Salicylate-induced growth arrest is associated with inhibition of p70s6k and down-regulation of c-myc, cyclin D1, cyclin A, and proliferating cell nuclear antigen.

Salicylate and its pro-drug form aspirin are widely used medicinally for their analgesic and anti-inflammatory properties, and more recently for their ability to protect against colon cancer and cardiovascular disease. Despite the wide use of salicylate, the mechanisms underlying its biological activities are largely unknown. Recent reports suggest that salicylate may produce some of its effects by modulating the activities of protein kinases. Since we have previously shown that the farnesyltransferase inhibitor l-744, 832 inhibits cell proliferation and p70(s6k) activity, and salicylate inhibits cell proliferation, we examined whether salicylate affects p70(s6k) activity. We find that salicylate potently inhibits p70(s6k) activation and phosphorylation in a p38 MAPK-independent manner. Interestingly, low salicylate concentrations (</=250 microm) inhibit p70(s6k) activation by phorbol myristate acetate, while higher salicylate concentrations (>/=5 mm) are required to block p70(s6k) activation by epidermal growth factor + insulin-like growth factor-1. These data suggest that salicylate may selectively inhibit p70(s6k) activation in response to specific stimuli. Inhibition of p70(s6k) by salicylate occurs within 5 min, is independent of the phosphatidylinositol 3-kinase pathway, and is associated with dephosphorylation of p70(s6k) on its major rapamycin-sensitive site, Thr(389). A rapamycin-resistant mutant of p70(s6k) is resistant to salicylate-induced Thr(389) dephosphorylation.

Salicylate has been used since ancient times for its analgesic and anti-inflammatory properties. More recently, aspirin has received attention because of its protective effects against colon cancer (1,2) and cardiovascular disease. Aspirin is a pro-drug form of salicylate that is rapidly hydrolyzed to salicylate in vivo (3). Aspirin is known to act by directly inhibiting cyclooxygenases 1 and 2 (COX1 and COX2), 1 thereby blocking the produc-tion of prostaglandins. Salicylate however, inhibits the synthesis of prostaglandins in vivo, but has little effect on COX1 and COX2 activities in vitro (3). Salicylate must therefore inhibit COX1 and COX2 activities in vivo through an alternate mechanism not involving direct effects on COX1 and COX2. It has recently been reported that salicylate potently inhibits transcription of the COX2 gene (4), although how salicylate abrogates COX2 transcription is unknown.
It is also becoming apparent that many of the biochemical effects of salicylate are independent of effects on cyclooxygenase activity. The increasing number of protein kinases reported to be modulated by salicylate might provide a potential explanation for the ability of salicylate to regulate COX2 transcription, as well as the cyclooxygenase-independent effects of salicylate. Previous studies in our laboratory demonstrating that the p38 MAPK inhibitor SB203580 potentiates PMA-induced p70 s6k activation (5), along with studies demonstrating that salicylate activates p38 MAPK (6) suggest that salicylate might regulate p70 s6k activity through a p38 MAPK-dependent mechanism.
Although there have been many recent discoveries involving the mechanisms of p70 s6k activation, the kinases that phosphorylate a number of the key regulatory sites of p70 s6k are unknown. Also unknown are the identities of the p70 s6k substrates responsible for mediating its biological effects. p70 s6k was among the first mitogen-activated protein kinases identified (7,8), and has subsequently been shown to be important for G 1 cell cycle progression (9 -11). Although the regulation of p70 s6k activity is complex, it is clear that the mammalian target of rapamycin, mTOR, plays a key role in p70 s6k activation. The immunosuppressant drug rapamycin acts by first binding FKBP12, and then forming a ternary complex with mTOR, resulting in mTOR inactivation. Through mechanisms not completely understood, mTOR regulates the phosphorylation of multiple sites on p70 s6k . Thr 389 was reported to be the major rapamycin-sensitive p70 s6k phosphorylation site because it is dephosphorylated with the most rapid kinetics, is required for p70 s6k activity, and a mutant in which Thr 389 is mutated to Glu displays partial rapamycin resistance (12).
