JOURNAL OF CELLULAR PHYSIOLOGY 151:623-629 (1992)

Lanthanum Influx Into Cultured Human Keratinocytes: Effect on Calcium Flux and

Terminal Differentiation SREEKUMAR PILLAI* AND DANIEL D. BIKLE

Department of Medicine, University of California, San Francisco (S.P., D.D.B), and Division of Endocrinology (D.D.B.1 and Department of Dermatology (S.P.), Veterans Administration

Medical Center, San Francisco, CA 94 12 1

Trivalent cation lanthanum (La) binds to calcium binding sites of cells and either mimics the properties of calcium or inhibits the effects of calcium by displacing calcium from i ts binding sites. Extracellular calcium induces differentiation of human epidermal keratinocytes in culture, in part by increasing the intracellular calcium levels (Ca,). Therefore, in this study we determined the effect of La on differentiation and intracellular calcium levels of keratinocytes. We observed that La inhibited the production of cornified envelopes, a marker for terminal differen- tiation of keratinocytes. La inhibited the calcium requiring envelope cross-linking enzyme, transglutaminase, in a direct manner, presumably, by displacing cal- cium from its binding site on the enzyme. La inhibited the influx and the cfflux of 45 Ca from keratinocytes. Paradoxically, extracellular La appeared to increase the Ca, levels of keratinocytes as measured by the fluorescent probe indo-I. However, subsequent experiments revealed that indo-I bound La with a higher affinity than Ca and emitted fluorescence in the same wavelength as the Ca bound form. Using this probe, we observed that La enters keratinocytes in a dose-dependent fashion and achieves concentrations exceeding 80 n M when the external La concentra- tion is raised to 300 pM. This fully accounted for the apparent increase in Ca, when La was added to the cells. Treatment of cells with ionomycin increased indo-1 fluorescence maximally in the presence of La indicating influx of La via this Ca specific ionophore. Our results indicate that La enters cells and inhibits cal- cium mediated keratinocyte differentiation both by blocking Ca influx and by blocking calcium regulated intracellular processes such as transglutaminase di- rected cornified envelope formation. o 1992 wilcy-Liss, Inc.

The trivalent cation lanthanum (La) exhibits several properties similar to that of calcium in biological sys- tems presumably by virtue of its comparable ionic ra- dius (Lettvin et al., 1964). For example, La displaces calcium from its binding site on the external surface of the plasma membrane and blocks calcium influx and efflux from cells (Weiss, 1974; Van Breeman et al., 1979; Langer et al., 1979). Lanthanum and other lan- thanide ions (such as terbium) can replace calcium and activate the calcium requiring cytosolic enzyme calpain I and I1 (Zimmerman and Schlaeffer, 1988). La blocks slow calcium channels (Katzyng et al., 19731, inhibits the Na-Ca exchanger (Barry and Smith, 1982) in myo- cardial tissue, and displaces superficially bound cal- cium at the external surface of the sarcolemma of myocardial cells (Larger and Frank, 1972). Low concen- trations of La (10-30 pM) increase calcium influx in rat thymocytes, but higher concentrations (0.1-10 mM) ap- pear to block calcium influx (Segal, 1986)

Differentiation of normal human keratinocytes in culture is regulated by the extracellular calcium con- centration (Ca,) (Hennings et al., 1980; Yuspa et al., 1989; Pillai et al., 1990. Ca, raises Ca, of kerati- nocytes and induces their differentiation (Hennings (L) 1992 WILEY-LISS, INC.

et al., 1989; Sharpe et al., 1989; Pillai and Bikle, 1989). In this study, we evaluated the interaction between La and Ca on keratinocyte differentiation. We found that La enters keratinocytes, blocks the influx and the ef- flux of Ca, and inhibits cornified envelope formation by directly inhibiting the activity of the calcium-requiring enzyme, transglutaminase.

METHODS AND MATERIALS Cell culture

First or second passage keratinocytes isolated from neonatal human foreskins were used in these studies. They were grown in serum-free Keratinocyte Growth Medium (KGM) obtained from Clonetics Corporation (San Diego, CA) as described previously (Pillai and Bikle, 1988). Briefly, keratinocytes were isolated from newborn human foreskins using 0.25% trypsin, and the primary cultures were grown to confluence in KGM containing 0.07 mM calcium before passaging them

Received October 15,1990; accepted January 15,1992. * To whom reprint requestsicorrespondence should be addressed.

