Mechanism of Palytoxin Action on the Epidermal Growth Factor Receptor

promoter of the non 12-0-tetradecanoylphorbol-13-acetate (TPA) type, .... washed in binding medium (DME containing 0.1% ovalbumin and 50 mM Hepes) ...
0 downloads 0 Views 1MB Size
Chapter 15 Mechanism of Palytoxin Action on the Epidermal Growth Factor Receptor 1

2

3,4

Elizabeth V. Wattenberg , Hirota Fujiki , and Marsha Rich Rosner 1

Department of Applied Biological Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139 Cancer Prevention Division, National Cancer Center Research Institute, Tsukiji 5-1-1, Chuo-ku, Tokyo 104, Japan The Ben May Institute and Department of Pharmacological and Physiological Sciences, University of Chicago, Chicago, IL 60637

2

3

Palytoxin is a potent marine toxin and mouse skin t u m o r promoter that is able to act o n a wide variety o f systems. W e have studied the mechanism o f palytoxin action i n the context o f growth c o n t r o l by analyzing the effect o f palytoxin o n the epidermal growth factor (EGF) receptor i n murine fibroblasts. O u r results indicate that picomolar levels o f palytoxin are able to down modulate the EGF receptor by reducing the number and affinity o f EGF binding sites. T h e mechanism o f palytoxin action differs from that o f 12-O-tetradecanoylphorbol-13-acetate (TPA) type t u m o r promoters i n several respects, including kinetics, dose-response, and the fact that it is not dependent u p o n protein kinase C . Further, under o u r conditions, palytoxin action is sodium-dependent rather than calcium dependent, and palytoxin causes sodium influx with a dose-response that parallels the effects o n E G F binding. These results suggest that palytoxin is able to activate a sodium p u m p or channel, resulting i n sodium influx that leads to loss o f epidermal growth factor binding sites.

M a n y environmental toxins interact with specific cellular receptors, including enzymes, i o n channels and i o n pumps, and thus provide natural tools for the study o f cellular signalling pathways. Palytoxin, a c o m p o u n d isolated from the coelenterate o f genus Palythoa, is one such useful and intriguing c o m p o u n d . T h e structure o f palytoxin was first determined i n 1981 independently by H i r a t a (7) and M o o r e (2). A s one o f the most potent marine toxins known, palytoxin has been studied i n a variety o f systems ranging from erythrocytes to neurons. A s a t u m o r promoter o f the n o n 12-0-tetradecanoylphorbol-13-acetate ( T P A ) type, palytoxin can also be studied i n the context o f a growth c o n t r o l system. Palytoxin is a relatively large ( M W 2681), hydrophilic c o m p o u n d (7, 2), unlike the prototypical T P A - t y p e tumor promoters, the p h o r b o l esters. A l t h o u g h palytoxin 4

Address correspondence to this author. Formerly at the Department of Applied Biological Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139. 0097-6156/90/0418-0204S06.00/0 o 1990 American Chemical Society

