Induction of Ornithine Decarboxylase Activity and ... - Cancer Research

6 downloads 0 Views 944KB Size Report
Induction of Ornithine Decarboxylase Activity and DMA Synthesis in. Hairless Mouse Epidermis by Retinoids1. Michael J. Connor2 and Nicholas J. Lowe.
[CANCER RESEARCH 43,5174-5177,

November 1983]

Induction of Ornithine Decarboxylase Activity and DMA Synthesis in Hairless Mouse Epidermis by Retinoids1 Michael J. Connor2 and Nicholas J. Lowe Division of Dermatology, UCLA School of Medicine, University of California. Los Angeles, California 90024

in normal epidermis (9,10,16),

ABSTRACT The ability of all-frans-retinoic acid (RA) and other retinoid derivatives to enhance DNA synthesis and to induce omithine decarboxylase [L-omithine carboxylyase; EC 4.1.1.17 (ODC)] activity has been investigated in normal and tape-stripped hair less mouse epidermis. Initial studies showed that the retinoids could inhibit the induction of epidermal ODC activity found 4.5 hr after tape stripping. Ten nmol RA, 13-c/s-retinoic acid (13-c/sRA), ethyl-all-frans-9-(4-methoxy-2,3,6-trimethylphenyl)-3,7-dimethyl-2,4,6,8-nonatetraenoate (aromatic retinoid), or ethyl-p[(E)-2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)-1-propenyi (benzoate (arotinoid ethyl ester) applied topically to the skin at 1 hr before tape stripping inhibited the induction of ODC activity. Induction of epidermal ODC activity was inhibited by arotinoid ethyl ester but not by RA, 13-c/s-RA, or aromatic retinoid when they were applied to the skin at 24 hr prior to tape stripping. RA applied topically to normal hairless mouse skin induced a dose-dependent increase in epidermal ODC activity, detectable 24 hr or more after treatment. RA induced epidermal ODC activity to levels only 15- to 30-fold less than found after treatment with the potent tumor promoter tetradecanoylphorbol-13-acetate. Ep idermal ODC activity was also induced by topical 13-c/s-RA, aromatic retinoid, and arotinoid ethyl ester at this time, although in lower amounts than after RA treatment. The induction of ODC activity by RA was itself inhibited by topical arotinoid ethyl ester treatment. RA, 13-c/s-RA, and aromatic retinoid induced ODC activity at doses below those required to enhance epidermal DNA synthesis. In summary, we have shown that, in common with other proliferative stimuli, retinoids can induce ODC activity in hairless mouse epidermis per se. Our results suggest that, because of their ability to also inhibit the expression of ODC activity, the induced ODC activity is found only after the retinoids have been depleted. The ability both to inhibit and to induce ODC activity may be related to the action of RA as a weak tumor promoter under certain conditions and as an inhibitor of promotion under others. INTRODUCTION Retinoids have paradoxical tissue proliferation. They have ative properties (11, 29) and treatment of hyperproliferative However, retinoids can induce

effects on the rates of cell and anticarcinogenic and ant iprolifer are used therapeutically in the diseases such as psoriasis (22). hyperproliferation and acanthosis

1This investigation was supported by USPHS Grant CA 25970, awarded by the National Cancer Institute, Department of Health and Human Services, and by grants from the Dermatologie Research Foundation of Calif. 2To whom requests for reprints should be addressed. Received December 20,1982;

5174

accepted August 2,1983.