624 PILLAI AND BIKLE

into KGM containing either 0.07 mM or 1.2 mM cal- cium. Cells were evaluated at confluence.

Measurement of Cai Keratinocytes were dispersed using 0.1% trypsini

0.01% EDTA, and trypsin was neutralized with Soy- bean Trypsin Inhibitor. After washing, the cells were suspended in 20 mM Hepes buffer (pH 7.4) containing 120 mM sodium chloride, 5 mM potassium chloride, 1 mM magnesium chloride, 1 mgiml pyruvate, 1 mgiml glucose and 0.03 mM calcium chloride (buffer A) at a cell concentration of 1-1.5 x lo6 cellsiml. Indo-1 AM was added to a final concentration of 2 pM from a stock solution in DMSO and incubated at 35°C for 1 h. Cells were then washed 3 times with buffer A and resus- pended in buffer A at the original cell concentration of 1-1.5 million cells per milliliter. Cells were washed and resuspended in fresh buffer A before each measure- ment, and all the measurements were completed within 2-2.5 hours after indo-1 loading. Cells studied in this fashion show little indo-1 leakage. Fluorescence was recorded with a Perkin-Elmer 650-40 fluorimeter using 350 nm and 405 nm for excitation and emission wave- lengths, respectively, with a thermostatted cuvette (37°C) equipped with a magnetic stirrer. Fluorescence signals were recorded by a chart recorder continuously after each addition, and calibrated for each sample by the addition of ionomycin (10 p.M final concentration) (Fmax) followed by 0.1% triton x 100 and 10 mM EGTA/Tris (pH 8.3) (Fmin). Ca, was calculated from the following formula: Ca, = Kd (F - Fmin)/(Fmax - F) where Kd for indo-1 for Ca is 250 nM (Grynkiewicz et al., 1985). Additions of La, Ca, ionomycin, or other agents were done from concentrated solutions such that changes in volume were minimal with each addition, except for caffeine where 50 pl of a stock solution of 200 mM was added to 1 ml cells in the cuvette. Ca or La and all other agents were added from a thousand fold con- centrated stock solution made up in distilled water. All these experiments were repeated 3 4 times with differ- ent cell preparations and representative profiles are shown in the figures.

Measurement of 45Ca influx and efflux To study the influx of Ca, confluent cultures of kera-

tinocytes in 12 well multiwell plates were incubated with 1 ml of the Hepes buffer used for Ca, measurement (buffer A) containing 0.1 mM Ca and 0.1 pCi of 4sCa either in the presence or in the absence of 0.1 mM La for various lengths of time. The plates were washed 3 times with 1 mM EGTA, cells dissolved in 0.1N NaOH, and an aliquot counted to quantitate the amount of 45Ca taken up into the cells. (Takuma et al., 1987). For studying the effect of La on Ca efflux, cells were incu- bated with 45Ca for 1 h under conditions used for the influx study, and then washed 3 times with buffer A to remove all 45Ca from outside the cells. These 45Ca- loaded cells were then exposed to buffer A containing either 0.1 mM Ca alone or 0.1 mM Ca and 0.1 mM La for different times. The amount of 45Ca remaining inside the cells and that released into the medium at various times was quantitated by scintillation spectroscopy. Each experiment was performed 2 times with different

cell preparations and the data from one experiment are shown in the figure.

Lanthanum and calcium binding to indo-1 To determine the binding of indo-1 to La or Ca, a 1

pM solution of indo-1 free acid in a calcium-free buffer (Hepes, pH 7.4, containing 140 mM KC1 treated with chelex-100) was incubated with increasing concentra- tions of Ca or La, and the change in emission fluorescence was scanned a t an excitation wavelength at 360 nm.

