JOURNAL OF CELLULAR PHYSIOLOGY 143:13-20 (19901

Extracellular Calcium Affects the Membrane Currents of Cultured Human Keratinocytes THEODORA M. MAURO,* PAMELA A. PAPPONE,

AND

R. RIVKAH ISSEROFF

Department of Dermatology School of Medicine (T.M.M., R.R.I.) and Department of Animal Physiology (P.A.P), University of California, Davis, Davis, California 956 I6 Electrophysiologic properties of cultured human keratinocytes were studied using the patch voltage-clamp technique. Undifferentiated, proliferative keratinocytes grown in low Ca2+ medium had an average resting membrane potential of -24 mV. Voltage-clampexperiments showed that these cells had two membrane ionic currents: a large voltage-independent leak conductance, and a smaller voltagedependent CI- current that activated with depolarization. Increasing the extracellular Ca2' concentration from 0.15 to 2 mM resulted in a doubling of the magnitude of the voltage-gated current and a shift in current activation to more negative potentials. Since levels of extracellular Ca'+ can alter the morphology and differentiation state of keratinocytes, the finding of a Ca2 +-activated CIcurrent in these cells suggests a role for this conductance in the initiation of differentiation As the keratinocytes of the epidermis in vivo move from the basal layer to the uppermost cornified layer, they acquire specific morphologic and biochemical characteristics that collectively constitute the differentiated phenotype. In vitro, many of these characteristics of differentiation can be induced by manipulation of culture conditions. Most notably, extracellular levels of the cation Ca2+ dramatically alter the differentiation state of mouse keratinocytes (Hennings et al., 1980). In human keratinocytes, levels of extracellular calcium also alter the cellular phenotype: at levels of -500

Fig. 2. Total membrane ionic currents. A Membrane currents from a patch-clamped keratinocyte recorded in the whole cell configuration in the presence of 0.15 mM external Ca" showing the large linear leak current. The cell membrane potential was stepped from the hold-

ing potential of -60 mV to potentials between -80 mV and + 60 mV in increments of 20 mV. Duration of the voltage steps was 200 msec. B:Peak current-voltage relation for the currents shown in A showing the ohmic nature of the leak.

RESULTS An example of neonatal human keratinocytes cultured for 3 days in low Ca2' medium (0.15 mM) and maintained under our standard recording conditions is shown in Figure 1. Keratinocytes plated on glass proliferate less vigorously than when grown on plastic, but do still grow and divide. The morphology of the cells is normal for cells maintained in low Ca2. medium, demonstrating large intercellular spaces, and the cells remain viable, as judged both by their appearance and by their ability to exclude trypan blue for more than 1 hr under these conditions (Fig. 1).Both isolated keratinocytes and keratinocytes in monolayer clusters like that shown in Figure 1 were used in patch-clamp experiments. Currents recorded from cells growing in clusters were indistinguishable from those of isolated cells, indicating that currents due to electrical coupling of cells do not contribute significantly to the patchclamp currents. Current-clamp measurements made immediately, after the whole-cell configuration was achieved, produced resting membrane potentials that averaged -24 mV ( 5 2 mV, n = 12, range - 16 to -36), in our normal bicarbonate-containing solutions. One of these cells was part of a two-cell cluster having a resting membrane potential of -32 mV. When cells in similar external solutions with nonbicarbonate buffers were equilibrated with room air rather than with 95%

02/5% COz, the resting membrane potentials were close to 0 mV, with no cell showing a potential negative to -8 mV under these conditions. Thus, our cultured cells seem to be normal and viable when perfused with oxygenated, bicarbonate-buffered solutions but require these conditions to maintain a negative membrane potential.

Membrane Currents in Keratinocytes Figure 2 shows whole-cell patch-clamp recordings from a n undifferentiated keratinocyte grown in low Ca2+ medium for 2 days recorded in the presence of oxygenated bicarbonate-buffered solution containing 0.15 mM Ca2i. All keratinocytes recorded from in these experiments had relatively leaky cell membranes, such as the cell shown in Figure 2. This is reflected in the low cell resistances measured, which averaged 0.2 GR (20.06 GlL, n = 19) for cells with an average membrane capacitance of 22.1 pF (54.3pF, n = 19). Proliferating keratinocytes had two major membrane ionic conductances. Figure 2A shows the total membrane ionic current in response to 200-msec hyperpolarizing or depolarizing voltage-clamp pulses. The current records are dominated by a large voltageindependent leak conductance. The magnitude of the leak conductance averaged 14.0 nS ( 2 3 nS, n = 16) and

MAIJRO ET AL.

