Protection against malignant conversion in SENCAR mouse skin by all Trans retinoic acid: Inhibition...

MOLECULAR CARC/NOGFN€S/S 17:13-22 (7996)

ARTICLES

Protection Against Malignant Conversion in SENCAR Mouse Skin by All Trans Retinoic Acid: Inhibition of the ras p21 -Processing Enzyme Farnesyltransferase and Ha-ras p21 Membrane Localization Rajesh Agarwall, Rajiv R. Mohan, Nihal Ahmad, and Hasan Mukhtar Department of Dermatology, Skin Diseases Research Center, University Hospitals of Cleveland, Case Westem Reserve University, Cleveland, Ohio

Many studies have shown that all trans retinoic acid (RA) exhibits significant protective effects against mouse skin tumor promotion and spontaneous as well as enhanced malignant conversion. In a recently completed study, we showed that under treatments in which papillomas on SENCAR mouse skin are induced a t low and high probabilities to convert to malignant carcinomas, RA affords significant protection against both tumor promotion and subsequent malignant conversion. More than 95% of these mouse skin papillomas and carcinomas have been shown to contain point mutation at the 61 codon of Ha-ras oncogene. The ras oncogene encodes a p21 protein that, in i ts mutated form, transforms mammalian cells only when p21 is at the inner surface of the plasma membrane, by a series of enzymatic reactions in which the initial step is catalyzed by farnesyltransferase (FTase). In this study, we assessed whether the protective effect of RA against malignant conversion involves the inhibition of ras p21 processing in those tumors that contain the activated ras oncogene. The FTase activity and the levels of cytosolic and membrane-bound Ha-ras p21 were determined in all papillomas and carcinomas obtained from acetone- or RA-treated animals. No matter how the data were analyzed and what comparisons were considered, in all the protocols used, compared with controls, papillomas and carcinomas obtained from RA-treated groups showed significantly decreased (P < 0.01-0.001) FTase activity. Furthermore, the tissue samples from RA-treated groups in different protocols also showed significantly diminished membrane localization of Ha-ras p21, with a concomitant increase in cytosolic Ha-ras p21 levels. The analysis of these data also showed that in all the protocols used, the increased Rase activity and membrane localization of Ha-ras p21 were associated with the induction of papillomas and their subsequent malignant conversion to squamous cell carcinomas. Taken together, these results indicate a strong correlation between the inhibition of ras p21 farnesylation because of a decrease in FTase activity by RA and i t s protective effect against malignant conversion of papillomas to carcinomas. Based on the results of this study, it is tempting to suggest that clinical trials evaluating the preventive or therapeutic potential of retinoids may be directed more toward those clinical malignancies that are known to contain the activated ras oncogene.

Key words: ras oncogene, ras p21, farnesyltransferase, retinoic acid, tumor progression

Q tMwiley-Liss, Inc.

INTRODUCTION Retinoids, specifically all trans retinoic acid (RA)

and 134s retinoic acid, have been used extensively as chemopreventive and therapeutic agents against a variety of clinical malignancies [ 1-61. Ongoing chemoprevention studies at the National Cancer In- stitute National Institutes of Health, using groups at high risk for cancers of the prostate, lung, breast, cer- vix, colon and rectum, oral cavity, and bladder include various retinoids [A. The test protocols with retinoids alone or in combination with other agents include all-trans-N-(4-hydroxyphenyl) retinamide alone and with tamoxifen or 2=difluoromethylomithine; a combination of 13-cis-retinoic acid, &carotene, and ret- inol; a combination of B-carotene and retinyl palmitate; and finasteride [7J. From these studies it is clear that retinoids, either alone or in combina-

tion, constitute a major part of all these clinical trials of cancer chemoprevention and that they are important cancer chemopreventive and chemotherapeutic agents. Retinoids are used worldwide for the prevention of skin cancer in patients with multiple cancers resulting from inherited disorders such as xeroderma pigmentosum and nevoid basal cell carcinoma syndrome [8-113. Several studies (reviewed by Chen and De Luca [ll]) have demonstrated the efficacy of various retinoids in inducing complete or

'Corresponding author: Department of Dermatology, Case Western Reserve University, 11 100 Euclid Avenue, Cleveland, OH

Abbreviations: DMBA, 7.12-dimethylbenz[a]anthracene; DTT, dithiothreitol; FTase, farnesyltransferase; FPP, farnesyl pyrophosphate; IgG. immunoglobulin G; PMSF. phenylmethylsulfonyl fluoride; RA, all trans retinoic acid; TEST, Tris-buffered saline, pH 7.5, with 0.1 % Tween-20; TPA, 12-O-tetradecanoylphorbol-13-acetate.

44106-5028.

0 1996 WLEY-USS, INC.

14 AGAR WAL ET AL.

partial remission of human skin cancers and their significance as chemopreventive agents against experimentally induced papillomas and carcinomas in mouse skin. In this context, the studies from our laboratory [12] and those conducted by De Luca and colleagues [13-151 showing that topical application or dietary administration of RA was highly effective against the malignant conversion of nonmalignant lesions in mouse skin, were extremely important because the progression of benign lesions into malignant lesions is the most critical event in cancer development, as the later lesions are capable of meta- static spread ultimately leading to the death of the host. In a recently completed study from our laboratory [16], with results similar to previous findings [12-151, we also demonstrated that RA affords highly significant protection against tumor promotion as well as the conversion of nonmalignant lesions to malignant lesions in SENCAR mouse skin in different protocols of cutaneous chemical carcinogenesis in which papillomas are induced at low and high probabilities to convert to carcinoma.

