Vol. 181, No. 3, 1991 December

BIOCHEMICAL

31, 1991

AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages '1378-1384

A PEPTIDE FROM THE GAP-BINDING DOMAIN OF THE ~~21 PROTEIN AS WELL AS AZATYROSINE BLOCK =INDUCED MATURATION OF XENOPUS OOCMES

Denise L. Chung*, Paul Brandt-RauP, Randall B. Murphy*, Susumu Nishimura”, Weinsteina, and Matthew R. Pincuss*’

2. Yamabumi’,

I. Bernard

*Department of Chemistry and Center for Neural Sciences, New York University, NY, NY 10003 “Division of Environmental Sciences and Department of Medicine, and Comprehensive Cancer Center, Columbll College of Physicians and Surgeons, NY, NY 10032 “National Cancer Center Research Institute, Tokyo, Japan sDepartment Received

of Pathology, SUNY Health Science Center, 780 East Adams St., Syracuse, NY 13210

September

3, 1991

The a-oncogene-encoded ~21 protein is known to produce malignant transformation of NIH 3T3 cells as well as maturation of Xerropus oocytes when microinjected into these cells. ~21 protein is known to bind a GTPase actiiating protein (GAP) intracellularly; residues 32-48 have been implicated in interacting with GAP. We demonstrate here that a peptide corresponding to residues 3547 of ~21 as well as the antibiotic azatyrosine inhibit the B-induced maturation of Xenopus oocytes in a dose-related manner upon microinjection. We have previously shown that this p21 peptide and azatyrosine could inhibii the effects of ~21 protein on cell transformation and pinocytosis in NIH 3T3 cells. In the present study, in which we have extended these results to the oocyte system, we also demonstrate that both partially inhibit insulin-induced oocyte maturation, a process which is thought to involve activation of endogenous p21 protein; on the other hand, both agents fail to inhibit oocyte maturation induced by progesterone, which is known not to act through ~21 protein activation. Control studies with other peptides and tyrosine analogues support the selective nature of these events. These results suggest that both the p21-related peptide and azatyrosine have potent anti-s 0 ~391 Academic mess, Inc. effects intracellularly.

The s-gene

family codes for a set of proteins of molecular mass 21 Kd, the ~21 proteins, that are

known to cause malignant transformation substitutions

of cells (1). The oncogenic forms of the protein contain amino acid

at critical positions (such as residues 12, 13, and 81) in the sequence.

protein, found in all eukaryotic cells, is a G-protein. growth factors, the protein is induced to exchange causes cell proliferation,

The “normal’ form of the

Presumably, as a result of the stimulus from extracellular GDP for GTP, resulting in activation of the protein that

or, in some cell lines, cell differentiation

(I ,2).

A number of dlfferent proteins have been identified as possible candidates for the downstream of m-action

1

(3-7). It is known that activated normal, or oncogenic,

To whom reprint

0006-291W91 Copyright All rights

requests

should

be addressed.

$1.50

0 1991 by Academic Press, Inc. of reproduction in any form reserved.

1378

effecters

p21 protein binds a GTPase activating

Vol.

BIOCHEMICAL

181, No. 3, 1991

AND BIOPHYSICAL

protein (GAP) which stimulates the GTPase activity of these proteins. suggest that the region of the ~21 protein comprising

RESEARCH COMMUNICATIONS

Site-specific mutagenesis

experiments

residues 32-45 may be critical in binding to GAP (8).

We have recently reported that a peptide containing residues from the 35-47 region of the m-oncogeneencoded p21 protein has been found to inhibit cell transformation blocks the enhanced related manner.

pinocytotic

(9). In particular, we found that this peptide

activity of NIH 3T3 cells when co-injected

with oncogenic

p21 in a dose

Enhanced pinocytotic activity is known to be associated with the transformed cell phenotype

in this cell line (10). It is known that the oncogenic those observed

p21 protein produces effects in Xenopus laevis oocytes which parallel

in NIH 3T3 cells (11).