16

B

A

I (ph) 500

400

300

200

-80 -60 -40’ -20

I 62.5

I

E (mV>

50 ms

Fig. 3. Voltage-gated outward currents are carried by C1 . A. The upper records show currents recorded in normal C1- containing external solution using the same voltage protocol as Figure 2. Linear components of the current have been subtracted from the records using the Pi4 procedure as described under Methods to show the voltagedependent component of the current in isolation. The lower tracing shows currents from the same cell aftcr changing the external solution to one in which the C1- has been replaced with glutamate. Replacement of external C1 abolishes the outward currents normally

associated with depolarization, suggesting that they are due to the inward movement of CI ions. B. The effects of C1- replacement are reversible. The bathing solution was switched back and forth between chloride and glutamate. Shown are peak currents a s a function of membrane potential measured in a single cell. Curves 1 and 3 are from currents measured in chloride solution and 2 and 4 in glutamate solution. The numbers (1-4) indicate the sequence in which the bathing solutions were applied.

constituted approximately 80% of the total membrane conductance a t depolarized membrane potentials. The leak conductance was ohmic (Fig. 2B); it was not affected by increasing the extracellular Ca2 concentration from 0.15 mM to 2.0 mM, by replacing extracellular C1- with glutamate or MSA, by replacing 100 mM of the extracellular Na’ with K t , or by adding 10 mM TEA to the extracellular solution. Nor was the resting conductance affected by changing the composition of the pipette solution. Currents obtained with different anions (C1 , F-), cyclic nucleotides [cyclic adenosine monophosphate (CAMP)or cyclic guanosine monophosphate (cGMP)], a n ATP-generating system, or with Ca2 ’ buffered to low levels with EGTA were similar to those measured using our standard pipette solution. We believe that this high leak conductance is normal for proliferating cultured keratinocytes for several reasons: (1) the results discussed above indicate that ke-

ratinocytes are viable under our recording conditions; (2) we never saw any evidence of keratinocytes with “tighter” cell membranes, and the leakiness was always apparent from the instant we achieved the whole cell recording configuration; and (3) cells patchclamped using the almost noninvasive “perforated patch” technique (Horn and Marty, 1988)had the same property. The second main component of the membrane current in keratinocytes is a n outward current activated by depolarization. The records shown in Figure 3A display currents recorded from the same cell as Figure 2, with the linear components of the current record subtracted from the traces as described under Methods. The outward current activated with depolarizations positive to -50 to -40 mV showed little or no inactivation during depolarizing pulses and deactivated rapidly following membrane repolarization. Chloride cur-

+

MEMBRANE CURRENTS IN HUMAN KERATINOCYTES

17

A B

I (PA)

2oool f

0.15 mM Calcium

2

Iy

1

500 pA

2 niM Calcium

50 MS

--r-40

%

'0

itiM

Ca

I

I

1

I

20

40

.( 60

E (mV)

Fig. 4. Increasing the external Ca'+ concentration increases the voltage-gated current. A. Membrane currents recorded in normal external solution with 0.15 mM Ca2+ (top) or 2 min after switching to solution containing 2 mM Ca'+ (bottom). There is an almost fourfold

increase in the voltage-gated currents in the high calcium solution. Pulse protocols as in Figures 2-3, linear components of the currents have been subtracted from the records. B. Peak current-voltage relations measured from the records in A.

