The Resorcylic Acid Lactone Hypothemycin Selectively Inhibits the Mitogen-Activated Protein Kinase Kinase-Extracellular Signal-Regulated Kinase Pathway in Cells

168 Biol. Pharm. Bull. 33(2) 168—173 (2010) Vol. 33, No. 2

,a a,b a b c Hidesuke FUKAZAWA,* Yoshimi IKEDA, Mari FUKUYAMA, Takeshi SUZUKI, Hiroshi HORI, d e Toru OKUDA, and Yoshimasa UEHARA a Department of Bioactive Molecules, National Institute of Infectious Diseases; 1–23–1 Toyama, Shinjuku-ku, Tokyo 162– 8640, Japan: b Division of Basic Biological Sciences, Faculty of Pharmacy, Keio University; 1–5–30 Shiba-koen, Minato- ku, Tokyo 105–8512, Japan: c Department of Life Science, Tamagawa University; 6–1–1 Tamagawa-gakuen, Machida, Tokyo 194–8610, Japan: d Mycology and Metabolic Diversity Research Center, Tamagawa University Research Institute; 6–1–1 Tamagawa-gakuen, Machida, Tokyo 194–8610, Japan: and e Department of Microbial Chemical Biology and Drug Discovery, Iwate Medical University School of Pharmacy; 2–1–1 Nishi-tokuda, Yahaba-cho, Shiwa-gun, Iwate 028–3694, Japan. Received August 31, 2009; accepted November 4, 2009; published online November 12, 2009

The resorcylic acid lactone hypothemycin has been shown to inactivate protein kinases by binding to a cysteine conserved in 46 protein kinases, including mitogen-activated protein kinase kinase (MEK), extracellular signal-regulated kinase (ERK) and platelet-derived growth factor receptor (PDGFR). We assessed the selectivity of hypothemycin in cellular contexts. Hypothemycin normalized the morphology and inhibited anchorage-independent growth of Ki-ras transformed normal rat kidney (NRK) cells with selectivity and potency comparable to or greater than that of the MEK inhibitor U0126. In Ki-ras-transformed and phorbol 12-myristate 13-acetate (PMA)-treated NRK cells, hypothemycin blocked ERK activation but showed a minimal effect on autophosphorylation of protein kinase D1 (PKD1), another kinase containing the conserved cysteine. Hypothemycin potently inhibited PDGFR autophosphorylation and activation of the MEK-ERK pathway in platelet-derived growth factor (PDGF)-treated NRK cells. However, the phosphoinositide-3-kinase (PI3K) pathway was only modestly atten- uated. Hypothemycin also inhibited growth factor- and anchorage-independent growth of human cancer cell lines with a constitutively active MEK-ERK pathway. Although hypothemycin has the potential to inactivate various protein kinases, the results indicate that in intracellular environments, hypothemycin can inhibit the MEK- ERK axis with sufﬁcient selectivity to normalize transformed phenotypes of cells dependent on this pathway. Key words hypothemycin; resorcylic acid lactone; protein kinase; signal transduction

