Farnesyltransferase Inhibitors Block the Neurofibromatosis Type I (NF1) Malignant Phenotype Ning Yan, Carolyn Ricca, Jonathan Fletcher, Thomas Glover, Bernd R

[CANCER RESEARCH 55, 3569-3575, August 15, 1995] Farnesyltransferase Inhibitors Block the Neurofibromatosis Type I (NF1) Malignant Phenotype Ning Yan, Carolyn Ricca, Jonathan Fletcher, Thomas Glover, Bernd R. Seizinger, and Veeraswamy Manne1

Department of Oncology Drug Discovery, Bristol-Mvers Squibb Pharmaceutical Research Institute. Princeton, New Jersey 08543-4000 Â¡N.Y., C. R., B. R. S., V. M.\: Department of Pathology, Brighum and Women's Hospital, Boston, Massachusetts 02115 fj. F.]; Department of Human Genetics, The University of Michigan Medical School, Ann Arbor, Michigan, 48109 Â¡T.G.Â¡

ABSTRACT and peripheral nervous system including neurofibrosarcomas, gli- omas, and schwannomas (1). The gene responsible for this disorder Neurofibromatosis type I (NF1) is a hereditary tumor and develop was mapped to the chromosome 17qll.2 region by linkage anal mental disorder whose defective gene was cloned previously. The pro ysis (2, 3). Subsequently, the NF1 gene was isolated (4-6) and tein product of the NF1 gene, neurofibromin, contains a domain that found to encode a protein of 2818 amino acids (7) named neuro shows significant sequence homology to the known catalytic domains of mammalian Ras GTPase-activating proteins (GAP) and the yeast IRAI fibromin (8-10). Neurofibromin contains a domain of approxi and IRA2 proteins. This homologous region of neurofibromin has been mately 360 amino acids that shows a significant sequence homol shown to exhibit GAP activity toward Ras proteins. Malignant schw- ogy to the known catalytic domains of mammalian GAP and the annoma cell lines from NF1 patients contain normal levels of GAP and yeast IRAI and IRA2 proteins (11-14). This homologous region, nonmutated Ras proteins but barely detectable levels of neurofibro called GAP-related domain, is responsible for its GAP activity, min, based on genetic mutations in the NF1 gene. Because these cells which involves the stimulation of the weak intrinsic GTPase ac contain constitutively activated Ras.GTP, it has been proposed that tivity of Ras proteins. Ras proteins, existing in inactive Ras.GDP neurofibromin may be the sole negative regulator of Ras in these cells. form in unstimulated cells, are activated to the Ras.GTP state by a Overall, these results have implied an important role of the Ras variety of stimuli (e.g., growth factors), and the resulting growth signaling pathway in NF1 malignant schwannomas. Recently, several signals unleash a cascade of phosphorylation events (Raf kinases, laboratories have developed small molecule inhibitors of Ras function that inhibit the enzyme farnesyltransferase (FT). FT-mediated posi MAP kinases, and others). The signaling is attenuated when Ras ti unshilional farnesylation of Ras proteins is absolutely necessary for .GTP returns to the Ras.GDP state, and GAP proteins including Ras function since this modification is required for the anchoring of neurofibromin play a major role in this process (15, 16). Ras proteins to the plasma cell membrane. Although previous studies Ras proteins are among the most frequently mutated oncoproteins have shown that FT inhibitors can block the growth of tumor cells in human cancers (15, 17), and such mutations contribute significantly carrying mutant Ras proteins, it remained unclear how this class of to the malignant properties of cancer cells. Oncogenic Ras proteins are inhibitors would affect tumor cells such as in NF1, whose malignant in the constitutively activated Ras.GTP state. This is due to the low growth appears to be mediated by up-regulation of wild-type Ras intrinsic rate of GTP hydrolysis coupled with the inability of GAPs to activity. Thus, in the current study, we investigated whether BMS- act on mutated versions of Ras and down-modulate its activity by 186511, a bisubstrate analogue inhibitor of FT, would inhibit the accelerating the hydrolysis of Ras bound GTP to GDP (16, 18). malignant growth properties of a cell line established from malignant schwannoma of an NF1 patient. Our results indicate that the malignant The constitutively activated Ras.GTP status as determined by high growth properties of STS8-14 cells, the most malignant cell line among Ras.GTP levels occurs in at least some tumor cell lines established several well-characterized NF1 cells, are inhibited by BMS-186511 in a from malignant schwannomas of NF1 patients (19, 20). These cells concentration-dependent manner. Following treatment with express normal levels of GAP and nonmutated Ras proteins but BMS-186511, ST88-14 cells became Hat, nonrefractile, were contact- reduced neurofibromin, which has been proposed to be a tumor inhibited, and lost their ability to grow in soft agar. In the drug- suppressor with GAP-like negative regulatory activity on Ras exposed cells, Ras proteins were prevented from FT-mediated mem (19, 20). The levels of neurofibromin negatively correlate with the brane association. BMS-186511 was found to specifically inhibit FT, levels of Ras.GTP in these cells. Hence, ras protein is constitutively but not geranylgeranyltransferase I, a closely related enzyme. Thus, it activated in NF1 cells, suggesting a positive growth role for ras is conceivable that FT inhibitors may ultimately become the first activity in NF1 malignancies. One cell line, ST88-14, had barely generation of drugs against the malignant phenotype in NF1 based on rational insights into the mechanism of action of neurofibromin. detectable NF1 protein and had 50% of the nonmutated Ras locked in the GTP-bound state. Transfection of the catalytic region of GAP or microinjection of Y13-259 anti-ras mAb into these cells inhibited INTRODUCTION their growth (19, 20). These results raised the possibility that antag NF12 (Von Recklinghausen neurofibromatosis) is one of the onists of Ras function might have therapeutic potential in NF1 most common dominantly inherited human genetic disorders, af malignancies (21). Recent progress in unraveling the crucial importance of farne- fecting 1 in 3500 people. NF1 is characterized by cafe-au-lait spots sylation-mediated membrane association for Ras function paved (flat pigmented skin lesions), Lisch nodules (abnormality of the the way for identification of several classes of FT inhibitors, iris), skeletal abnormalities, learning disabilities, neurofibromas, including CAAX box-based tetrapeptide analogues (22-32), bisub and increased risk of developing malignant tumors of the central strate analogues (33), FPP analogues (34, 35), and natural products (36) as Ras antagonists. Several of these Ras antagonists were Received 4/13/95; accepted 6/14/95. shown to block the growth of oncogenic Ras-transformed rodent The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with cells and human tumor cells as determined by several criteria, 18 U.S.C. Section 1734 solely to indicate this fact. 1To whom requests for reprints should be addressed, at Department of Oncology, including inhibition of anchorage-dependent and -independent Bristol-Myers Squibb Pharmaceutical Research Institute, P.O. Box 4000, Princeton, NJ growth, reversal of transformed morphology, and restoration of 08543. actin cytoskeleton (23, 24, 37-39). However, it is not known how 2 The abbreviations used are: NF1, neurofibromatosis type I; GAP, GTPase activating protein; FT, farnesyltransferase; GOT, geranylgeranyltransferase; FPP, farnesyl PPÂ¡; FT inhibitors would affect tumor cells such as in NF1, whose GGPP, geranylgeranyl PPÂ¡. malignant growth appears to be mediated by up-regulation of 3569