can enhance DMBA3 skin carcin-

ogenesis (27), and can act as tumor promoters after DMBA initiation (15). ODC converts L-ornithine into putrescine, from which spermidine and spermine can be formed. These polyamines are thought to be essential to normal cell division, differentiation, and growth (23). In 1976, O'Brien (20) suggested that the induction of ODC activity may be an essential step in chemical tumor promotion. The enzyme is also induced by mechanical stimuli such as wounding and abrasion (8), by tape stripping (3, 5), and by complete carcinogens such as DMBA (21) and UV (17). The increase in ODC activity occurs at a time dependent upon the particular stimulus used, but usually before the gross appearance of a hyperproliferative response or the accompanying increase in DNA synthesis. Retinoids can inhibit the expression of ODC activity induced by a variety of stimuli, and one proposed mechanism of action of the retinoids is by their specific inhibition of the induction of ODC activity (14, 29). When different retinoids were applied to mouse skin shortly before or after TPA treatment, their inhibition of the induction of ODC activity correlated with their anti-tumorpromoting properties (28). However, although systemic retinoid treatment reduced DMBA carcinogenesis in tissues such as breast (26) and rabbit ear-skin (18), topical RA enhanced both DMBA skin carcinogenesis and the induction of epidermal ODC activity by DMBA (27). We found that RA treatment regimens that inhibited the induction of ODC activity in hairless mouse epidermis by UVB (17) also reduced UVB skin carcinogenesis (9). Other workers using different treatment regimens found inhibition or no effect on UV skin carcinogenesis (12) or found an enhancement of UV skin carcinogenesis by RA (13). Generally, ODC activity is elevated in proliferating tissues and cancers and in cells stimulated to divide (reviewed in Refs. 23 and 24), but ODC is reportedly not induced by retinoids in the epidermis (28), although this has been observed in mouse mel anoma cells in culture (25). Thus, retinoids appear to have the ability to induce proliferation without requiring the induction of ODC activity in some circumstances. In the present study, we have reinvestigated the effects of topical retinoids on the epidermal ODC activity of normal and tape-stripped hairless mice. Our experiments show that single doses of RA and other retinoids, even at doses below those required to enhance DNA synthesis and in the range used in carcinogenesis studies, are able both to induce and to inhibit the induction of ODC activity in mouse epidermis in vivo. 9The abbreviations used are: DMBA, 7.12-dimethylbenz(a)anthracene; ODC, omithine decarboxylase (L-omithine carboxy-lyase; EC 4.1.1.17); TPA, tetradeca noylphorbol-13-acetate; RA, alkrans-retinoic acid; UVB, ultraviolet-B; aromatic retinoid, ethyl-all-frans-9-(4-methoxy-2,3,6-trimethylphenyl)-3,7-dimethyl-2,4,6,8nonatetranoate; 13-c/s-RA, 13-c/s retinole acid; arotinoid ethyl ester, ethyl-p-|(E>2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)-1-propenyl]benzoate.