Resorcylic acid lactones (RALs) constitute a family of polyketide mycotoxins with a variety of biological activi- ties.1) Representative examples are the heat shock protein 90 inhibitor radicicol and the estrogen agonist zearalenone. Hy- pothemycin (Fig. 1), a RAL containing a cis-enone moiety that initially did not reveal any particularly interesting activity, was later shown to inhibit ras-transformation and T cell activation.2—4) Hypothemycin and other closely related Fig. 1. Structure of Hypothemycin cis-enone RALs gained attention as compounds that irreversibly inhibit certain protein kinases such as mitogen-acti- ras-induced transformation was associated with accelerated vated protein kinase kinase (MEK), extracellular signalf-reg- cyclin D1 degradation,3) and a MEK inhibitor caused a loss ulated kinase (ERK) and TAK1, but not RAF, protein kinase of cyclin D expression in BRAF mutants.11) C (PKC) or protein kinase A (PKA).5—8) From their struc- Here, we compared the effects of hypothemycin and MEK tures and irreversible mode of action, cis-enone RALs were inhibitors on signaling pathways in cells. Our results show predicted to form stable Michael addition products with cys- that even though the RAL target cysteine is not confined to teine residues of protein kinases. Bioinformatics studies re- the components of the MEK-ERK pathway, in cellular set- vealed that a cysteine residue that immediately precedes the tings hypothemycin preferentially blocks the MEK-ERK axis highly conserved Asp-Phe-Gly (DFG) motif is the cis-enone with sufficient selectivity to impair transformed phenotypes RAL binding site.9) This cysteine is conserved in 9% (46/ of cancer cells requiring this pathway. 510) of the protein kinases in the human kinome. Although different kinases contain the RAL target cys- MATERIALS AND METHODS teine, cis-enone RALs have been isolated as selective inhibitors of MEK.7,8) Furthermore, cis-enone RALs share var- Materials Hypothemycin and 7,8-dihydrohypothe- ious biological properties with MEK inhibitors. Sonoda et al. mycin were isolated from culture broth of Hypomyces subicu- identified hypothemycin as a ras-signaling inhibitor.3) We re- losus TAMA 117. U0126 was purchased from Promega (Madi- ported that the MEK inhibitor U0126 reverses ras-mediated son, WI, U.S.A.) and PD98059 was obtained from Cal- transformation.10) Hypothemycin and MEK inhibitors showed biochem (San Diego, CA, U.S.A.). Rat platelet-derived strong activity in cell lines with the activating BRAF V600E growth factor (PDGF)-BB is a product of R&D Systems mutation.11) Sonoda et al. reported that the suppression of (Minneapolis, MN, U.S.A.). Anti-phosphotyrosine antibody