Here we show that salicylate inhibits p70 s6k activity and phosphorylation in a p38 MAPK-independent manner. Salicy-late inhibits PMA-induced p70 s6k activation at low concentrations (Ն250 M), and both PMA and growth factor-induced p70 s6k activation at higher concentrations (Ն5 mM). Salicylateinduced p70 s6k dephosphorylation at Thr 389 is not mediated through effects on the PI 3-kinase pathway. Finally, we demonstrate that salicylate and FTI induce the same effects on the levels of cell growth-associated proteins as the mTOR inhibitor rapamycin, specifically, down-regulation of c-Myc, cyclin D1, cyclin A, and PCNA protein levels.

EXPERIMENTAL PROCEDURES
Cell Culture, [ 3 H]Thymidine Incorporation Assays, and [ 35 S]Methionine/Cysteine Incorporation Assays-Balb/MK cells were maintained as described previously (13). HEK 293 cells were obtained from the American Type Culture Collection and maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. Cells were treated as indicated in the figure legends with 1 M PMA (Sigma), 20 M SB203580 (Alexis), or 10 M LY294002 (Sigma) dissolved in dimethyl sulfoxide. Sodium salicylate (Sigma) was prepared as a 3 M stock solution in water and diluted into medium immediately before use. One molar stock solutions of acetaminophen (Sigma) and indomethicin (Sigma) were prepared in ethanol and diluted into medium immediately before use.
For [ 3 H]thymidine incorporation assays, cells were plated at 20,000 cells/well in 24-well plates and incubated overnight. Cells were then pretreated as indicated in the figures and pulsed for 1 h with 20 Ci/well of [ 3 H]thymidine (PerkinElmer Life Sciences), and the results were analyzed as described previously (13). [ 35 S]methionine/cysteine incorporation was performed as with the [ 3 H]thymidine incorporation assays, except that the cells were pulsed with 20 Ci/well Tran 35 S-label (ICN) and the cell lysates were prepared in 0.1 N NaOH containing 0.1% sodium dodecyl sulfate.
Preparation of Cell Extracts-Cell lysates were prepared as described previously (13) except in Fig. 6A. Since in Fig. 6A some of the proteins analyzed were nuclear, the extraction buffer was supplemented with 0.1% sodium dodecyl sulfate, and the lysates were sonicated prior to centrifugation. The protein concentrations of the extracts were quantitated using the Bradford assay (Bio-Rad), with bovine serum albumin as the standard.
cDNAs and Transient Transfections-A plasmid encoding HA-tagged p85 s6k was kindly provided by Joseph Avruch (Harvard Medical School, Cambridge, MA). The flag-⌬NT/⌬CT construct was described previously (5). Flag-tagged wild type p70 s6k was amplified from the plasmid encoding HA-p85 s6k using PCR with the primer sets: 5Ј-TTTTGGATCCA-TGGACTATAAGGACGATGATGACAAAGCAGGAGTGTTTGACATA-G-3Ј and 5Ј-TTTTGAATTCTCATAGATTCATACGCAGGTG-3Ј. Underlined portions represent regions complementary to the plasmid template. PCR was performed using Pfu polymerase (Promega) according to the manufacturer's instructions. The PCR product was digested with EcoRI and BamHI and subcloned into the pcDNA3.1 expression vector (Invitrogen). The insert was verified to be correct by DNA sequencing. For transfections, HEK 293 cells were seeded at 800,000 cells/100-mm dish and incubated overnight. The cells were transfected with 6 g/plate of either the flag-p70 s6k or flag-⌬NT/⌬CT constructs using LipofectAMINE (Life Technologies, Inc.) according to the manufacturer's instructions. Cell extracts were prepared as described above.

RESULTS
We previously observed that the p38 MAPK inhibitor SB203580 is able to potentiate the activation of p70 s6k by PMA (5). These results are similar to those reported by others (14,15) in which SB203580 activates MAPK by blocking a negative regulatory pathway mediated by p38 MAPK. Since sodium salicylate was shown to activate p38 MAPK (6), we hypothesized that salicylate might inhibit p70 s6k by activating a negative regulatory pathway involving p38 MAPK. As expected, salicylate treatment of Balb/MK cells potently abrogated PMAinduced p70 s6k activation (Fig. 1). Inhibition of p70 s6k by salicylate was associated with a shift in electrophoretic mobility and decreased phospho-Thr 389 staining, consistent with salicylateinduced p70 s6k dephosphorylation.