CANCER

RESEARCH

VOL. 43

Retinólas and Epidermal ODC and DNA Synthesis MATERIALS AND METHODS Materials. L-[1-14C)Omithine hydrochloride (specific radioactivity, 57 Ci/mol), [mefriy/-3H]thymidine (specific radioactivity, 2 Ci/mmol), and NCS Solubilizer were bought from Amersham (Arlington Heights, III.). L-Omithine, RA, pyridoxal-5-phosphate, dithiothreitol, and calf thymus DNA were bought from Sigma Chemical Co. (St. Louis, Mo.). Aromatic retinola (Etretinate; Ro 10-9359), 13-c/s-RA (Acculane; Ro 4-3780), and arotinoid ethyl ester (Ro 13-6298) were gifts kindly provided by Hoffman LaRoche Inc., (Nutley, N. J.). All experiments were performed on 8- to 10-weekold hairless female SKH/hr1 mice originating from the Skin Cancer Hospital (Temple University, Philadelphia, Pa.). Preparation of Mouse Epidermis. Mice were killed by cervical dislo cation, and the back skin was removed. To separate the epidermis, the whole skin was immersed in a water bath at 55°for 30 sec, cooled in ice-water, blotted dry with filter paper, and placed flat on a cooled glass plate, and the epidermis (0.08 to 0.15 g) was scraped off the dermis with a scalpel blade. For ODC assay, each epidermis was homogenized in ice-cold 50 HIM-sodium phosphate buffer, pH 7.2 (1 ml), containing 0.4 mw pyridoxal 5-phosphate and 5 rriM dithiothreitol, with a Brinkmann homogenizer, at setting 6, for 15 sec. The homogenates were centrifuged at 30,000 x g for 30 min, and the supematants were recovered and stored at -70° until used. The stored extracts retained their activity for several months under these conditions. To measure DNA synthesis, mice were given injections 1 hr prior to sacrifice with 30 ¿(Ci[3H]thymidine, the epidermis was recovered as above, and the DNA was extracted and assayed according to the method of Baird ef al. (2). Each epidermis was homogenized in 1-ml ice-cold 0.4 M perchloric acid with a Brinkmann homogenizer, at setting 6, for 30 sec. After homogenates had stood on ice for 15 min, they were centrifuged (1500 rpm), and the pellets were recovered and washed free of unbound [3H]thymidine by resuspending in a further 1-ml 0.4 M perchloric acid and recentrifuging, resuspending, and recentrifuging twice in absolute ethanol. The pellets were then suspended in 2 ml of 0.4 M perchloric acid, and the DNA was hydrolyzed by heating at 90° for 20 min in a block heater. Tape Stripping. Mice were given i.p. injections of chloral hydrate (300 mg/kg body weight) 15 min before tape stripping. Since ODC induction by tape stripping was "dose dependent," each mouse was stripped the same number of times (10 times, unless otherwise stated) with Scotch Magic Tape. During tape stripping, strips of tape about 10 cm long were placed along the back of the mice, pressed down firmly to make contact with the skin, and massaged on the skin for about 10 sec before being removed by a single pull. Each piece of tape was used once only. Retinoid Treatments. Retinoids were applied topically in an acetone vehicle in all experiments, and the solutions were made up immediately before use. The retinoid solutions (0.2 ml) were applied to the dorsal skin of the mice as evenly as possible, using micropipets with disposable tips, over an area of about 20 sq cm. All of the procedures were performed under subdued light, and the mice were housed in the dark for 4 to 5 hr following treatment to minimize photodegradation of the retinoids. ODC Assay. ODC activity in epidermal samples was determined by measuring the release of 14CO2from L-[1-14C]ornithine. Incubations were performed in 0.05 M sodium phosphate buffer, pH 7.2, containing 5 mw dithiothreitol, 0.4 ÕTIM pyridoxal 5-phosphate, 1 mw i.-ornithine, 0.5 nCi L[1-14C]omithine, and 0 to 0.2 ml sample in a final volume of 0.3 ml. The reaction mixtures were incubated at 37°for 1 hr in sealed tubes fitted with side-arms. The released "CO2 was collected directly into scintillation vials, containing 0.15 ml of NCS Solubilizer, attached to the side-arms by holed rubber stoppers. The reactions were terminated by injecting 2 M citric acid (0.5 ml) through a sleeve stopper in the mouth of each tube. Total protein in the epidermal extracts was determined using the BioRad protein assay (Bio-Rad Laboratories, Richmond, Calif.). ODC activ ities are expressed as nmol CO? released per mg soluble protein per hr. DNA Synthesis. Total epidermal DNA was determined on aliquots of NOVEMBER

1983

hydrolyzed samples by the diphenylamine reaction (6) using calf thymus DNA as the standard. The specific radioactivity of the DNA was deter mined by counting 0.2-ml aliquots of each hydrolyzed sample in a scintillation counter. DNA synthesis was expressed as cpm/^g DNA. Histology. Representative skin biopsies were taken from the mice at 24 to 48 hr after treatment and were fixed, sectioned, and stained as described previously (9).

RESULTS Properties of ODC Induced by Tape Stripping. The time course of the induction of ODC activity in SKH/hr1 mouse epidermis after tape stripping was similar to that in guinea pig ear epidermis (3). Peak ODC activity occurred at 4 to 5 hr after tape stripping and returned to control levels after 24 hr (Chart 1). The induction of ODC was dose-dependent, i.e., the greater the insult the higher the ODC activity (Chart 1, inset). Kinetic parameters, determined with crude extracts of epidermis from mice stripped 15 times as the enzyme source, were similar to those of ODC induced by the tumor promoter TPA or by UVB (Kmapppyridoxal 5-phosphate, 0.026 mw; of L-omithine, 0.08 mM). The ODC activity in the crude extracts was not inhibited by any of the retinoids when they were added in the concentrations used in this study. Effects of Topical Retinoids on Epidermal ODC Activity after Tape Stripping. Mice (5/group) were painted with 10 nmol retinoid in 0.2 ml acetone at 1 or 24 hr before being tape stripped (10 tape strippings). The mice were killed at 4.5 hr after tape stripping, and the epidermal ODC activity was determined. All of the retinoids significantly inhibited the induction of ODC activity when applied at 1 hr before tape stripping (Table 1). Only arotinoid ethyl ester significantly inhibited the induction of ODC activity when applied at 24 hr before tape stripping. These results suggest that RA, 13-c/s-RA, and aromatic retinoid remain active in the epidermis for a shorter period of time than does arotinoid ethyl ester. Effects of Topical Retinoids and Tape Stripping on DNA Synthesis in Hairless Mouse Epidermis. A single dose of 10nmol RA had no effect on DNA synthesis when measured at 0, 4, 24, or 48 hr later. A single dose of 10 nmol arotinoid ethyl