In cutaneous chemical carcinogenesis protocols involving tumor initiation with 7,lZdimethylbenz- [alanthracene (DMBA) and tumor promotion with 12-0-tetradecanoylphorbol-13-acetate (TPA), mezerein, or benzoyl peroxide in mouse skin, several studies have shown that > 95% of papillomas and carcinomas thus formed have elevated expression of and point mutation at codon 61 of the Ha- rm oncogene [17-221. The r m oncogenes encode a 21-kDa protein known as ras p21, which, in its mutated form (mutated or oncogenic ras p21), transforms mammalian cells (leading to abnormal growth and proliferation) only when localized at the inner side of the plasma membrane [23-251. This membrane anchoring of ras p21 is specifically initiated by a cytosolic enzyme, famesyltransferase (FTase), followed by other enzymatic reactions [23-251. Recent studies from our laboratory have shown that compared with untreated normal epidermis, the levels of FTase activity and membrane-associated Ha-ras p21 are significantly higher in DMBA-TPA- and ultraviolet B radiation-induced papillomas in SENCAR and SKH-1 hairless mice, respectively [26,27]. In other studies we also demonstrated that inhibition of Ha-ras p21 membrane localization arrests the growth of DMBA-TPA-induced papillomas and leads to their regression in SENCAR mice [28]. Taken together, these studies strongly suggested that FTase and the membrane localization of Ha-ras p21 play a key role in at least papilloma formation and in their subsequent growth in these mouse skin carcinogenesis protocols [29].

In the study reported here, we therefore assessed whether the protective effect of RA against tumor progression is associated with the inhibition of ras p21 processing. To accomplish this goal, the papillomas and carcinomas obtained from acetone-treated

positive controls and RA-treated experimental groups in low-risk TPA and high-risk TPA and mezerein protocols [16] were analyzed for FTase activity and the levels of both cytosolic and membrane-bound Ha- ras p21. The normal skin from the age- and sex- matched untreated control SENCAR mice as well as uninvolved skin from tumor-bearing mice was also used in these studies.

MATERIALS AND METHODS Purified recombinant Ha-ras p21g'y-'z was a kind

gift from Dr. Veeraswamy Manne (Squibb Institute for Medical Research, Princeton, NJ). [3H]farnesyl pyrophosphate (FPP) (15 Ci/mmol) was obtained from American Radiolabeled Chemicals Inc. (St. Louis, MO). Phosphocellulose paper (P-81) was purchased from Whatman Biosystems Ltd. (England). Nonimmune immunoglobulin G (IgG) was obtained from Cappel Research Products (Durham, NC). For immunoprecipitation, anti-Ha-ras monoclonal IgG antibody (clone Y 13-238), protein A-agarose beads, and protein G-agarose beads were purchased from Oncogene Science (Uniondale, NY). Ten percent pre- cast sodium dodecyl sulfate-polyacrylamide slab gels were obtained from Schleicher & Schuell (Keene, NH). The Ha-ras monoclonal antibody used for de- tecting ras p21 on the membrane was from Trans- duction Laboratories (Lexington, KY). Conjugated antimouse 1gG:horseradish peroxidase and the ECL kit were from Amersham (Arlington Heights, IL). All other chemicals and reagents used were of the high- est purity commercially available.

Induction of Papillomas and Carcinomas in SENCAR Mice

The details of papilloma induction in SENCAR mouse skin by low-risk TPA, high-risk TPA, and high- risk mezerein protocols were described by Hennings et al. [30] and detailed elsewhere [16]. The papilloma- bearing mice in three different protocols were subjected to malignant conversion of papillomas to carcinomas beginning at week 20 and then treated twice weekly either with acetone as a vehicle in positive-control groups or with 10 pg of RA in acetone in experimental groups as detailed elsewhere [16]. These treatments were performed to week 49, when the experiment was terminated.

Tissue Collection and Preparation of Cytosolic and Soluble-Membrane Fractions

At the end of the experiment at week 49, we collected the uninvolved skin and histopathologically verified papillomas and carcinomas from the tumor- bearing mice under three protocols without and with RA treatment. These samples were also collected randomly at week 31 of treatment. Normal skin from the age- and sex-matched untreated control animals of the same lot was also collected. Each collected sample was vertically divided into two equal parts,

lNHlB/T/ON OF RAS P21 PROCESSING BY RETINOIC ACID 15

snap frozen in liquid nitrogen, and stored at -1 70°C until further use.

One half of each tissue sample was removed from the -170°C storage, thawed at 4"C, cut into small pieces, and homogenized at 4°C with a Polytron tissue homogenizer in 0.1 M HEPES buffer, pH 7.4, containing 25 mM MgC12, 10 mM dithiothreitol (DTT), 1 mM phenylmethylsulfonyl fluoride (PMSF), 5 pg/ mL leupeptin, and 5 pg/mL pepstatin. A 20% tissue homogenate was thus prepared and centrifuged at 1000 xg at 4°C for 10 min, and the supernatant obtained was centrifuged at 100 OOO xg at 4°C for 60 min. The 100 000 xg supernatant was stored as the cytosolic fraction, and the pellet thus obtained as the membrane fraction was dissolved in lysis buffer. In brief, to each membrane pellet we added 1 mL of 50 mM HEPES buffer, pH 7.4, containing 150 mM NaCl, 50 mM NaF, 25 mM P-glycerophosphate, 10 mM sodium pyrophosphate, 5 mM p-nitrophenyl phosphate, 0.2% (w/v) NP-40, 1 mM sodium vana- date, 1 mM EDTA, 1 mM benzamidine, 1 mM PMSF, 5 pg/mL leupeptin, and 5 pg/mL pepstatin. The recipe for this lysis buffer was recently reported by James and colleagues [31]. Each tube was sonicated for 20 s, vortexed, and recentrifuged at 100 OOO xg at 4°C for 60 min. The supernatant thus obtained was used in further studies as the soluble-membrane fraction. The protein concentration of both the cytosolic and soluble-membrane fractions was determined by the method of Bradford [32] by using bovine serum al- bumin as the reference standard.