rents at +30 mV averaged 411.5 pA ( ~ 4 3 . 3n, = 13, range 217-706) for single cells and 438.0 pA (t119.2, n = 3, range 309-676) for cells in clusters. The voltagedependent current contributed 17%(+6%, n = 8) of the total current in cells patch-clamped using the standard whole-cell configuration and 10% (*2%, n = 4) of the current in cells clamped using the "perforated patch" technique. The voltage-gated outward current seemed t o be carried by an influx of C1- ions. Figure 3 shows the effects of replacing all the extracellular C1- with glutamate. Removal of C1- from the external solution reduced the outward voltage-dependent current amplitude measured at + 30 mV by 65% in this cell and by an average of 86%(+-7%,IZ = 7). Replacement of extracellular C1with MSA had similar effects on the outward current, reducing its amplitude by 84% (+6%,n = 3). To test whether residual current after glutamate substitution was carried by K' ions, TEA was added, or extracellular K + was increased. The addition of 10 mM TEA to the extracellular solution had no effect on the current magnitude, nor did substitution of 100 mM Na+ in the

bathing solution with K' . These results all indicate that the voltage-gated outward current is carried by C1- ions. Since the remaining outward current is not carried by K + , it is most likely that the residual current results from glutamate or HCO, passing through C1- channels (Gray et al., 1989; Stea and Nurse, 1989). It was not possible to measure reversal potentials as a function of C1- concentration directly because the current showed strong outward rectification. Effects of extracellular Ca2+ ions. Since Ca2+ ions affect the differentiation of cultured keratinocytes, we investigated the effects of increased Ca2+ concentrations on the membrane currents of patch-clamped keratinocytes. All but one of these experiments were done on single isolated cells to exclude the possibility that increasing Ca2+would result in increased current secondary to increased electrical coupling of cells (reviewed by Spray and Bennett, 1985). Figure 4 shows membrane currents recorded from a keratinocyte in low Ca2' (0.15 mM, Fig. 4A, upper tracing) solution and in the same cell 2 min after switching to a solution containing 2 mM Ca'+ (Fig. 4A, lower tracing). There

18

MAURO ET AL.

is an almost fourfold increase in the voltage-gated currents in the high Ca" solution. Increasing the external Ca2+concentration from 0.15 mM to 2 mM resulted in an average increase in the voltage-gated current magnitude of 241% (*61%, n = 10, range 17-411%). C1- currents from the one cell in contact with other cells increased 188%.In addition, the currents activate at more negative membrane potentials in high Ca2 , as shown in the current-voltage plot of Figure 4B. The direction of this shift in voltage dependence is in the opposite direction to that expected from Ca2 screening of negative surface potentials and so must be attributable to some other action of Ca2+ ions. The effects of added Ca2 were apparent immediately (within 5-15 sec) upon changing the bathing solution and were maintained for as long as 45 min in the presence of high Ca2+.The increased current in high Ca2+ solution seemed to be due to a selective increase in the voltage-gated C1- current. In the high Ca' solution, replacement of extracellular CIF with glutamate resulted in an 84% (&6%,n = 3) decrease in the current amplitude, similar to the effect seen in low Ca2+. In addition, the increased current in high Ca2+ was insensitive to TEA (n = 3). These results suggest that increasing the extracellular Ca2+ concentration in the solution that bathes cultured keratinocytes modulates their C1K conductance by increasing its magnitude and shifting its voltage dependence such that the current is activated at more negative membrane potentials. Experiments using Ca2+ channel blockers were not performed, since the initial rise in intracellular Ca' ' seen in response to increased extracellular Ca2+ is not blocked by these agents (J.A. Fairley, personal communication), and these compounds do not affect keratinocyte differentiation (Hennings et al., 1983a). +

+

+

+

DISCUSSION The electrophysiologic properties of human neonatal keratinocytes maintained in low Ca" (0.15 mM) medium were examined. Keratinocytes so maintained are proliferative and appear to be morphologically undifferentiated, although a small fraction do express some markers of differentiation (Dover and Watt, 1987). Under these conditions, a voltage-independent, nonspecific leak comprised 80% of the membrane conductance of the keratinocytes. The remaining membrane current is due to a specific C1- conductance that can be activated by depolarization and by increasing the external Ca2 concentration. The high leak conductance was a consistent feature of keratinocyte membranes. It appeared immediately upon going to the whole-cell configuration, using either suction to break into the cell or when permeabilizing the patch using the "slow-patch" technique (Horn and Marty, 1988). Since our cells appeared viable as assessed by their appearance, trypan blue exclusion, and their ability to continue to grow and divide when maintained in culture under the conditions used for these experiments, we believe that this large leak conductance may be normal for undifferentiated keratinocytes. Some cells become leaky in low Ca2+ solution because K' channels can lose their selectivity and voltage dependence when Ca2+ is removed from the external solutions (Armstrong and Lopez-Barneo, 1987). It is unlikely, however, that, in the case of ke+