15. WATTENBERG ET AL.

205

Epidermal Growth Factor Receptor

is a t u m o r promoter i n the two stage mouse skin assay, palytoxin does not activate all o f the cellular systems activated by the T P A - t y p e t u m o r promoters (3). F o r example, palytoxin does n o t induce ornithine decarboxylase i n mouse skin (4). M o s t distinctively, palytoxin does not bind to o r activate protein kinase C and is therefore classified as a n o n - T P A - t y p e tumor promoter (3, 4). O n a cellular level it has been shown that palytoxin can act synergistically with T P A - t y p e t u m o r promoters, further suggesting that these different types o f t u m o r promoters activate different cellular pathways. F o r example, palytoxin acts synergistically w i t h T P A - t y p e tumor promoters and other activators o f protein kinase C to stimulate prostaglandin release (5, 6). Palytoxin also synergizes w i t h T P A - t y p e tumor promoters i n the stimulation o f histamine release from rat peritoneal mast cells and superoxide release from neutrophils (7, 8). Recent evidence indicates that a variety o f growth control systems are regulated by m u l t i p l e pathways. F o r example, the induction o f O D C and the proto-oncogenes c-myc and c-fos i n cell culture can be mediated by protein kinase C-independent as well as protein kinase C-dependent pathways (9-11). Therefore, it is o f interest to determine whether palytoxin can modulate a growth control system and, i f so, whether palytoxin activates alternate signal transduction pathways. Evidence from a number o f systems suggests that i o n flux plays a role i n palytoxin action. I n a wide range o f systems, palytoxin effects are accompanied by a change i n intracellular cation levels. F o r example, the influx o f N a and/or C a is associated w i t h palytoxin-stimulated contraction o f cardiac and s m o o t h muscle, the release o f norepinephrine by rat pheochromocytoma (PC12) cells, and the depolarization o f excitable membranes (12—15). Palytoxin also induces K efflux from erythrocytes and thus alters i o n flux i n a nonexcitable membrane system as well (16-19). I n b o t h excitable and nonexcitable membranes, the ultimate action o f palytoxin has been shown to be dependent o n extracellular cations. T h e palytoxininduced effects o n smooth muscle and erythroctyes can be inhibited by removing Ca from the media, and the palytoxin-induced release o f norephinephrine from P C 1 2 cells can be blocked i n N a + free media (13, 14, 18, 20, 21) T h e role o f i o n flux i n signal transduction has been studied extensively, especially i n the context o f neurological and growth c o n t r o l pathways. I n cell growth systems, N a influx and cellular alkalinization appear to correlate with fertilization i n sea urchin eggs, and growth factor-induced mitogenesis i n cell culture (22, 23). Ca appears to be a virtually universal second messenger, acting i n regulatory roles as diverse as neurotransmitter release, short term memory, a n d chemotaxis (24). Ca also appears to play a role i n growth factor systems. P r i o r to mitogenesis, epidermal growth factor ( E G F ) stimulates the influx o f extracellular C a , a n d platelet derived growth factor ( P D G F ) causes release o f C a from intracellular stores (22, 25-27). O n e well-characterized model for studying growth c o n t r o l pathways, and i n particular the interaction o f foreign compounds with growth c o n t r o l systems, is the E G F receptor system. T h e E G F receptor is subject to transmodulation by a number o f agents including other growth factors and tumor promoters. W e and others have shown that t u m o r promoters inhibit E G F binding to a class o f high affinity receptors and block E G F - s t i m u l a t e d tyrosine kinase activity (28-30). Since these effects appear to be mediated by protein kinase C , we determined whether the n o n - T P A type tumor promoter palytoxin also alters E G F receptors and, i f so, whether a similar mechanism is involved. T h e results indicate that palytoxin, like T P A - t y p e t u m o r promoters, inhibits E G F binding. However, the mechanism by which palytoxin alters E G F binding differs significantly from that o f the T P A - t y p e t u m o r promoters. I n particular, o u r results suggest a role for N a but not C a i n palytoxin action o n the E G F receptor. +

2 +

+

2 +

+

2 +

2 +

2 +

2 +

+

MARINE TOXINS: ORIGIN, STRUCTURE, AND MOLECULAR PHARMACOLOGY

206

Experimental Materials.

1 2 5

I - E G F was either made by iodinating mouse E G F (Biomedical Technologies Inc.) by the chloramine T method, to a specific activity o f approximately 1-2 C i / / i m o l , using N a - I (Amersham) o r purchased from N e w E n g l a n d Nuclear. P h o r b o l diterpene esters were purchased from Sigma. Palytoxin was isolated from Palythoa tuberculosa as previously described (7). 1 2 5

Cultures. Swiss 3T3 and A 4 3 1 cells were grown i n a gassed (5.5% C 0 ) , h u m i d i fied incubator i n Dulbecco's M o d i f i e d Eagle M e d i a ( D M E ) supplemented w i t h 10% heat inactivated fetal calf serum ( F C S ) . 2

Quantitation of I - E G F binding to high affinity E G F receptors. A 4 3 1 cells and Swiss 3T3 cells were plated i n 24 well dishes at a concentration ranging from 50,000 to 100,000 cells/ml. A 4 3 1 cells were assayed the day after plating. Swiss 3T3 cells were grown to confluence (appr. 105 cells/well). T h e medium was removed and replaced with D M E containing 0.1% F C S o r C R - I T S premix (purchased from Collaborative Research) the night before assaying. T h e cells were washed i n binding medium ( D M E containing 0.1% o v a l b u m i n and 50 m M Hepes) o r incubation medium (130 m M N a C l , 50 m M Hepes, 1.8 m M C a C l , 0.8 m M M g S 0 , 5.5 m M glucose, 0.5 m M boric acid, 0.1% B S A , p H 7.4) and the appropriate agents added i n a total v o l u m e o f 0.4 m l binding m e d i u m o r incubation medium as indicated i n the figure legends and the text. E a c h variable was tested i n triplicate. Cells were then placed o n ice and washed w i t h b i n d i n g media. IE G F (0.05-0.1 n M ) i n binding media was added for 4 - 6 hr at 4 ° C . This concentration o f I - E G F binds primarily to high affinity E G F receptors, as demonstrated by the 8 0 % reduction i n cpm b o u n d after 37 C p h o r b o l dibutyrate ( P D B u ) treatment o f cells that had not been depleted o f protein kinase C . Finally, cells were washed, lysed, and quantitated for specific I - E G F binding. D a t a were normalized to the amount o f specific I - E G F binding to cells that were treated with binding medium o r incubation medium alone. In general, the b o u n d cpm ranged from 1000 to 4000 c p m w i t h a standard deviation o f 5 - 1 0 % . T h e nonspecific E G F binding was 100—250 c p m . T h e specifics for each experiment presented here are given i n the figure legends. In order to generate cells having different levels o f protein kinase C , cultures were treated for an additional 72 hr w i t h 0, 20, 200, o r 2000 n M P D B u i n D M E / 0 . 1 % F C S as previously described (31, 32). T h e cultures were washed four times with binding medium ( D M E containing 0.1% ovalbumin) over a period o f 2 hr at 37 ° C prior to incubation with the appropriate agents. Scatchard analysis was done as previously described (33). 1 2 5