The inhibition of p70 s6k by salicylate, however, was independent of effects on p38 MAPK activity because SB203580 failed to block the inhibitory effect of salicylate on p70 s6k activation even though SB203580 was still able to induce an increase in phospho-Erk1/2 staining in the presence of salicylate (Fig. 1). The results indicate that, under these experimental conditions, SB203580 blocked the effect of p38 MAPK on Erk1/2 phosphorylation. In addition, salicylate treatment did not induce a significant increase in the phosphorylation of p38 MAPK on sites required for activity. Together, these results demonstrate that salicylate inhibits p70 s6k in a p38 MAPKindependent manner.
Because salicylate is an aspirin metabolite and its ability to inhibit p70 s6k may play a role in the physiological responses elicited by aspirin and other salicylates, the ability of salicylate to inhibit p70 s6k and cell proliferation was explored. To determine whether aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) and analgesic drugs affect p70 s6k activity, Balb/MK cells were treated with various NSAIDs and cell extracts were prepared and analyzed for p70 s6k activity, p70 s6k phosphorylation, and Erk1/2 phosphorylation (Fig. 2). Salicylate and aspirin both inhibited p70 s6k activity while acetaminophen and indomethacin had little effect. As with salicylate, the inhibition of p70 s6k by aspirin was associated with a shift in p70 s6k electrophoretic mobility and a decrease in the phospho-FIG. 1. Salicylate inhibits p70 s6k in a p38 MAPK-independent manner. Balb/MK cells were plated at 800,000 cells/100-mm dish and incubated overnight. The cells were pretreated with serum-free medium with or without 20 mM sodium salicylate for 24 h, pretreated with SB203580 for 2 h, and stimulated for 30 min with 1 M PMA in serum-free medium with the continued presence of the inhibitors. Cell extracts were prepared and assayed for p70 s6k activity (top panel) and analyzed by immunoblotting (lower panels) using antibodies specific for p70 s6k (p70 s6k IB), p70 s6k phosphorylated on Thr 389 (P-T389 IB), phosphorylated, active MAPK (P-Erk1/2 IB), or phosphorylated, active p38 MAPK (P-p38 IB) as described under "Experimental Procedures." Results are presented as percentage of control kinase activity, with the control normalized to 100%. rylation of p70 s6k on Thr 389 . The phosphorylation of Thr 389 is required for p70 s6k catalytic activity (12). Decreased phosphorylation of Thr 389 suggests that salicylate inhibits kinases upstream of p70 s6k , resulting in p70 s6k inhibition.
To determine the salicylate concentration range required to inhibit p70 s6k , Balb/MK cells were stimulated with PMA or EGF ϩ IGF-1 in the presence of increasing concentrations of salicylate (Fig. 3). Both treatments stimulated p70 s6k activity to similar levels in the absence of salicylate. p70 s6k was inhibited by almost 40% in the presence of 250 M salicylate when PMA was used as the stimulus (Fig. 3A). For comparison, administration of a single analgesic antipyretic dose of salicylate to patients yields a serum salicylate concentration of about 400 M (3). Near complete inhibition of PMA-stimulated p70 s6k activity, however, required 5 mM salicylate. Immunoblots performed using antibodies to p70 s6k (second panel) or phospho-Thr 389 p70 s6k (third panel) demonstrated that inhibition is associated with p70 s6k dephosphorylation. Dephosphorylation was apparent for both the p70 (upper bands) and p85 (lower bands) alternative translational products of the S6 kinase gene.
Interestingly, activation of p70 s6k by EGF and IGF-1 was only inhibited by higher (Ն5 mM) concentrations of salicylate (Fig. 3B). The results of the kinase assays in Fig. 3 (A and B) were normalized to control and plotted in Fig. 3C. These data suggest that salicylate may act on two distinct targets during p70 s6k activation. One putative target would be sensitive only to high salicylate concentrations and would be required for activation of p70 s6k by both PMA and EGF ϩ IGF. The other putative target would be sensitive to lower concentrations of salicylate and would be required only for activation of p70 s6k by PMA. Thus, at low concentrations (Յ1 mM), salicylate mimics the down-regulation of PKCs by chronic PMA treatment, which specifically blocks p70 s6k activation in response to PMA treatment (5). At high concentrations (Ն5 mM), salicylate acts in a manner similar to rapamycin by inhibiting p70 s6k activity irrespective of the stimulus. The millimolar concentrations of salicylate required to inhibit p70 s6k activation regardless of the stimulus are similar to those reported by others to induce effects on p38 MAPK activity (6), IB phosphorylation (6, 16), and cellular growth (17).