However, in Xenopus

oocytes, oncogenic

p21 protein causes cell

maturation, as opposed to cell transformation

in NIH 3T3 cells. It is thus desirable to explore the effects of the

~21 35-47 peptide on E-induced

of oocytes.

maturation

In addition, it is known that the antibiotic, azatyrosine, causes reversion of m-transformed human pancreatic adenocarcinoma through induction of the a-gene

NIH 3T3 and

cells in culture to a normal phenotype (12). This agent appears to function (13) in NIH 3T3 cells (Nishimura, S., et al, unpublished

data). Thus, it is also

further desirable to investigate the effects of this antibiotic on cell maturation. In a preliminary communication (14) we found that both of these agents inhibited the effects of oncogenic now report that both the 35-47 peptide and azatyrosine statistically significant dose-dependent

~21 protein on cell maturation. We

inhibit p21-induced

maturation

of oocytes in a

manner; that the effect of the 3547 peptide is specific since another,

unrelated peptide failed to inhibit this maturation effect: that both agents partially inhibit the maturation signal from insulin, indicating that insulin may transmit growth

signals using other pathways in addition to s-

pathways; and that neither agent blocks the effects of progesterone

which is known to act *w-independent

pathways (11).

MATERIALS AND METHODS p21 Protein Oncogenic human (I/al 12) ~21 protein was overexpressed in E. Co/i cells by using an expression vector (pGH-L9), containing the chemically synthesized c-Ha-ras gene, and cell lysates were purified as previously described (15). Normal (Gly 12) ~21 protein was obtained in a similar manner.

The peptide corresponding to the 35-47 residue sequence of ~21, Thr-lle-Glu-Asp-Ser-Tyr-Arg-Lys-GlnVal-Val-lIeAsp, was prepared using standard solid phase methodology (16) by Dr. Davkl Schlessinger of the Cancer Center, New York Unfversity. We prepared a control peptide from the CD4 receptor, Ser-Leu-Thr-LeuThr-Leu-Glu-Ser-Pro-Pro-Cys-Ser-Ser-PreSer, by similar methodology. After purification by HPLC, peptides were judged to be “99% pure by amino acid analysis. Azatvrosine [L-&atyrosine; L-t%(shydroxy-Z-pyddyt) atanfne] was fsofated trOm Sfrepfomyces chibenensis and was purified and characterized as previously described (12). 1379

Vol.

181, No. 3, 1991

Oocvte Mfcroiniection

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

and Maturation

Xenopus laevis females were obtained from Hamamatsu Biilogical Research Service, Inc. (Tokyo, Japan). The frogs were anesthetized by hypothermia, and the ovarian fragments were surgically removed from the animals and placed in Barth medium (88 mM NaCI, 1 mM KCI, 2.4 mM NaHCO,, 15 mM HEPES, 0.3 mM Ca(N0&4 HsO, 0.41 mM CaCI,B HsO, 0.8 mM MgSC4.7 HsO, 10 pg/mL Na Penicillin, and 20 &mL streptomycin sulfate). The ovarian follicle cells were removed by incubatton of the ovarian fragments for 2 hr at room temperature wtth 2 mg/ml colfagenase (Type 7, Hiih Purity, Sigma, St. Louis, MO) in Barth medium. At the end of this incubation period, the fragments were washed five times in 66 mL of cold Barth medium, and the oocytes were stored overnight at 16’ C. Groups of 20, stage-VI oocytes were subjected to cytoplasmic microinjection (11) with a 66 nL volume of the desired solution per oocyte. The microinjected oocytes were incubated at room temperature (26’ C) in Barth medium for 12-16 hrs. Docyte maturation was assessed by the appearance of a white spot on the pigmented animal pole. Germinal vesicle (nuclear) breakdown (GVBD) was veritied by manual dissection of the oocytes, after fixation in 16% (w/v) trichloroacetic acid (Sigma, St. Louis, MO). In experiments in which insulin (Sigma, St. Louis, MO) was used in the incubation medium, a final concentration of 10 pg/mL was found to yield an optimal effect. The same concentration of progesterone (Calbiochem, La Jolla, CA) was found to be effective and was used in all subsequent experiments. For each set of conditions, at least four separate experiments were performed on different sets of oocytes. Means and standard errors for each time point for each set of experimental conditions were determined.

Xenopw

RESULTS AND DISCUSSION Effect of pas-p21 Proteins on Oocvte Maturation.

In Figure 1, the effect of the oncogenic

upon oocyte maturation is shown, when the protein was microinjected can be seen in this figure, full maturation of oocytes was achiied observed (11). In other experiments

at a concentration

in a 16hour

Val 12-p21 protein of 0.05 mg/mL. As

period, as has been previously

(c.f. Figure 4) the time in which 166% oocyte maturation was achieved

was found to vary between 4-10 hours.