5 10 Number of »trippings U

o o o 24 Tim«,hours Chart 1. Induction of ODC activity by tape stripping. Groups of mice were tape stripped (x 10) and sacrificed at 0 to 48 hr later. Points, mean from at least 3 mice. Inset, dose dependency of the induction of ODC activity by tape stripping; points, mean epidermal ODC activity determined at 4 hr after mice (5 per group) were tape stripped for the appropriate number of times.

5175

M. J. Connor and N. J. Lowe ester significantly enhanced DNA synthesis to 3- to 4-fold higher compared to the acetone-treated controls. Treatment with 100 nmol RA also enhanced DNA synthesis (Table 2). A single treatment with 100 nmol 13-c/s-RA or 100 nmol aromatic retinoid failed to alter epidermal DNA synthesis at this time. Retinoid treatments, which increased [3H]thymidine incorporation into DNA, induced epidermal acanthosis by 24 to 48 hr, with an increase from the control 2 to 3 epidermal cell layers to 3 to 4 cell layers after 100 nmol RA, and 4 to 5 cell layers after 10 nmol arotinoid ethyl ester, confirming that the enhanced thymidine incorporation observed represented a real increase in DNA syn thesis. Tape stripping evoked a significant increase in DNA synthesis which peaked at 24 hr and slowly returned to control levels by 72 hr. The time course of DNA synthesis after tape stripping was unchanged when mice were treated with 10 nmol RA 1 hr before tape stripping. Treatment with 10 nmol arotinoid ethyl ester significantly elevated epidermal DNA synthesis (Table 2) to above that after tape-stripping or arotinoid treatment alone. The experiment produced no support for a differential effect of retinoids on DNA synthesis in normal compared to hyperproliferative epidermis. Effects of Topical Retinoids on ODC Activity in Normal Hairless Mouse Epidermis. Table 3 shows the ODC activity in epidermis from mice treated with RA (3.4,10, or 100 nmol), 13c/s-RA (100 nmol), aromatic retinoid (100 nmol), or arotinoid ethyl ester (10 nmol) and killed at 24 hr later. There was a dosedependent induction of ODC activity by RA at 24 hr which was Table 1 Effect of retinoids on the induction of epidermal ODC activity by tape stripping Groups of 5 mice were tape stripped (10 times) and killed 4.5 hr later. The mice were treated with acetone or retinoid in acetone at 1 or 24 hr prior to tape stripping. ODC activity (nmol CO2/mg soluble protein/ hr) at following time of treatment with reti noid prior to tape stripping TreatmentAcetone

hr3.25 hr4.30 ±1.93* (1OOf (0.2 ml) ±1.59 (100) 0.29 ±0.37° (8.9) RA (10 nmol) 5.02 ±3.72 (107) 0.70±0.35C (21.5) 13-c/s-RA(10nmol) 4.39 ±1.55 (102) 1.78±1.08C (55) Aromatic retinoid (10 nmol) 3.70 ±1.50 (86) 0.31 ±0.25C (5.6)24 1.59±1.06C (37) Arotinoid ethyl-ester (10 nmol)1 8 Mean ±S.D. '' Numbers in parentheses, percentage of control (acetone-treated) activities. c Significantly lower than that of acetone-treated controls; p < 0.05.