Determination of Rase Activity The FTase activity was determined in all the cyto-

solic fractions obtained with different tissue samples by using a method recently developed in our laboratory [33]. In brief, a reaction mixture (10 pL) containing 10pg of the cytosolic protein, 9.5 pM purified recombinant Ha-ras p21g'y-'2, and 2 pM (0.16 pCi) rH]FPP in 0.1 M HEPES buffer, pH 7.4, containing 10 mM DTT and 25 mM MgClz was incubated at 3 7°C for 60 min. Thereafter, it was cooled on ice for 2 min and quickly centrifuged, and the entire reaction mixture was spotted onto phosphocellulose paper (2.5 cm'). After air drying for 5 min, each paper was sepa- rately soaked in l% aqueous trichloroacetic acid solution for 10 min, and thereafter, all the papers were washed under running distilled water for 20 min. The papers were then air dried, and the radioactivity associated with [3afarnesylated Ha-ras p21g1y~1z was determined by liquid scintillation counting. The FTase activity was calculated after subtracting the counts obtained from the enzyme blank and is rep- resented as pmol of [3H]farnesyl transferred to purified Ha-ras p21g'y-'2/h/mg protein.

Determination of Ha-ras p21 Levels The levels of both cytosolic and membrane-bound

Ha-ras p21 were detected by the previously published

procedure from our laboratory [26-281 with some modifications as described by James et al. [31]. In brief, 100 pg of total protein from either the cytosolic or soluble-membrane fractions prepared from different tissue samples was diluted with 0.1 M HEPES buffer, pH 7.4, containing 25 mM MgCl', 10 mM DIT, 1 mM PMSF, 5 pg/mL leupeptin, and 5 pg/mL pepstatin to a total volume of 500 pL and was added to 10 pg of nonimmune IgG plus 25 pL protein A agarose bead solution. This solution was incubated at room temperature for 30 min and centrifuged at 12 OOO xg for 10 min, and the supernatant obtained was added to 1 pg of monoclonal IgG antibody Y13- 238 (Oncogene Science) + 25 pL protein G-agarose solution and incubated at room temperature for 60 min. The solution was centrifuged at 12 000 xg for 10 min. The protein G-agarose beads were collected and washed three times with the same buffer, and then the Ha-ras p21 associated with the beads was dissolved in 20pL of l x sodium dodecyl sulfate-polyacrylamide gel electrophoresis sample buffer [34], boiled for 5 min at lOO"C, and centrifuged. The clear supernatant was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis on a 1Wo gel, and the resolved proteins were transferred onto a nitrocellulose membrane as described by Towbin et al. [35]. The nonspecific sites on the membrane were blocked by incubation at room temperature for 1 h with Tris-buffered saline, pH 7.5, with 0.1% Tween- 20 (TBST) containing 5% nonfat dry milk (blocking buffer). The membrane was then incubated at room temperature for 1 h with fresh blocking buffer containing 1 pg/mL of Ha-ras monoclonal antibody IgG (Transduction Laboratories). Thereafter, the membrane was washed in TBST for 30 min at room temperature with agitation and a change of buffer every 5 min and incubated at room temperature for 1 h with blocking buffer containing antimouse 1gG:horseradish peroxidase (1:2000 dilution). Themembrane was then washed again in TBST for 30 min at room temperature with agitation and a change of buffer every 5 min, and the blots were developed by using the ECL kit according to the vendor's protocol.

RESULTS Protective Effect of RA Against Tumor Promotion and Malignant Conversion

In a recent study [16] we showed that topical application of RA to DMBA-initiated SENCAR mouse skin treated with the tumor promoter TPA or mezerein results in highly significant protection against skin tumor promotion in (i) low-risk TPA, (ii) high-risk TPA, and (iii) high-risk mezerein protocols. The protective effects of RA were 35-8W0, 67- 83%, and 70-98'240 in terms of tumor incidence, tumor multiplicity, and tumor volume/mouse, respectively, in these three treatment groups. In addition, we also showed that topical application of RA onto the papilloma-bearing mice in these three protocols results

16 AGARWAL ETAL.

in 3545% protection in terms of percentage of mice with carcinomas and 50-63Yo protection in terms of both carcinomas/mouse and rate of malignant conversion [16]. In the study reported here, we concen- trated our efforts on assessing whether the protective effect of RA against malignant conversion in mouse skin is associated with the inhibition of ras p21 processing. The target for such studies, inhibition of FTase and thereby ras p21 processing, followed by membrane localization, was selected based on the fact that (i) most of these tumors have Ha-ras oncogene activated by point mutation at codon 61 [17- 221, (ii) ras p21 exerts its cell-transforming potential only after localization to the inner side of the plasma membrane [23-251, and (iii) in earlier studies, FTase and Ha-ras p21 were shown to be associated with induction [26] as well as inhibition and regression [28] of such papillomas in SENCAR mice. It is important to mention here that all papillomas and carcinomas obtained from acetone-treated positive control and RA-treated experimental groups in the three different protocols analyzed were found to contain point mutations at codon 61 of the Ha-ras oncogene (A1’’ + T transversion) [16]. This finding further strengthens the argument that Ha-ras p21 processing may be involved in the induction of such tumors and that the protective effect of RA may be associated with the inhibition of Ha-ras p21 processing.