2

4

1 2 5

1 2 5

W

1 2 5

1 2 5

DNA synthesis assays.

3

D N A synthesis was m o n i t o r e d by incorporation o f H thymidine into T C A precipitable material as described by McCaffrey and R o s n e r (32). +

Na Influx Studies.

+

N a influx was monitored according to the procedure o f O w e n and V i l l e r e a l (34), w i t h some modifications. Cells were seeded o n t o 60-mm culture dishes, grown, and serum starved as described for the assays above. T h e cells were washed w i t h incubation media and incubated i n 3 m l o f the appropriate agent at 37 C . A f t e r incubation the cells were rapidly washed i n ice c o l d 0.1 m M M g C L and extracted w i t h 5% T C A / 0 . 5 % K N 0 for sodium determination o r 0.2% S D S for protein determination. S o d i u m concentration was measured using a V a r i a n Model 275 A t o m i c A b s o r p t i o n Spectrophotometer. P r o t e i n was determined fluorimetrically. W

3

15.

WATTENBERG ET AL.

207

Epidermal Growth Factor Receptor

Results T o determine whether palytoxin could modify E G F receptors, the effect o n E G F binding o f palytoxin and a T P A - t y p e tumor promoter, P D B u , were compared. Swiss 3T3 cells, a murine fibroblast cell line which is mitogenically responsive to E G F , were treated at 37 ° C for 1 5 - 1 2 0 m i n with 1-11 p M palytoxin o r 1 5 - 6 0 m i n w i t h 2 - 2 0 0 n M P D B u . T h e cells were then washed and incubated at 4 ° C w i t h I-EGF and cell associated radioactivity was determined. T h e results indicate that palytoxin, like P D B u , causes i n h i b i t i o n o f E G F binding (Figure 1). However, the kinetics o f the palytoxin effect differ significantly from that o f the P D B u effect. A t 11 p M palytoxin, the i n h i b i t i o n o f E G F binding begins to plateau by 3 0 - 6 0 m i n whereas at the lower doses, up to 120 m i n or longer is required to obtain a similar degree o f i n h i b i t i o n . In contrast, P D B u inhibits E G F binding rapidly, and the maximum level o f i n h i b i t i o n is strictly dose-dependent. Thus, i n h i b i t i o n o f E G F binding by palytoxin occurs at a slower rate than P D B u and has a different dose-dependence. 1 2 5

T o ensure that the i n h i b i t i o n o f E G F binding by palytoxin was not a consequence o f cell toxicity, the effect o f palytoxin o n D N A synthesis i n Swiss 3T3 cells was monitored. W h e n cells were incubated i n the presence o f palytoxin, 10% fetal calf serum, and H - t h y m i d i n e for 19.5 hr, no depression i n the extent o f H thymidine incorporation into D N A was detected up to 3.7 p M palytoxin (Table I). A l t h o u g h 11 p M palytoxin was toxic when present for a prolonged period, under the conditions o f the assays described above no toxicity was detected (Table I). W h e n cells were incubated i n the presence o f palytoxin, 0.1% fetal calf serum, and H - t h y m i d i n e , palytoxin did not stimulate significant incorporation o f H - t h y m i d i n e into D N A Thus, although it can modulate the E G F receptor system under these conditions, palytoxin alone does not appear to be mitogenic for Swiss 3T3 cells. 3