Differentiation markers Activity of membrane-bound transglutaminase was

determined as described (Pillai et al., 1988). One hun- dred microliters of keratinocyte homogenate was incu- bated for 1 h with 600 pg dimethyl casein and 2 pCi 'H putrescine in a buffer containing 50 mM Tris-HC1 and 5 mM DTT (pH 8.1). Homogenizing buffers and assay buffers were made without EDTA or Ca, and indicated amounts of Ca or La were added a t the time of the assay. Control incubations were done in the absence of the cations and in the presence of 1.0 mM EDTA. After termination of the reaction, the amount of 'H-pu- trescine incorporated into casein was determined by TCA precipitation and scintillation spectroscopy. En- zyme activity is expressed as cpm 'H putrescine incor- porated into caseidkg protein. Cornified envelope con- tent was determined by the method of Rice and Green (1979). Briefly, the cells grown in 0.07 mM Ca, were switched to 1.2 mM Ca, for 24 h with or without 100 pM La. Cells were treated with a calcium ionophore, iono- mycin, for the last 2 h l o induce cornified envelope formation. Some cells were exposed to La only during the last 2 h with ionomycin. Cells were then harvested by trypsinization, total cell number determined, and an aliquot dissolved in 2% SDS/20 mM DTT by boiling for 5 min a t 100°C. The envelopes in the boiled cell aliquot were counted using phase-contrast microscopy, and the results are expressed as the percentage of total cells that formed cornified envelopes. Assays were done in duplicate for cornified envelope formation and in tripli- cate for transglutaminase activity. Each experiment was performed twice with different cell preparations to confirm reproducibility.

RESULTS Effect of La on keratinocyte differentiation

When keratinocytes grown in low calcium medium (0.03 mM) were switched to 1.2 mM Ca,, cornified enve- lope formation was stimulated (Table 1). Twenty-four hour treatment with 1.2 mM Ca, increased the ionomy- cin induced CE formation from 24% for cells grown in 0.03 mM Ca, to 43% for cells grown in 1.2 mM Ca,. The presence of 200 pM La during the 24 h calcium switch period inhibited the Ca, induced envelope formation from 43% to 20%. The envelope formation of cells in 0.03 mM Ca, was also inhibited by La. This suggests that La inhibited the production of envelope precursors and/or inhibited the Ca induced cross-linking of the precursors. The presence of La during the last 2 h of the ionomycin treatment inhibited the cornified envelope formation of calcium switched keratinocytes (Table 2) from 43% for controls to 10% for cells treated with 1 mM La. This suggests a n ionomycin-mediated entry of La

EFFECT OF LANTHANUM ON KERATINOCYTES

TABLE 1. Effect of lanthanum on cornified envelope formation of keratinocytes'

Cornified envelopes Ca switch (mM) from to (% of total cells)

0.03 0.03 0.03 0.03

0.03 0.03 + 200 pM La 1.2 1.2 + 200 ctM La

24 * 1 10 * 2 43 * 2 20 i 1

'Keratinocytes were grownin KGM/0.03 mM Ca tillconfluence. At confluence, cells were divided into 4 groups as shown above and incubated for a further 24 h. 10 pM ionomycin was added tothecellsforthelast2 hoftheincubation, and thecells were then dissolved in 2% SDS/BO mM DTT for quantitation of cornified envelopes. Data given are the average of duplicate determinations i range.

TABLE 2. Acute effect of lanthanum on cornified envelooe formation'

Treatment Cornified envelopes 6% of total cells)

6000 1

5000

625

A

0 1 I

0 2 0 4 0 6 0 8 0

time (min)

1.2 mM Ca, 1.2 mM Ca, + 200 pM La 1.2 mM Can + 1.0 mM La

43 * 2 25 * 3 10 * 1

B

'Keratinocytesgrown toconfluencein0.03mMCa0 wereswitched to l.2mMCa0for 24 h. Cells were exposed to no La, 200 rM La, or 1.0 mM La for the last 2 h of incubation with 10 pM ionomycin. Cornified envelope content of the control group and the La exposed group was determined as before. Average values of duplicate determinations i range are given for each sample.

120wo

5.