wild-type Ras activity. In the present study, we report the effects of time, the monolayers were rinsed twice with ice-cold PBS, and cells were of BMS-186511, a bisubstrate prodrug FT inhibitor, on the collected by scraping. Total lysates were prepared by sonication of the cell ST88-14 cell line derived from a malignant schwannoma of an NF1 pellets in buffer A [20 mM HEPES (pH 7.4), 1 mM EDTA, 1 mM MgCl2, 1 mM patient, the most malignant cell line among several well-charac DTT, 1 mM PMSF, 2 JJ.Mpepstatin, and 200 units/ml aprotinin]. Samples were terized NF1 schwannomas tested (19, 20). Our results demonstrate clarified by centrifugation at 1000 X g and fractionated to obtain soluble that the anchorage-dependent and -independent growth of these (S100) and particulate (P100) fractions (41). Protein was determined by Bio-Rad reagent, and equal amounts of total proteins were analyzed by 15% NF1 cells can be effectively inhibited, associated with distinct SDS-PAGE. Proteins were transferred to Immobilen P membrane (Millipore) reversion of the transformed morphology by BMS-186511. The in a wet transfer using Transblot (Bio-Rad), probed with the mouse mAb effects of BMS-186511 are specifically mediated by FT inhibition, 146-03E4, and detected with enhanced chemiluminescence reagents thereby blocking the processing and membrane localization of Ras. (Amersham).

MATERIALS AND METHODS RESULTS

Chemicals and Solutions. Unless specified, chemicals and solutions used Of the four NF1 cell lines (NF88-3, NF88-14, NF-90-8, and ST88- for the maintenance of cell culture were obtained from GIBCO/BRL. Sterile 14) that have been well characterized (19, 20, 39, 40, 44), ST88-14 tissue culture ware was obtained from Corning Laboratories (Corning, NY). cells are the most malignant, as established by their ability to grow in [3H]FPP (22.5 Ci/mmol) and [3H]GGPP (19.3 Ci/mmol) were supplied by soft agar, to grow in low serum concentrations (19), and to grow in New England Nuclear. BMS-186511 (37) is soluble in PBS but requires refractile multilayers (20). Three of the cell lines (NF88-3, NF88-14, vigorous vortexing and sonication. and NF90-8) grew very poorly in our hands. Moreover, these cells are Cell Lines and Culture. The ST88-14, NF88-3, NF88-14, and NF90-8 tumor lines were established from malignant schwannomas (neurofibrosarco- not as malignant as ST88-14 in terms of their morphology and growth mas) of NF1 patients (40, 41). The NF88-3 and NF88-14 cell lines were in soft agar. The poor growth characteristics and the lack of well- derived from samples from the same tumor site of the same patient (NF88-14 defined transformed morphology in NF88-3, NF88-14, NF90-8 cells was taken 6 months later). Unless specified otherwise, the cells were grown in precluded their study with BMS-186511. Because the potential rever RPMI 1640-based medium supplemented with 15% PCS in a humidified CO2 sal of the transformed phenotype could be monitored, we selected (5%) incubator maintained at 37Â°C. ST88-14 cells for our studies aimed at understanding the effects of FT Anchorage-independent Growth in Soft Agar Suspension. Anchorage- inhibitors on NF1 tumor cells. independent growth was measured in soft agar suspension as described earlier BMS-186511 Inhibits Anchorage-dependent Growth of (37). Briefly, single-cell suspensions of ST88-14 cells were obtained, mixed ST88-14 Cells. To determinewhetherFT inhibitorswould inhibitthe into a top soft agar mixture (0.3%), and laid on a 0.6% bottom agar layer. growth of NF1 malignant schwannomas with wild-type Ras, but Initially, inhibitors at the indicated concentrations were incorporated into the elevated Ras.GTP levels, ST88-14 cells were treated with BMS- soft agar mixture. FT inhibitors were replenished every 2 days for 8 days by overlaying 100 jj,l of 20-fold concentrated stock in PBS on the surface of the 186511 for a period of 10 days (compound replenished every 48 h) top agar layer. All the assays were carried out in parallel duplicate wells. while their growth was monitored. The anchorage-dependent growth Colonies larger than 0.1 mm size in diameter were counted. of ST88-14 cells was significantly suppressed in a concentration- Anchorage-dependent Growth. On day â€”1,cells were plated in a 6-well dependent manner by BMS-186511 (Fig. 1). The percentage of dish at 3 X IO4 cells/well in 3 ml culture medium and were allowed to attach growth inhibition increased gradually to reach 53 and 80% by day 8 overnight. The treatment was initiated on day 0, and fresh inhibitors were replenished every 48 h along with the change of the culture medium. BMS- 186511 was dissolved directly into the culture medium at the indicated con centrations. Viable cell counts were determined at the indicated times using 3e+06 O OuM BMS-186511 trypsinized cultures stained with trypan blue (GIBCO). Each growth data point shown is the mean of duplicate cultures carried out in parallel. In cases where â€¢ 10 (lM BMS-186511 2e+06 the drug treatment was withdrawn to study reversibility, the residual compound V 100 uM BMS-186511 was removed by washing cells twice for 3 h in culture medium without drug. In Vitro Prenylation Assays. Farnesyltransferase and geranylgeranyltrans- T 100|iM BMS-186511 (1 dose on day 0) ferase I activities were determined by quantitating the amount of 3[H] radio O 2e+06 activity transferred from [3H]FPP or [3H]GGPP onto the protein substrates, Â¿a Ras or Ras(CVLL), respectively (42). Ras and Ras(CVLL) are substrates for FT and GGTI, respectively. The standard reaction mixture consisted of 100 fl 2e+06 mM HEPES (pH 7.4), 10 mM DTT, l M.M[3H]FPP (22.5 Ci/mmol; NEN), or 0.5 H.M[3H]GGPP (19.3 Ci/mmol; NEN), protein substrate, and the appropri g ate cytosolic fraction in a final volume of 10 /j.1.In addition, various amounts 1e+06 of zinc and/or magnesium were added, depending on the specific enzyme, i.e., for FT, 25 mM MgCl2; for GGTI, 5 mM MgCl2, and 5 JUMZnCl2 Reaction mixtures contained Ras-CVLL or Ras at concentrations of 11 and 7 /J.M, 56+05 respectively. Assays were incubated for l h at 37Â°C and stopped by the addition of Laemmli sample buffer. Reaction products were separated by SDS-PAGE. Gels were stained for 5 min with Coomassie blue and destained; Oe+00 Iâ€”Â» then the p21Ras band was excised to determine the amount of 3H incorporated into each substrate (43). 10 12 14 16 Analysis of Ras Protein Processing by Western Blots. Unprocessed Ras proteins migrate slower compared to fully processed Ras in SDS-PAGE gels. Days This mobility shift is particularly striking in gels containing low concentrations Fig. 1. Inhibition of ST88-14 schwannoma cell growth by BMS-186511. The culture of bisacrylamide cross-linker. Mouse mAb 146-03E4 (Quality Biotech, Cam- conditions for growth curves were described in "Materials and Methods." Each data point den, NJ) recognizes H-Ras proteins specifically and may be used to detect Ras value represents the average of duplicate cultures treated in parallel. Fresh inhibitor was added to the cultures every 48 h along with a change of medium. On day 2 (single proteins in Western blots without prior concentration of Ras in the samples by dose-treated cells) and day 10 (arrow), drug treatment was terminated, and cultures were immunoprecipitation or metabolic labeling. After treatment for various periods washed free of the residual drug and cultured. 3570