To determine whether salicylate acts rapidly to inhibit p70 s6k , consistent with a relatively direct effect on the p70 s6k signaling pathway, time course experiments were performed. Balb/MK cells were treated with 20 mM salicylate in normal growth medium (Fig. 4A). At 5 min, salicylate induced a 53% inhibition of p70 s6k activity, indicating a rapid effect of salicylate on p70 s6k . p70 s6k inhibition reached a plateau at 20 min and was associated with a shift in p70 s6k electrophoretic mobility and a loss of phospho-Thr 389 immunostaining. The observation that salicylate rapidly inhibited p70 s6k activity suggests p70 s6k inhibition was independent of effects on COX2, because COX2 regulation at the transcriptional level (4) could not ex-  Fig.  1 and incubated overnight. The cells were treated for 1 h with the indicated inhibitors or vehicle controls diluted in normal growth medium, and cell extracts were prepared. Cell extracts were analyzed as described in Fig. 1 by kinase assay (top panel) or by immunoblotting (lower panels) with antibodies specific for p70 s6k (p70 s6k IB), phospho-Thr 389 p70 s6k (P-T389 IB), or phosphorylated, active Erk1/2 (P-Erk1/2). plain the rapid kinetics of p70 s6k inactivation. Under the same conditions, salicylate also induced dephosphorylation of Erk1/2. The effect of salicylate on Erk1/2 phosphorylation varied from experiment to experiment and was only observed at high concentrations of salicylate (Ն20 mM) and was therefore not investigated further.
Because of our past observations that FTI rapidly inhibited both p70 s6k activity and cell proliferation, we examined whether salicylate could rapidly inhibit DNA synthesis by performing [ 3 H]thymidine assays (Fig. 4B). Although salicylate had a rapid effect on DNA synthesis, the magnitude of the effect was small at early time points, and gradually increased to approximately 50% inhibition of [ 3 H]thymidine incorporation after a 2 h salicylate pretreatment. One of the earliest observations that may relate to the ability of salicylate to affect cell proliferation involves the ability of salicylate to inhibit protein synthesis without inhibiting the uptake of amino acids (16).
In order to determine whether the rapid inhibition of p70 s6k by salicylate correlates with a rapid inhibition of protein synthesis, the rate of protein synthesis was determined after various salicylate pretreatment intervals by monitoring the incorporation of 35 S-labeled methionine and cysteine into cellular protein (Fig. 4C). Salicylate induced a greater than 50% inhibition of protein synthesis after a 5-min pretreatment, followed by a 1 h pulse with [ 35 S]Met/Cys in the continued presence of salicylate. This inhibition was maximal because no further inhibition occurred after a 2-h pretreatment. Since protein synthesis is required throughout the cell cycle in order for cells to proliferate (17), the inhibition of protein synthesis by salicylate may play a role in salicylate-induced growth arrest.
