Farnesyltransferase Inhibitor Induces Rapid Growth Arrest and Blocks p70s6k Activation by Multiple Stimuli*

We have previously shown that the peptidomimetic farnesyltransferase inhibitor L-744,832 (FTI) inhibits p70s6k activation and cell growth in a mouse keratinocyte cell line but only at concentrations of FTI signifi-cantly higher than those required for the inhibition of Ras farnesylation. Here we show that the rapid kinetics of FTI inhibition of DNA synthesis (within 1.5 h) in both normal and v -K-Ras transformed keratinocytes matches the rapid kinetics of p70s6k inhibition observed previously. It is further shown that FTI inhibits p70s6k activation in response to serum, phorbol myristate acetate, and increased amino acid levels. The phosphatase inhibitor calyculin A partially reverses the FTI-induced dephosphorylation of p70s6k, suggesting that FTI may act upstream of a protein phosphatase. A rapamycin-resist-ant mutant of p70s6k is shown to be resistant to FTI-induced dephosphorylation of the major rapamycin-sen-sitive phosphorylation site of p70s6k, Thr 389 . Together, these data demonstrate that FTI rapidly inhibits DNA synthesis irrespective of the presence of v -K-Ras and that FTI inhibits p70s6k activation in response to multiple stimuli. Because the FTI L-744,832 mimics many of the effects of rapamycin, this FTI may prove effective to immunoblots using monoclonal (Sigma).

the relevant target(s) of FTIs but that FTIs may act through other farnesylated proteins or through a farnesylation-independent mechanism to inhibit cell growth.
We have previously shown that the peptidomimetic FTI L-744,832, induces tumor regression associated with a G 1 cell cycle arrest and increased apoptosis in transgenic mice (6). Tumor regression was associated with p70s6k dephosphorylation. In a mouse keratinocyte cell line (Balb-MK) inhibition of DNA synthesis by FTI correlated closely with p70s6k inhibition, and FTI was shown to inhibit p70s6k activity as early as 15 min (7). Under the same conditions, FTI did not affect Erk1/2 activity, even though Erk1/2 activity is completely dependent on Ras function in these cells. These observations indicated that Ras is not the only target of FTI. It has been suggested that farnesylated proteins other than Ras may be important targets for FTI (8,9). Although this is likely, such a mechanism cannot explain the rapid inhibition of p70s6k by FTI. To be consistent with the rapid time course, the putative farnesylated target would have to exhibit a very short half-life. The rapid time course observed may indicate that FTI acts by farnesylation-independent as well as farnesylation-dependent mechanisms. The observations that FTI concentrations higher than those required to completely inhibit Ha-Ras farnesylation are necessary to inhibit the growth of cells in culture (7) also suggests a farnesylation-independent FTI target. Regardless of the identity of the FTI target involved in p70s6k inhibition, blockade of this pathway correlates closely with tumor regression in transgenic mice and with growth inhibition in cell culture.
p70s6k is situated downstream of the evolutionarily conserved kinase mTOR (mammalian target of rapamycin) in mammalian cells. mTOR regulates protein synthesis and cell cycle progression (10 -14) and is the target of the immunosuppressant rapamycin. Rapamycin acts by binding its intracellular receptor FKBP12 to form an inhibitory complex that then binds and inactivates mTOR. Recently, the mTOR/p70s6k pathway was shown to be inappropriately activated in several tumor cell lines (15,16) and activated in response to the Gli oncogene (17), implicating the mTOR pathway as a potential target for anti-cancer drugs.
To further clarify the mechanism(s) by which FTI inhibits mitogen-dependent cell growth we have examined the kinetics and Ras dependence of DNA synthesis inhibition by FTI and investigated the possible involvement of the mTOR pathway and protein phosphatases in the FTI-induced inhibition of p70s6k. The results of these studies demonstrate that FTI rapidly inhibits DNA synthesis irrespective of the presence of v-K-Ras and that FTI inhibits p70s6k activation in response to multiple stimuli. We also present evidence using a rapamycinresistant mutant of p70s6k and the phosphatase inhibitor calyculin A that FTI and rapamycin actions are consistent with the involvement of a protein phosphatase.