Effect of RA Treatment on FTase Activity During Different Protocols of Malignant Conversion of Papillomas to Carcinomas

We first determined FTase activity in the non-tu- morous skin of tumor-bearing mice (hereafter re- ferred to as uninvolved skin) and in papillomas and carcinomas obtained from both acetone-treated controls and RA-treated experimental groups by low- and high-risk protocols. These tissue samples included those obtained at the end of the experiment at week 49 and those obtained randomly from different animals at week 31 of the experiment. The FTase activity data obtained from all these samples were compared with those from normal skin samples from age- and sex-matched, untreated control SENCAR mice from the same lot. As shown in Figures 1 4 , no matter how the data were analyzed and what comparisons were made, the uninvolved skin, papillomas, and carcinomas obtained from RA-treated groups showed statistically significant inhibition of FTase activity compared with their acetone-treated control counterparts.

As shown by the data in Figure 1, in the low-risk TPA protocol at week 31 of the experiment, specific Rase activity was found to be significantly lower in uninvolved skin, papillomas, and carcinomas by 37%, 22%, and 47% (P< 0.01-0.001, Student’s t test), respectively, in the RA-treated group compared with acetone-treated controls. Further decreases in FTase activity in these tissue samples from the RA-treated

100

.- i? 31st week

Figure 1. FTase activity in normal skin, uninvolved skin, papillomas, and carcinomas in SENCAR mouse skin at week 31 (top panel) and week 49 (bottom panel) in the low-risk TPA protocol. The details of the experimental protocols, tissue sample collection and cytosolic fraction preparation, and FTase assay are described in Materials and Methods. The data shown for week 31 are the means * SE of four tissue samples in each case. The data shown for week 49 are the means * SE of eight normal skin samples and 15 samples each of uninvolved skin, papillomas, and carcinomas. Each enzyme assay for each sample was performed in duplicate. Uninvol. skin, uninvolved skin from tumor-bearing mice.

group were observed at the end of the experiment at week 49 (46Y0, 46940, and 61% decreases (P < 0.01- 0.001, Student’s t test), respectively) (Figure 1). As with the low-risk TPA protocol, when the data for FTase activity were analyzed for both the high-risk TPA and the high-risk mezerein protocols, as shown in Figures 2 and 3, a significant decrease in Rase activity in tumor tissues derived from RA-treated animals was observed. In the case of the high-risk TPA protocol, compared with acetone-treated controls, the decreases in enzyme activity were 38%, 39%, and 54% (P c 0.01-0.001, Student’s t test) in uninvolved skin, papillomas, and carcinomas, respectively, obtained from the RA-treated group at week 31 of the experiment (Figure 2). Similarly, at the end of the experiment (at week 49), the RA-treated tissue

INHlBlTlON OF RAS P21 PROCESSING BY RETlNOlCAQD 17

,

.- > I 31stweek I 31 st week

Figure 2. !Taw activity in normal skin, uninvolved skin, papillomas, and carcinomas in SENCAR mouse skin at week 31 (top panel) and week 49 (bottom panel) in the high-risk TPA protocol. The data shown for week 31 are the means * SE of four tissue samples in each case. The data shown for week 49 are the means 2 SE of eight normal skin samples and 10 samples each of uninvolved skin, papillomas, and carcinomas. See the Figure 1 legend for further details.

samples showed 29%, 45%, and 36% decreases (P < 0.01-0.001, Student's t test) in FTase activity for these three respective tissue groups (Figure 2). In the case of the high-risk mezerein protocol, compared with acetone-treated controls, the uninvolved skin, papillomas, and carcinomas obtained from RA-treated groups showed 24Oh, 44%, and 43% decreases (P < 0.01-0.001, Student's t test), respectively, in Rase activity at week 31 of the experiment, and 28% 37%, and 49% decreases (P < 0.01-0.001, Student's t test), respectively, at the end of the experiment (Figure 3). The data shown in Figures 1-3 also clearly demon- strate that in both the low- and high-risk protocols, the increase in Rase activity in acetone-treated controls was in the order carcinomas > papillomas > uninvolved skin > normal skin for all the tissue samples obtained at both weeks 31 and 49 of the experiment; the only exception was the high-risk TPA group, in which uninvolved skin showed enzyme

Figure 3. Ffase activity in normal skin, uninvolved skin, papillomas, and carcinomas in SENCAR mouse skin at week 31 (top panel) and week 49 (bottom panel) in the high-risk mezerein protocol. The data shown for week 31 are the means * SE of four tissue samples in each case. The data shown for week 49 are the means i SE of eight normal skin samples and 10 samples each of uninvolved skin, papillomas, and carcinomas. See the Figure 1 legend for further details.

activity higher than that of both carcinomas and papillomas at week 31 of the study. Another inter- esting observation from the FTase activity data shown in Figures 1-3 for the three different protocols was that the enzyme activity in papillomas and carcinomas at week 31 of the experiment was higher than that at week 49.

By analyzing the specific enzyme activity data for FTase shown in Figures 1-3 in terms of percent inhibition of enzyme activity by RA in uninvolved skin, papillomas, and carcinomas at weeks 31 and 49 in the three different protocols, at week 31 of the experiment, we observed maximum inhibition in FTase activity in the carcinomas obtained from RA-treated mice in the high-risk TPA protocol followed by low- risk TPA protocol (Figure 4). In the case of the high- risk mezerein protocol, the percent inhibition of enzyme activity was almost identical in RA-treated papillomas and carcinomas (Figure 4). When these

AGAR WAL ET AL.

31st week

Figure 4. Inhibitory effect of topical application of RA on Rase activity in uninvolved skin, papillomas, and carcinomas in SENCAR mouse skin at week 31 (top panel) and week 49 (bottom panel) in three different protocols. The percent inhibition data shown are those obtained from Figures 1-3 for low- risk TPA, high-risk TPA, and high-risk mezerein protocols, respectively.

data were analyzed at the end of the experiment at week 49, the maximum inhibition of FTase activity was evident in carcinomas obtained from RA-treated mice in the low-risk TPA protocol, followed by the high-risk mezerein protocol (Figure 4). In the high- risk TPA protocol, the papillomas obtained from RA- treated mice showed a much higher inhibition of FTase activity than did the carcinomas (Figure 4). Taken together, these results show that RA may be more effective in inhibiting FTase activity in the high- risk TPA protocol during the early phase of malignant conversion of papillomas to carcinomas. A more defined kinetic study with shorter time intervals for tissue collection in this protocol is warranted to test this assumption.