3

3

3

Inhibition o f E G F binding by palytoxin could be due to a decrease i n receptor affinity, as i n the case o f T P A - t y p e tumor promoters, and/or a decrease i n receptor number. In Swiss 3T3 cells there are two classes o f E G F receptors. T h e dissociat i o n constants for the two E G F receptor classes were determined to be approximately 2 x 1 0 " M and 2 x 1 0 ' M , corresponding to approximately 1 x 1 0 and 1 x 1 0 receptor molecules per cell, respectively (33). Scatchard analysis revealed that treatment o f Swiss 3T3 cells with palytoxin, like P D B u , caused an apparent loss i n high-affinity binding (Figure 2). However, i n contrast to P D B u , palytoxin also caused a significant (approximately 50%) loss o f l o w affinity E G F binding. 10

8

4

5

T h e differences between palytoxin and P D B u with respect to kinetics, temperature dependence, and effect o n low affinity binding suggest that these two different types o f tumor promoters may be acting through different mechanisms. Further, i n contrast to P D B u , the effect o f palytoxin is not readily reversible (33). T o determ i n e where the two pathways differ, we compared the relative ability o f palytoxin and P D B u to inhibit E G F binding i n protein kinase C depleted cells. Swiss 3T3 cells were depleted o f protein kinase C to different extents by exposing confluent quiescent cells to 0, 20, 200, or 2000 n M P D B u for 72 hr. Previous results indicate that this treatment depletes cells o f protein kinase C activity i n a dose-dependent manner (37). Changing cellular levels o f protein kinase C i n Swiss 3T3 cells did not significantly affect the ability o f palytoxin to inhibit E G F binding (Figure 3). Whereas P D B u action is highly dependent u p o n cellular levels o f protein kinase C , palytoxin inhibits E G F binding to approximately the same extent regardless o f the cellular level o f protein kinase C . This experiment demonstrates that, u n l i k e the T P A - t y p e t u m o r promoters, palytoxin modulates E G F receptor properties i n an apparently protein kinase C independent manner. Further studies indicated that palytoxin is not inhibiting E G F binding through direct competition with E G F for the receptor,

MARINE TOXINS: ORIGIN, STRUCTURE, AND MOLECULAR PHARMACOLOGY

208

I

0

i

30

i

60

i

I

I

i

I

90

120

0

30

60

TIME

(min)

I

90

1 2 5

1

120

Figure 1. Effect o f t u m o r promoter treatment o n binding o f I-EGF to Swiss 3T3 cells. Confluent quiescent Swiss 3T3 cells ( A and B ) were treated for the indicated times at 37 C with P D B u [ A : 2 n M (o), 20 n M ( A ) , 200 n M ( • ) ] or palytoxin [B: 1.1 p M (•), 3.7 p M (A), 11 p M ( • ) ] . Cultures were subsequently washed and assayed for * I - E G F binding at 4 ° C as described i n the Experimental section. T h e data is expressed as the percent o f I - E G F binding i n cells treated w i t h either P D B u or palytoxin relative to control cells. The specific binding o f I - E G F constituting 100% for control cells was appr. 800 cpm for Swiss 3T3 cells. E a c h point represents the mean o f triplicate samples ± S. D . (Reproduced with permission from Ref. 33. Copyright 1987 Cancer Research, Inc.) #

I 2

1 2 5

1 2 5

15. WATTENBERG ET AL.

Epidermal Growth Factor Receptor

209

Table I. Effect of Palytoxin on DNA Synthesis in Swiss 313 Cells Treatment**

3

3

H-Thymidine Incorporated(cpm )

b

C

10% F C S 10% F C S + 10% F C S 10% F C S + 10% F C S + 10% F C S + 0.1% F C S 0.1% F C S + 0.1% F C S +

P T X (11 p M )

86312 126999 82287 105780 99502 32067 27631 25915 18106

c

P T X (1 p M ) P T X (3.7 p M ) P T X (11 p M ) P T X (3.7 p M ) P T X (11 p M )

± ± ± ± ± ± ± ± ±

2537 9288 626 9735 3457 5207 609 5238 1156

a

H-Thymidine Incoroporated(% ) 100 147 100 129 121 39 100 94 65

Confluent Swiss 3 1 3 cells were serum-starved by incubation for 48 hr i n D M E containing 0.1% F C S . Cells were then incubated with the indicated c o m p o u n d at 37 ° C i n the presence o f H - t h y m i d i n e , 0.1 o r 10% F C S for 19.5 hr, washed, and then assayed for H - t h y m i d i n e i n c o r p o r a t i o n into D N A as described i n Methods. T h e average o f duplicate determinations ± range. Cells were incubated with palytoxin for 2.5 hr at 37 ° C , washed, and then incubated i n the presence o f H - t h y m i d i n e and 10% F C S for 19.5 hr. S O U R C E : R e p r o d u c e d with permission from R e f . 33. Copyright 1987 Cancer Research, Inc. 3