Cell Culture and [ 3 H]Thymidine Incorporation
Assays-Balb-MK cells were grown as described previously (7). v-K-Ras transformed Balb-MK cells (KC) were grown under the same conditions as the Balb-MK cells. For amino acid deprivation, Balb-MK cells were incubated overnight in minimum essential medium containing 5 mg/ml calcium chloride and lacking L-cysteine, L-isoleucine, L-leucine, and L-tryptophan. For amino acid stimulation, cells were treated for 30 min with the same minimum essential medium containing the complete complement of amino acids. For stimulation by growth factors, Balb-MK cells were treated with 8% dialyzed fetal bovine serum and 4 ng/ml EGF in minimum essential medium for 15 min. Cos1 cells obtained from the American Type Culture Collection were grown in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. For [ 3 H]thymidine incorporation assays Balb-MK or KC cells were plated at 20,000 cells/well in 24-well plates and incubated for 24 h. Cells were treated as described in the figures and pulsed for 1 h with [ 3 H]thymidine. Results were quantitated as described previously (7). The compounds calyculin A, PD98059, and SB202190 were obtained from Alexis Biochemicals, New England Biolabs, and Calbiochem, respectively. LY294002, ionomycin, wortmannin, and PMA were obtained from Sigma. FTI was obtained as a gift from Merck.
cDNA Constructs and Transfection-A plasmid encoding hemagglutinin antigen-tagged p85s6k was kindly provided by Joseph Avruch (Harvard Medical School). FLAG-tagged wild type p85s6k and a FLAGtagged p85s6k mutant in which 30 amino acid residues were truncated from the N terminus (7 residues truncated from the corresponding p70s6k sequence) and 101 amino acid residues were deleted from the C terminus (⌬NT/⌬CT) were amplified from the plasmid encoding hemagglutinin antigen-p85s6k using polymerase chain reaction with the primer sets: 5Ј-TTTTGGATCCATGGACTATAAGGACGATGATGACA-AAAGGCGACGACGGAGGCGG-3Ј and 5Ј-TTTTGAATTCTCATAGAT-TCATACGCAGGTG-3Ј, and 5Ј-TTTTGGATCCATGGACTATAAGGAC-GATGATGACAAAGACCTGGACCAGCCAGAG-3Ј and 5Ј-TTTTG-AATTCTCATTCTTTCACACTTTCAAG-3Ј,respectively.Underlinedportions represent regions complementary to the plasmid template. Polymerase chain reaction was performed using Pfu polymerase (Promega) according to the manufacturer's instructions. Polymerase chain reaction products were restricted with EcoRI and BamHI and subcloned into the pcDNA3.1 expression vector (Invitrogen). Both inserts were verified to be correct by DNA sequencing. Cos1 cells were transfected in 100-mm dishes with 2 g/plate of wild type FLAG-p85s6k or 3 g/plate of ⌬NT/⌬CT p70s6k using 2.5 l of Fugene 6 (Roche Molecular Biochemicals)/g of DNA.
Cell Lysis, Immunoblotting, and Kinase Assays-Cell lysis, protein quantitation, and immunoblot analysis were performed as described previously (7). Antibodies specific for p85/70s6k, phospho-Erk1/2, and Ser(P) 473 PKB were described previously (7). Phospho-specific antibodies to Thr 389 of p70s6k were obtained from New England Biolabs. FLAG-tagged proteins were detected on immunoblots using the M2 monoclonal antibody (Sigma). Kinase assays were performed as described previously (7) using the S6-peptide (Santa Cruz) as the substrate. The results of kinase assays were analyzed using a PhosphorImager (Molecular Dynamics) and quantitated using ImageQuant software.

FTI Induces Rapid Growth Arrest in Both Normal and K-Ras
Transformed Keratinocytes-FTI rapidly inhibits p70s6k (7), and p70s6k is thought to play a role in G 1 cell cycle progression (18 -20). If FTI inhibits cell growth by inhibiting the p70s6k pathway, then FTI might also rapidly inhibit cell cycle progression. As shown in Fig. 1, a 30-min pretreatment of Balb-MK cells with FTI inhibited DNA synthesis by over 60%, consistent with a role for the inhibition of the p70s6k pathway in FTI action. Similar analyses of the effect of FTI on cell proliferation were carried out on KC cells. KC cells (21) display constitutively high levels of active Erk1/2 compared with the parental Balb-MK cells, and this high level of Erk1/2 activity is not altered by FTI (data not shown). This is consistent with our earlier observations that Erk1/2 is insensitive to FTI in the parental Balb-MK cells. As shown in Fig. 1, FTI inhibited DNA synthesis in KC cells and Balb-MK cells to approximately the same extent. Because oncogenic K-Ras is geranylgeranylated and retains its biological activity in the presence of FTIs (3,4), these data further support the hypothesis that Ras is not a relevant target in rapid FTI-induced growth arrest.