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1995 American Association for Cancer Research. FARNESYLTRANSFERASE INHIBITORS AND NF1 with 10 and 100 JJ.Mof BMS-186511, respectively. By day 10 in the experimental section. Inhibition of colony growth in 10 and 100 LIM continuous presence of 100 JU.MBMS-186511, ST88-14 cells stopped BMS-186511-treated cells is shown in Fig. 3. Although a modest growing. The doubling times for 0, 10, and 100 JULMBMS-186511- inhibition was observed at 10 /LIM,it was complete at 100 LIMBMS- treated cells were not significantly different for the first 4 days (18.4, 186511. These results are similar to those obtained in cell culture and 22, and 24.4 h, respectively). The growth of untreated cells then suggest that BMS-186511 inhibits anchorage-dependent as well as slowed in the next 4 days (average doubling time, 40 h). Although anchorage-independent growth of ST88-14 NF1 cells. there was a modest 2-fold increase in the average doubling time (80 h) BMS-186511 Affects Farnesyltransferase but not Geranylgera- of 10 JAMBMS-186511 treated samples over days 4-8, the 100 nyltransferase I Activity. BMS-186511 is a methyl ester prodrug of (UM-treatedcells practically stopped growing (1200 h/doubling). How a potent FT inhibitor (33) and by itself is a poor FT inhibitor in vitro ever, following termination of the drug treatment on day 10 (Fig. 1, (IC5(), 2400 nM). However, it is cleaved by cellular esterases into arrow) and removal of the residual compound by washing, drug- BMS-185878 (33, 37), a potent FT inhibitor (IC50, 6 nin). To obtain treated cells recovered rapidly and started to grow again. Finally, direct evidence that BMS-186511 inhibited FT in ST88-14 cells, we continuous drug treatment appears to be necessary for growth inhibi measured the FT activity in treated versus untreated cells. Following tion because treating cells with a single dose for 48 h exerted no treatment for various times, cells were collected and fractionated into significant growth inhibition (Fig. 1). soluble cytosolic (S100) and paniculate (P100) fractions. The FT and When plated at Â«10% confluence, ST88-14 cells grow in 10 days GGTI activities in S100 were then determined. The activity of FT in to high density by piling up and are slightly refractile (Fig. 2A). treated cells was inhibited in a dose-dependent manner (Fig. 4/4). By Growth inhibition upon treatment with BMS-186511 was accompa day 2 of the treatment, the levels of FT activity were inhibited by 50 nied by distinct morphological changes in that the cells became flat and 77% with 10 and 100 LIMBMS-186511, respectively. After 10 and nonrefractile and grew in a monolayer (Fig. 2C). Although the days of treatment, the FT activity was inhibited by 70 and 90% with morphological reversion was partial at 10 Â¿AM(Fig. 2B), it was 10 and 100 JAMBMS-186511, respectively. However, when the treat complete at 100 /AMBMS-186511 (Fig. 2C). Microscopic observa ment period was short (48 h), inhibition of FT was transient. After tions at high (X200) magnification revealed more striking changes. withdrawal of BMS-186511 from the culture medium, FT activity ST88-14 cells treated with 100 /AMBMS-186511 became extremely recovered rapidly and returned to the control levels in 2 days. Pro flat and the cell body enlarged and spread out, appearing as a very thin longed treatment followed by withdrawal of the inhibitor from the sheet structure and seemingly fused with adjacent cells at places (data culture medium resulted in slow recovery in FT activity. Six days after not shown). The changes in morphology were detectable as early as 48 the drug withdrawal, the FT activity was still inhibited in cells treated h after the first dose of the compound, were pronounced by day 4, and continuously for 10 days with 100 JAMBMS-186511 (Fig. 4/1). We were complete by days 6-8. Interestingly, a single-dose treatment also determined the activity of GGTI, an enzyme closely related to (100 LIM)starting at day 0 for a period of 48 h was enough to initiate FT, which transfers a geranylgeranyl group from GGPP to the termi the morphological reversion but could not sustain it to result in nal cysteines of several proteins that terminate in CAAL (where C is complete reversal by days 6-8 as occurred in the continuously treated a cysteine, A is an aliphatic amino acid, and L is leucine; Refs. 45 and cells. 46). In contrast to the FT activity, the GGT I activity was not BMS-186511 Inhibits Anchorage-independent Growth of significantly affected by BMS-186511 (Fig. 4B), supporting our ear ST88-14 NF1 Cells. Anchorage-independent growth, as measured by lier conclusion that BMS-186511 is a specific FT inhibitor (33, 41). colony formation in soft agar, is typically associated with malignant Curiously, the levels of FT activity in untreated ST88-14 cells growth, i.e., untransformed cells do not grow in soft agar. ST88-14 increased 3-fold from day 2 to 10 (Fig. 4/4). However, we did not cells have been shown to form 1-2-mm diameter colonies when detect any such increase in GGTI activities in the same S100 samples seeded in soft agar suspension (20). We tested whether soft agar used for FT activity determination (Fig. 4ÃŸ). colony growth of ST88-14 cells, like the growth in cell culture, was BMS-186511-induced Redistribution of Ras Protein. Farnesyla- also inhibited by BMS-186511. Cells were seeded into the soft agar tion of Ras protein is a mandatory posttranslational modification for layer and treated with BMS-186511 on a schedule described in the its insertion into the plasma membrane and hence for its function in

