81

Molecular and Ceiluiar Endocrinology, 12 (1990) 81-87 Elsevicr Scientific Publishers Ireland, Ltd.

MOLCEL

02325

Down-regulation of tyrosinase mRNA levels in melanoma by tumor promoters and by insulin B-B. Fuller ‘, I. Niekrasz

’

and G.E. Hoganson

cells

2

’ Department of BioehemistT and Molecuiar Biology* The Uniuerstt,v of Oklahoma Health Sciences Cenier, Okiuhoma City, OK? U.S.A., and ’ Department of Pediairics,

The Unwersig

(Received

Kq w~xw!x Phorbol

ester: Protein

of Illinois College of Medicine, Chicago. IL, U.S.A.

9 May 1990; accepted

kinase C; Tyrosinase;

Insulin;

11 May 1990)

Meian~yte-stimulating

hormone

Summary

Mouse melanoma cells in culture respond to melanocyte-stimulating hormone (MSH) or to cyclic AMP analogues by demonstrating an increase in tyrosinase activity. In this study the effect of the tumor promoter, 12-O-tetradecanoyiphorbol 13-acetate (TPA), on the hormonal induction of tyrosinase was examined. TPA was found to lower basal levels of tyrosinase activity in melanoma cells and to reduce tyrosinase levels in cells treated with either MSH (lo-’ M), dibutyryl CAMP (10m4 M), isobutylmethylxanthine (IBMX, low4 M), or with the potent MSH analogue, [Nle4,D-phe7]-ar-MSH. The phorbol ester, phorbol 12,13-dibutyrate was also effective in lowering tyrosinase activity levels, while 4a-phorbol 12,13_didecanoate, which does not bind protein kinase C, was ineffective. In order to determine how TPA may reduce tyrosinase activity in melanoma cells, the levels of tyrosinase mRNA in untreated or TPA-treated cells were determined by Northern blot analysis. A marked down-regulation of constitutive levels of tyrosinase mRNA was observed in cells treated with the tumor promoter. Tyrosinase mRNA levels in cultures exposed to TPA for 48 h were only 7% of control levels. Tyrosinase mRNA levels in cells treated with both MSH and TPA were also lower than in cells treated with MSH alone. Previous studies from this laboratory have shown that insulin both lowers basal tyrosinase activity in melanoma cells and antagonizes the MSH stimulation of the enzyme. We have now determined that this inhibition is also due to reduced levels of tyrosinase mRNA. These results implicate a role for protein kinase C in the phorbol ester and insulin-mediated down-regulation of tyrosinase gene activity in melanoma cells and suggest the possible presence of both inhibitory and stimulatory DNA control elements in the tyrosinase gene which may control transcriptional rates.

Introduction The activity of tyrosinase, zyme for melanin synthesis,

the rate-listing enis markedly stimu-

Address for correspondence: Bryan B. Fuller, Ph.D., Department of Biochemistry and Molecular Biology, The University of Oklahoma Health Sciences Center, P.O. Box 26901, Oklahoma City, OK 73190, U.S.A.

0303-7207/90/$03.50

0 1990 Elsetier

Scientific

Publishers

Ireland,

lated in mouse melanoma cells by treatment with either MSH (melanocyte-stimulating hormone), cyclic AMP analogues such as dibutyryl CAMP, or by compounds which elevate CAMP levels in cells, such as theophylline or IBMX (isobutylmethylxanthine) (Wong et al., 1974; Fuller and Viskochil, 1979; Fuller et al., 1987). This hormonally induced increase in tyrosinase activity apparently involves both an increase in enzyme synthesis as Ltd.