4G10 was purchased from Millipore (Billerica, MA, U.S.A.). per well, cultured for 2 d, and then serum starved for 24 h. Phospho-Akt (Ser473), phospho-PKD/PKCm (Ser 916), phos- Cells were treated with inhibitors for 3 h and stimulated with pho-ERK1/2 (Thr202/Tyr204), phospho-p90RSK (Thr359/ 100 nM PMA for 30 min or 50 ng/ml rat PDGF-BB for 10 min S363), phospho-S6 ribosomal protein (Ser 235/236) and and then processed for immunoblotting as described above. eukaryotic initiation factor 4E antibodies were acquired from To observe multiple signaling pathways downstream of Cell Signaling Technology, Inc. (Danvers, MA, U.S.A.). PDGFR simultaneously, a cocktail was used that contained Measurement of Anchorage- and Growth Factor-Inde- antibodies against phosphorylated forms of the components pendent Growth Anchorage-independent growth was of the PKC pathway (phospho-PKD), the phosphoinositide- measured on poly-2-(hydroxyethyl methacrylate) (poly- 3-kinase (PI3K)-AKT pathway (phospho-AKT, phospho- HEMA)-coated plates as described previously.12) Cells were S6R) and the MEK-ERK pathway (phospho-ERK and phos- inoculated in polyHEMA-coated culture plastic 96-well pho-p90RSK). plates in a volume of 135 ml at a density of 1000 cells per well for Ki-ras transformed normal rat kidney (NRK) cells (Ki-ras/ RESULTS NRK) and 5000 per well for HCT116 and HT29. For measurement of growth in serum-free medium, cells were sus- Hypothemycin Normalizes Growth Properties of Ki- pended in Dulbecco’s Modified Eagle Medium: Nutrient ras-Transformed Rat Fibroblasts We have previously re- Mixture F-12, 1 : 1 supplemented with 1 mg/ml fatty acid- ported that the MEK inhibitor U0126 normalizes growth free bovine serum albumin and 5.5 mg/ml transferrin, and properties of Ki-ras-transformed rat fibroblasts (Ki-ras/ seeded in wells of collagen-coated 96-well plates in a volume NRK).10) Hypothemycin also has been shown to reverse ras- of 135 ml at a density of 5000 cells per well. Compounds dis- mediated transformation.3,4) We compared the effects of hy- solved in 15 ml of medium were added to the wells, and the pothemycin with those of two MEK inhibitors, U0126 and cells were cultured. After 4 d, 15 ml of 3-[4,5-dimethylthia- PD98059; we focused on the morphology and anchorage-in- zol-2-yl]-2,5-diphenyltetrazalium bromide (MTT) solution dependent growth of Ki-ras/NRK cells. Treatment of Ki-ras/ (5 mg/ml in phosphate buffered saline) was added and further NRK with 2 m M hypothemycin caused distinct morphological incubated for 4 h. The resulting MTT formazan was solubi- reversion to a flat, normal appearance (Fig. 2). U0126 and lized with 100 ml of sodium dodecyl sulfate (SDS) solution PD98059 also induced cell flattening, but higher concentra- (20% in 10 mM HCl), and the absorbance was measured after tions were required. The morphology of Ki-ras/NRK cells 24 h at 570 nm and a reference wavelength of 690 nm. did not change when treated with 7,8-dihydrohypothemycin, Immunoblotting Analysis Ki-ras/NRK cells were indicating that the cis-enone system is necessary for mor- grown in 24-well plates to approximately 80% confluency phology-reversing activity. and treated with inhibitors for 24 h. Cells were fixed for 10 We next examined the effects of hypothemycin and MEK min with 10% cold trichloroacetic acid and lysed with 9 M inhibitors on anchorage-independent growth of Ki-ras/NRK urea, 2% Triton X-100 and 2% lithium dodecyl sulfate. Lysates cells using polyHEMA-coated 96-well plates. Hypothemycin were neutralized with 2 M Tris and passed through a syringe suppressed the growth of Ki-ras/NRK cells on the nonadhe- to reduce viscosity. Proteins were subjected to SDS-poly- sive polyHEMA surface at concentrations lower than those acrylamide gel electrophoresis and analyzed by immunoblot- required on normal tissue culture plastic (Fig. 3). U0126 and ting. To observe effects on phorbol 12-myristate 13-acetate PD98059 also inhibited anchorage-independent growth of (PMA)- and PDGF-signaling, NRK cells were seeded in 96- Ki-ras/NRK cells, but at higher concentrations. The results well plates in a volume of 100 ml at a density of 10000 cells indicate that hypothemycin reverses Ki-ras-induced pheno-

Fig. 2. Morphological Reversion of Ki-ras/NRK Cells Induced by MEK Inhibitors Cells were treated for 24 h with the indicated concentrations of inhibitors. HTM, hypothemycin; d-HTM, 7,8-dihydrohypothemycin. 170 Vol. 33, No. 2