3

b

c

3

210

MARINE TOXINS: ORIGIN, STRUCTURE, AND MOLECULAR PHARMACOLOGY 140

n

Control

201

8

6 BOUND

10

4

6

BOUND

(fmoles) 1 2 5

12

(fmoles)

8

10

12

Figure 2. Scatchard analysis o f I - E G F binding to Swiss 3T3 cells treated w i t h P D B u o r palytoxin. Confluent quiescent Swiss 3T3 cells were treated at 37 ° C with solvent (o) or 200 n M P D B u (•) for 15 m i n ( U p p e r panel); or with solvent (o) or 11 p M palytoxin (•) for 60 m i n (Lower panel). Cells were assayed as i n Figure 1. (Reproduced w i t h permission from Ref. 33. Copyright 1987 Cancer Research, Inc.)

WATTENBERG ET AL.

A. PDBu

0

30

211

Epidermal Growth Factor Receptor

B. PALYTOXIN

60

90

120

0

30

60

90

TIME (min) Figure 3. R e l a t i o n s h i p between levels o f protein kinase C and i n h i b i t i o n o f E G F binding to Swiss 3T3 cells by P D B u and palytoxin. C o n fluent quiescent Swiss 3T3 cells were pretreated for 72 hr w i t h 0 . 1 % F C S plus solvent (•), 20 n M P D B u (o), 200 n M P D B u ( A ) , o r 2000 n M ( • ) . Cells were then washed and treated w i t h either 200 n M P D B u ( A ) o r 11 p M palytoxin (B) for the times shown and assayed as i n Figure 1. O t h e r conditions were as described i n the legend to Figure 2. *The data for 200 n M P D B u are composites o f two independent experiments. (Reproduced with permission from Ref. 33. Copyright 1987 Cancer Research, Inc.)

120

212

MARINE TOXINS: ORIGIN, STRUCTURE, AND MOLECULAR PHARMACOLOGY

and that the mechanism is temperature sensitive and may reflect an energydependent process (33). Since i o n flux is implicated i n palytoxin action i n a number o f systems, we investigated the role o f C a and N a i n palytoxin action o n the E G F receptor. F o r these experiments the cells were incubated i n the simpler incubation medium instead o f D M E (see the discussion o f the incubation medium i n the Experimental section). W e have found that there is some variability i n the potency o f palytoxin from assay to assay, and the use o f this borate containing incubation medium has helped to reduce this variability. T o determine the effect o f C a o n palytoxin action, Swiss 313 cells were incubated i n the presence or absence o f palytoxin i n either a C a containing medium or a C a deficient medium that also included 100 / i M E G T A . T h e results show that external C a is not necessary for the i n h i b i t i o n o f E G F binding by palytoxin (Figure 4), although i n some experiments we have noted a slight shift i n the dose-response curve. E G F binding does appear to be slightly depressed i n the absence o f C a , suggesting that external C a may somewhat enhance E G F binding. In concurrence with the result that extracellular Ca does not appear to be necessary for palytoxin action o n the E G F receptor, we found that palytoxin does not cause C a influx or the release o f C a from internal stores at the doses used i n these experiments, as measured using the photosensitive protein aqueorin (42). However, palytoxin does appear to cause C a entry at toxic doses, higher than those used i n the assays described above. 2 +

2

2 +

2 +

2 +

2 +

2 +

2 +

2 +

2 +

2 +

2 +

+

Na also appears to play a role i n palytoxin action i n some systems. T o determ i n e i f there is a N a requirement for palytoxin action o n the E G F receptor, Swiss 3T3 cells were assayed for palytoxin activity i n N a containing medium versus Na deficient medium. W h e n N a C l is replaced by c h o l i n e C l , palytoxin can no longer inhibit E G F binding i n Swiss 3T3 cells (Figure 5). B y contrast, P D B u is equipotent i n both N a containing and N a free media (data not shown). Because these results suggest that extracellular N a is required for i n h i b i t i o n o f E G F binding by palytoxin i n these cells, we determined i f palytoxin caused N a influx i n Swiss 3T3 cells. W h e n N a influx was monitored at an early time point (7 min), it was found that palytoxin causes an influx o f N a and that the rate o f N a influx is dose dependent (Figure 6). In parallel with its effect o n E G F binding, palytoxin at different doses increases intracellular N a to the same final level (42). A l t h o u g h N a influx occurs prior to the i n h i b i t i o n o f E G F binding, these results and the apparent N a dependence o f the palytoxin effect suggest a role for N a i n the action o f palytoxin o n the E G F receptor. +