82

well as an increase in the catalytic activity of the enzyme (Fuller et al., 1987; Jimenez et al., 1988). Further, either MSH or dibutyryl CAMP will promote an increase in the levels of tyrosinase mRNA in melanoma cell cultures (Kwon et al., 1988; Hoganson et al., 1989). In contrast to the stimulatory effect of MSH or CAMP on tyrosinase activity and melanin synthesis, the tumor promotor, 12-O-tetradecanoylphorbol 13-acetate (TPA) has been shown to mediate a decrease in melanogenesis in mouse melanoma cells (Mufson et al., 1979; Laskin et al., 1983) and in chick-embryo melanocytes (Oetting et al., 1985). The expression of several proteins, including tyrosinase, was found to be inhibited in chick melanocytes cells exposed to TPA (Oetting et al., 1985). TPA has been recently shown to be an effective melanocyte mitogen and is now used extensively for the culture of human melanocytes (Eisinger et al., 1979; Eisinger and Marko, 1982; Lerner et al., 1988). However, the possible effects of TPA on melanogenesis in these cultures has not been examined. In this study we have investigated the effect of TPA on constitutive levels of tyrosinase and on the hormonally induced increase in tyrosinase activity in melanoma cells. We have found that TPA reduces tyrosinase activity in melanoma cells and does so by down-regulating tyrosinase mRNA levels. A similar inhibitory effect is also mediated by insulin, suggesting that both of these compounds may act through protein kinase C to lower tyrosinase gene activity. Materials and methods Materials The MEL-11A clone of Cloudman S-91 mouse melanoma cells used in these studies was derived in our laboratory as previously described (Fuller et al., 1988). Horse serum was purchased from Sterile Systems (Logan, UT, U.S.A.) and Ham’s F-10 nutrient medium was obtained from Gibco (Grand Island, NY, U.S.A.). Tyrosine, L-dopa (dihydroxyphenylalanine), 12-O-tetradecanoylphorbol 13-acetate (TPA), phorbol 12,13-dibutyrate (PDBu), 4a-phorbol 12,13_didecanoate (4a-PDD), insulin, isobutylmethylxanthine (IBMX), dibutyryl CAMP (dbcAMP), ar-melanocyte stimulating hormone (a-MSH), and [Nle4,D-phe’]-cu-melano-

cyte stimulating hormone (Nle4,D-phe’-a-MSH) were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). [ 32P]Deoxycytidine triphosphate (specific activity 3000 Ci/mmol) was purchased from Amersham Corp. and [ring-3,53H]tyrosine (specific activity 47 Ci/mmol) was obtained from DuPont NEN Products (Boston, MA, U.S.A.). Cell cultures Melanoma cells (MEL-11A) were grown in Ham’s F-10 nutrient medium fortified with 10% horse serum. Penicillin (100 units/ml) and streptomycin (100 pg/ml) were also present in the medium. Stock cultures were maintained in 150 cm2 Corning culture flasks and subcultured weekly. Cells were removed from flasks with Tyrode’s salt solution containing 5 mM EDTA. Cell counts were made with a Coulter counter. For experiments, cells were seeded into either 25 cm2 or 75 cm2 flasks, as indicated, at low densities to assure hormone responsiveness (Fuller and Lebowitz, 1980). Cells were allowed to attach for 6 h and then exposed to the various compounds indicated in the text for either 24 or 48 h. At the end of this time, cells were removed from the flasks and either processed for RNA or homogenized and assayed for tyrosinase activity. Tyrosinase assays To determine in situ tyrosinase activity in cells, growth medium was supplemented with 1 pCi/ml of [3H]tyrosine and added to cells for various time periods (usually 24 h). The medium removed from the flasks was assayed for the presence of 3H20 formation using a modification of the charcoal absorption method (Fuller et al., 1987). Assays to determine tyrosinase activities in cell homogenates were carried out as previously described (Fuller and Viskochil, 1979; Fuller et al., 1987; Hoganson et al., 1989). Briefly, lo5 cells were homogenized by sonication on 0.5 ml of phosphate-buffered saline (PBS) containing 1% Triton X-100 and 1 mM phenylmethylsulfonyl fluoride (PMSF). Aliquots of homogenate (100 ~1) in triplicate were incubated in 1 ml of a reaction mixture consisting of 0.1 mM tyrosine, 2 pCi/ml of [3H]tyrosine and 0.1 mM r_-dopa in 0.1 M sodium phosphate buffer, pH 6.8. Incubations