p70s6k Activation in Mouse Keratinocytes-To our knowledge the regulation of p70s6k activity in keratinocytes by various stimuli has not been reported. Thus, we tested the ability of several agents to activate p70s6k in Balb-MK cells. EGF, IGF-I, and PMA were able to activate p70s6k in Balb-MK cells (Fig. 2). Ionomycin, however, did not appreciably activate p70s6k, although ionomycin activates p70s6k in mouse fibroblasts (22). Activation of p70s6k by EGF and IGF-I was associated with a shift in electrophoretic mobility and increased phosphorylation of Thr 389 . Thr 389 is the major rapamycin-sensitive site on p70s6k, and phosphorylation of this site is required for catalytic activity (23). PMA, like EGF and IGF-I, activated p70s6k, induced a shift in electrophoretic mobility, and increased phosphorylation of Thr 389 as determined by staining with a Thr(P) 389 -specific antibody (Fig. 2). IGF-I induced PKB phosphorylation on Ser 473 , an event required for catalytic activity, but did not induce Erk1/2 phosphorylation on sites required for activity. In contrast, PMA induced Erk1/2 phosphorylation but not PKB phosphorylation. The data indicate that p70s6k, PKB, and Erk1/2 are each differentially regulated in mouse keratinocytes.
Growth factors and PMA activate p70s6k by distinct mechanisms that are, respectively, sensitive and insensitive to wortmannin (24). Activation of p70s6k by serum in Balb-MK cells was blocked by inhibition of PI 3-kinase with wortmannin ( Fig.  3A) but not by down-regulation of classical and novel protein kinase C isoforms by chronic PMA pretreatment (Fig. 2). In contrast, PMA-induced activation of p70s6k was inhibited by chronic PMA treatment but not by wortmannin. These data indicate that in mouse keratinocytes serum and PMA activate p70s6k by different mechanisms.
As we have shown previously (7), expression of the N17 dominant negative mutant of Ha-Ras in Balb-MK cells is able to partially inhibit p70s6k activation in response to serum and EGF while completely abolishing Erk1/2 phosphorylation. Interestingly, N17-Ha-Ras expression did not inhibit PMA-induced Erk1/2 phosphorylation as reported by others (25)(26)(27) but still partially inhibited p70s6k activation (Fig. 2). This observation suggests that the requirements of p70s6k and Erk1/2 for Ras activity are fundamentally different.
FTI Abrogates p70s6k Activation in Response to Distinct Stimuli-Because serum and PMA activate p70s6k through distinct mechanisms, we examined whether FTI could inhibit p70s6k activation by both serum and PMA. As shown in Fig. 3A (top panel), FTI inhibited p70s6k activation by both serum and PMA to levels similar to that observed in unstimulated cells. FTI inhibition of p70s6k activation by both stimuli was associated with a decrease in phosphorylation of Thr 389 (Fig. 3A,  second panel) and a loss of the slowly migrating hyperphosphorylated forms of p70s6k (Fig. 3A, third panel). We also observed FTI inhibition of p70s6k activation by amino acids (Fig. 3B). Studies performed in a number of laboratories indicate that amino acids stimulate p70s6k through the regulation of mTOR activity (28 -30). Together, our data raise the possibility that FTI may act on the mTOR pathway to inhibit p70s6k activation by multiple stimuli.