Effect of RA Treatment on Ha-ras p21 Levels in Cytosol and Membrane Fractions During Different Protocols of Malignant Conversion of Papillomas to Carcinomas

To assess the effect of RA on Ha-ras p21 membrane localization during different protocols of malignant conversion, we first verified the levels of both cyto-

solic and membrane-bound Ha-ras p21 in normal skin, uninvolved skin, papillomas, and carcinomas obtained from acetone-treated positive controls in different protocols. As shown in Figures 5-7, in western blot analysis, a single band was observed at ap- proximately 22 kDa in both cytosolic and membrane fractions prepared from these tissues in different protocols. However, compared with the levels in normal epidermis (Figures 5-7), elevated levels of Ha-ras p21 were observed in both cytosol and membrane fractions prepared from the uninvolved skin of tumor-bearing mice. These levels were further increased in papillomas, followed by carcinomas. As with the Rase activity data, the levels of both cytosolic and membrane-bound Ha-ras p21 in all the protocols were in the order carcinomas > papillomas > uninvolved skin > normal skin. With regard to the modulatory effect of RA on the levels of membrane- bound versus cytosolic Ha-ras p21 in different tissue samples, as shown in Figures 5-7, treatment of mice with RA during three different protocols of malignant conversion produced significantly diminished membrane localization of Ha-ras p21 and an increase in cytosolic Ha-ras p21 compared with the corresponding acetone-treated controls. The modulatory effect of RA on Ha-ras p21 levels was clearly evident in all cases (i.e., uninvolved skin from tumor-bearing mice, papillomas, and carcinomas) and was also evident at both the time periods used in these studies (i.e., at week 31 of treatment as well as at the end of the experiment at week 49) (Figures 5-7). From these results and the inhibition of FTase activity by RA in different tissue samples during low-risk TPA, high-risk TPA, and high-risk mezerein protocols, it is logical to conclude that topical RA treatment of papilloma-bearing mice results in the inhibition of FTase activity in the tumors, which ultimately results in decreased membrane localization of Ha-ras p21 with a concomitant increase in cytosolic Ha-ras p21. The inhibition of Ha-ras p21 farnesylation by RA observed in the different tissues correlated strongly with its protective effect against malignant conversion of papillomas to carcinomas in all the protocols used in these studies.

DISCUSSION

With regard to genetic changes associated with cancer induction, the frequent detection of Ha-, Ki-, and N-rus oncogenes activated by point mutations in clinical human malignancies in different tissues, as well as in experimentally induced neoplasms in animals, suggests that rus oncogenes make an important contribution to cancer induction [17-22,36- 431. The three rus oncogenes encode a group of structurally related 21-kDa 188-189 amino acid proteins designated p21. These proteins are believed to function as molecular switches in signaling events of cell growth and differentiation [23-251. The mutated (oncogenic) ras p21, encoded by the mutated

lNHlBlTlON OF RAS P21 PROCESSING BY RETlNOlCAClD 79

Figure 5. Modulatory effect of topical application of RA on cytosolic and membrane-bound Ha-ras p21 levels in uninvolved skin, papillomas, and carcinomas in SENCAR mouse skin at weeks 31 and 49, in the low-risk TPA protocol. The details of the experimental protocols, tissue sample collection, cytosolic and soluble-membrane fraction preparation, and detection of Ha-ras p21 levels are described in Materials and Methods. Only representative western blot data are shown for each group of tissue samples; except for normal

skin, each set of two lanes has two independent samples. NS, normal skin from age- and sex-matched, untreated control mice from the same lot as used in the tumor experiments; US, uninvolved skin from tumor-bearing mice; P, papilloma; C, carcinoma; -RA, samples prepared from acetone-treated control groups; +RA, samples prepared from RA-treated experimental groups. The position of Ha-ras p21 shown was determined by comparison t o wild-type Ha-ras p21g’y-’2 protein and the standard molecular weight marker proteins.

Figure 6. Modulatory effect of topical application of RA on cytosolic and membrane-bound Ha-ras p21 levels in uninvolved skin, papillomas, and carcinomas in SENCAR mouse skin at weeks 31 and 49 in the high-risk TPA protocol. See the Figure 5 legend for further details.

20 AGARWAL ETAL.

Figure 7. Modulatory effect of topical application of RA on cytosolic and membrane-bound Ha-ras p21 levels in uninvolved skin, papillomas, and carcinomas in SENCAR mouse skin at weeks 31 and 49 in the high-risk mezerein protocol. See Figure 5 legend for further details.

rus oncogene, transforms mammalian cells only when localized at the inner side of the plasma membrane [23-251. The membrane localization of p21 requires at least three post-translational modifications, which mainly occur at the carboxyl terminal of p21, which has the consensus motif Cys-Aaa-Aaa- Xaa (where Aaa is aliphatic amino acid and Xaa is any amino acid) and is termed the CAAX box [23- 25,44,45]. The major post-translational modification of p21 required for its membrane localization is the transfer of a farnesyl group to the cysteine residue of the CAAX box (farnesylation of p21), a reaction catalyzed by a specific cytosolic enzyme known as FTase [23-25,44,45]. The importance of post-translational processing in promoting membrane association of p21 followed by cell-transforming activity has been verified by experimental mutagenesis of the p21 CAAX box [25].