Table 2 Comparative effects of topical retinoid treatment on DNA synthesis in hairless mouse epidermis Groups of 5 mice were dosed topically with retinoid and sacrificed at 48 hr later or were dosed, tape stripped after 1 hr, and sacrificed at 48 hr later. Treatment DNA synthesis (cpm/pg) 48.0 ± 2.75" (1OOf Acetone (0.2 ml) RA (10 nmol) in acetone (0.2 ml) 50.3 ±12.5 (105) 90.7 ± 2.48e (189) RA (100 nmol) in acetone (0.2 ml) 155.5 ±25.6e (324) Arotinoid ethyl ester (10 nmol) in acetone (0.2 ml) 13-cis-RA (100 nmol) in acetone (0.2 ml) 48.7 ± 4.62 (101) Aromatic retinoid (100 nmol) in acetone (0.2 46.8 ± 5.76 (98) ml) 75.4 ±17.7°(157) Acetone (0.2 ml), 1 hr prior to tape-stripping 172.0 ±33.7" (359) Arotinoid ethyl ester (10 nmol). 1 hr prior to tape-stripping " Mean ±S.D. of the specific radioactivity (cpm/^g DNA). '' Numbers in parentheses, percentage of the controls treated with acetone only. c Significantly higher than in acetone-treated controls (p < 0.05). " Significantly higher than in acetone-treated tape-stripped controls (p < 0.05).

5176

Tables ODC activity induced in hairless mouse epidermis at 24 hr after treatment with topical retinoids

TreatmentAcetone (0.2 ml) RA (3.4 nmol) in acetone (0.2 ml) RA (10 nmol) in acetone (0.2 ml) RA (100 nmol) in acetone (0.2 ml) Arotinoid ethyl ester (10 nmol) in acetone (0.2 ml) RA (100 nmol) and arotinoid ethyl ester (10 nmol) in acetone (0.2 ml) 13-c/s-RA (100 nmol) in acetone (0.2 ml) Aromatic retinoid (100 nmol) in acetone (0.2 ml)ODC

activity (nmolCOz/mg soluble protein/hr)0.081 ±0.048a

rela tive to control1

0.1 39 ±0.080 0.34 ±0.15 2.07 ±1.41 ±0.060.13 0.18

1.7 4.4 25.5 2.21.65.2

±0.040.41 8 ±0.24 0.1 26 ±0,07Activity

1.5

Mean ±S.D. of the activity in the epidermis from 5 mice.

the time of maximum induction of activity. Topical 13-c/s-RA, aromatic retinoid, and arotinoid ethyl ester treatment also in duced ODC activity, although in smaller amounts. ODC activity at 24 and 48 hr after treatment with 10 nmol arotinoid ethyl ester was significantly lower than after 10 nmol RA, possibly reflecting the longer half-life of arotinoid in the epidermis'4 and its greater inhibitory potency. Ten nmol arotinoid ethyl ester applied with 100 nmol RA inhibited the induction of ODC activity by RA. DISCUSSION In this study, we have shown that retinoids can both induce ODC activity and inhibit the expression of ODC activity in mouse epidermis. The induction of ODC activity by RA in the epidermis was dose dependent; 10 nmol RA induced a 4-fold increase, and 100 nmol RA induced a 25-fold increase in ODC activity 24 hr later. Epidermal ODC activity was induced 60-fold by 1.7 nmol and 130-fold by 17 nmol of the potent tumor promoter TPA (21); thus, on a molar basis, RA is only 15 to 30 times less potent than TPA at inducing ODC activity. The ability to induce ODC activity was shared by 13-c/s-RA, aromatic retinoid, and arotinoid ethyl ester, although these were less potent than RA. The ODC activity induced 24 hr after 10 nmol arotinoid ethyl ester was much lower than after 10 nmol RA. This may reflect the more potent inhibitory properties and ability to persist longer in the epidermis of the arotinoid, as indicated by its ability to inhibit the induction of ODC activity by RA or tape stripping when applied 24 hr earlier. The mechanism by which retinoids inhibit the expression of ODC is obscure. Results from other studies have indicated that the retinoids may act at a late stage in the induction process, exerting a specific metabolic control at the level of transcription or translation (4). The evidence for this is indirect and rests on the failure either to find specific ODC inhibitors induced by retinoids, or to find any significant alteration in the affinity of ODC for its substrate, or to observe changes in the time course of induction and decay of ODC after retinoid treatment. Other workers have suggested that the site of action of the retinoids may be posttranslational. Posttranslational modification of ODC by proteases (19), phosphorylation (1), antizyme action (7), and transglutaminase (25) has been documented. The latter example may be of particular interest, since transglutaminase uses pu4 M. J. Connor, M. Lincte, and N. J. Lowe, unpublished observations.