B Fig. 2. Effect of BMS-186511 on morphology of ST88-14 NF1 cells. The initial plating conditions and the treatment protocol with inhibitor have been described in "Materials and Methods." Representative areas of the wells were photographed on day 10; X 40. A. no treatment; 0, 10 fiM BMS-186511; C, 100 jaw BMS-186511.

3571

B Fig. 3. Effect of BMS-186511 on growth of ST88-14 NF1 cells in soft agar. Experimental conditions used to assay the anchorage-independent growth in soft agar were described in "Materials and Methods." After 21 days of growth at 37Â°C,colonies were photographed at X 40 using phase contrast ring 2. Colonies larger than t), l mm were counted in the untreated (A) and treated samples (ÃŸand C) to determine the percentage of inhibition of colony growth. Colony growth was inhibited by 12 and 91% with BMS-186511 at 10 (B) and 100 (J.M (O, respectively. The total number of colonies/well in untreated samples were 810.

signal transduction. Inhibition of FT would prevent the farnesylation FT inhibitors results in a permanent reversal of the malignant pheno- and association of Ras to the cell membrane. Thus, we performed type. subcellular fractionation studies to show that this redistribution of Ras Our in vitro FT assays and Ras subcellular distribution studies on protein to the cytosolic fraction occurs in ST88-14 cells treated with extracts from drug-treated ST88-14 cells indicate that BMS-186511 BMS-186511. Extracts of ST88-14 cells were separated into soluble inhibits farnesylation and membrane association of Ras. Ras proteins S100 and paniculate P100 fractions and analyzed by SDS-PAGE. in ST88-14 cells are not mutated (19, 20); therefore, if the growth Proteins were transferred to a nylon membrane and probed with an inhibition observed in these cells is due to inhibition of normal Ras anti-H-Ras antibody. The Western blot (Fig. 5) shows that Ras pro activity, our results represent the first demonstration of FT inhibitors teins from the untreated samples were present exclusively in the inhibiting the growth of cells with normal Ras proteins. Although it is particulate membrane fraction, whereas Ras proteins from BMS- conceivable that the growth-inhibitory effects of BMS-186511 on 186511-treated samples appeared in the soluble fractions. In addition, ST88-14 cells could also be due to inhibition of farnesylation of Ras proteins in the soluble fractions migrated slower compared to nuclear lamins or other farnesylated small molecule GTPases, we those in the particulate fractions, and such a slower mobility is consider it unlikely because BMS-186511 was shown previously to characteristic of unprocessed Ras proteins (23-25, 37, 47, 48). A have no effect on NIH3T3 cells transformed by geranylgeranyl-Ras or significant amount of Ras was processed and was in the particulate myristoyl-Ras (37). fraction in cells treated with 10 JLLMBMS-186511 (Fig. 5, Lane 4). In Since farnesylation is required for normal Ras function, which in comparison, the amount of processed Ras in the particulate fraction turn is essential for growth responses of normal fibroblasts, FT inhib was negligible in cells treated with 100 /U.Mcompound (Fig. 5, Lane itors may potentially be harmful to normal cells. Quite surprisingly, 6). Unprocessed Ras began to appear in the cytosolic fractions as BMS-186511 at 100 JLLMdid not affect the monolayer growth of early as 48 h after treatment, and the resumption of processing was untransformed NIH 3T3 cells, while severely inhibiting the anchor seen 48 h after drug withdrawal (data not shown). age-dependent and anchorage-independent growth of both H-ra.v- and K-ra.v-transformed cells (37). Our results with BMS-186511 and the results from several other laboratories (23, 24) with different FT DISCUSSION inhibitors indicated that FT inhibitors had no significant effects on the Our results show that both the anchorage-dependent and anchorage- growth of untransformed cells despite near-complete inhibition of independent growth of the NF1 malignant schwannoma cell line, (normal) Ras farnesylation. These results raised questions about the ST88-14, was inhibited by the FT inhibitor BMS-186511 in a dose- importance of normal Ras function for cell growth. Although several dependent manner. In anchorage-dependent growth, ST88-14 cells possible reasons for these seemingly contradictory effects of FT grow to high density piling up in multilayers, and the cells are inhibitors on normal cells versus oncogenic Ras-transformed cells refractile. These growth properties were dramatically altered when have been discussed (23, 24, 37, 42), our results indicate that the cells were treated continuously with 100 JLLMBMS-186511 for 10 malignant growth of tumor cells that depend on elevated wild-type days. Cells became nonrefractile and grew to a contact-inhibited Ras activity is also inhibited by FT inhibitors such as BMS-186511. monolayer. However, these BMS-186511-induced changes are revers BMS-186511 is a methyl ester prodrug of BMS-185878, a potent ible; once the drug is removed, the contact-inhibited monolayer re FT inhibitor (IC5(), 6 HM;Refs. 33 and 37). Since BMS-186511 is a poor inhibitor of FT (I'Mvitro IC50, 2400 nM), it is unlikely that sumes growth, and eventually cells become refractile again and pile up in multilayers. These results suggest that normal growth control BMS-186511 itself is responsible for FT inhibition. Thus, hydrolysis mechanisms are restored to NF1 malignant schwannoma cells in the of BMS-186511 by intracellular esterases should provide presence of FT inhibitor, and that the effects are reversible upon drug BMS-185878. Similar prodrug strategies with FT inhibitors have been withdrawal. It remains to be determined if long-term treatment with reported (23-25, 29). Our in vitro assays indicate that GOTI, an 3572