p70s6k becomes phosphorylated on multiple proline-directed phosphorylation sites during activation, and MAP kinases, such as p38, can phosphorylate such sites. Therefore, we tested the ability of the p38 inhibitor SB202190 to inhibit p70s6k activation. Although SB202190 had no effect on p70s6k activation, electrophoretic mobility, or Thr 389 phosphorylation in response to serum stimulation, it markedly increased Erk1/2 phosphorylation on sites required for catalytic activity (Fig.  3A). Interestingly, SB202190 induced "hyperactivation" of p70s6k stimulation by PMA, which was associated with a marked decrease in p70s6k electrophoretic mobility and increased Erk1/2 phosphorylation. This hyperactivation of p70s6k and increased Erk1/2 phosphorylation likely relates to studies in which SB203580 was shown to activate Raf (31) and SB202190 was shown to activate Erk1/2 (32) presumably by blocking an inhibitory pathway involving p38. Because oncogenic Raf induces p70s6k activation (33), and Raf is activated by PMA (34), the increased PMA-induced activation of p70s6k in the presence of SB202190 likely resulted from an increased activation of Raf that led to increased p70s6k and Erk1/2 activation. The difference in the effect of SB202190 on PMA and serum-induced p70s6k activation suggests that Erk1/2 and p70s6k share a common upstream activator during PMA stimulation but are regulated by different upstream signaling elements during serum stimulation.
FTI Induces Dephosphorylation of Thr 389 of Wild Type p85s6k, but Not Rapamycin-insensitive p70s6k-To more fully

FIG. 2. p70s6k is activated in Balb-MK cells by multiple stimuli.
Balb-MK cells were treated with or without an adenovirus encoding N17-Ha-Ras in medium containing serum and EGF and incubated for 24 h. Cells were fed with serumless medium with or without 16 M PMA and incubated another 24 h. Cells were stimulated with serum and/or growth factors for 15 min or with PMA and/or ionomycin for 30 min in the presence of FTI and the other inhibitors. Cell extracts were prepared and subjected to p70 s6 kinase assays (upper panel) or subjected to immunoblot analysis using antibodies specific for Thr(P) 389 of p70s6k, total p70s6k, active, dually phosphorylated Erk1/2, and Ser(P) 473 PKB (lower panels).
FIG. 3. FTI abrogates p70s6k activation in response to serum, PMA, and amino acid stimulation. A, Balb-MK cells were treated as in Fig. 2 except that prior to stimulation, cells were pretreated as indicated with 60 M FTI for 24 h or with 20 M SB202190, 100 nM wortmannin, or 20 M PD98059 for 1 h. Cell extracts were analyzed as described in the legend to Fig. 2. B, Balb-MK cells were quiesced with a 14-h treatment in amino acid-deficient medium lacking serum and EGF with or without FTI. Cells were stimulated for 30 min with medium containing all essential amino acids and/or serum with EGF in the absence or continued presence of FTI. Cell extracts were analyzed for p70 s6 kinase activity as in Fig. 2. explore the specificity and mechanisms of action of FTI on the p70s6k pathway, we examined the effect of FTI on wild type p85s6k and on a rapamycin-insensitive deletion mutant of p70s6k termed ⌬NT/⌬CT. P85s6k is a variant of p70s6k expressed from the same gene but containing an additional 23 amino acids at the N terminus. P85s6k activity is required for cell proliferation (35), but p85s6k is expressed at much lower levels than p70s6k. The ⌬NT/⌬CT deletion mutant is similar to previously characterized mutants (36 -38) and was prepared by deleting the 30 N-terminal and the 101 C-terminal residues of p85s6k.
To examine the effect of FTI on p85s6k and ⌬NT/⌬CT, FLAG-tagged p85s6k or FLAG-tagged ⌬NT/⌬CT was expressed in Cos1 cells. The cells were treated with FTI, stimulated with serum, and analyzed for the phosphorylation level of the rapamycin-sensitive site Thr 389 using a phospho-specific antibody or analyzed for the level of protein expression using an anti-FLAG antibody (Fig. 4, top and bottom panels, respectively). FTI decreased the level of Thr 389 phosphorylation of p85s6k but did not affect the level of Thr 389 phosphorylation of the ⌬NT/⌬CT mutant. Treatment with rapamycin, like FTI, induced dephosphorylation of p85s6k at Thr 389 but did not affect the level of Thr 389 phosphorylation of the ⌬NT/⌬CT mutant.
We next investigated the dependence of Thr 389 phosphorylation of p85s6k and ⌬NT/⌬CT on the PI 3-kinase pathway using the PI 3-kinase inhibitor, LY294002. PI 3-kinase inhibition abolished the phosphorylation of both p85s6k and ⌬NT/ ⌬CT on Thr 389 , indicating that neither FTI nor rapamycin act through effects on the PI 3-kinase pathway. Because these results are more consistent with FTI and rapamycin activating a Thr 389 phosphatase than inhibiting a Thr 389 kinase, we investigated the ability of the phosphatase inhibitor calyculin A to antagonize the FTI and rapamycin-induced dephosphorylation of Thr 389 . As shown in Fig. 4, 10 nM calyculin A was able to partially reverse FTI and rapamycin-induced dephosphorylation of p85s6k at Thr 389 . Higher concentrations of calyculin A in combination with FTI were cytotoxic in Cos1 cells, precluding their use.