+

+

+

+

+

+

+

+

+

+

+

+

+

Discussion T h e results discussed here show that the n o n - T P A - t y p e t u m o r promoter palytoxin, like the T P A - t y p e t u m o r promoters, can inhibit E G F binding i n Swiss 3T3 cells and A 4 3 1 cells. However, significant differences i n time course, dose response, reversibility, effect o n receptor number, dependence u p o n protein kinase C , and dependence o n extracellular N a indicate that the mechanism o f action o f palytoxin is one that has not been previously described (Figure 7). T h e effect o f T P A - t y p e tumor promoters o n E G F binding is rapid, reversible, and strictly dose dependent; the extent o f i n h i b i t i o n o f E G F binding is directly proportional to the level o f protein kinase C , consistent with direct activation o f the enzyme by these agents. B y contrast, the n o n - T PA - t y p e t u m o r promoter palytoxin inhibits E G F binding w i t h slower kinetics and is not readily reversible; the extent o f i n h i b i t i o n o f E G F binding reaches a maximal level almost independent o f the initial dose o f palytoxin, consistent w i t h an ion-flux-dependent mechanism. These results indicate that there are at least two +

15.

WATTENBERG ET AL.

Epidermal Growth Factor Receptor

110

100 TIME (MIN) 2 +

Figure 4. Effect o f C a o n palytoxin action i n Swiss 3T3 cells. C o n fluent quiescent Swiss 3T3 cells were incubated for the indicated times at 37 ° C w i t h C a - f r e e incubation media containing 100 uM E G T A (+), complete incubation media containing 3.7 p M P T X ( • ) , o r C a free incubation media containing 100 pM E G T A plus 3.7 p M P T X ( • ) . Cells were assayed as i n Figure 1. D a t a expressed as percentage o f I - E G F binding i n cells treated w i t h complete incubation media alone. 2 +

2 +

i 2 5

MARINE TOXINS: ORIGIN, STRUCTURE, AND MOLECULAR PHARMACOLOGY

214

2000

E

1500-

Na

Na

-

PTX

-

PTX

+

Figure 5. Effect of Na on palytoxin action in Swiss 3T3 cells. Confluent quiescent Swiss 3T3 cells were incubated for 120 min at 37 ° C with incubation media containing media alone (Na/-), incubation media plus 3.7 pM PTX (Na/PTX), Na -free incubation media containing 130 mM cholineCl (-/-) or Na -free incubation media containing 130 mM cholineCl plus 3.7 pM palytoxin (-/PTX). Cells were assayed as in Figure 1. Data expressed as specific cpm of I-EGF bound per 10 cells. +

+

125

5

50

[PALYTOXIN] (pM) +

Figure 6. Effect of palytoxin on the rate of Na influx in Swiss 3T3 cells. Confluent quiescent Swiss 313 cells were incubated for 37 ° C for 7 min in incubation media containing 0.1 pM PTX, 1.1 pM PTX, or 11 pM PTX. Intracellular Na was determined as described in the Experimental section. Data points represent the mean of quadruplicate points. +

WATTENBERG ET AL.

Epidermal Growth Factor Receptor

Palytoxin

Kinetics Activity at 4 ° C

PDBu

slow

rapid

-

+

High Affinity Binding —

Low Affinity Binding

i

Reversibility

-

+

-

+

Dependence on Protein Kinase C

Figure 7. C o m p a r i s o n o f effects o f palytoxin and p h o r b o l o n E G F receptor binding.

dibutyrate

216

MARINE TOXINS: ORIGIN, STRUCTURE, AND MOLECULAR PHARMACOLOGY

very different mechanisms by which tumor promoters modulate cellular growth regulatory systems. T h e results i n o u r system suggest a role for N a i n palytoxin action o n the E G F receptor. Palytoxin causes N a influx i n Swiss 3T3 cells at a rate which is dose-dependent. A l t h o u g h the effect o n N a flux occurs m u c h m o r e rapidly than the effect o n the E G F receptor, the results indicate that extracellular N a is required for i n h i b i t i o n o f E G F binding by palytoxin and thus suggest a l i n k between the two events. A correlation between N a influx and i n h i b i t i o n o f E G F binding is further supported by the observation that, i n b o t h cases, a similar endpoint is reached with time almost independent o f dose. T h e N a - i n d u c e d loss o f E G F binding may also occur under physiological conditions, since N a influx is an early event i n the action o f many growth factors. +