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1995 American Association for Cancer Research. FARNESYLTRANSFERASE INHIBITORS AND NF1 enzyme closely related to FT, is not affected in ST88-14 cells treated with 100 H,MBMS-186511. These results are consistent with the demonstrated 3000-fold selectivity of BMS-185878 for FT over GGTI 123456 (33). There is a delay between the time course of onset of the FT inhibition and the inhibition of growth. The kinetics of FT inhibition were rapid, occurring as early as 2 days following drug treatment. However, the inhibition in growth rate and morphological changes were not significant until day 6. These results are consistent with the 24-h half life of the Ras protein and the time required to remove >95% of membrane-associated Ras proteins to the cytosolic fraction. When cells were treated for short periods of time, the recovery of FT activity following drug withdrawal to the levels seen in untreated cells was rapid and occurred in 48 h. However, when the treatment was prolonged (10 days), the recovery was only partial after 6 days of drug withdrawal. Because the wash procedures used to remove drug from cultures in short and long treatments were the same, the slow recovery after long treatment may be due to the effects of BMS- 186511 on other factors that govern the levels of FT activity. Despite

â€¢untreated controls 0 10|iM BMS-186511 D 100 HM BMS-186511 100 HM BMS-186511 (single dose) ol/mg/hr)E"â€¢^

Fig. 5. Inhibition of Ras protein processing by BMS-186511 in ST88-14 cells. ST88-14 cells were treated with 0 /XM(Lanes I and 2), W /IM (Lanes 3 and 4) and 100 /AM(Lanes 5 and 6) BMS-186511 under conditions described in "Materials and Methods." Cell-free extracts were prepared, fractionated into soluble SKX) (Lanes 1, 3. and 5) and

-â€¢><Â°20000 particulate PIOO (Lanes 2, 4, and 6) fractions by centrifugation, analyzed by 15% SDS-PAGE, transferred to immobilon P membrane, probed with 146-03E4 antibody, and detected by ECL detection system. The amount of protein loaded was 20 /ng in each case.

10000-t0 the partial recovery, the resumption of Ras processing and growth has occurred in 100 p,M-treated cells. Iv/rv% â€¢-\â€¢-, It is noteworthy that FT activity in growing cells seemed to increase -1^KV1 w/Tz.,ix.-//'/TT_L,II'///XI'//li^?:;/Ã¯\/,/-.-â€¢with cell density. The levels of FT activity were 3-fold higher in ST88-14 cells near confluence. In contrast, the levels of FT activity Days in culture were low (untreated cells, days 2-4) when cells were subconfluent. â€¢Untreated controls While FT activity has been reported from normal and Ras-trans- a 1O^MBMS-186511 formed cell lines (43), its relationship to malignant growth properties G 100 HM BMS-186511 Q 100 uM BMS-186511 (single dose) is not known. A recent study reported that FT activity was signifi cantly higher in chemical carcinogen-induced benign skin papillomas in SENCAR mice than in the epidermis of control animals (49). Because no significant changes were observed in initiated or initiated o plus early stages of promotion, FT has been speculated to play a role in the late stages of tumorigenesis. Another study found that FT activity was 3-fold higher in UVB radiation-induced tumors in SKH-1 hairless mice compared to epidermis from controls (50). In any case, a the significance of increased FT activity in ST88-14 cells near con- fluency is not clear at present. O Although the potential usefulness of FT inhibitors for treating o human tumors with Ras mutations has been discussed previously by us and others (27, 34, 35, 51), our present results suggest that these inhibitors may also be useful for human tumors such as NF1 malig Days in culture nant schwannomas. Because the effects of FT inhibitors are still reversible upon termination of short-term treatment (10 days), it is Fig. 4. Prenyltransferase activities were determined as described in "Materials and Methods." FT activity (A) was determined using Ras protein and [3H]FPP as the likely that longer term treatment will be required to achieve efficacy substrates, whereas GGTI activity (B) was determined using Ras-CVLL protein and in vivo. Although a number of formidable hurdles still have to be [3H]GGPP as the substrates. The activities shown were the average values of 3 and 2 overcome, it is conceivable that FT inhibitors may ultimately become independent experiments for FT and GGTI, respectively, with standard errors as indicated; bars. SEM. In each experiment, the activities were determined in triplicate for each the first generation of rationally designed drugs against the malignant sample. phenotype in NF1. It cannot be expected, however, that FT inhibitors 3573