In order to gain some insight into how salicylate might inhibit p70 s6k , we made use of a rapamycin-resistant deletion mutant of p70 s6k (⌬NT/⌬CT). We have previously shown that wild-type p85 s6k becomes dephosphorylated on Thr 389 in response to treatment with FTI, rapamycin, and the PI 3-kinase inhibitor LY294002. The Thr 389 site of ⌬NT/⌬CT, however, is unaffected by FTI or rapamycin treatment, and is only dephosphorylated in response to LY294002 treatment (5). To determine the effect of salicylate on Thr 389 phosphorylation of both p70 s6k and ⌬NT/⌬CT, HEK 293 cells were transfected with either construct and left untreated or treated with 20 mM salicylate, 5 nM rapamycin, or 10 mM LY294002 (Fig. 5). Analysis of the resulting cell extracts by immunoblotting with antibodies specific for phospho-Thr 389 indicated that salicylate, rapamycin, and LY294002 all potently inhibited Thr 389 phosphorylation of p70 s6k . Salicylate induced only minor dephosphorylation of ⌬NT/⌬CT on Thr 389 , while LY294002 potently FIG. 4. Salicylate rapidly inhibits p70 s6k activity and phosphorylation, DNA synthesis, and general protein synthesis. A, Balb/MK cells were plated at 800,000 cells/100-mm plate and incubated overnight. The cells were then treated with 20 mM salicylate in normal growth medium for the indicated periods and cell extracts prepared. Cell extracts were analyzed as described in Fig. 1 by kinase assay (top panel) or by immunoblotting (lower panels) with antibodies specific for p70 s6k (p70 s6k IB), phospho-Thr 389 p70 s6k (P-T389 IB), or phosphorylated, active Erk1/2 (P-Erk1/2). inhibited ⌬NT/⌬CT phosphorylation on Thr 389 . As expected, rapamycin had no effect on ⌬NT/⌬CT phosphorylation at Thr 389 . Immunoblotting with antibodies specific for the flag epitope tag present on both proteins indicated that the expression levels of the constructs were similar in each of the four treatments. These results suggest that salicylate did not primarily inhibit Thr 389 phosphorylation of p70 s6k by a mechanism involving the PI 3-kinase pathway. Similar results were also obtained in COS 7 cell transfections, and the splice variant p85 s6k behaved identically to p70 s6k with respect to salicylateinduced inhibition and dephosphorylation (data not shown). To further demonstrate that salicylate does not act through the PI 3-kinase pathway, we examined the effect of salicylate on the phosphorylation of Akt on Thr 308 using a phosphospecific antibody. Phosphorylation of Akt on Thr 308 is catalyzed by the kinase PDK1 through a PI 3-kinase-dependent mechanism (18) and thus serves as a marker for in vivo PDK1 activity. Salicylate treatment had no effect on Thr 308 phosphorylation, while LY294002 caused a marked reduction in Thr 308 phosphorylation. Moreover, salicylate abrogated PMA-induced activation of p70 s6k (Figs. 1 and 3), which occurs through a PI 3-kinaseindependent pathway (5,19). Together these results suggest that salicylate acts like rapamycin and FTI rather than PI 3-kinase inhibitors to block p70 s6k activity.
There is an increasing body of evidence implicating the mTOR signaling pathway (20,21), and more specifically p70 s6k (9 -11), in regulating cell proliferation. Little is currently known regarding the identity of the downstream targets of p70 s6k involved in cell cycle regulation or how these targets regulate cell cycle events. In order to more fully characterize the effects of salicylate, rapamycin, and FTI on the cell cycle, rapidly growing Balb/MK cells were treated for 24 or 48 h with each agent and cell extracts were prepared. Analysis of the extracts was performed with antibodies specific for c-Myc, cyclin D1, PCNA, or cyclin A (Fig. 6A). The levels of these proteins were previously reported to be regulated by either rapamycin (22)(23)(24)(25)(26)(27)(28) or salicylate (29), and are known to play a role in cell proliferation. The results revealed that the most rapid and dramatic effect of the three agents was c-Myc down-regulation. c-Myc down-regulation at 24 h correlated closely with Thr 389 dephosphorylation of p70 s6k , suggesting that the mTOR/p70 s6k pathway may play a role in regulating c-Myc levels in Balb/MK cells as was reported in Epstein-Barr virus immortalized B-cell lines (22). At 48 h, rapamycin, salicylate, and FTI treatment continued to suppress c-Myc levels and also induced a downregulation in the levels of cyclin D1, PCNA, and cyclin