were carried out at 37 o C for 3 h. To terminate the reaction, 1 ml of charcoal (10% w/v, in 0.1 N HCl) was added to each assay tube, the samples centrifuged at 2000 x g for 10 min and processed for liquid scintillation counting as described elsewhere (Fuller et al., 1987). Preparation of melanoma RNA and Northern blot analysis Total cellular RNA was prepared from melanoma cultures using a guanidinium isothiocyanate method (Chomczynski and Sacchi, 1987) as previously described (Hoganson et al., 1989). Total cellular RNA was isolated from duplicate 75 cm’ flasks (approximately 2 X lo6 cells/flask) with a typical preparation yielding 20-40 pg of RNA. For Northern blot analysis, approximately lo20 pg of total cellular RNA was electrophoresed on a 1% formaldehyde agarose gel and transferred to nitrocellulose. The nitrocellulose filter was hybridized to a tyrosinase cDNA probe previously labeled by an oligo-labeling method (Prime Time kit, International Biotechnologies, New Haven, CT, U.S.A.). Hybridization conditions were as previously described (Hoganson et al., 1989). After removal of the tyrosinase cDNA, the filters were reprobed under identical conditions with a glyceraldehyde 3-phosphate dehydrogenase (GAPDH) probe (Fort et al., 1985) to control for any variability in the amount or quality of mRNA loaded onto the gels. Filters were exposed for at least 12 h with a DuPont Cronex intensifying screen at - 80 o C. Messenger RNA levels from Northern blot films were analyzed by densitometric scanning on a Molecular Dynamics model 300A Computing Densitometer. All tyrosinase mRNA levels were normalized to the amount of GAPDH present in the same sample. Tyrosinase mRNA levels in cells treated with the compounds under investigation were then expressed as a percentage of control tyrosinase mRNA values. Results Since previous studies have found that TPA can reduce the level of melanin content in melanoma cells (Mufson et al., 1979) and cause a disappearance of tyrosinase in chick melanocytes (Oetting et al., 1985) we initiated these studies to

04 10-g

10-e TPA concentration

10-7 (M)

Fig. 1. Dose-response of TPA inhibition of tyrosinase activity in mouse melanoma cell cultures. Melanoma cells (2x10’ cells/flask) were seeded into 25 cm2 flasks and allowed to attach for 6 h. At this time the medium was replaced with Ham’s F-10 nutrient medium containing 1 nCi/ml of [3H]tyrosine, 10% horse serum and antibiotics as described in Materials and Methods. TPA was added at the concentrations shown and, at the same time, cells were either exposed to MSH (lo-’ M) (triangles) or left untreated (circles). The medium was replaced at 24 h intervals and the spent medium assayed for ‘H,O as described in Materials and Methods. Data shown is from 48 h exposure of cells to TPA and, where indicated MSH. Values are the averages of six determinations f SD.

determine the molecular basis for this inhibitory effect. For this study we selected a mouse melanoma cell line which had high basal levels of tyrosinase activity and melanin content and which were responsive to MSH (Fuller et al., 1988). We first determined the sensitivity of these cells to inhibition by TPA by carrying out the dose-response studies shown in Fig. 1. Basal tyrosinase levels were found to be inhibited by concentrations of TPA greater than lo-’ M while the induction of tyrosinase by MSH was found to be reduced in cells treated simultaneously with TPA at concentrations greater than lop9 M. At none of the concentrations used was cell growth inhibited by TPA, and growth rates were actually somewhat elevated in cells exposed to lo-’ M TPA (1.3 times the growth rate of untreated cultures, data not shown). In no experiments was 1O-9 M TPA ever found to have an effect on either tyrosinase activity or growth rates. The concentration of TPA needed to produce a maximal inhibition of tyrosinase activity (lo-’ M) is comparable to other data with either melanocytes (inhibition of tyrosinase was noted at lo-’