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1995 American Association for Cancer Research. FARNESYLTRANSFERASE INHIBITORS AND NF1 will ever be a "cure" for NF1, since this disorder is associated with 22. Goldstein, J. L., Brown, M. S., Stradley, S. J., Reiss, Y., and Gierasch, L. M. Nonfarnesylated tetrapeptide inhibitors of protein farnesyltransferase. J. Biol. Chem., a variety of clinical manifestations, only one of which is tumor 266: 15575-15578, 1991. formation. 23. Kohl, N. E., Mosser, S. D., deSolms, S. J., Giuliani, E. A., Pompliano, D. L., Graham, S. L., Smith, R. L., Scolnick, E. M., Oliff, A., and Gibbs, J. B. Selective inhibition of ras-dependent transformation by a farnesyltransferase inhibitor. Science (Washington DC), 260: 1934-1937, 1993. ACKNOWLEDGMENTS 24. James, G. L., Goldstein, J. L., Brown, M. S., Rawson, T. E., Somers, T. C., McDowell, R. S., Crowley, C. W., Lucas, B. K., Levinson, A. D., and Marsters, J., Jr. We thank Drs. David Floyd and Eric M. Gordon for their great interest and Benzodiazepine peptidomimetics: potent inhibitors of Ras farnesylation in animal support of our work, Dr. D. Patel for providing us with BMS-186511, and cells. Science (Washington DC), 260: 1937-1942, 1993. Dr. J. Hunt for critical review of the manuscript. 25. Garcia, A. M., Rowell, C., Ackermann, K., Kowalczyk, J. J., and Lewis, M. D. Peptidomimetic inhibitors of Ras farnesylation and function in whole cells. J. Biol. Chem., 268: 18415-18418, 1993. 26. Nigam, M., Seong, C. M., Qian, Y., Hamilton, A. D.. and Sebti, S. M. Potent REFERENCES inhibition of human tumor p21ras farnesyltransferase by AlA2-lacking p21ras CA1A2X peptidomimetics. J. Biol. Chem., 268: 20695-20698, 1993. 1. Riccardi. V. M. Neurofibromatosis: Phenotype, Natural History and Pathogenesis, Ed. 2, pp. 1-179. Baltimore: Johns Hopkins University Press, 1992. 27. Gibbs, J. B., Oliff, A., and Kohl, N. E. Farnesyltransferase inhibitors: Ras research yields a potential cancer therapeutic. Cell, 77: 175-178, 1994. 2. Barker, D., Wright. E., Nguyen, K., Cannon, L., Fain, P., Goldgar, D., Bishop, D. T., Carey, J., Baty, B.. Kivlin, J., Willard, H., Waye, J. S., Greig, G., Leinwand, L., 28. Kohl, N. E., Wilson, F. R., Mosser, S. D., Giuliani, E., deSolms, S. J., Conner, M. W.. Nakamura, Y., O'Connell, P., Leppert. M., Lalouel, J. M., White, R., and Skolnick, Anthony, N. J., Holtz, W. J., Gomez, R. P., Lee, T., Smith, R. L., Graham, S. L., M. Gene for von Recklinghausen ncurofibromatosis is in the pericentromeric region Hartman, G. D., Gibbs, J. B., and Oliff, A. Protein farnesyltransferase inhibitors block of chromosome 17. Science (Washington DC), 236: 1100-1102, 1987. the growth of ras-dependent tumors in nude mice. Proc. Nati. Acad. Sci. USA, 91: 9141-9145, 1994. 3. Seizinger, B. R., Rouleau, G. A., Ozelius, L. J., Lane, A. H., Faryniarz, A. G., Chao, M. V., Huson, S., Korf, B. R., Parry, D. M., Pericak-Vance, M. A., Collins, F. S., 29. Vogt, A., Qian, Y. M., Blaskovich, M. A., Possum, R. D., Hamilton, A. D., and Sebti, Hobbs, W. J., Falcone, B. G., lannazzi, J. A., Roy, J. C, St. George-Hyslop, P. H., S. M. A non-peptide mimetic of Ras-CAAX: selective inhibition of farnesyltrans ferase and ras processing. J. Biol. Chem., 270: 660-664, 1995. Tanzi. R. E., Bothwell, M. A., Upadhyaya. M., Harper, P., Goldstein, A. E., Hoover, D. L., Bader, J. L., Spence, M. A., Mulvihill, J. J., Aylsworth, A. S., Vance, J. M., 30. Wai, J. S., Bamberger, D. L., Fisher, T. E., Graham, S. L., Smith, R. L., Gibbs, J. B., Rossenwasser, G. O. D., Gaskell, P. C., RosÃ©s,A. D., Martuza, R. L., Breakefield, Mosser, S. D., Oliff, A. I., Pompliano, D. L., Rands, E., and Kohl, N. E. Synthesis and X. O., and Gusella, J. F. Genetic linkage of von Recklinghausen neurofibromatosis to biological activity of Ras farnesyl protein transferase inhibitors: tetrapeptide analogs the nerve growth factor receptor gene. Cell, 49: 589-594, 1987. with amino methyl and carbon linkages. Bioorg. Med. Chem., 2: 939-947, 1994. 4. Cawthon, R. M., Weiss, R., Xu, G. F., Viskochil, D., Culver, M., Stevens, J., 31. Masters, J. C., McDowell, R. S., Reynolds. M. E., Oare, D. A., Somers, T. C., Stanley, Robertson, M., Dunn, D., Gesteland, R., O'Connell, P., and White, R. A major M. S., Rawson, T. E., Struble, M. E., Burdick, D. J., Chan, K. S., Duarte, C. M., Paris, segment of the neurofibromatosis type 1 gene: cDNA sequence, genomic structure, K. J., Tom, J. Y. K., Wan, D. T., Xue, Y., and Burnier, J. P. Benzodiazepine and point mutations. Cell, 62: 193-201, 1990. peptidomimetic inhibitors of farnesyltransferase. Bioorg. Med. Chem., 2: 949-957, 5. Viskochil, D., Buchberg, A. M., Xu, G.. Cawthon, R. M., Stevens, J., Wolff, R. K., 1994. Culver, M., Carey, J. C., Copeland, N. G., Jenkins, N. A., White, R., and O'Connell, 32. Leftheris, K., Kline, T., Natarajan, S., DeVirgilio, M. K., Cho, Y. H., Pluscec, J., P. Deletions and a translocation interrupt a cloned gene at the neurofibromatosis type Ricca, C., Robinson, S., Seizinger, B. R., Manne, V., and Meyers, C. A. Peptide based 1 locus. Cell, 62: 187-192, 1990. p21 ras farnesyl transferase inhibitors: systematic modification of the tetrapeptide 6. Wallace, M. R., Marchuk. D. A., Andersen, L. B., Letcher, R., Odeh, H. M.. Saulino, CA,A2X motif. Bioorg. Med. Chem. Lett., 4: 887-892, 1994. A. M., Fountain, J. W.. Brereton, A., Nicholson, J., Mitchell, A. L., Brownstein, 33. Patel, D. V., Gordon, E. M., Schmidt, R. J., Weiler, H. N., Young, M. G.. Zahler, R., B. H., and Collins, F. S. Type l neurofibromatosis gene: identification of a large Barbacid, M., Gullo-Brown, J. L., Ricca, C., Robinson, S., Seizinger, B. R., Tuomari, transcript disrupted in three NF1 patients. Science (Washington DC), 249: 181-186, V. A., and Manne, V. Phosphinyl acid based bisubstrate analog inhibitors of farnesyl 1990. protein transferase. J. Med. Chem., 38: 435-442, 1995. 