ratinocytes, the leak conductance is passing through K' channels, since the size of the leak conductance and its response to 10 mM TEA, a K+-channel blocker, did not change when the external Ca2+ concentration was increased. Resting membrane potentials of these cells were low, averaging only -24 mV. To our knowledge, resting membrane potentials of cultured keratinocytes have not been reported previously. However, intracellular resting membrane potentials in this range have been reported in other epithelial cells: -34.6 mV in sweat duct cells (Reddy and Quinton, 1987) and -25.6 mV in cultured human lens epithelial cells (Stewart et al., 1988). There are a number of possible mechanisms by which extracellular Ca2+could regulate C1- conductance. Increased intracellular Ca2' (Fairley et al., 1988; Hennings et al., 1989) could bind directly to the C1- channel and modulate its conductance. Alternatively, activated protein kinase C resulting from a rise in extracellular Ca2+ (Ziboh et al., 1984) could change the C1- channel conductance by phosphorylating sites on the channel or a regulatory protein, as seen in airway epithelium (Barthelson et al., 1987; Li et al., 1989). Extracellular Ca2+ could act directly by screening the negative surface potentials present at the outer portion of the C1- channel. This is unlikely in keratinocytes, as raising extracellular Ca2 shifts the C1 current activation to more negative potentials (Fig. 4B),a direction opposite to what would be seen if Ca2 ' screening were taking place. The physiologic role of the voltage and Ca'+-activated C1- current observed in keratinocytes is unclear. The magnitude of the voltage-sensitive C1- current doubled in response to raising extracellular Ca2 from 0.15 mM to 2.0 mM, and the current became active at potentials closer to the resting potential of the cell. Both changes will tend to increase C1F fluxes through the keratinocyte membrane in the presence of increased Ca2+.The magnitude of the C1- current response to extracellular Ca' increased within a range of 17-411% (average 241%). This variation is not surprising, given the marked variation among individual non-neoplastic keratinocytes in the rise of intracellular Ca2 noted in response to increased extracellular Ca2 ' (Hennings et al., 1989). Since intracellular C1- levels in keratinocytes are unknown, the normal direction of C1- movement through the voltage-gated conductance cannot be predicted. The finding of a Ca"-activated CIF current in keratinocytes does not prove that this C1- current has a causal role in keratinocyte differentiation. However, several characteristics of this conductance suggest that it may be linked to the differentiation of these cells. First, this current is increased at levels of extracellular Ca2 that lead to keratinocyte differentiation (Hennings et al., 1983a). Second, the C1- current increases rapidly in response t o extracellular Ca2+,which corresponds to the rapid increases in intracellular Ca2+and IP, seen after raising extracellular Ca" (Fairley et al., 1988; Hennings et al., 1989; Tang et al., 1988; Jaken and Yuspa, 1988). Third, the increase in C1- conductance continued for at least 45 min, in agreement with the sustained increase in intracellular Ca2 ' seen with Ca' ' -sensitive dye studies (Hennings et al., 1989). +