84

M) or with melanoma cells (lo-’ M showed the highest inhibition). However, to provide additional evidence that the inhibitory response seen witfi TPA is, in fact, a specific response (involving protein kinase C) and not due to some non-specific binding of this highly lipophilic phorbol ester, two additional phorbol esters were studied for their inhibitory effects on tyrosinase. The highly specific phorbol ester, 12,13_dibutyrate (PDBu) (Blumberg, 1988) reduced basal tyrosinase activity approximately 50% (at lo-’ M) by 24 h and antagonized the MSH induction of the enzyme. On the other hand, the phorbol ester, 4cy-phorbol 12,13-didecanoate (4a-PDD, lo-’ M), which does not bind protein kinase C (Blumberg, 1988), had no inhibitory effect on either basal tyrosinase activity or on the MSH stimulation of the enzyme (Fig. 2). These results suggest that TPA is functioning specifically through kinase C to exert its inhibitory effects. Since the dose-response data showed that the MSH-induced increase in tyrosinase activity could be reduced by TPA, and because MSH is known to exert its effects through CAMP (Wong and Pawelek, 1973; Fuller and Viskochil, 1979; Fuller et al., 1987), we next investigated the capacity of TPA to inhibit the action of compounds which elevate intracellular CAMP levels. We found that the level of induction of tyrosinase by either dibutyryl CAMP, or isobutylmethylxanthine (which inhibits phosphodiesterase activity) was inhibited by TPA (Fig. 3). We also examined the effect of TPA on cells treated with a potent MSH analogue, Nle4,D-phe’-a-MSH which reportedly binds almost irreversibly to the MSH receptor and is 10 times more potent than (u-MSH (Sawyer et al., 1980, 1982). Again TPA was found to be effective in antagonizing the stimulation of tyrosinase by this MSH analogue (Fig. 3). The inhibitory effect of TPA could be due either to: (1) inhibition of tyrosinase activity, (2) inhibition of tyrosinase synthesis from pre-existing mRNA, or (3) inhibition of tyrosinase mRNA levels. Since earlier studies on chick melanocytes suggested that TPA acts to lower the cellular content of tyrosinase protein (Oetting et al., 1985), we investigated the effect of this phorbol ester on both the basal level of tyrosinase mRNA in melanoma cells and on the induction of tyrosinase

a L E

300

5

250

6 n

200

control MSH TPA MO-‘Ml TPA llO-61 PDBU (low PDBu llO-%M)

0 I EZI 0 m EZI

untreated

250

200

150

MSH

IlO-‘MI

MSH

(10~‘MI

I

0 control MSH I TPA IlO-‘Ml m TPA (lo-@MI, = 4aPDD ilO- M)ESJ 4aPDD IlO-%4)EXl

100

50

-

0

Fig. 2. Effects of different phorbol esters on tyrosinase activity in melanoma cell cultures. Melanoma cells (2 x 10’ cells/flask) were seeded into 25 cm2 flasks and allowed to attach for 6 h. At this time the medium was replaced with Ham’s F-10 nutrient medium containing 1 pCi/ml of [‘HJtyrosine, 10% horse serum and antibiotics as described in Materials and Methods. The phorbol esters, TPA and PDBu (panel A) or TPA and 4a-PDD (panel B) were added at the concentrations shown and, at the same time, cells were either exposed to MSH (lo-’ M) or left untreated. At 24 h, the medium was removed and assayed for ‘H,O as described in Materials and Methods. Panel A: control, open bars; MSH with no TPA, solid bars; TPA (lo-’ M), left diagonal; TPA (10-s M), horizontal lines; PDBu (lo-’ M), right diagonal; PDBu (IOU’ M), crosshatch. Panel B: control, open bars; MSH with no TPA, solid bars; TPA (lo-’ M), left diagonal; TPA (lo-* M), horizontal lines; 4a-PDD (lo-’ M), right diagonal; 4a-PDD (lo-” M), crosshatch. Values are the averages of six determinations &SD.

mRNA by MSH. Fig. 4 shows a Northern blot of tyrosinase mRNA from cells treated with TPA in the presence or absence of MSH for 24 and 48 h. It is clear from lanes 3 (24 h exposure to TPA) and 7 (48 h exposure to TPA) that tyrosinase mRNA levels are markedly depressed with a 48 h exposure to TPA resulting in a 93% reduction in the level of tyrosinase mRNA present in the cell. Fig. 5 shows the comparison of the Northern blot

85

I

0 5

lYROSlNASE

.I

control

6.0 FZd

TPA

data from Fig. 4 to tyrosinase activity levels measured in the same cells. Tyrosinase activity levels generally correlate with the level of tyrosinase mRNA in the cell cultures. Of interest is the observation that while TPA is almost totally effective in reducing basal levels of tyrosinase mRNA, the cells can still respond to MSH by increasing tyrosinase mRNA levels and activity. Apparently the principle inhibitory effect of TPA resides at the level of reducing constitutive gene activity.