One promising pharmacological approach for inhibiting oncogenic ras p21 activity in malignancies is interfering with ras p21 membrane localization by inhibiting the farnesylation reaction, which could inhibit the function of oncogenic ras p21 without completely inhibiting overall cellular function [23]. Studies using cell-culture systems have shown that anticarcinogens like dehydroepiandrosterone and d- limonene and its metabolites inhibit ras protein p21 farnesylation as well as its membrane association [46,47]. The authors of these studies interpreted their results as a plausible explanation for the cancer chemopreventive and chemotherapeutic effects of

these agents and suggested that the inhibition of farnesylation of p21 by dehydroepiandrosterone and d-limonene may contribute to their anticancer effects [46,47]. It is important to mention here that dehydroepiandrosterone has been found to prevent experimental carcinogenesis in colon, lung, liver, pancreas, and breast, and d-limonene possesses sub- stantial chemopreventive and chemotherapeutic activity against chemically induced mammary, lung, and stomach cancer in rodents [46,47]. Most of these experimentally induced tumors have been shown to contain rus oncogenes activated by point mutations [36-431. In other studies, Hara et al. [48], using purified FTase, have shown that UCF1-A, -B, and -C, active compounds produced by Streptornyces, inhibit FTase activity by competing with one of the substrates, FPP. Gibbs et al. [49] showed that (a- hydroxyfarnesy1)-phosphonic acid, chaetomellic acid, and zaragozic acid inhibit FTase by competing with FPP; (a-hydroxyfarnesyl) phosphonic acid also inhibits ras processing in Ha-rus-transformed NIH 3T3 cells [49]. Studies from the laboratories of Gibbs [SO] and Brown [Sl] have also identified the selective inhibitors of p21 farnesylation by using purified Rase and cell-culture systems. One of these FTase inhibitors was also shown to block the growth of ras-dependent tumors in nude mice [52]. Studies from the laboratory of Sebti [53,54] reported potent peptidomimetic inhibitors of ras p21 farnesylation. More recently, Manne et al. [SS] reported that the bisubstrate inhibitors of FTase are a novel class of

lNHlBlTlON OF RAS P21 PROCESSING BY RETlNOlC ACID 21

specific inhibitors of ras-transformed cells. It is important to note at this point that whereas there are intense efforts to synthesize inhibitors of oncogenic ras p21 farnesylation for use as cancer therapeutic agents, very little effort has been made to assess the inhibitory effects of the known cancer chemopreventive agents against ras p21 farnesylation in experimental carcinogenesis as a plausible mechanism of their cancer-chemoprevention potential. We are aware of only one such study in which difluoro- methylornithine and piroxicam were shown to inhibit the membrane association of ras p21 during chemoprevention of azoxymethane-induced colon carcinogenesis in rats [56]; these tumors contain a r m oncogene activated by point mutation [56].

With regard to the retinoids, many studies have assessed and demonstrated their chemopreventive and chemotherapeutic potential against clinical human malignancies, including non-melanoma human skin cancer [&11,57-601. Most of the experimental studies, however, have been performed with the mouse skin multistage carcinogenesis model to show the protective effects of retinoids against tumor promotion as well as progression [ll]. Because many human cancers at different body sites contain a ras oncogene activated by point mutation [36-40] and because of the study reported here showed the inhibition of ras p21 processing by RA in malignant conversion of non-malignant lesions, it is tempting to suggest that the clinical trials evaluating the preventive or therapeutic potential of retinoids should be directed more toward those clinical cancers that are known to contain activated rm oncogenes. This sug- gestion, however, does not mean that in malignancies with activated ras oncogenes, retinoids exert their effect only by inhibiting ras p21 processing, or that retinoids are not effective in the prevention or treatment of malignancies without ras oncogene activation. There are several other mechanisms associated with the action of retinoids against malignancies aside from the results of the present study [ll].

In summary, the results of the study presented here clearly demonstrated that RA inhibited the FTase activity and the membrane localization of Ha-ras p21 in carcinomas induced in the skin of SENCAR mice by different protocols and that this RA effect was strongly related to its protective effect against malignant conversion of papillomas to squamous carcinomas. More detailed studies, however, are needed to further de- fine whether the inhibitory effect of RA on Rase activity is due to inhibition of FTase mRNA expression and whether the selectivity of RA and other retinoids in inhibiting the growth of various human malignant cells containing different rm oncogene mutations is due to the inhibition of ras p21 processing.

ACKNOWLEDGMENTS

This work was supported by American Institute for Cancer Research grant 93A30 and U.S. Public Health

Service grants CA 64514 and P-30-AR-39750. RA was also partially supported by an Upjohn Company Foundation Career Development Award through the Dermatology Foundation.

Received December 7, 1995; revised May 2, 1996; accepted May 15, 1996.

REFERENCES 1. Lippman SM, Meyskens Jr, FL. Results of the use of vitamin A

and retinoid in cutaneous malignancies. Pharmacol Ther

2. Sporn MB, Newton DL. Chemoprevention of cancer with retinoid. Fed Proc 38:2528-2534, 1979.

3. Lotan R. Effects of vitamin A and its analogs (retinoid) on normal and neoplastic cells. Biochim Biophys Acta 605:33-91, 1980.

4. De Luca LM, Darwiche N, Celli G, et al. Vitamin A in epithelial differentiation and skin carcinogenesis. Nutr Rev 52545-552. 1994.

5. Smith MA, Parkinson DR, Cheson BD, Friedman MA. Retinoids in cancer therapy. J Clin Oncol 10:839-864, 1992.

6. Sporn MB, Roberts AB. Cervical dysplasia regression induced by all-trans-retinoic acid. J Natl Cancer lnst 86:476-477, 1994.