CANCER

RESEARCH

VOL. 43

Retinoids and Epidermal ODC and DNA Synthesis trescine as the substrate, and transglutaminase may be induced by retinoids (25). It seems unlikely that the retinoids both induce and inhibit the expression of ODC activity by the same mechanism. The tran sient nature of the inhibitory effect of the retinoids and their ability to induce ODC activity per se may suggest that the more likely site of inhibitory action is posttranslational, rather than the induction process itself. Until more direct evidence of the site of action is available, it may be more correct, therefore, to refer to retinoids as inhibitors of ODC expression rather than induction. If retinoids do induce a posttranslational modification of ODC, the failure to see ODC activity induced by topical retinoids at times earlier than 24 hr may simply reflect their induction of the modification. The ability of arotinoid to inhibit the expression of ODC activity by RA supports this. This possibility could be readily examined using a monospecific antisera to ODC. The absence of induced ODC protein in the epidermis after retinoid treatment would tend to support a pretranslational site of action for the retinoids. The ability both to inhibit and to induce the expression of ODC activity may be relevant to understanding some of the apparently contradictory reports of the effects of retinoids on ODC activity. If a low dose of a retinoid is applied immediately before or after a stimulus such as TPA, which induces peak ODC activity within 4 to 5 hr, the net effect would be a reduction in the overall amount of ODC activity expressed since, even if the retinoid dose were sufficient to induce ODC activity at later times, the induction of ODC by the more potent inducer TPA would be largely inhibited. However, if the retinoid is applied such that it induces ODC activity at a similar time and magnitude to that of a carcinogen, the combined effect on the level of ODC activity induced may be additive. DMBA induces lower levels of ODC activity in the epidermis than does TPA, and the activity peaks at 24 to 48 hr after application (27), a time at which RA can induce ODC activity. RA enhanced the induction of ODC activity (and carcinogenesis) by DMBA in mouse epidermis (27). The ability of retinoids to induce ODC activity raises questions on the specificity of this response for tumor promoters. The increased epidermal ODC activity after retinoids occurs in the same time frame as that of the enhanced DNA synthesis; how ever, 10 nmol RA and 100 nmol 13-c/s-RA both induced ODC activity without affecting DNA synthesis. Arotinoid ethyl ester was a potent stimulus of DNA synthesis but induced lower levels of ODC activity than equimolar amounts of RA. These observa tions suggest that the induction of ODC activity by retinoids is not simply a consequence of a wave of retinoid-induced prolif eration, but may represent a more specific response. REFERENCES 1. Atmar, V. J., and Kuehn, G. D. Pnosphorylatkxi of omithine decarboxylase by a polyamme dependent protein kinase. Proc. Nati. Acad. Sci. U. S. A., 78: 5518-5522,1981. 2. Baird, W. M., Sedgwick, J. A., and Boutwell, A. K. Effects of phorbol and four diesters of phorbol on the incorporation of tritiated precursors into DNA, RNA,