+

+

+

+

19

MEMBRANE CURRENTS IN HUMAN KERATINOCYTES

Ca2+-activated Cl- currents have been described in various other activated C1- epithelial cells, such as rat lacrimal gland (Evans and Marty, 1986), canine tracheal epithelium (Welsh, 1983), as well as human (Welsh, 1986; Frizzell et al., 1986a) tracheal epithelium, although in these cells the chloride current appears to be associated with secretory activities (reviewed by Frizzell et al., 1986b). The keratinocytes of the epidermis form a stratified squamous and nonsecretory epithelium. However, since C1- channel activity is important in regulating epithelial cell volume (Hazama and Okada, 19881, and an increase in keratinocyte volume may be one of the earliest signals of keratinocyte differentiation (Watt, 19881, Ca2 ' activated C1- entry into the keratinocyte may be a trigger mechanism for terminal differentiation. Alternatively, if the C1- conductance, Na/C1cotransporter, and Na/K ATPase activity are linked in the keratinocyte, as they are in tracheal epithelium (Welsh, 1983; reviewed by Ziyadeh and Agus, 1988), Ca2 - induced increases in C1- conductance could increase intracellular levels of Na+ and K ' in response to increases in extracellular Ca2 ' levels, concomitant with induction of the differentiated phenotype. Thus, althou h the in' -induced crease in C1- conductance observed with Ca+ differentiation may be a co-observed phenomenon, there are two possible mechanisms by which the Ca2+activated C1- current of keratinocytes may function to induce differentiation in these cells. Finally, possible interactions of the C1- conductance and the phosphoinositidelprotein kinase C signal transduction system should be examined. Increases in extracellular levels of Ca2+sufficient to generate a differentiative response to keratinocytes induce both a rapid (30-sec) and sustained (30-min to 4-hr) increase in intracellular levels of Ca2' (Fairley et al., 1988; Hennings et al., 1989). Since the initial intracellular rise in Ca2+ is not blocked by Ca2 ' -channel blockers (J.A. Fairley, personal communication) it is probably the result of release of intracellular Ca2+stores by the second messenger inositol triphosphate UP3), which has been shown to be rapidly generated in keratinocytes induced t o differentiate by increases in intracellular Ca2+ (Tang et al., 1988; Jaken and Yuspa, 1988).The driving force for the more sustained phase of increased intracellular Ca2+may be provided by Ca2+activated C1- conductance. Such a role for C1- currents has been suggested in rat mast cells, in which the C1current can clamp the membrane to negative values and drive a sustained calcium influx in agonist-stimulated cells (Penner et al., 1988). The hydrolysis of phosphatidylinositol by keratinocytes treated with increased levels of extracellular Ca2' yields not only a rapid increase in IP, but also an additional hydrolytic product, diacylglycerol (Ziboh et al., 1984) which acts as an activator of protein kinase C (Nishizuka, 1984).A number of reports now present evidence implicating the activation of protein kinase C as an early event in keratinocyte differentiation (Dunn et al., 1983; Jeng et al., 1985; Snoek et al., 1987; Isseroff et al., 1989), although the precise sequence of events resulting in terminal differentiation is unknown. Since activation of protein kinase C has been shown to increase C1- conductance in airway epithelium (Barthelson et al., 1987; Li et al., 1989), it is reasonable to suggest that analo+

gously, in the keratinocyte, Ca2 -induced PI hydrolysis and subsequent diacylglyceride formation activate protein kinase C and enhance C1- conductance, thereby amplifying the initial differentiative signal. These possible interactions await further exploration. However, the finding of a voltage-sensitive C1- channel in keratinocytes that can be modulated by extracellular Ca"', a cation that plays a key role in regulating terminal differentiation in this cell, suggests an important physiologic role for this ion conductance. +