A. The two c-Myc bands likely represent the 67-and 64-kDa isoforms, termed c-Myc1 and c-Myc2, respectively (30). The three cyclin D1 bands observed may result from alternative splicing (31) and/or phosphorylation. Although rapamycin up-regulates p27 in T cells (32), neither rapamycin, salicylate, nor FTI up-regulated p27 in Balb/MK cells. The decrease in the levels of c-Myc, cyclin D1, PCNA, and cyclin A was highly reproducible and not due to unequal protein loading because equal amounts of total protein were loaded into each lane based on protein assays, and based on immunoblotting with a ␤-tubulin antibody as a loading control. The ability of salicylate and FTI to induce the same decreases in the levels of c-Myc, cyclin D1, PCNA, and cyclin A as rapamycin may indicate a possible role for the mTOR/p70 s6k signaling pathway in the growth inhibitory actions of these drugs. Importantly, the changes observed in the levels of proliferation-associated proteins correlated well with the inhibition of DNA synthesis (Fig. 6B). Rapamycin inhibited DNA synthesis by approximately 50% and 75% at 24 and 48 h, respectively, while high concentrations of salicylate and FTI were able to inhibit DNA synthesis to a greater extent. The increased inhibition induced by high salicylate and FTI concentrations may result from the inhibition of pathways in addition to the mTOR/p70 s6k pathway. FTI inhibits the farnesylation of a number of proteins, so its effects on cell proliferation represents an integration of the effects of FTI on all farnesylated proteins. Similarly, salicylate inhibits NF-B activity (33) as well as affecting the activity of several different protein kinases (6,34,35). Indeed, we have observed inhibition of NF-B DNA binding activity in Balb/MK cells at salicylate concentrations similar to those required to inhibit p70 s6k (data not shown). Thus, salicylate, like FTI, probably inhibits cell proliferation through actions on multiple signaling pathways. Since high concentrations of salicylate (Ն5 mM) are required to FIG. 6. Salicylate, rapamycin, and FTI induce similar changes in the levels of proliferation-associated proteins. A, Balb/MK cells were plated at 700,000 cells/100-mm dish and incubated 24 h. Cells were treated with inhibitors at the indicated concentrations in normal growth medium for either 24 or 48 h, and cell extracts were prepared. Cell extracts were analyzed as described in Fig. 1 by immunoblotting with antibodies specific for phospho-Thr 389 p70 s6k (P-T389 IB), c-Myc, cyclin D1, PCNA, cyclin A, p27, or ␤-tubulin. B, Balb/MK cells were plated at 20,000 cells/well in 24-well plates and incubated 24 h. Cells were incubated with the same treatments used in A and incubated for either 24 or 48 h and pulsed for 1 h with [ 3 H]thymidine. Results were analyzed and presented as described in Fig. 4B. inhibit p70 s6k activation in response to serum (Fig. 2) or EGF ϩ IGF-1 (Fig. 3), it is unclear whether salicylate-induced inhibition of p70 s6k activity and cell proliferation is physiological because serum salicylate concentrations of ϳ2.5 mM are the highest achieved clinically (3). Chronic salicylate treatment may allow inhibition of p70 s6k at lower salicylate concentrations (Fig. 6A). Further studies will be necessary to determine whether salicylate-induced p70 s6k inhibition and growth arrest occur at physiologically relevant concentrations of salicylate. DISCUSSION We report the first evidence that the mTOR/p70 s6k pathway may play a role in mediating some of the biological effects of salicylate. The effect of salicylate on p70 s6k is apparent within 5 min of treatment. p70 s6k inhibition is not general to all NSAIDs and analgesic antipyretic drugs because acetaminophen and indomethacin have no effect on p70 s6k activity.
Mechanistically, low concentrations of salicylate (Յ1 mM) act similarly to the down-regulation of PKCs by inhibiting p70 s6k activation by PMA, but not by EGF ϩ IGF-1. High concentrations of salicylate (Ն5 mM), however, appear to act similarly to rapamycin and FTI to inhibit p70 s6k irrespective of the stimulus used. The ability of salicylate to preferentially inhibit PMAinduced p70 s6k activation at low concentrations, but to inhibit p70 s6k activation regardless of the stimulus at higher concentrations, suggests that salicylate may prove to be a useful tool to help identify and distinguish different upstream activators of p70 s6k .