0

!3$&-

1

2

3

4

5 6 7 6 Time @ours)

9

10

111213

~ooovte-~~by-p2110~wmg/ml)pratsh~ew prewncOofdiauent-dttn~~m&k4lae36-47m3mtbeGAP-

bindingregionofthep2ipmteinco-mtcrdnje&edtntooccyks. The -umsofpeptideusedwereo.1 mg/ml (W), 0.25 mg/ml (P-V) and 0.50 mg/ml (p-0. Resutts frcm control e In which the peptide alone (0.5 mg/ml. v--v) was microinjected into oocytes rrnd in which 0.06 m@nd oncogenic p21 and 0.5 ma/ml CD4 rsceptor peptide were co-microlnjected into eccytes (e-e) are also Ulustmtsd .6tsnciard ermr bars are shown except for pdnts where the standard error was too smell to be represented. 1380

Vol.

BIOCHEMICAL

181, No. 3, 1991

When the normal (Gly 12-containing) concentration,

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

p21 protein was microinjected

into the oocytes at the same

oocyte maturation occurred at significantty lower rates (results not shown).

of the oocytes were found to mature over a twelve-hour cause an enhanced

maturation

Effect of the 3547

~21 Peotide on p21-Induced

period.

Thus, oncogenic

Only about 50%

p21 protein was found to

signal in these cells.

maturation of the peptide (35-47)

Oocvte Maturation.

In Figure 1, the effect on oocyte

present at several different concentrations,

is shown as a function of time.

The peptide from the effector region of the p21 protein appears to inhibit a ~-specific

effect in the oocyte cell

line. Further to explore the specificii

of inhibition of the effector peptide on m-induced

incubated with insulin after microinjection y& a E-dependent

of the peptide.

events, oocytes were

Insulin is thought to induce differentiation

of oocytes

pathway (11). As shown in Figure 2, oocytes exposed to insulin present in the bathing

medium at a concentration

of 10 pg/mL undergo

effector peptide when present in a concentration higher concentrations

differentiation

(11). This differentiation

is suppressed

by

maturation (Figure 2). However,

found to inhibit einduced

of peptiie (e.g., 1 mg/ml) resulted in no further inhibition of maturation (not shown). The

greatest extent of inhibition appears to be achieved at a level of about 30 percent maturation of cells. Part of this effect may be due to the intracellular of peptlde during extracellular

degradation

of the peptide, an event that would lower the inhibition

insulin stimulation. Howver, this phenomenon

03

Time (hours)

0

1

2

3

would not explain the inability of

4

5 6 7 6 Time (hours)

9

10111213

w Rapressntati resuks for the time-dependence of cocyts dii induced by ins&ii alone (c-o) and in the presence d the tridecspqtide containii m&dues 35-47 (0.5 mg/mt) frcm ths GAP-binding region of the p21 protein (0-O). insulin alone (V-V) ir also shown. Fia. Effects by progesterone

concentration

in incubation

of the ras peptide (ijaoted at a concentration present in a ccncentrstion of 10 pglml.

medium

was 10 &ml.

of 0.5 mglml)

on oocyte

The effect ofthe

peptide

diierentiation

induced

The symbols used are as fdlcws: O-0, effects of injection of pqresterone alone; O-8, effects of injec&g rap-p@de and progesterone; ~Qeffscts of raspeptlde atone; v--v, CD4 psptide alone: v-P,CD~ pqtlde + progesterone. Error bars are omitted tram b% points where the standard errcr was too low to be represented.

1381

Vol.

181, No. 3, 1991

BIOCHEMICAL

AND BIOPHYSICAL

the peptide to inhibit the insulin effect more completely inhibit

RESEARCH COMMUNICATIONS

at higher concentrations.

Thus, while the peptide

the maturation effects of insulin, this inhibition is partial, raising the possibility that insulin can transmit

maturation

signals by a w-pathway

and, concurrently,

by alternate pathways. We have recently obtained

further data that support this conclusion; we have found that a specific inhibitor of protein kinase c completely blocks p21-induced manuscript

oocyte maturation but only partially inhibits insulin-induced

maturation

(Chung, DL et al,

submitted).

There

WES

some evidence that there may be differences between the effects of the 3547 peptide in

oocytes and in NIH 3T3 cells.