+

+

+

+

+

+

T h e mechanism by which palytoxin stimulates i o n flux is not yet k n o w n . Palytoxin alone does not appear to be an i o n o p h o r e since the c o m p o u n d does not cause i o n flux i n artificial liposomes (75, 16). T h e high potency o f palytoxin and the apparent dependence o f its action o n a cellular component suggest that palytoxin may act through a specific cellular receptor. In various systems, tetrodotoxin does not appear to block the palytoxin-induced monovalent cation flux, diminishing the possibility that palytoxin activates a tetrodotoxin-sensitive N a channel (14, 15, 36). Results obtained with amiloride and m o r e specific N a / H antiporter inhibitors i n the E G F receptor system (Wattenberg, Cragoe, and Rosner, data not shown) and i n the rat erythrocyte system (37) suggest that palytoxin does not cause N a influx by activation o f this i o n p u m p . It has been proposed that the N a , K - A T P a s e is the cellular receptor for palytoxin based o n the observation that oubain can antagonize palytoxin action i n some systems and that palytoxin can block H - o u b a i n binding (38, 39). A l t h o u g h these studies suggest that palytoxin may b i n d to the N a , K - A T P a s e , there is n o correlation between the sensitivity o f a system to oubain and its sensitivity to palytoxin (38); Wattenberg and Rosner, data not shown). In addition, there is n o direct evidence that this i o n p u m p mediates palytoxin action. It appears that the sugar moiety o f cardiac glycosides is necessary for blocking palytoxin action, but inhibition o f the i o n p u m p itself does not affect the activity o f the toxin (19, 37). Palytoxin has been shown to inhibit the N a , K - A T P a s e i n vitro, but at concentrations at least 10,000 times that required to elicit a biological effect such as K release from erythrocytes o r i n h i b i t i o n o f E G F binding, thus raising into question the significance o f these results (40, 41). It remains possible that palytoxin generates o r activates a N a channel that has not yet been characterized. T h e system that we have described should be a useful o n e for investigating the mechanism o f action o f palytoxin and determining the biochemical link between i o n flux and the modulation o f a growth c o n t r o l system. +

+

+

+

+

+

3

+

+

+

+

+

+

Acknowledgments W e w o u l d like to thank M i t c h V i l l e r e a l and his colleagues for their assistance and helpful discussions. This work was supported by N a t i o n a l Cancer Institute Grants C A 3 5 5 4 1 and C A 4 0 4 0 7 to M . R . R . and N a t i o n a l Institute o f H e a l t h Toxicology G r a n t T 3 2 - E S 0 7 0 2 0 to E . V . W .

Literature Cited 1. Uemura, D.; Ueda, K.; Hirata, Y. Tetrahedron Letters 1981, 22, 2781-2784. 2. Moore, R. E.; Bartolini, G. J. Am. Chem. Soc. 1981, 103, 2491-2494. 3. Fujiki, H.; Suganuma, M.; Nakayasu, M.; Hakii, H.; Horiuchi, T.; Takayama, S.; Sugimura, T. Carcinogenesis 1986, 7, 707-710. 4. Fujiki, H.; Suganuma, M.; Tahira, T.; Yoshioka, A.; Nakayasu, M.; Endo, Y.;

15.

5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

WATTENBERG ETAL.