Fig. 3. Inhibition of Anchorage-Independent Growth of Ki-ras/NRK Cells by MEK Inhibitors Anchorage-independent growth in polyHEMA or anchorage-dependent growth in uncoated 96-well plates was measured as described under Materials and Methods. Each col- umn represents the meanS.D. of quadruplicate wells. HTM, hypothemycin. types with potency greater than or at least comparable to that of U0126 and PD98059. Hypothemycin Preferentially Inhibits the MEK-ERK Pathway in Rat Fibroblasts cis-Enone RALs including hypothemycin have been reported to irreversibly inhibit a subset of protein kinases such as MEK and TAK1.6,8) The irreversible inhibition was shown to be a result of Michael addition to a cysteine residue that immediately precedes the conserved DFG-motif.9) This cysteine is conserved in 46 of the 510 identiﬁed protein kinases in the human kinome. We investigated whether hypothemycin can discriminate among the possible targets in cellular contexts. Because previous reports raised the possibility that hypothemycin selectively blocked the MEK-ERK pathway, we compared the phosphorylation of ERK, which is mediated by MEK, and autophosphorylation of PKD1 (PKCm), another protein kinase containing the conserved cysteine. Cell lysates were prepared from hypothemycin-treated Ki-ras/NRK cells shown in Fig. 2 and analyzed by immunoblotting. Phosphorylation of ERK was completely inhibited in hypothemycin-treated cells with normalized morphology (Fig. 4a). U0126 and PD98059 also reduced ERK phosphorylation, but 7,8-dihydrohypothemycin had no effect. Schirmer et al. demonstrated 94% inhibition of PKD 9) Fig. 4. Hypothemycin Inhibits Phosphorylation of ERK but Not PKD1 in with 2 m M hypothemycin in an in vitro assay. However, hy- Ki-ras-Transformed or PMA-Stimulated NRK Cells pothemycin at the same concentration did not have any effect (a) Ki-ras/NRK cells shown in Fig. 1 were processed as described under Materials on autophosphorylation of PKD1 in Ki-ras/NRK cells. Like- and Methods and immunoblotted with the indicated antibodies. (b) NRK cells were wise, hypothemycin inhibited PMA-induced ERK phospho- starved of serum for 24 h, treated with the indicated concentrations of hypothemycin for 3 h, and then stimulated with 100 nM PMA for 30 min. Cells were lysed and analyzed by rylation in NRK cells (97% at 4 m M and 81% at 2 m M), with a immunoblotting using the indicated antibodies: HTM, hypothemycin; d-HTM, 7,8-di- modest effect on autophosphorylation of PKD1 (Fig. 4b). hydrohypothemycin; PD, PD98059; p-ERK, phosphorylated ERK; p-PKD1, phospho- The results demonstrate that in cells, hypothemycin does not rylated PKD1. inhibit every protein kinase with the RAL-target cysteine and in vitro inhibition does not always predict inhibition in a cel- contains antibodies against phosphorylated forms of the com- lular environment. ponents of the PKC pathway (PKD), the PI3K pathway (AKT PDGFRa and PDGFRb are two of the seven tyrosine and S6 ribosomal protein), and the MEK-ERK pathway kinases that contain the RAL target cysteine.9) NRK cells (ERK and p90RSK) to examine the effects of hypothemycin were stimulated with PDGF-BB in the presence of various on signaling pathways downstream of PDGFR. concentrations of hypothemycin and the cell lysates were Inhibition of the downstream targets was not completely subjected to immunoblotting using an anti-phosphotyrosine parallel with the reduction in PDGFR autophosphorylation antibody. As expected, hypothemycin inhibited the PDGF-in- (Fig. 5, bottom). Surprisingly, although ligand-stimulated duced tyrosine phosphorylation in NRK cells (Fig. 5, top). phosphorylation of PDGFR appeared to be fully blocked by The cell lysates were also immunoblotted with a cocktail that 10 m M hypothemycin, phosphorylation of S6 ribosomal pro- February 2010 171

Fig. 5. Hypothemycin Preferentially Inhibits the MEK-ERK-RSK Path- way in PDGF-Stimulated NRK Cells NRK cells that were serum starved for 24 h were then treated with the indicated concentrations of hypothemycin for 3 h and stimulated with 50 ng/ml rat PDGF-BB for 10 min. Cells were lysed and analyzed by immunoblotting using anti-phosphotyrosine (top) or a cocktail containing indicated phospho-speciﬁc antibodies (bottom). Arrow- head, position of PDGFR. p-PKD1, phosphorylated PKD1; p-p90RSK, phosphorylated 90 kDa ribosomal S6 kinase; p-AKT, phosphorylated AKT; p-ERK, phosphorylated ERK; p-S6R, phosphorylated S6 ribosomal protein; eIF4E, eukaryotic initiation factor 4E; HTM, hypothemycin. The eIF4E antibody detects total eIF4E to ensure equal load- ing. Positions and sizes (kDa) of markers are shown on the right.