7. Marchuk, D. A., Saulino, A. M., Tavakkol, R., Swaroop, M., Wallace, M. R., 34. Tamanoi, F. Inhibitors of Ras farnesyltransferases. Trends Biochem. Sci., 18: Andersen, L. B., Mitchell, A. L., Gutmann, D. H., Boguski, M., and Collins, F. S. 349-353, 1993. cDNA cloning of the type l neurofibromatosis gene: complete sequence of the NF1 35. Manne, V., Ricca, C. S., Cullo Brown, J., Tuomari, A. V., Yan, N., Patel, D. V., gene product. Genomics, /// 931-940, 1991. Schmidt, R., Lynch, M. J., Ciosek, C. P., Carboni, J. M., Robinson, S., Gordon, E. M., 8. Daston, M. M., Scrable, H., Nordlund. M., Sturbaum, A. K., Nissen, L. M., and Barbacid, M., Seizinger, B. R., and Biller, S. A. Ras farnesylation as a target for novel Ratner, N. The protein product of the neurofibromatosis type 1 gene is expressed at antitumor agents: potent and selective farnesyl diphosphate analog inhibitors of highest abundance in neurons, Schwann cells, and oligodendrocytes. Neuron, 8: farnesyltransferase. Drug Dev. Res., 34: 121-137, 1995. 415-428, 1992. 36. Crowell, P. L., Ren, Z. B., Lin, S. Z., Vedejs, E., and Gould, M. N. Structure-activity 9. DeClue, J. E., Cohen, B. D., and Lowy, D. R. Identification and characterization of relationships among monoterpene, inhibitors of protein Â¡soprenylation and cell the neurofibromatosis type 1 protein product. Proc. Nati. Acad. Sci. USA, 88: proliferation. Biochem. Pharmacol., 47: 1405-1415, 1994. 9914-9918, 1991. 37. Manne, V., Yan, N., Carboni, J. M., Tuomari, A. V., Ricca, C. S., GÃ¼iloBrown, J., 10. Gutmann, D. H., Wood, D. L., and Collins, F. S. Identification of the neurofibroma Andahazy, M. L., Schmidt, R. Jâ€žPatel, D., Zahler, R., Der, C. J., Cox, A. D., tosis type 1 gene product. Proc. Nati. Acad. Sci. USA, 88: 9658-9662, 1991. Weinmann, R., Hunt, J. T., Barbacid, M., and Seizinger, B. R. Bisubstrate inhibitors 11. Xu, G. F., Lin, B., Tanaka, K., Dunn, D., Wood, D., Gesteland, R., White, R., Weiss, of farnesyltransferase: a novel class of specific inhibitors of Ras transformed cells. R., and Tamanoi, F. The catalytic domain of the neurofibromatosis type 1 gene Oncogene, 10: 1763-1779, 1995. product stimulates ras GTPase and complements ira mutants of 5. cerevisiae. Cell, 63: 38. Prendergast, G. C., Davide, J. P., Desolms, S. J., Giuliani, E. A., Graham, S. L., 835-841, 1990. Gibbs, J. B., Oliff, A., and Kohl, N. E. Farnesyltransferase inhibition causes mor 12. Xu, G. F., O'Connell, P., Viskochil, D., Cawthon, R., Robertson, M., Culver, M., phological reversion of ras-transformed cells by a complex mechanism that involves Dunn, D., Stevens. J., Gesteland, R., White, R., and Weiss, R. The neurofibromatosis regulation of the actin cytoskeleton. Mol. Cell. Biol., 14: 4193-4202, 1994. type 1 gene encodes a protein related to GAP. Cell, 62: 599-608, 1990. 39. Cox, A. D., Garcia, A. M., Westwick, J. K., Kowalczyk, J. J., Lewis, M. D., Brenner, 13. Martin, G. A., Viskochil, D., Bollag, G., McCabe, P. C., Crosier, W. J.. Haubruck, H., D. A., and Der, C. J. The CAAX peptidomimetic compound b581 specifically blocks Conroy, L., Clark, R.. O'Connell, P., Cawthon, R. M., Innis, M. A., and McCormick, farnesylated, but not geranylgeranylated or myristylated, oncogenic ras signaling and F. The GAP-related domain of the neurofibromatosis type 1 gene product interacts transformation. J. Biol. Chem., 269: 19203-19206, 1994. with ras p21. Cell, 63: 843-849, 1990. 40. Fletcher, J. A., Kozakewich, H. P., Hoffer, F. A., Lage, J. M., Weidner, N., Tepper, 14. Ballester, R., Marchuk, D., Boguski, M., Saulino, A., Letcher, R., Wigler, M., and R., Pincus, G. S., Morton, C. C., and Corson, J. M. Diagnostic relevance of clonal Collins, F. The NF1 locus encodes a protein functionally related to mammalian GAP cytogenetic aberrations in malignant soft-tissue tumors. N. Engl. J. Med., 324: and yeast IRA proteins. Cell, 63: 851-859, 1990. 436-442, 1991. 15. Barbacid, M. ras genes. Annu. Rev. Biochem., 56: 779-827, 1987. 41. Glover, T. W., Stein, C. K., Leagues, E., Anderson, L. B., Brereton, A., and Johnson, 16. McCormick, F. The world according to GAP. Oncogene, 5: 1281-1283, 1990. S., Molecular and cytogenetic analysis of tumors in von Recklinghausen neurofibro 17. Bos, J. L. ras oncogenes in human cancer: a review. Cancer Res., 49: 4682â€”4689. matosis. Genes Chromosomes Cancer, 3: 62-70, 1991. 1989. 42. Carboni, J. M., Yan, N., Cox, A. D., Bustelo, X., Graham, S. M., Lynch, M. J., 18. McCormick, F. ras GTPase activating protein: signal transmitter and signal termina Weinmann, R., Seizinger, B. R., Der, C. J., Barbacid, M., and Manne, V. Farnesyl tor. Cell, 56: 5-8, 1989. transferase inhibitors arc inhibitors of Ras, but not R-Ras/TC21 transformation. 19. Basu, T. N., Gutmann, D. H., Fletcher, J. A., Glover, T. W., Collins, F. S., and Oncogene, 10: 1905-1913, 1995. 43. Manne, V., Roberts, D., Tobin, A., O'Rourke, E.. De-Virgilio, M., Meyers, C., Downward, J. Aberrant regulation of ras proteins in malignant tumour cells from type 1 neurofibromatosis patients. Nature (Lond.), 356: 713-715, 1992. Ahmed, N., Kurz, B., Resh, M., Kung, H. F., and Barbacid. M. Identification and 20. DeClue, J. E., Papageorge, A. G., Fletcher, J. A., Diehl, S. R., Ratner, N., Vass, W. C., preliminary characterization of protein-cysteine farnesyltransferase. Proc. Nati. Acad. and Lowy, D. R. Abnormal regulation of mammalian p21ras contributes to malignant Sci. USA, 87: 7541-7545, 1990. tumor growth in von Recklinghausen (type 1) neurofibromatosis. Cell, 69: 265â€”273, 44. Nigro, J. M., Baker, S. J., Precinger, A. C., Jessup, J. M., Hostetter, R., Cleary, K., 1992. Bigner, S. H., Davidson, N., Baylin, S., Devilee, P., Glover, T., Collins, F. S., 21. Seizinger, B. R. NF1: a prevalent cause of tumorigenesis in human cancers? Nat. Westom, A., Modali, R., Harris, C. C., and Vogelstein, B. Mutations in the p53 gene Genet., 3: 97-99, 1993. occur in diverse human types. Nature (Lond.), 342: 705-708, 1989.