Epidermal Growth Factor Receptor

217

Shudo, K.; Takayama, S.; Moore, R. E.; Sugimura, T. Cellular Interactions by Environmental Tumor Promoters; Japan Sci. Soc. Press, Tokyo/VNU Science Press, Utrecht, 1984, pp. 37-45. Levine, L.; Fujiki, H . Carcinogenesis 1985, 6, 1631-1634. Levine, L.; Xiao, D.; Fujiki, H. Carcinogenesis 1986, 7, 99-103. Ohuchi, K.; Hirasawa, N.; Takahashi, C.; Watanabe, M.; Tsurufuji, S.; Fujiki, H.; Suganuma, M.; Hakii, H.; Sugimura, T.; Christensen, B. Biochim. Biophys. Acta. 1986, 887, 94-99. Kano, S.; Izuka, T.; Ishimura, Y.; Fujiki, H.; Sugimura, T. Biochem Biophys Res Commun. 1987, 143, 672-677. Hovis, J. G.; Stumpo, D. J.; Halsey, D. L.; Blackshear, P. J. J. Biol. Chem. 1986, 261, 10380-10386. Ran, W.; Dean, M.; Levine, R. A.; Henkle, C.; Campisi, J. Proc. Natl. Acad. Sci. USA 1986, 83, 8216-8220. Tsuda, T.; Hamamori, Y.; Yamashita, T.; Fukumoto, Y.; Takai, Y. FEBS Lett. 1986, 208, 39-42. Rayner, M. D.; Sanders, B. J.; Harris, S. M.; Lin, Y. C.; Morton, B. E . Res. Commun. Chem. Pathol. Pharmacol. 1975, 11, 55-64. Ito, K.; Karaki, H.; Urakawa, N. Eur. J. Pharmacol. 1977, 46, 9-14. Tatsumi, M.; Takahashi, M.; Ohizumi, Y. Molecular Pharmacology. 1984, 25, 379-383. Lauffer, L.; Stengelin, S.; Beress, L.; Hucho, F. Biochim. Biophys. Acta. 1985, 818, 55-60. Ahnert-Hilger, G.; Chhatwal, G. S.; Hessler, H . -J.; Habermann, E . Biochim. Biophys. Acta. 1982, 688, 485-494. Nagase, H.; Ozaki, H.; Karaki, H.; Urakawa, N. FEBS Lett. 1986, 195, 125-128. Nagase, H.; Ozaki, H.; Urakawa, N. FEBS Lett. 1984, 178, 44-46. Ozaki, H.; Nagase, H.; Urakawa, N. FEBS Lett. 1984, 173, 196-198. Ohizumi, Y.; Shibata, S. J. Pharmacol. Exp. Ther. 1980, 212, 209-214. Ishida, Y.; Satake, N.; Habon, J.; Kitano, H.; Shibata, S. J. Pharmacol. Exp. Ther. 1985, 232, 557-560. Rozengurt, E. Science 1986, 234, 161-166. Bell, J.; Nielsen, L.; Sariban-Sohraby, S.; Benos, D. Current Topics in Membranes and Transport. Vol. 27. Academic Press: Orlando, 1986, pp. 129-162. Rasmussen, H. New. Eng. J. Med. 1986, 314, 1164-1170. Moolenaar, W. H.; Aerts, R. J.; Tertoolen, L. G.; de Laat, S. W.J.Biol. Chem. 1986, 261, 279-284. Macara, I. G. J. Biol. Chem. 1986, 261, 9321-9327. McNeil, P. L.; McKenna, M. P.; Taylor, D. L. J. Cell Biol. 1985, 101, 372-379. Friedman, B.; Frackelton, A. R.; Ross, A. H.; Connors, J. M.; Fujiki, H.; Sugimura, T.; Rosner, M. R. Proc. Natl. Acad. Sci. USA 1984, 81, 3034-3038. McCaffrey, P. G.; Friedman, B.; Rosner, M . R. J. Biol. Chem. 1984, 259, 12502-12507. Foulkes, J. G.; Rosner, M. R. Molecular Mechanisms of Transmembrane Signalling; Elsevier Scientific Publishing Co.; Inc.: New York, 1985, pp. 217-252. Friedman, B.; Rosner, M. R. J. Cell Biochem. 1987, 34, 1-11. McCaffrey, P. G.; Rosner, M. R. Cancer Res. 1987, 47, 1081-1086. Wattenberg, E . V.; Fujiki, H.; Rosner, M . R. Cancer Research. 1987, 47, 4618-4622. Owen, N. E.; Villereal, M. L. Cell 1983, 32, 979-985. McCaffrey, P. G.; Rosner, M. R.; Kikkawa, U.; Sekiguchi, K.; Ogita, K.; Ase, K.; Nishizuka, Y. Biochem. Biophys. Res. Commun. 1987, 146, 140-146.

218 36. 37. 38. 39. 40. 41. 42.

MARINE TOXINS: ORIGIN, STRUCTURE, AND MOLECULAR PHARMACOLOGY Sauviat, M.; Pater, C.; Berton, J. Toxicon 1987, 25, 695-704. Ozaki, H.; Nagase, H.; Urakawa, N. Eur. J. Biochem 1985, 152, 475-480. Habermann, E.; Chhatwal, G. S. Naunyn Schmiedebergs Arch Pharmacol 1982, 319, 101-107. Bottinger, H.; Beress, L.; Habermann, E. Biochim. Biophys. Acta. 1986, 861, 165-176. Ishida, Y.; Takagi, K.; Takahashi, M.; Satake, N.; Shibata, S. J. Biol. Chem. 1983, 258, 7900-7902. Bottinger, H.; Habermann, E. Naunyn Schmiedebergs Arch Pharmacol 1984, 325, 85-87. Wattenberg, E.V.; McNeil, P.L.; Fujiki, H.; Rosner, M.R. J. Biol. Chem. 1989, 264, 213-219

RECEIVED June 6, 1989