12345678 B -28S-

NZlMlY

(X of control)

24hn

24 hn

4.Yhn

48hn

TPA (W7M)

Fig. 3. Effect of MSH, dbcAMP, IBMX and [Nle4,D-phe’]-aMSH on tyrosinase activity in TPA-treated cells. Melanoma cells were seeded into 25 cm2 flasks as described in Materials and Methods and treated with either MSH (lo-’ M; solid bars), dibutyryl CAMP (dbcAMP, 1O-4 M; narrow crosshatch), isobutylmethylxanthine (IBMX, lo-4 M; wide crosshatch), or with [Nle4,n-phe7]-a-MSH (lo-’ M; diagonal). Medium and hormones were replaced every 24 h and the amount of 3H,0 produced determined as described in Materials and Methods. Values are the averages of six determinations f SD.

12345678

IYROSIME

Nle4.D-phe7MSH

0.0 without

mRNA (X of control)

e

-l8S-

Fig. 4. Northern blot of tyrosinase mRNA from melanoma cells treated with TPA and MSH. Melanoma cells were seeded into 75 cm2 flasks (lo6 cells/flask) and either left untreated (lanes 1 and 5), treated with TPA (lo-’ M; lanes 3 and 7), MSH (lo-’ M; lanes 2 and 6) or with TPA and MSH together (lanes 4 and 8) for either 24 h (lanes l-4) or for 48 h (lanes 5-8). Cells were then removed from flasks, lo5 cells were used for tyrosinase assays, and the remainder of the cells processed for total RNA. Northern blots were carried out as described in Materials and Methods. Panel A is tyrosinase mRNA and panel B is GAPDH mRNA.

MSH/-rPA TPA

350 300

250 200

150

100

50

50

100

150

200

250

4

Fig. 5. TPA effects on tyrosinase mRNA and activity levels in melanoma cell cultures. Tyrosinase activities in the cell cultures used for the Northern analysis shown in Fig. 4 were determined in cell homogenates as described in Materials and Methods. The tyrosinase mRNA levels were quantitated from the Northern blots shown in Fig. 4 by a computing densitometer. All RNA values were normalized to the GAPDH levels in order to control for any variabilities in the amount of RNA loaded onto the gels. Tyrosinase values are from triplicate assays k SD.

Previous studies from our laboratory have shown that tyrosinase activity can be inhibited in melanoma cells treated with either insulin (Fuller and Ehlers, 1984) or with calcium ionophore A23187 (Fuller, 1987). To determine if the inhibition of tyrosinase activity by insulin, like that shown by TPA, is the result of an insulin-mediated down-regulation of tyrosinase mRNA levels, we have carried out Northern analysis of RNA prepared from insulin-treated cell cultures. The results show that cells treated with insulin also had markedly depressed levels of tyrosinase mRNA, a finding which suggests that insulin may exert its inhibitory effects through the involvement of a C-kinase-mediated pathway (Fig. 6). Discussion

We have shown, in this study, that the phorbol ester, TPA is a potent inhibitor of tyrosinase mRNA in melanoma cells. Previous studies from other laboratories have shown that either

86 TYROSINASE

350 300

mRNA

250 200

(% of control)

150

100

50

TYROSINASE ACTIVITY (% of co&olf

50

100

150

200

250

Fig. 6. Insulin effects on tyrosinase mRNA and activity levels in melanoma ceil cultures. Melanoma cells were exposed to either insulin (lo-’ M) alone or to insulin plus MSH (10.. ’ M) for 48 h. Cells were removed from flasks, 10’ cells were homogenized and assayed for tyrosinase, and the remainder were processed for RNA. RNA quantitation and tyrosinase assays were carried out as described in Materials and Methods. Tyrosinase values are from triplicate assays + SD.