> 0 ,

0)

c .- .- -

.E 0 5 80000

: z

5:

.E cn E a Q -

40000

2 +

0 0 1 0 0 2 0 0 300 4 l

ion concentration (uM) Fig. 1. Effect of Ca and La on transglutaminase activity of kerati- nocytes. Keratinocyte homogenates were incubated with increasing concentrations of Ca (01, La (0) or Ca + La ( A ) in the incubation mixture. The amount of 3H putrescine incorporated into the substrate dimethyl casein was quantitated after precipitation with TCA. The results are expressed as cpm of 3H putrescine incorporated into caseinipg of the total hornogenate protein * standard deviation of triplicate determinations.

into keratinocytes and direct inhibition of transglutam- inase. To evaluate this possibility, we determined the effect of La on transglutaminase activity in an in vitro assay as shown in Figure 1. Transglutaminase was ac- tivated by Ca with an approximate Kact of 30 pM. La inhibited the basal activity of the enzyme as well as the

10 1 0 2 0 4 0 6 0 8 0

Time (min)

Fig. 2. Effect of La on the influx and efflux of 45Ca from kerati- nocytes. A. * T a influx. Keratinocytes in 12 well multiwell plates were incubated in buffer A containing 0.1 mM calcium and 0.1 pCiiml of ""Ca either in the presence (0) or in the absence (0) of 100 p M La for various lengths of time. The amount of 45Ca retained in the cells at each time point was determined by scintillation spectroscopy. Each point is the mean of triplicate determinations k SD. B. 45Ca efflux. Keratinocytes in the multiwell plates were preloaded with 45Ca for 1 h before incubation in the presence of 100 pM La (0 ) or in its absence (0). The amount of 45Ca in the medium and cells after the different time periods was quantitated and the fractional efflux ratio was calculated. Each point represents the mean of triplicate determinations f SD.

Ca activated activity at higher concentrations (above 30 pM), A t concentrations beyond 300 pM, La precipi- tates the substrate in the assay system. These studies suggest that La directly inactivates transglutaminase.

Effect of La on the Cai of keratinocytes Cultured human keratinocytes take up 45Ca in a

rapid and saturating fashion (Fig. 2a). This uptake of 45Ca was inhibited by the presence of 100 pM La. La also inhibited 45Ca efflux when 45Ca preloaded kerati- nocytes were incubated with 100 pM of La (Fig. 2b). The effect of La on influx and efflux of 45Ca was ob- served both at the initial rate as well as the final levels of 45Ca (60 min) after equilibrium was achieved. When indo-1 loaded keratinocytes in a calcium-free medium

626 PILLAI AND BIKLE

Ca (2 mMJ

La (100 uM)

EGTA (1 mM) Ls (100 uM)

c_

C -- I I

1 min U

Fig. 3. Fluorescence changes of indo-1 loaded keratinocytes in re- sponse to La. Keratinocytes were loaded with 2 pM indo-1 AM, washed, and resuspended in buffer A containing 0 Ca. The cells were exposed to 2 mM Ca by the addition of 2 p1 of a 1 M CaCl, solution to 1 ml of cells in the cuvette (A) or to 100 pM La by adding 10 p1 of a 10 mM stock solution of LaCl, prior to the addition of 2 mM Ca (B). In C, cells were first exposed to 1 mM EGTA before the addition of La. The fluorescence changes at 405 nm were noted for PF, min. Each trace is representative of experiments repeated a t least 4 times with different cell Preparations. The basal fluorescence of all the cell preparations (A,B,C) are approximately the same. The tracings are redrawn to the same scale of fluorescence units (Y axis).

were exposed to 2.0 mM Ca,, an acute and sustained increase in Ca, was observed (Fig. 3a) which was blocked by 100 pM La when added prior to the Ca, addition (Fig. 3b). However, La alone increased the flu- orescence of indo-1 loaded keratinocytes (Fig. 3b). This increase in fluorescence by La was blocked by exposure of cells to 1 mM EGTA prior to La addition (Fig. 3c). La evoked a large, sustained increase in fluorescence of indo-1 loaded keratinocytes in the absence of Ca, and in the presence of ionomycin (Fig. 4a) similar to that ob- served by Ca, (Fig. 4b), while ionomycin alone in the absence of either cation evoked only a transient eleva- tion of Cai (Fig. 4c). La also increased the fluorescence of indo-1 loaded cells whose intracellular calcium stores were depleted with 10 mM caffeine (Fig. 4d). The in- crease in fluorescence was proportional to the amount oiLa added (Fig. 5).These experiments suggest that the effect of La in increasing indo-1 fluorescence is not by changing Cai levels, but by a direct interaction of La with indo-1 after influx into keratinocytes. To test this possibility, we incubated indo-1 (free acid form) with La and measured the fluorescence changes.

