pancreas. ]. Membr. Biol. 8 1 , 2 4 1 - 5 3 . Tang, C.-S. R., Peterson-Yantorno, K. and Civan, M.M. (1989). Coupling of volume and Na + transport in frog skin epithelium. Biol. Ceil 66, 183-90. Wangemann, P., Wittner, M., Di Stefano, A., Englert, H. C., Lang, H.J., Schlatter, E. and Greger, R. (1986). C1 channel blockers in the thick ascending limb of the loop of Henle. Structure activity relationship. Pflfigers Arch. 407 (Suppl. 2), S128-S141. Yada, T., Rose, B. and Loewenstein, W.R. (1985). Diacylglycerol downregulates junctional membrane permeability. TMB-8 blocks this effect. ]. Membr. Biol. 88, 217-32. Yang, X. C. and Sachs, F. (1988). Stretch-activated (SA) ion channel in Xenopus oocytes: permeation and block. Biophys. J. 53,412a. Yantorno, R.E., Carr4, D. A., Coca-Prados, M., Krupin, T. and Civan, M. M. (1992). Whole-cell patch clamping of ciliary epithelial cells, during anisosmotic swelling. Am. J. Physiol. : Cell (in press). Yantorno, R.E., Coca-Prados, M., Krupin, T. and Civan, M.M. (1989). Volume regulation of cultured, transformed, non-pigmented epithelial cells from human ciliary body. Exp. Eye Res. 4 9 , 4 2 3 - 3 7 . Yantorno, R.E., Krupin, T. and Civan, M.M. (1987). Selective and non-selective ion channels in intact rabbit ciliary epithelium. Invest. Ophthalmol. Vis. Sci. 28 (Suppl.), 374.