Salicylate does not act on the PI 3-kinase pathway to inhibit p70 s6k because: (a) salicylate inhibits p70 s6k activation by PMA, which occurs independently of the PI 3-kinase pathway (19); (b) salicylate does not inhibit the phosphorylation of Akt on Thr 308 , an event catalyzed by PDK1 in a PI 3-kinase-dependent mechanism (18); (c) salicylate induces only a weak dephosphorylation of ⌬NT/⌬CT at Thr 389 , while the PI 3-kinase inhibitor LY294002 induces potent dephosphorylation of ⌬NT/ ⌬CT at Thr 389 . We (5) and others (36,37) have previously argued that rapamycin acts through a protein phosphatase to induce p70 s6k dephosphorylation, based on the observation that rapamycin-resistant deletion mutants of p70 s6k remain phosphorylated on Thr 389 in the presence of rapamycin. In addition, the phosphatase inhibitor calyculin A can partially reverse the effects of rapamycin on p70 s6k (38,39). One interpretation of these results is that the kinase that normally phosphorylates Thr 389 is still active in the presence of rapamycin, but the phosphatase that normally dephosphorylates Thr 389 does not dephosphorylate the truncation mutants. Recent reports demonstrating that both mTOR (40) and PDK1 (41) are able to phosphorylate Thr 389 , however, raise an alternate possibility. mTOR might normally phosphorylate Thr 389 , rendering the site rapamycin-sensitive. Thr 389 of the rapamycin-resistant deletion mutants, however, may be a better substrate for PDK1 than for mTOR, rendering the mutants rapamycin-insensitive, but still sensitive to PI 3-kinase inhibitors. This model is consistent with our results involving FTI (5, 13) and salicylateinduced inhibition of p70 s6k . The idea that salicylate (1, 2), FTI (42,43), and rapamycin analogs (23, 44) may be useful anticancer drugs, and the observation that these agents all inhibit p70 s6k , might indicate a common role for p70 s6k in some of the actions of these drugs.
The observation that salicylate and FTI mimic rapamycin in their ability to induce growth arrest and c-Myc, cyclin D1, cyclin A, and PCNA down-regulation suggest the possibility that inhibition of the mTOR/p70 s6k pathway may play an important role in the cytostatic effects of salicylate and FTI. These results are consistent with those of others showing that rapamycin can induce down-regulation of c-Myc (22), cyclin D1 (23,24), cyclin A (27,28), and PCNA (25,26), and that induction of growth arrest by salicylate is associated with cyclin D1 down-regulation (29). Although salicylate can inhibit p70 s6k phosphorylation, inhibit cell proliferation, and induce downregulation of c-Myc, cyclin D1, PCNA, and cyclin A at relatively high concentrations (Ͼ2 mM), it is unclear whether clinically relevant concentrations of salicylate (Յ2.5 mM) are able to produce these effects in vivo. It may be possible, however, to design salicylate analogs that would more potently and specifically inhibit p70 s6k activity in vivo.
Interestingly, the maximal growth inhibition induced by salicylate exceeds that observed in response to rapamycin, as we observed previously with FTI (13). This increased inhibition probably results from the ability of salicylate to inhibit other targets besides p70 s6k . Salicylate, for example, inhibits NF-B (33), which is involved in regulating cell proliferation and is implicated in tumorigenesis (45). Salicylate may inactivate NF-B by inhibiting the protein kinases IB kinase-␤ (34) and RSK2 (35). Salicylate inhibition of IB kinase-␤ was shown to result from competition for ATP binding (34). Competitive inhibition of ATP binding to IB kinase-␤ by salicylate provides a potential explanation for the ability of salicylate to inhibit the activity of multiple protein kinases. Although the ability of FTI and salicylate to inhibit multiple signaling pathways makes it more difficult to decipher which pathways are mediating the effects observed, the ability of signal transduction inhibitors to simultaneously inhibit multiple signaling pathways may be important for their ability to inhibit tumorigenic cell growth. The mitogen-activated protein kinase/extracellular signal-regulated kinase kinase inhibitor U0126 for example, which inhibits both the MAPK and p70 s6k pathways, blocks the anchorage-independent growth of Ki-Ras-transformed fibroblasts (46). The mitogen-activated protein kinase/extracellular signal-regulated kinase kinase inhibitor PD98059, however, which also inhibits the MAPK pathway, but only weakly inhibits the p70 s6k pathway, must be combined with rapamycin to block the anchorage-independent growth of Ki-Ras-transformed fibroblasts.
Together, these observations represent the first report that p70 s6k is a potential target of salicylates. Further studies will be necessary to determine the role of salicylate inhibition of p70 s6k in the many biological effects of salicylate, and whether clinically relevant concentrations of salicylate are able to inhibit p70 s6k activity in vivo.