Figure 6A shows the emission spectra of indo-1 in the presence of Ca, La, and EGTA at an excitation wave- length of 355 nm. The fluorescence intensity increased a t 400 nm (indo-1 calcium bound form) with a corre- sponding decrease at 470 nm (indo-1 free acid form) with the addition of Ca. A similar pattern of change was observed with La. In the presence of EGTA, maxi- mum fluorescence emission was observed at 470 nm which is the fluorescence maxima for the free indo-1 acid form. The data suggest that the affinity of indo-1 to La is approximately three to ten fold greater than that of Ca since the concentrations of La required for compa-

QJ V C 0 U w W CI 0 s ir, c

c E E

9 F

1 min U

Fig. 4. Ionomycin response of keratinocytes in the presence of Ca or La. Indo-1 loaded keratinocytes in calcium free buffer A were exposed to 10 pM ionomycin subsequent to addition of 100 p M La (A), 2.0 mM Ca (B), or 1 .O mM EGTA (C) and the fluorescence changes were noted. In D, 10 mM caffeine is added prior to 100 pM La. A decrease in fluorescence observed upon addition of caffeine is due to the dilution effect and a slight quenching of indo-1 fluorescence by caffeine. Repre- sentative profiles of 3 separate experiments are shown.

rable increase in indo-1 fluorescence are three- to ten- fold lower than that of Ca (300 nM for Ca vs. 100 nM for La after normalizing for Fmax or 300 nM Ca vs. 30 nM La for 40 units of fluorescence without normalizing for Fmax). (Fig. 6b). Considering that the Kd for indo-1 binding of Ca is 250 nM, this suggests a Kd of approxi- mately 25-80 nM for indo-1 to La. Precise measure- ment of the Kd of indo-1 for La would require La buff- ered solutions similar to Ca-EGTA buffers used for measuring accurately the free Ca concentration of a solution. Means of determining the free La in such buff- ers are not currently available.

To determine whether the failure of La to increase indo-1 fluorescence in the presence of EGTA (Fig. 3C) was due to chelation of La by EGTA, we verified the binding of La to EGTA and EDTA. We found that both EDTA and EGTA chelated La with apparent affinities even greater than that for Ca (data not shown). These studies demonstrate that compounds such as EGTA and indo-1 which bind Ca also bind La with apparently higher affinity.

Mechanism of La uptake by keratinocytes To determine whether La influx occurs via Na-Ca

exchange, we carried out the experiments described in Figure 7. We examined the rate of increase of indo-1 fluorescence of Na loaded and Na-depleted cells ex- posed to 100 pM La. Na loading of cells was achieved by pretreatment of cells with 0.1 mM ouabain (inhibitor of NaiK ATPase); Na depletion was achieved by suspend-

EFFECT OF LANTHANUM ON KERATINOCYTES 627

1 niin U

Fig. 5 . Concentration dependence of La response of keratinocytes. Indo-1 loaded keratinocytes in Ca free buffer A were exposed to 3 (A), 10 (B), 30 (C) , 100 (D), or 300 (E) pM La and the changes in fluores- cence a t 405 nm were noted. La, concentrations were estimated using an approximate Kd of La to indo-1 of 25-80 nM (calculated from Fig. 6B). La, of cells varied from 10-28 nM for cells exposed to 10 pM La to 50-144nM for cells exposed to 300 p M La (assuming a Kd of 25-80 nM for La binding to indo-1). Basal fluorescence was the same, and the tracings are redrawn to the same scale of fluorescence units.

ing the cells in a sodium-free medium (buffer in which NaCl was replaced by choline chloride). As seen in the figure, both these treatments failed to alter the La in- duced increase in indo-1 fluorescence. To determine whether La influx occurs via dihydropyridine in- hibitable, voltage dependent calcium channels, indo-1 loaded keratinocytes were incubated for 2 min with 10 pM verapamil before the addition of 100 pM La. Rate of change in fluorescence (i.e., influx of La) was similar to that of controls suggesting that verapamil sensitive, voltage dependent calcium channels were not responsi- ble for La influx into keratinocytes. These studies sug- gest that La enters keratinocytes via a mechanism in- dependent of voltage dependent calcium channels or Na-Ca exchange.