Calyculin A Partially Blocks FTI and Rapamycin-induced Dephosphorylation of Endogenous p70s6k-Because calyculin A inhibited FTI and rapamycin-induced dephosphorylation at Thr 389 of p85s6k overexpressed in Cos1 cells, we sought to determine whether calyculin A also inhibited FTI and rapamy-cin-induced dephosphorylation of endogenous p70s6k. As shown in Fig. 5 (second panel), calyculin A alone induced a marked decrease in p70s6k electrophoretic mobility relative to untreated cells, whereas FTI induced a substantial increase in electrophoretic mobility relative to untreated cells. In combination with FTI, calyculin A treatment antagonized the ability of FTI to induce p70s6k dephosphorylation. Similar results were observed with the combination of rapamycin and calyculin A treatments.
We also analyzed the effect of the various treatments on the degree of Thr 389 phosphorylation using phospho-specific antibodies (Fig. 5, top panel). NIH Image 1.62 was used to estimate the relative staining intensities in the various lanes. Calyculin A treatment alone increased Thr 389 phosphorylation by about 12% over control. Rapamycin nearly abolished Thr 389 staining, whereas calyculin A co-treatment was able to restore about 18% of the control Thr(P) 389 staining. Calyculin A in combination with FTI increased Thr(P) 389 staining about 3-fold over that observed with FTI alone. The partial inhibition of FTIinduced dephoshorylation by calyculin A resulted from effects on both total p70s6k levels and from effects on p70s6k phosphorylation. The high concentrations of FTI required to induce more complete p70s6k dephosphorylation (Ͼ60 M) caused some down-regulation of total p70s6k protein without having a noticeable effect on the levels of Erk1/2 or other proteins. Calyculin A was able to partially reverse the observed p70s6k down-regulation. Out of a total of six experiments, calyculin A in combination with FTI produced consistently greater Thr(P) 389 staining than FTI treatment alone. Calyculin A-induced reversal of FTI effects on endogenous p70s6k resulted from a combination of the reversal of p70s6k down-regulation (Fig. 5, second panel) and a reversal of FTI-induced dephosphorylation (Fig. 4, top panel). Importantly, neither FTI nor rapamycin were able to reverse the marked p70s6k mobility shift induced by calyculin A, presumably indicating that calyculin A has a dominant effect on total levels of p70s6k phosphorylation when combined with FTI or rapamycin treatments.
Calyculin A-induced hyperphosphorylation of p70s6k showed some selectivity. In the six experiments performed, calyculin A had little effect on Erk1/2 phosphorylation in three experiments (as in Fig. 5, third panel)  FTI and rapamycin on p70s6k may be mediated by a calyculin A-sensitive protein phosphatase. These results are supported by recent reports in which calyculin A partially reversed rapamycin-induced p70s6k inhibition (39) and reversed rapamycin and osmotic shock induced p70s6k inactivation and dephosphorylation (40). DISCUSSION FTI potently inhibits the growth of tumors in mouse models (6,(41)(42)(43)(44) and inhibits the growth of tumor cells (45) and nontransformed cell lines in culture (7). Although FTIs were originally designed to inhibit Ras-dependent tumors by blocking the farnesylation of Ras, it is now apparent that other targets are necessary to explain the growth inhibitory properties of FTI. Previously we showed that the p70s6k pathway represents a novel target of FTI and that FTI inhibits p70s6k activity as early as 15 min. Here we show that FTI also rapidly inhibits DNA synthesis, suggesting a possible link between the rapid inhibition of p70s6k and the rapid inhibition of cell proliferation. Because farnesylation is irreversible, the effect of an FTI would only become apparent after a substantial portion of the target farnesylated protein has been turned over. Thus, the rapid effect of FTI on DNA synthesis and p70s6k argue for the existence of an FTI target that is unrelated to protein farnesylation.