3574

45. Cox. A. D., and Der. C. J. The ras/cholesterol connection: implications for ras ali/ation leads to disruption of p21 ras membrane association in whole cells. Bioorg. oncogenicity. Crii. Rev. Oncog., J: 365-400. 1992. Med. Chem. Lett.. 4: 2579-2584, 1994. 46. Clarke. S. Protein isoprenylation and methylation at carhoxy-terminal residues. Annu. 49. Agarwal. R.. Khan. S. G.. Athar. M.. Zaidi. S. I.. Bickers. D. R.. and Mukhlar. H. ras Rev. Biochem.. 61: 355-386. 1992. protein p21 processing enzyme farnesyltransferase in chemical carcinogen-induced 47. Graham. R. L., deSolms. S. J., Giuliani, E. A., Kohl, N. E., Mosser, S. D.. Oliff, A., murine skin tumors. Mol. Carcinog.. Â«:290-298. 1993. Pompliano, D. L.. Rands, E.. Brcslin, M. J.. Deana, A. A.. Garsky, V. M.. Scholz, 50. Khan, S. G.. Bickers. D. R., Mukhtar, H., and Agarwal. R. Ras p21 farnesylalion in T. H.. Gibbs. J. B.. and Smith. R. L. Pseudopcptide inhibitors of Ras famcsyl-prolcin ultraviolet B radiation-induced tumors in the skin of SK1I-I hairless mice. J. Invest, transferase. J. Med. Chem.. 37: 725-732., 1994. Dermatol., 102: 754-758. 1994. 48. Qian. Y. M.. Blaskovich. M. A.. Seong. C. M.. Vogt. A.. Hamilton. A. D.. and Sebli. 51. Gibhs. J. B. Ras C-terminal processing enzymesâ€”new drug targets? Cell, 65: 1-4, S. M. Peptidomimetic inhibitors of p21ras farnesyltransferase: hydrophobic function- 1991.

3575

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 1995 American Association for Cancer Research. Farnesyltransferase Inhibitors Block the Neurofibromatosis Type I (NF1) Malignant Phenotype

Ning Yan, Carolyn Ricca, Jonathan Fletcher, et al.

Cancer Res 1995;55:3569-3575.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/55/16/3569

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/55/16/3569. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.