melanoma cells or chick melanocytes have reduced melanogenic capacity and tyrosinase activity following treatment with TPA (Mufson et al., 1979; Oetting et al., 1985). In the chick melanocyte study, the synthesis of nine proteins was found to be inhibited by 10m7 M TPA and one of these was tyrosinase (Oetting et al., 1985). Results from one study with B16 mouse melanoma cells demonstrated the inhibition of two proteins associated with the appearance of melanogenesis (Laskin et al., 1983). In this study TPA was used at a concentration of 1.6 x lo-’ M. In another study concerned with the inhibition of melanogenesis in melanoma cells by TPA, concentrations ranging from 3.2 x 10e7 M to 0.16 nM were used with the greatest in~bition of melanogenesis observed at 1.6 x lo-’ M TPA (La&in et al., 1983). The results from our study have now shown that the TPA-mediated inhibition of tyrosinase and melanin synthesis in melanoma cells and in meianocytes is due largely to the lowering of tyrosinase mRNA levels. The role of phorbol esters in altering gene activity has been investigated in other cell systems and TPA-responsive DNA elements, APl and AP2, have been identified (Imagawa et al., 1987; Deutsch et al., 1988; Roesler et al., 1988). The API element (TRE) has the consensus sequence, TGA(C or G)TCA, and is responsive to TPA alone, while the AP2 element (consensus sequence

CCCCAGGC) responds to both TPA and to cyclic AMP (Chiu et al, 1987). Analysis of the 5’ DNA sequence in the mouse tyrosinase gene reveals three possible AP2 binding sites which show 75% homology with the AP2 consensus sequence. The 5’ ends of these sequences are located at positions -917, -911, and - 291 from the transcriptional start site reported by Yamamoto and coworkers (1989). Recent studies have identified regulatory proteins (trams-acting factors) which interact with these sites to alter transcriptional activity of associated genes (Imagawa et al., 1987). A cyclic AMP-responsive element, CRE, has also been identified in a number of CAMP-responsive genes and the octamer consensus sequence of this element, TGACGTCA, is quite similar to the heptamer sequence of the TRE (see Roeslet et al., 1988 for review). An examination of the 5’ upstream sequences of the tyrosinase gene reveals the presence of one possible CRE (87% homologous) at positions - 148 to - 141. This element, however, contains a CTCA sequence instead of the invariant GTCA sequence characteristic of CAMP-responsive genes (Roesler et al., 1988). Thus, it is questionable whether or not this putative CRE functions as a CAMP-responsive element. Of particular interest is the presence of a TRE-related sequence, TTAGTCA at positions -108 to - 102. Whether this particular sequence is involved in the TPA-mediated down-regulation of tyrosinase gene activity remains to be determined. To our knowledge, in all those systems where the function of the TRE has been examined, TPA has been shown to stimulate transcriptional activity through this element (Angel et al., 1987; Chiu et al., 1987). In the studies described here, however, TPA was found to be a potent inhibitor of tyrosinase gene expression. Whether or not TPA acts through the same TRE sequence and trans-acting factors in melanoma cells to down-regulate tyrosinase gene activity as it does in other systems to up-regulate genes remains to be determined. We do not know, at present, whether the inhibition of tyrosinase mRNA levels is due to reduced tyrosinase gene activity or due to increased turnover of tyrosinase mRNA. Given the extensive evidence that TPA exerts its effects primarily on transcription and does so fairly rapidly, it is likely that the TPA-induced down-

87

regulation of tyrosinase mRNA levels occurs at the transcriptional level. Of additional interest in this study is our finding that TPA does not completely prevent the hormonally induced increase in tyrosinase mRNA. Thus the CAMP-responsive area of the tyrosinase gene (if, in fact, CAMP promotes increased transcription and not mRNA stabilization) must be able to act independently from the TPA-sensitive area and promote increased transcription even in the presence of TPA binding its site and reducing promoter activity. Our observation that insulin also reduces tyrosinase activity and tyrosinase mRNA levels suggests that this hormone may exert its action by cellular mechanisms which are similar to those involved in TPA action, namely, protein kinase C. The phorbol ester, TPA was found by Eisenger and coworkers to promote the growth of human melanocytes in culture, and this important discovery has made the long-term culture of human melanocytes possible (Eisinger et al., 1979; Eisinger and Marko, 1982; Lerner et al., 1988). However, the data presented in this paper and that from other laboratories which shows that phorbol esters inhibit melanogenesis in both melanoma cells and in melanocytes suggest that human melanocytes cultures grown continuously in the presence of TPA may not be appropriate models for the study of the regulation of human pigmentation. Studies are now in progress to identify the TPA- and CAMP-sensitive regulatory elements on the tyrosinase gene and also to isolate and characterize the regulatory proteins which interact with these elements to control the expression of tyrosinase in melanocytes. References Angel, P., Imagawa, M., Chiu, R., Stein, B., Imbra, R.J., Rahmadorf, H.J., Jonat, C., Herrlich, P. and Karin, M. (1987) Cell 49, 729-739. Blumberg, P.M. (1988) Cancer Res. 48, 1-8. Chiu, R., Imagawa, M., Imbra, R.J., Bockoven, J.R. and Karin, M. (1987) Nature 329, 648-651.