NOVEMBER

1983

and protein in mouse epidermis. Cancer Res., 37:1434-1439,1971. 3. Bolton, L. L, Constantino, B. E., and Rovee, D. T. The kinetics of omithine decarboxylase activity as a function of wounding in guinea pig ear. J. Invest. Dermatol, 76:480-483,1981. 4. Boutwell, R. K. Retinoids and the inhibition of omithine decarboxylase activity. J. Am. Acad. Dermatol., 6: 796-798,1982. 5. Breeding, J., Sakamoto, M., Lowe, N., and Puhvel, M. Epidermal omithine decarboxylase and DNA synthesis activity following different stimuli in the hairless mouse. Clin. Res., 30: 156,1982. 6. Burton, K. A study of the condition and mechanism of the diphenylamine reaction for the colorimetrie estimation of deoxyribosenucleic acid. Brachem. J., 62: 315-325, 1955. 7. Canellakis. E. S., Kynakidis, D. A., Heller, J. S., and Pawlak, J. W. The complexity of regulation of omithine decarboxylase. Med. Biol. (Helsinki),59: 279-285,1981. 8. Clarke-Lewis, I., and Murray, A. W. Tumor promotion and the induction of epidermal omithine decarboxylase activity in mechanically stimulated mouse skin. Cancer Res., 38: 494-497,1978. 9. Connor, M. J., Lowe, N. J., Breeding,J., and Chalet, M. Inhibition of uitravioletB skin carcinogenesis by all-frans-retinoic acid regimens that inhibit omithine decarboxylase induction. Cancer Res., 43:171-174,1983. 10. Elias, P. M., Fritsch, P. O., Lampe, M., Williams, M. L., Brown, B. E., Nemanic, M., and Grayson, S. Retinoid effects on epidemial structure, differentiation and permeability. Lab. Invest., 44:531-540,1981. 11. Elias,P. M., and Williams,M. L Retinoids,cancer and the skin. Arch. Dermatol., 117:160-180, 1981. 12. Epstein, J. H. Effects of retinoids on ultraviolet-induced carcinogenesis. J. Invest. Dermatol., 76:144-146,1981. 13. Forbes, P. D., Urbach, F., and Davies. R. E. Enhancement of experimental photocaroinogenesisby topical retinoic acid. Cancer Lett., 7:85-90,1979. 14. Haddox, M. K., and Russell, D. H. Cell cycle-specificlocus of vitamin A inhibition of growth. Cancer Res., 39: 2476-2480,1979. 15. Hennings, H., Wenk, M. L., and Donahoe, R. Retinoid acid promotion of papillomaformation in mouse skin. Cancer Lett., 6:1-5,1982. 16. Lawrence, D. J., and Bern, H. A. On the specificity of the response of mouse epidermis to vitamin A. J. Invest. Dermatol., 31: 313-325,1958. 17. Lowe, N. J., and Breeding, J. Retinoic acid modulation of ultraviolet lightinduced epidermal omithine decarboxylase activity. J. Invest. Dermatol.. 78: 121-124,1982. 18. Mährte,G., and Berger, H. DMBA-induced tumors and their prevention by aromatic retinoid (Ro 10-9359). Arch. Dermatol. Res., 272: 37-47,1982. 19. Murray, A. W., and Froscio, M. Induction of omithine decarboxylase activity by a tumor promoter: decreased response in mice maintainedon a vitamin B6 deficient diet. Btochem. Biophys. Res. Commun., 77:693-699,1977. 20. O'Brien, T. G. The induction of omithine decarboxylase as an early, possibly obligatory, event in mouse skin carcinogenesis.Cancer Res., 36:2644-2653, 1976. 21. O'Brien, T. G., Simsiman,R. C., and Boutwell, R. K. Induction of the polyamine biosynthetic enzymes in mouse epidermis by tumor-promoting agents. Cancer Res., 35:1662-1670,1975. 22. Orfanos, C. E. Oral retinoids—presentstatus. Br. J. Dermatol., 703:473-481, 1980. 23. Russell, D. H., and Dune, B. G. M. Polyamines as biochemical markers of normal and malignantgrowth. Progress in Cancer Researchand Therapy, Vol. 8, pp. 1-13. New York: Raven Press, 1978. 24. Scalabrino, G., and FerioJi, M. F. Polyamines in mammalian tumors. Adv. Cancer Res., 35:151-268,1982. 25. Scott, K. F. F., Meyskens, F. L., and Russell, D. H. Retinoids increase transglutaminaseactivity and inhibit omithine decarboxylase activity in Chinese hamster ovary cells and melanomacells stimulated to differentiate. Proc. Nat). Acad. Sei. U. S. A., 79: 4093-4097,1982. 26. Thompson, H. J., Meeker, L. D., Tagliaferro, A. R., and Baca, P. I. Effect of retinylacetate on the occurrence of ovarian hormone-responsive and -nonre sponsive mammary cancers in the rat. Cancer Res., 42:903-905,1982. 27. Verma.A. K., Conrad, E. A., and Boutwell, A. K. Induction of mouse epidermal omithine decarboxylase activity and skin tumors by 7,12-dimethylbenz[a]anthracene: modulation by retinoic acid and 7,8-benzoflavone.Carcinogenesis (Lond.), 7:607-611, 1980. 28. Verma, A. K., Shapas, B. G., Rice, H. M., and Boutwell, R. K. Correlation of the inhibitionby retinoidsof tumor promoter-inducedmouseepidemial omithine decarboxylase activity and of skin tumor tumor promotion. Cancer Res., 39: 419-425, 1979. 29. Voorhees, J. Polyaminesand psoriasis. Arch. Dermatol., f 75:943-944,1979.

5177