DISCUSSION The mechanism by which Ca, induces keratinocyte

differentiation remains unclear. Ca, clearly raises Cai of keratinocytes in an acute fashion (Hennings et al., 1989; Sharpe et al., 1989; Pillai and Bikle, 1989). Ca, may regulate a variety of intracellular events such as signal transduction pathways and enzymes involved in cell growth and differentiation (Pillai et al., 1990). La can block these calcium regulated cellular events ei- ther by blocking Ca influx into cells and/or by entering into cells and directly inhibiting the calcium requiring enzymatic steps. In this study, we found that La acts by both these mechanisms. La blocked 45Ca uptake by ke-

" d

300 4 0 0 5 0 0 6 0 0

emission (nm)

120 - 100 - 80 -

60 -

40 -

20 -

B La

Ca

0 0 30 100 300 1000 3000 10000

cation concentration (nM)

Fig. 6. A. Emission spectra of La or Ca bound form of indo-1. The fluorescence emission of indo-1 free acid in a calcium free buffer was measured from 380-520 nm a t an excitation wavelength of 355 nm in the presence of 500 nR'l EGTA (a), 1000 nM CaC1, (b), or 300 nM LaC1, (c). The free acid form of the dye fluoresces maximally at 470 nm while the cation bound form of the dye fluoresces maximally at 405 nm. B. Concentration dependence of indo-1 fluorescence in the pres- ence of Ca (lo) or La (0). One pM of indo-1 acid was exposed to 30- 10,000 nM concentrations of either Ca or La and the fluorescence a t 405 nm was noted (excitation a t 355 nm). The relative fluorescence was plotted against concentration of the 2 cations. La binds indo-1 with approximately three fold higher affinity than Ca.

ratinocytes and directly inhibited transglutaminase ac- tivity. Calcium binding to the catalytic subunit of transglutaminase induces dissociation of its subunits (Folk, 1980). Calcium binding should occur before sub- strate binding for enzyme activation. La presumably blocks this. Transglutaminase cross-links soluble sub- strates like involucrin into the insoluble cornified en- velope. The fact that La inhibits cornified envelope for- mation acutely (2 h with ionomycin) suggests that one of the mechanisms by which La inhibits cornified enve- lope formation is by inhibition of transglutaminase. The possibility that La competes with Ca for its binding to ionomycin and entry into the cells may also contrib- ute, at least in part, to the effect of La on inhibition of keratinocyte differentiation. The ability of La to block

f'

-- I A

C

1 min

PILLAI AND BIKLE

D

Fig. 7. Indo-1 loaded keratinocytes in buffer A or in a medium in which sodium chloride was replaced by 130 mM choline chloride (Na free medium) were incubated with the appropriate agents or vehicle alone for 2 min before challenging with 200 pM La. Other experiments using longer incubations with ouabain and verapamil (up to 60 min) gave similar results (data not shown). The fluorescence changes were measured at 405 nm, and the representative profile of one of three experiments is given. A Control. B: 10 pM verapamil. C: Cells in Na free medium. D: 1 mM ouabain.

the calcium-induced induction of involucrin synthesis was not measured in this study.

The La, levels inside keratinocytes estimated in this study (200-300 nM) is at least 2 orders of magnitude less than the La concentration required for inhibition of transglutaminase activity in the in vitro assay (Kd = 30 pM). This apparent paradox, which is true for Ca, levels also (Ca, of 100-200 nM, but Kd of Ca for trans- glutaminase activation is 30 pM), may be explained by the fact that the Ca or La concentrations at the mi- croenviornment of the membranes encountered by the membrane-bound enzyme transglutaminase may far exeed the nM range. It is also likely that, as for other Ca requiring membrane-bound proteins such as cal- modulin and protein kinase C, association of transglu- taminase with membrane lipids may increase the affin- ity of the enzyme for calcium ions. In addition to transglutaminase, La, can also affect (activate or in- hibit) other calcium and calmodulin dependent en- zymes such as protein kinases and phosphodiesterases (Schulman and Greengard, 19781, adenylate cyclases (Cheung et al., 1978), phospholipases (Daniel, 1985), or thymidylate synthetases (Whitefield e t al., 1976). All of these enzymes are directly or indirectly involved in cell proliferation and differentiation.