Chomczynski, P. and Sacchi, N. (1987) Anal. B&hem. 162. 156-159. Deutsch, P.J., Hoeffler, J.P., Jameson, J.L. and Habener, J.F. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 7922-7926. Eisinger, M. and Marko, 0. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 2018-2022. Eisinger, M., Lee, J.S., Hefton, J.M., Darzynkiewicz, 2.. Chiao, J.W. and deHarven, E. (1979) Proc. Natl. Acad. Sci. 76, 5340-5344. Fort, P., Marty, L., Piechaczyk, M., Sabrouty. S.E., Dani, C., Jeanteur, P. and Blanchard, J.M. (1985) Nucleic Acids Res. 13, 1431-1442. Fuller, B.B. (1987) Pigment Cell Res. 1, 176-180. Fuller, B.B. and Ehlers, SE. (1984) Endocrinology 114, 222226. Fuller, B.B. and Lebowitz, J. (1980) J. Cell. Physiol. 103, 279-287. Fuller, B.B. and Viskochil, D.H. (1979) Life Sci. 24, 2405-2416. Fuller, B.B., Lunsford, J.B. and Iman, D.S. (1987) J. Biol. Chem. 262, 402444033. Fuller, B.B., Iman, D.S. and Lunsford, J.B. (1988) J. Cell. Physiol. 134, 149-154. Hoganson, G.E., Ledwitz-Rigby, F., Davidson, R.L. and Fuller, B.B. (1989) Somatic Cell Mol. Genet. 15, 255-263. Imagawa, M., Chiu, R. and Karin, M. (1987) Cell 51, 251-260. Jimenez, M.. Kameyama, K., Maloy, W.L., Tomita, Y. and Hearing, V.J. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 3830-3834. Kwon, B.S., Wakulchik, M., Haq, A.K., Halaban, R. and Kestler, D. (1988) B&hem. Biophys. Res. Commun. 153, 1301-1309. Laskin, J.D., Piccinini, L., Engelhardt, D.L. and Weinstein, I.B. (1983) J. Cell. Physiol. 114, 68-72. Lemer, A.B., Halaban, R., Klaus, S.N. and Moellmann, G.E. (1988) J. Invest. Dermatol. 1987, 219-224. Mufson, R.A., Fisher, P.B. and Weinstein, LB. (1979) Cancer Res. 39, 3915-3919. Getting, W.. Langner, K. and Brumbaugh, J.A. (1985) Differentiation 30, 40-46. Roesler, W.J., Vandenbark, G.R. and Hanson, R.W. (1988) J. Biol. Chem. 263, 9063-9066. Sawyer, T.K., Sanfilippo, P.J., Hruby, V.J., Engel, M.H., Heward, C.B., Burnett, J.B. and Hadley, M.E. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 5754-5758. Sawyer, T.K., Hruby, V.J., Wilkes, B.C., Draelos, M.T., Hadley, M.E. and Bergsneider, M. (1982) J. Med. Chem. 25, 1022-1027. Wong, G. and Pawelek, J. (1973) Nature New Biol. 241, 213-215. Wong, G., Pawelek, J., Sansone, M. and Morowitz, J. (1974) Nature 248, 351-354. Yamamoto, H., Takeuchi, S., Kudo, T., Sato, C. and Takeuchi, T. (1989) Jpn. J. Genet. 64, 121-135.