Peptidomimetic FTIs are effective in the treatment of cancers in animal models, and several pharmaceutical companies have FTIs in clinical trials (46). Our data indicate that the FTI L-744,832 may act, in part, by inhibiting the p70s6k pathway. The p70s6k pathway is inappropriately up-regulated in tumor cells such as the MiaPaCa-2 cell line (16) and small cell lung cancer cell lines (15). The mTOR/p70s6k pathway is also acti-vated by the Gli oncogene (17) and the E1A viral oncoprotein (47). Thus, it is important to elucidate the precise mechanism of action of FTIs as a class of compounds. Our contention that inhibition of the p70s6k pathway is important in the anticancer properties of FTI is further supported by studies demonstrating that agents that inhibit the mTOR/p70s6k pathway, such as rapamycin (15,16) and asparaginase (28), potently inhibit tumor cell growth in vitro and in vivo, respectively.
To more fully understand the mechanisms of action of FTI, we compared its properties with those of the well known inhibitor of the mTOR/p70s6k pathway, rapamycin. Like rapamycin, FTI inhibits p70s6k in response to serum, PMA, and increased amino acid levels, which activate p70s6k by distinct mechanisms. Another similarity between FTI and rapamycin is that both induce dephosphorylation of Thr 389 in wild type p85s6k but fail to inhibit Thr 389 dephsophorylation of rapamycin-insensitive p70s6k deletion mutants. In the analysis of a p70s6k deletion mutant similar to the one described here, Thr 389 was shown to remain phosphorylated in the presence of rapamycin, although Thr 389 was shown to be dephosphorylated in response to the PI 3-kinase inhibitor wortmannin (48). Dennis et al. hypothesize that the kinase that phosphorylates Thr 389 is still active in the presence of rapamycin, suggesting that rapamycin may activate a phosphatase that dephosphorylates Thr 389 . Recent reports showing that wild type p70s6k binds protein phosphatase 2A (39,49), a rapamycin-insensitive deletion mutant does not bind protein phosphatase 2A and that rapamycin activates a PHAS-1 (phosphorylated heat and acid stable protein 1) phosphatase (39) support this hypothesis. Our results with the ⌬NT/⌬CT mutant and the phosphatase inhibitor calyculin A are consistent with FTI acting through the regulation of a protein phosphatase situated in the mTOR pathway rather than on the stimulatory pathways initiated by serum or PMA (Fig. 6). Other possible explanations for the effect of FTI on p70s6k exist, however. For example, FTI might inhibit an as yet unknown pathway that is required for p70s6k activation.
Although we still do not know precisely how FTI inhibits p70s6k activity, our results presented here and elsewhere (7) demonstrate that FTI mimics the effect of rapamycin in: (a) its ability to induce p70s6k and PHAS-1 dephosphorylation, (b) p70s6k inactivation, (c) inhibition of p70s6k in response to FIG. 5. Calyculin A inhibits p70s6k dephosphorylation in response to FTI and rapamycin. Balb-MK cells were quiesced by treatment with serumless medium for 14 h. Cells were pretreated with FTI for 24 h or with 50 nM calyculin A or 20 nM rapamycin for 1 h. Cells were stimulated as described in the legend to Fig. 2 in the continued presence of FTI, calyculin A, and rapamycin during the stimulation. Cell extracts were analyzed either directly for Erk1/2 or phospho-Erk1/2 levels or subjected to immunoprecipitation with a p70s6k-specific antibody, followed by immunoblot analysis using antibodies specific for p70s6k or Thr(P) 389 p70s6k.
FIG. 6. Hypothetical model for p70s6k activation in keratinocytes and the blockade of p70s6k activation by FTI. PMA and serum signal through protein kinase C and PI 3-kinase-dependent pathways, respectively, to activate a kinase that phosphorylates p70s6k on Thr 389 . Our data are consistent with a model in which FTI inhibits p70s6k by inhibiting the mTOR pathway and may involve a calyculin A-sensitive p70s6k-phosphatase. Solid lines represent well characterized signaling events, whereas dashed lines represent less well characterized events. distinct stimuli, (d) insensitivity of a rapamycin insensitive mutant, (e) partial reversal of p70s6k dephosphorylation by calyculin A treatment, and, most importantly, (f) inhibition of the growth of cells in culture. The elucidation of the exact target and mechanism of FTI-induced growth arrest will allow the development of a new generation of mechanism based anticancer drugs.