It is often assumed that La does not cross the plasma membrane and enter cells. Electron microscopy of in-

tact myocardial tissue and cultured cells exposed to La (up to 1.0 mM) have failed to show the presence of electron dense La inside cells (Langer et al., 1979; Langer and Frank, 1972). However, perfusion of rat hearts with 1-15 mM La prior to fixation resulted in La deposits inside the cells (Weihe et al., 1977). A more sensitive X-ray microprobe technique has revealed de- tectable intracellular La levels (250 nM) inside guinea pig ventricular muscles after exposure to 0.1 mM La for 1 h (Wendt-Gallitelli and Isenberg, 1985). Influx of La into chick embryonic ventricular cells has also been demonstrated using the indo-1 binding properties of La (Peters et al., 1989). Exposure of ventricular cells to 0.1-1.0 mM La resulted in a rapid influx of La via a Na-La exchange mechanism and achieved a La, concen- tration of approximately 250 nM. This study showed that changes in indo-1 fluorescence are more sensitive than electron microscopy for measuring the influx of La into cells. In addition, indo-1 measures the intracellu- lar concentration of free La of cells while the other techniques measure total cellular La. Our results con- firm that La is not confined to the extracellular space, but enters cells. Therefore, the mechanism by which La blocks keratinocyte differentiation may involve both the inhibition of calcium influx and interference with calcium requiring intracellular mechanisms.

The mechanism by which La enters the keratinocyte is not clear. In ventricular cells La appears to bind and enter cells via a Na-Ca exchanger (Peters et al., 1989). This does not appear to be the mechanism in kerati- nocytes since the presence of ouabain or the absence of Na, does not alter La influx. The rapid inhibition of calcium influx by La in keratinocytes suggests that La binds to keratinocyte cell membranes and blocks the entry of Ca. Electron micrographic studies have shown that La binds preferentially to keratinocyte cell mem- branes (data not shown). As La enters cells and La, increases, La may also bind to the plasma membrane Ca-ATPase to account for its inhibition of Ca efflux.

In summary, our results indicate that La inhibits calcium-mediated keratinocyte differentiation both by blocking calcium influx and by blocking calcium regu- lated intracellular processes such as transglutaminase directed cornified envelope formation.

ACKNOWLEDGMENTS We appreciate the support of Dr. Peter Elias for help-

ful discussions and Mara Hincenbergs and Mark Bogan for providing keratinocyte cultures. This work was sup- ported by grants from the NIH (AR-383860 and AR- 39448) and a merit review from the Veterans Adminis- tration.

LITERATURE CITED Barry, W.H., and Smith, T.W. (1982) Mechanisms of transmembrane

calcium movement in cultured chick embryo ventricular cells. J. Physiol. (Lond.), 325243-260.

Cheung, W.Y., Lynch, T.J., and Wallace, R.W. (1978) An endogenous calcium-dependent activator protein of brain adenylate cyclase and cyclic nucleotide phosphodiesterase. Adv. Cyclic Nucl. Res., 9233- 251.

Daniel, L. W. (1985) Phospholipases, In: Biochemistry of Arachidonic acid Metabolism. W.E.M. Lands, ed. Martinus Nijhoff Publishing, Boston, pp. 175-189.

Folk, JE. (1980) Transglutaminases. Ann. Rev. Biochem., 49:517-31.

629 EFFECT OF LANTHANUM ON KERATINOCYTES

Grynkiewicz, G., Poenie, M., and Tsien, R.Y. (1985) A new generation of calcium indicators with greatly improved fluorescence properties. J . Biol. Chem., 260:34403447.

Hennings, H., Michael, D., Cheng, 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., Kruszewski,F.H, Yuspa, S.H., andTucker, R.W. (1989) Intracellular calcium alterations in response to increased external calcium in normal and neoplastic keratinocytes. Carcinogenesis 10:777-780.