As shown in Figure 1, the peptide alone does not stimulate cell maturation

significantly above that found in a buffer control and is therefore a purely competitive antagonist In NIH 3T3 cells, the peptide alone was found to stimulate some ceffularpinocytotic of a&vii

was much lower (about 10%) than that found with oncogenic

activity although this level

p21 protein (9). Thus in NIH 3T3 cells,

the effector peptide appears to be, functionally, a partial agonist by competing effector(

of w action.

for the same downstream

It is possible that in NIH 3T3 cells, the 35-47 peptide is capable, alone, of producing

a weak

transforming signal. This difference between the two cell fines may reflect differences in the effector molecules to which the activated p21 protein can bind. Specificitv of Pedide

Inhibition:

To examine further whether the inhibiiion

of m-induced

maturation

by the

35-47 ~21 peptide occurred in a specific manner, a series of experiments was performed in which a peptide of unrelated

sequence

but comparable

length and hydrophylic

sequence from an extrinsic portion of the CD4 receptor.

character

was used.

This peptide

This control peptide was either coinjected

was

a

with the

~21 protein into oocytes or was injected by itself in the presence of insulin. The rest&s indicated that there was no inhibition maturation

of p21-induced

(not shown).

maturation

(Fig. 1). The same results were found for insulin-induced

These findings thus support the conclusion

that the inhibition

of B-

and insulin-

in concentrations

N 10 pg/mL

induced maturation of the oocytes is specific to the 35-47 p21 peptide. Effect of the 35-47 Peptide on Prooesterone-Induced has been shown to produce maturation which was independent

Progesterone

of oocytes (11). This effect was found to occur through a pathway

of that utilized by s-~21,

failed to block progesterone-induced

Maturation:

since microinjection

into oocytes of an antii@2 antibody

maturation (11). In order to test further the specificity of the action of the

35-47 peptide upon oocyte maturation, the effects of this peptide on progesterone-induced explored.

maturation were

Figure 3 illustrates that the ras 3547 peptide as well as the CD4 sequence peptide failed to inhibit

progesterone-induced

maturation.

This result further supports the conclusion

maturation through a pathway which is selective for E-related

1382

pathways.

that the 35-47 peptide inhibits

Vol.

181,

No.

BIOCHEMICAL

3, 1991

BIOPHYSICAL

reversion of malignant cells which have been ~transformed

to a normal phenotype

induction of the ~-recision

events involved in E-oncogeneinduced

cells in culture (12). It appears that this effect

gene (rrg) (13) whose effect is essentially to inhibit the cellular cell transformation

explore the effects of azatyrosine further, we co-injected azatyrosine.

Initlllly, oocytes were preincubated

were subsequently conditions

injected with oncogenic

(Nishimura, S. et al, unpublished

oocytes with oncogenic

~21 protein.

functional or is not developed in the Xenoous oocyte.

antibiotic.

It is clear

concentrations,

that

this

compound

mechanism

acid analogue

for this amino

However, when co-microinjected

is not

with oncogenic

To demonstrate

of azatyrosine

m-induced

oocyte

at several concentrations maturation.

At lower

of the

azatyrosine

of this amino acid analogue.

that the effect of azatyrosine

which L-tyrosine was substituted inhibitory effect on m-induced

for azatyrosine.

is selective, addlional Figure

4 illustrates

experiments

that

the

were performed

normal

amino

acid

has

cellular maturation.

100 90 5 60

.j

60 70

'g 70 2 60

g 60

g

z

50

g 40

50

8 40 b n 30

0. 30

20

4

10 0

1

2

3

4

5 6 7 6 Time (hours)

9

loll

0

1213

5

0

1

2

3

4

5 6 7 Time(hours)

6

9

1011

1213

m Time-dependence of oocyte differentiation induced by ras (0.05 me/ml) alone (O-O) and in the presence of different concentrations of azatyrosine co-microinjectfito oocytes (V-V) 250 &ml; (v-3 500 &ml; (f-J--f-J)750 pg/ml). The effects of injection of L-Tyr + ras (O-0) and of L-Tyr alone (a-a) are also shown. Error bars are omitted from points to avoid cluttering. Representative errors are shown to illustrate major differences between mauration and maturation inhibition effects. Effects of azatyrosine (0.5 mglml) injected into oocytes on insulin and progesterone-induced maturation of oocytes. Both insulin and progesterone were present at concentrations of 10 pg/ml. The following data

&+

are shown: azatyrosine

p21

The results, shown in Figure 4, illustrate the time-

by microinjection inhibits

under these

this inhibition appears to be overcome after a 16-hour period (not shown), an effect possibly

due to cellular degradation

0

To

p21 protein together wfth

However, no effect of the azatyrosine

evidenced significant inhibition.

course of the inhibition of cell maturation

results).

with azatyrosine in the bathing medium, and then the oocytes

could be observed; presumably, the active uptake

protein, the azatyrosine

COMMUNICATIONS

that the tyrosine analogue

both in NIH 3T3 cells and in human pancreatic adenocarcinoma occurs through

RESEARCH

Recently, it has been demonstrated

Effects of Azatvrosine on Oocvte Maturation: azatyrosine produces

AND

progesterone alone (O-O); progesterone (~J); azatyrosine alone (l-J-0).