ACKNOWLEDGMENTS This work was supported by grants AR34766 (PAP), R01 AR39031, and 1K04-AR01803(RRI) from the National Institutes of Health. We thank Georgia Brown for her expert secretarial assistance and Dr. Mortimer Civan for his careful reading of the manuscript. Dr. Mauro expresses gratitude to Dr. Edward C. Gomez for his continued encouragement and support. Dr. Mauro was a Medical Scholar supported by the School of Medicine. LITERATURE CITED Armstrong, C.M., and Bezanilla, F. (1974) Charge movement associated with the opening and closing of the activation gates of the Na channels. J. Gen. Physiol., 63.533-552. Armstrong, C.M., and Lopez-Barneo, J. (1987) External calcium ions are required for potassium channel gating in squid neurons. Science, 236:7 12-714. Barthelson, R.A., Jacoby, D.B., and Widdicombe, J.H. (1987)Regulation of chloride secretion in dog tracheal epithelium by protein kinase C. Am. J. Physiol., 253:C802-C808. Boyce, S.T., and Ham, R.G. (1985) Normal human epidermal keratinocytes. In: In Vitro Models for Cancer Research. M. Webber and L. Sekely, eds. CRC Press, Boca Raton, FL, pp 245-274. Dover, R., and Watt, F.M. (1987) Measurement of the rate of epidermal terminal differentiation: Expression of involuerin by S-phase keratinocytes in culture and in psoriatic plaques. J . Invest. Dermatol., 89:349-352. Dunn, J.A.. and Blumberg, P.M. (1983) Specific binding of IZO-"HI12-deoxyphorbol13-isobGtyrateto phorbd ester receptor subclasses in mouse skin particulate preparations. Cancer Res., 43: 4632-4637. Evans, M.G., and Marty, A. (1986) Calcium-dependent chloride currents in isolated cells from rat lacrimal glands. J. Physiol. (Lond.), 378:437-460. Fairley, J.A., Ewing, N.M., and Keng, P.C. (1988) Extracellular calcium-induced terminal differentiation of keratinocytes is accompanied by a n increase in intracellular free calcium. J . Invest. Dermatol., 90:556 (abst). Frizzell, R.A., Rechkemmer, G., and Shoemaker, R.L. (1986a)Altered regulation of airway epithelial cell chloride channels in cystic fibrosis. Science, 233558-560. Frizzell, R.A., Halm, D.R., Rechkemmer, G., and Shoemaker, R.L. (1986b) Chloride channel regulation in secretory epithelia. Fed. Proc., 45:2727-2731. Gray, M.A., Harris, A,, Coleman, L., Greenwell, J.R., and Argent, B.E. (1989) Two types of chloride channel on duct cells cultured from human fetal pancreas. Am. J. Physiol., 257:C240-C251. Hamill, O.P., Marty, A., Neher, E., Sakmann, B., and Sigworth, F.J. (1981)Improved patch-clamp techniques for high resolution current recordings from cells and cell-free membranes. Pflugers Arch., 392: 85-100. Hazama, A. and Okada, Y. (1988) CaZ+-sensitivity of volume-regulatory K * and C1 channels in cultured human epithelial cells. J. Physiol. (Lond.), 402:687-702. Hennings, H., Michael, D., Chengs, C., Steinert, P., Holbrook, K. and Yuspa, S.H. (1980) Calcium regulation of growth and differentiation of mouse epidermal cells in culture. Cell, 19:245-254. Hennings, H., Holbrook, K.A., and Yuspa, S.H. (1983a) Factors influencing calcium-induced terminal differentiation of epidermal cells in culture. J. Cell Physiol., 116:265-287. Hennings, H., Holbrook, K.A., Yuspa, S.H. (1983133 Potassium mediation of calcium-induced terminal differentiation of epidermal cells in culture. J. Invest. Dermatol., 80:50s55s.

20

MAURO ET AL.

Hennings, H., Kruszewski, F.H., Yuspa, S.H., and Tucker, R.W. (1989) Intracellular calcium alterations in response to increased external calcium in normal and neoplastic keratinocytes. Carcinogenesis, 10:777-780. Horn, R., and Marty, A. (1988) Muscarinic activation of ionic currents measured by a new whole-cell recording method. J . Gen. Physiol., 92:145-159. Isseroff, R.R., Enders-Stephens, L., and Gross, J.L. (1989) Subcellular distribution of protein kinase Ciphorbol ester receptors in differentiating mouse keratinocytes. J. Cell Physiol. 141t235-242. Jaken S., and Yuspa, S.H. (1988) Early signals for keratinocyte differentiation: role of CaZ' -mediated inositol lipid metabolism in normal and neoplastic epidermal cells. Carcinogenesis, 9:1033-1038. Jeng, A.Y., Lichti, U., Strickland, J.E., and Blumberg, P.M. (1985) Similar effects of phospholipase C and phorbol ester tumor promoters on primary mouse epidermal cells. Cancer Res., 455714-5721. Li, M., McCann, J.D., Anderson, M.P., Clancy, J.P., Liedtke, C.M., Nairn, A.C., Greengard, P., and Welsh, M.J. (1989) Regulation of chloride channels by protein kinase C in normal and cystic fibrosis airway epithelia. Science, 244:1353-1356. Mauro, T.M., Pappone, P.A., and Isseroff, R.R. (1988) Membrane currents in cultured human keratinocytes. J. Cell Biol., 107:355a (abst). Menon, G.K., Grayson, S., and Elias, P.M. (1985) Ionic calcium reservoirs in mammalian epidermis: Ultrastructural localization by ion-capture cytocherAstry. J. Invest. Dermatol., 84:508-512. Nishizuka, Y. (1984) Turnover of inositol phospholipid and signal transduction. Science, 225t1365 -1370. Penner, R., Matthews, G., and Neher, E. (1988) Regulation of calcium influx by second messengers in rat mast cells. Nature (Lond.),334: 499-504. Pillai, S., Bikle, D.D., and Hincenbergs, M. (1988) Alterations in intracellular free calcium influences the differentiation of normal but not malignant keratinocytes. J. Cell Biol., 10Zr73a (abst). Praeger, F.c., Stanulis-Praeger, B.M., and Gilchrest, B.A. (1987) Use of strontium to separate calcium-dependent pathways for proliferation and differentiation in human keratinocytes. J. Cell. Physiol., 132t81-89. Read, J., and Watt, F.M. (1988) A model for in vitro studies of epidermal homeostasis: Proliferation and involucrin synthesis by cultured human keratinocytes during recovery after stripping off the suprabasilar layers. J . Invest. Dermatol., 9Ot739-743. Reddy, M.M., and Quinton, P.M. (1987) Intracellular potentials of