Katzyng, B.G., Reuter, H., and Porzig, H. (1973) Lanthanum inhibits calcium inward current but not Na-Ca exchange in cardiac muscle. Experientia, 29: 1073-1075.

Langcr, G.A., and Frank, J.S. (1972) Lanthanum in heart cell culture. J. Cell Biol. 54:441455.

Langcr, G.A., Frank, J.S., and Nudd. LM. (1979) Correlation of cal- cium exchange, structure and function in myocardial tissue culture. Am. J . Physiol., 237 (Heart Circ. Physiol., 6):H239-H246.

Lettvin, J.Y., Pickard, W.F., McCulloch, W.S., and Pitts, W. (1964) A theory of passive ion flux through membranes. Nature, 202:1338- 1339.

Peters, G.A., Kohmoto, O., and Barry, W.H. (1989) Detection of La influx in ventricular cells by indo-1 fluorescence. Am J. Physiol., 256:C351-C357.

Pillai, S., Bikle, D.D., Hincenbergs, M., and Elias, P.M. (1988) Bio- chemical and morphological characterization of growth and differ- entiation of normal human neonatial keratinocytes in a serum-free medium. J. Cell. Physiol, 134:229-237.

Pillai, S., Bikle, D.D. (1989) A differentiation dependent, calcium sensing mechanism in normal human keratinocytes. J. Invest. Der- matol., 92500a.

Pillai, S., Bikle, D.D., Mancianti, M.L., Cline, P., andHincenbergs, M. (1990) Calcium regulation of growth and differentiation of normal human keratinocytes: Modulation of differentiation competance by stages of growth and extracellular calcium. J. Cell. Physiol., 143294302.

Rice, R.H., and Green, H. (1979) Presence in human epidermal cells of a soluble protein precursor of the cross-linked envelope: Activation of the cross-linking by calcium. Cell, 18:681-694.

Schulman, H., and Greengard, P. (1978) Calcium dependent protein phosphorylating system in membranes from various tissues and its activation by “calcium dependent regulator.” Proc. Natl Acad. Sci. U.S.A., 75:5432-5436.

Segal, J. (1986) Lanthanum increases the rat thymocyte cytoplasmic free calcium concentration by enhancing calcium influx. Biochem. Biophys. Acta, 886267-271.

Sharpe, G.R., Gillespie, J.I., and Greenwell, J.R. (1989) An increase in intracellular free calcium is an early event during differentiation of cultured human keratinocytes. FEBS Lett., 254:25-28.

Takuwa, N., Takuwa, Y., Rollag, W.E., and Rasmussen, H. (1987) The effect of Bombesin on polyphosphoinositide and calcium metabolism in Swiss 3T3 cells. J. Biol. Chem., 262:182-188.

Van Breeman, C., Aaronson, P., and Loutzenhiser, R. (1979) Sodium- calcium interactions in mammalian smooth muscle. Phannacol. Rev., 30: 167-208.

Weihe. E., Haartschuh, W., Metz, J. and Bruhl, U. (1977) The use of ionic lanthanum as a diffusion tracer and as a marker of calcium binding sites. Cell Tissue Res., 178:28%3O2.

Wendt-Gallitelli, M.F., and Isenberg, G. (1985) Extra and intra cellu- lar lanthanum: Modified calcium distribution, inward currents and contractility in guinea pig ventricular preparations: Pfluegers Arch. 405:310-322.

Weiss, G.B. (1974) Cellular pharmacology of Ianthanum. Annu. Rev. Pharmacol., 14:343-354.

Whitefield, J.F., MacManus, J.P., Dixon, R.H., Boynton, A.L., You- dale, T., and Swierenga, S. (1976) The positive control of cell prolif- eration by the interplay of calcium ions and cyclic nucleotides. In Vitro, 22:l-18.

Yuspa, S.H., Kilkenny, A.E., Steinert, P.M., and Roop, D.R. (1989) Expression of murine epidermal differentiation markers is tightly regulated by restricted extracellular calcium concentration in vitro. J. Cell Biol., 109:1207-1217.

Zimmerman, U.P., and Schlaeffer, W.W. (1988) Activation of calpain I and calpain 11: A comparative study using terbium as a fluorescent probe for calcium-binding sites. Arch. Biochem. Biophys., 266:462- 469.