+ azatyrosine

1383

(v-p);

insulin

alone

(O-o);

insulin

+

in no

Vol.

181, No. 3, 1991

BIOCHEMICAL

AND BIOPHYSICAL

To define further the nature of the azatyrosineinduced co-microinjected

theamino

insulin or progesterone,

inhibition of minduced

oocyte maturation, we

acid analogue into oocytes which were incubated in a bathing medium containing respectively.

observed to inhibit the insulin-induced analogue.

RESEARCH COMMUNICATIONS

The results of this study are illustrated in Figure 5. Azatyrosine oocyte maturation at a concentration

is

of 0.5 mg/mL of the amino acid

As found with the 3547 pepbde, this inhibition was partial, a result possibly suggesting that insulin

may transmit a maturation signal by several distinct pathways. Conversely, as illustrated in the same ftgure, azatyrosine exerted no inhibitory effect on progesterone-induced conclusion

that azatyrosine-induced

oocyte maturation.

These data support the

inhibition of oocyte maturation is specific to the m

pathway.

In summary, both the 32-47 peptide from p21 and the antibiotic azatyrosine exhibit strong anti-=

effects

in Xenopus oocytes which parallel their effects in NIH 3T3 cells. These findings strongly suggest that both agents are strong antagonists of m-p21

cellular events, and may be of utility in investigating other effects

upon cellular function.

ACKNOWLEDGMENTS: suggestions

of m-induced

and discussions.

Drs. M. Ueffing, S. Fisch, and M. Gottesman This work was supported

02111 (IBW), and an award from the National Foundation

are thanked

for many helpful

in part by NIH grants NCI ROI CA 42566 (MRP), CA for Cancer Research (IBW).

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

Barbacid, M. (1987) Annu Rev Biihem 56,779-827. Guerrero, I., Wong, H., Peilicer, A. and Burstein, D.E. (1986) J Cell Physial 129, 71-76. Vogel, U.S., Diion, RAF., Shaber, M.D., Diehl, R.E., Marshall, MS., Scolnick, EM., Sigal, I.S., and Gibbs, J.B. (1988) Nature (London) 335, 96-93. West, M., Kung, H-F., and Kamata, T. (1996) FEBS Lett =,245-248. Wolfman, A. and Macara, I.G. (1996) Science =,6769. Delohery, T. M., Nishknura, S., Lee, G., Ronai, Z.A., Pincus, MR., Brandt-Rauf, P.W., Murphy, R.B., Yamabumi, i!., and Weinstein, I.B. (1989) Proc Natl Acad Sci USA e, 66768662. Tsai, M-H, Yu, C-L and Stacey, D.W. (1996) Science 250,982~965. Adari, H., Lowy, D. R., Willumsen, B.M., Der, C.J., and McCormick, F. (1986) Science 240,516521. Lee, G., Ronai, 2. A., Pincus, M.R., Murphy, R.B., Delohery, T.M., Miiimura, S., Yamaizumi, Z., Weinstein, I.B., and Brandt-Rauf, P.W. (1996) Med Sci Res 18, 771-772. Bar-Sargi, D. and Feramisco, J. B. (1986) Sciince 233, 1661-1656. Deshpande, A. K. and Kung, H-F. (1967) Mot Cell Biol 7, 1265-1288. ShindoOkada, N., Makabe, O., Nagahara, H. and Nthimura, S. (1989) Mol Carcinogen&s 2,159167. Contente, S., Kenyon, K., Rimoldi, D., and Friedman, R.M. (1996) Science 249, 796-798. Weinstein, I.B., and Chung, D.L., Brandt-Rauf, P.W., Murphy, R.B., Niihimura, S., Yamaizumi, Z., Pincus, M.R. (1991) Anticancer Res., in press. Inouye, Y., Nakamori, H., Iwai, S., Ohtsuka, E., Ikehara, M., Miura, K., Noguchi, S., and Niihimura, S. (1966) Jpn J Cancer Res (Gann) z, 45-51. Merrifteld, R. B. (1963) J Am Chem Sot 65,21492154.

1384