7. Wattenberg LW, Terada M. Molecular biology of cancer: Impli- cations for prevention and therapy. Third Joint Meeting of the American Association for Cancer Research and the Japanese Cancer Association. Cancer Res 55:5472-5481, 1995.

8. Kraemer KH, DiGiovanna JJ, Moshell AN, Tarone RE, Peck GL. Prevention of skin cancer in xeroderma pigmentosum with the use of oral isotretinoin. N Engl J Med 318:1633-1637, 1988.

9. DiGiovanna JJ, Zech LA, Ruddel ME, Gantt G, Peck GL. Etretinate. Arch Dermatol 125246251, 1989.

40:107-122, 1989.

10. Futoryan T, Gilchrest BA. Retinoids and the skin. Nutr Rev

11. Chen L-C. De Luca LM. Retinoid effects on skin cancer. In: 521299-310, 1994.

Mukhtar H (ed), Skin Cancer: Mechanism and Human Relevance. CRC Press, Boca Raton, FL, 1994, pp. 401-424.

12. Athar M, Agarwal R, Wang ZY, Lloyd JR, Bickers DR, Mukhtar H. All trans retinoic acid protects against free radical generating compounds-mediated conversion of chemicallv and ultraviolet B

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

radiation-induced skin papillomas to carcinomas. Carcinogen- esis 12:2325-2329, 1991. De Luca LM, Sly L, Jones CS, Chen L-C. Effects of dietary retinoic acid on skin papilloma and carcinoma formation in female SEN- CAR mice. Carcinogenesis 14539-542, 1993. Chen L-C, Sly L, De Luca LM. High dietary retinoic acid prevents malignant conversion of skin papillomas induced by a two-stage carcinogenesis protocol in female SENCAR mice. Carcinogenesis

Chen L-C, Tarone R, Huynh M, De Luca LM. High dietary retinoic acid inhibits tumor promotion and malignant conversion in a two-stage skin carcinogenesis protocol using 7,12-dimethylbenz[a]anthracene as the initiator and mezerein as the tumor promoter in female SENCAR mice. Cancer Lett 95:113-118, 1995. Mohan RR, Katiyar SK, Khan SG, Zaim MT, Mukhtar H, Agarwal R. Protective effect of all trans retinoic acid against tumor promotion and progression in low- and high-risk protocols of mouse skin chemical carcinogenesis. Int J Oncology 8:1079-1088, 1996. DiGiovanni J. Multistage carcinogenesis in mouse skin. Pharma- col Ther 54:63-128, 1992. Brown K, Buchmann A, Balmain A. Carcinogen-induced mutations in the mouse c-Ha-ras gene provide evidence of multiple pathways for tumor progression. Proc Natl Acad Sci USA

Bowden GT, Finch J, Domann F, Krieg P. Molecular mechanisms involved in skin tumor initiation, promotion, and progression. In: Mukhtar H (ed), Skin Cancer: Mechanism and Human Rel- evance, CRC Press, Boca Raton, FL, 1994, pp. 99-1 11. Bremner R, Kemp CJ, Balmain A. Induction of different genetic changes by different classes of chemical carcinogens during progression of mouse skin tumors. Mol Carcinog 11:90-97, 1994. Pelling JC, Ernst SM, Strawhecker JM, Johnson JA, Nairn RS, Slaga TJ. Elevated expression of Ha-ras is an early event in two- stage skin carcinogenesis in SENCAR mice. Carcinogenesis

Pelling JC, Fischer SM, Neades R, Strawhecker J, Schweickert L. Elevated expression and point mutation of the Ha-ras proto-on-

1512383-2386, 1994.

87:538-542, 1990.

7~1599-1602, 1986.

22 AGARWAL ETAL.

cogene in mouse skin tumors promoted by benzoyl peroxide and other promoting agents. Carcinogenesis 8.1 481-1484, 1987.

23. Gibbs JB, Oliff A, Kohl NE. Farnesyltransferase inhibitors: Ras research yields a potential cancer therapeutic. Cell 77:175-178, 1994.

24. Lowy DR, Zhang K, Declue JE, Willumsen BM. Regulation of p21 ras activity. Trends Gene 7:346-351, 1991.

25. Der CJ, Cox AD. lsoprenoid modification and plasma membrane association: Critical factors for ras oncogenicity. Cancer Cells

26. Agarwal R, Khan SG, Athar M, Zaidi SIA, Bickers DR, Mukhtar H. Ras Drotein 021 Drocessina enzvme farnesvltransferase in

3:331-340, 1991.

chemical carcindgen-induced thin; skin tumor; Mol Carcinog

27 Khan SG, Bickers DR, Mukhtar H, Aclarwal R Ras ~ 2 1 8 290-298, 1993

farnesylation in ultraviolet B radiation-induced tumors in the'skin of SKH-1 hairless mice. J Invest Dermatol 102:754-758, 1994.

28. Khan SG, Saxena R, Bickers DR, Mukhtar H, Agarwal R. Inhibi- tion of ras p21 membrane localization and modulation of protein kinase C isozyme expression during regression of chemical carcinogen-induced murine skin tumors by lovastatin. Mol Carcinog 12:205-212, 1995.

29. Agarwal R, Khan SG, Mukhtar H. Ras p21 post-translational processing: An overview and studies in skin cancer. In: Mukhtar H (ed), Skin Cancer: Mechanism and Human Relevance, CRC Press, Boca Raton, FL, 1994, pp. 293-304.

30. Hennings H, Shores R, Mitchell P, Spangler EF, Yuspa SH. Induc- tion of papillomas with a high probability of conversion to ma- lignancy. Carcinogenesis 6: 1607-1 61 0, 1985.

31. James GL, Brown MS, Cobb MH, Goldstein JL. Benzodiazepine peptodomimetic BZA-5B interrupts the MAP kinase activation pathway in H-ras-transformed Rat-1 cells, but no t in untransformed cells. J Biol Chem 269:27705-27714, 1994.

32. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248-254, 1976.

33. Khan SG, Mukhtar H, Agarwal R. A rapid and convenient filter- binding assay for ras p21 processing enzyme farnesyltransferase. J Biochem Biophys Methods 30:133-144, 1995.

34. Laemmli UK. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227:680-685, 1970.

35. Towbin H, Staehlin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets. Proc Natl Acad Sci USA 76:4350-4354, 1979.

36. Barbacid M. Ras genes. Annu Rev Biochem 56:779-827, 1987. 37. Bos JL. Ras oncogenes in human cancer: A review. Cancer Res

49:4682-4689, 1989. 38. Bishop JM. Molecular themes in oncogenesis. Cell 64:235-248,

1991. 39. Bos JL, Fearon ER, Hamilton SR, et al. Prevalence of ras gene mu-

tations in human colorectal cancers. Nature 327:293-297, 1987. 40. Almoguera C, Shibata D, Forrester K, Martin J, Arnheim N,

Perucho M. Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 53:549-554, 1988.

41. Balmain A, Brown K. Oncogene activation in chemical carcinogenesis. Adv Cancer Res 51 : 147-1 82, 1988.

42. Miyamoto S, Sukumar S, Guzman RC, Osborn RC, Nandi S. Trans- forming c-Ki-ras mutation is a preneoplastic event in mouse mammary carcinogenesis induced in vitro by N-methyl-N- nitrosourea. Mol Cell Biol 10:1593-1599. 1990.

43. Feig LA. Strategies for suppressing the function of oncogenic ras protein in tumors. J Natl Cancer lnst 85:1266-1268, 1993.

44. Schaber MD, O'Hara MB, Garsky VM, et al. Polyisoprenylation of ras in vitro by a farnesyl-protein transferase. J Biol Chem

45. Willumsen BM, Christensen A, Hubbert NL, Papageorge AG, Lowy DR. The p21 ras C-terminus is required for transformation and membrane association. Nature 31 0:583-586, 1984.

46. Schulz S, Nyce JW. Inhibition of protein isoprenylation and p21 ras membrane association by dehydroepiandrosterone in human colonic adenocarcinoma cells. Cancer Res 51 :6563-6567, 1991.

47. Crowell PL, Chang RR, Ren Z, Elson CE, Gould MN. Selective inhibition of isoprenylation of 21-26-kDa proteins by the anticarcinogen d-limonene and its metabolites. J Biol Chem

48. Hara M, Akasaka K, Akinaga S, e t al. Identification of ras farnesyltransferase inhibitors by microbial screening. Proc Natl Acad Sci USA 90:2281-2285, 1993.

49. Gibbs JB, Pompliano DL, Mosser SD, et al. Selective inhibition of farnesyl-protein transferase blocks ras processing in vivo. J Biol Chem 268:7617-7620, 1993.

50. Kohl NE, Mosser SD, deSolms SJ, et al. Selective inhibition of ras- dependent transformation by a farnesyltransferase inhibitor. Sci- ence 260: 1934-1 937, 1993.

51. James GL. Goldstein JL. Brown MS. et al. BenzodiazeDine

265:14701-14704, 1990.

266: 17679-1 7685, 1991.

peptidomimetics: Potent inhibitors of ras farnesylation in animal cells. Science 260: 1937-1 942, 1993.

52. Kohl NE, Wilson FR, Mosser SD, et al. Protein farnesyltransferase inhibitors block the growth of ras-dependent tumors in nude mice. Proc Natl Acad Sci USA 91:9141-9145, 1994.

53. Nigam M, Seong C-M, Qian Y, Hamilton AD, Sebti SM. Potent inhibition of human tumor p21raS farnesyltransferase by AlA2- lacking p21raS CAlAzX peptidomimetics. J Biol Chem

54. Qian Y. Blaskovich MA, Saleem M, et al. Design and structural requirements of potent peptidomimetic inhibitors of p21 ras farnesyltransferase. J Biol Chem 269:12410-12413, 1994.

55. Manne V, Yan N, Carboni JM, et al. Bisubstrate inhibitors of farnesyltransferase: A novel class of specific inhibitors of ras transformed cells. Oncogene 10: 1763-1 779, 1995.

56. Singh J, Reddy BS. Effect of chemopreventive agents on post- translational plasma-membrane association of ras-p21 during chemoprevention of azoxymethane-induced colon carcinogenesis. Int J Oncol 6:301-306. 1995.

57. Lancillotti F, Darwiche N, Celli G, De Luca LM. Retinoid status and the control of keratin expression and adhesion during the histogenesis of squamous metaplasia of tracheal epithelium. Cancer Res 52:6144-6152, 1992.

58. Lotan R, Dawson MI, Zou C-C, Jong L, Lotan D, Zou C-P. En- hanced efficacy of combination of retinoic acid- and retinoid X receptor-selective retinoids and a-interferon in inhibition of cervical carcinoma cell proliferation. Cancer Res 55:232-236, 1995.

59. Widschwendter M, Daxenbichler G, Dapunt 0, Marth C. Effects of retinoic acid and y-interferon on expression of retinoic acid receptor and cellular retinoic acid-binding protein in breast cancer cells. Cancer Res 55:2135-2139, 1995.

60. Lotan R. Retinoids and apoptosis: Implications for cancer chemoprevention and therapy. J Natl Cancer lnst 87: 1655-1 657, 1995.

268:20695-20698, 1993.

Protection against malignant conversion in SENCAR mouse skin by all Trans retinoic acid: Inhibition...

Documents

Transcript of Protection against malignant conversion in SENCAR mouse skin by all Trans retinoic acid: Inhibition...