microperfused human sweat duct cells. Pflugers Arch., 41 O:471475. Reverdin, E.C., Birchall, N., and Boulpaep, E. (1988) Human keratinocytes have voltage-dependent ion channels and a cationic conductance. J. Invest. Dermatol., 90t601 (abst). Reverdin, E.C., Cohen, G., Birchall, N.M., and Boulpaep, E. (1989) Ca2 ' current in human epidermal cells. Biophys. J., 55:603 (abst). Rheinwald, J.G., and Green, H. (1975) Serial cultivation of strains of human keratinocytes: The formation of keratinizing colonies from single cells. Cell, 6:331-344. Snoek, G.T., Boonstra, J., Ponec, M., and delaat, S. (1987) Phorbol ester binding and protein kinase C activity in normal and transformed human keratinocytes. Exp. Cell Res., 17:146-157. Spray, D.C., and Bennett, M.V.L. (1985) Physiology and pharmacology of gap junctions. Annu. Rev. Physiol., 47:281-303. Stea, A., and Nurse, C.A. (1989) Chloride channels in cultured glomus cells of the rat corotid body. Am. J. Physiol., 257tC174-Cl81. Stewart, S., Duncan, G., Marcantonio, J.M., and Prescott, A.R. (1988) Membrane and communication properties of tissue cultured human lens epithelial cells. Invest. Opthalmol. Vis. Sci., 29:1713-1725. Tang, W., Ziboh, V.A., Isseroff, R., and Martinez, D. (1988) Turnover of inositol phospholipids in cultured murine keratinocytes: Possible involvement of inositol triphosphate in cellular differentiation. J. Invest. Dermatol., 90:37-43. Tennenbaum, T., Kapitulnik, J., and Yuspa, S.H. (1988) Addition of Mg" to a low Ca"-containing culture medium improves the growth and proliferation of primary mouse keratinocytes. J. Invest. Dermatol., 90t612 (abst). Watt, F.M. (1988) Keratinocyte cultures: An experimental model for studying how proliferation and differentiation are coordinated in the epidermis. J. Cell Sci., 90525-529. Watt, F.M., Mattey, D.L., and Garrod, D.R. (1984) Calcium-induced reorganization of desmosomal components in cultured human keratinocytes. J. Cell Biol., 99.2211-2215. Welsh, M.J. (1983) Intracellular chloride activities in canine tracheal epithelium. J . Clin. Invest., 71:1392-1401. Welsh, M.J. (1986) An apical-membrane chloride channel in human tracheal epithelium. Science, 232:1648-1650. Ziboh, V.A., Isseroff, R.R., and Pandey, R. (1984) Phospholipid metabolism in calcium-regulated differentiation in cultured murine keratinocytes. Biochem. Biophys. Res. Commun., 222:1234-1240. Ziyadeh, F.N., and Agus, Z.S. (1988) Role of calcium in chloride-secreting epithelia. Mineral Electrolyte Metab., 14:78-85.