tein only decreased 38%. At 4 m M hypothemycin, PDGFR autophosphorylation was still inhibited about 85%, but the phosphorylation of S6 ribosomal protein was no longer im- paired. Phosphorylation of AKT and PKD1 also displayed hypothemycin resistance to some extent and substantial phosphorylation was observed after 10 m M treatment. On the other hand, components of the MEK-ERK pathway Fig. 6. Hypothemycin (a) and U0126 (b) but Not Paclitaxel (c) Selectively appeared to be more susceptible. The PDGF-stimulated Inhibit the Growth of HT29 and HCT116 Cells in Serum-Free Medium phosphorylation of ERK and p90RSK was inhibited by 8 m M Cells were grown in serum free (closed circles) or serum-supplemented medium (open circles) with inhibitors and quantified using the MTT assay as described under hypothemycin to levels indistinguishable from that of the un- Materials and Methods. Each plot represents the mean of triplicate wells; bars, SD. stimulated control. The phosphorylation of p90RSK was the HTM, hypothemycin. most sensitive, and apparent inhibition was observed at 2 m M hypothemycin. These results demonstrate that hypothemycin concentrations lower than those required in serum-supple- inhibits PDGFR activation, but the inhibitory effect de- mented medium. The IC50s for hypothemycin in serum-free creases as signals are transduced downstream, and signaling medium and serum-supplemented medium were 0.078 m M pathways that diverge from PDGFR are not affected uniformly. and 0.90 m M, respectively, for HT29 cells, and 0.83 m M and Hypothemycin Impairs the Transformed Growth Prop- 2 m M, respectively, for HCT116 cells. The MEK inhibitor erties of Human Cancer Cells with a Constitutively Ac- U0126 also selectively inhibited the growth factor-indepen- tive MEK-ERK Pathway Cancer cell lines with mutations dent growth of HT29 and HCT116 cells (Fig. 6b), but the cy- that activate the MEK-ERK pathway, the BRAF V600E mu- totoxic anti-cancer agents paclitaxel (Fig. 6c) or doxorubicin tation in particular, have been shown to be exceedingly sensi- (not shown) did not display any enhanced toxicity against the tive to hypothemycin.9,11,13) We tested whether hypothemycin two cell lines in serum-free medium. The growth of cancer could normalize the transformed phenotypes of human can- cells without MEK-ERK activating mutations was inhibited cer cell lines that harbor MEK-ERK activating mutations. We at similar concentrations of hypothemycin in the two media. chose two colon cancer cell lines, HT29, a BRAF V600E For example, the IC50s for the MCF7 breast cancer cell line mutant, and HCT116, a KRAS G13D mutant, and examined in serum-free and serum-supplemented medium were 5.3 m M the effects of hypothemycin on their growth factor-indepen- and 5.4 m M, respectively. dence and anchorage-independence, the two major in vitro Hypothemycin also selectively inhibited the anchorage-in- parameters of transformation. If hypothemycin is a selective dependent growth of HT29 and HCT116 cells on polyHEMA inhibitor of the MEK-ERK pathway in cells, it should show over anchorage-dependent growth on plastic (Fig. 7a), with enhanced activity against HT29 and HCT116 cells deprived an efficacy comparable to or greater than that of U0126 (Fig. of growth factors or anchorage. 7b). Hypothemycin did not show preference for polyHEMA As shown in Fig. 6a, hypothemycin inhibited the growth of to tissue culture plastic in cells without a constitutively acti- HT29 and HCT116 cells in serum-free defined medium at vated MEK-ERK pathway, and cytoxic anti-cancer reagents 172 Vol. 33, No. 2

PDGFRs in the screen of a 124 kinase panel.9) However, PKD1 was not inhibited in any of the cellular contexts we tested. These results demonstrate that effects of protein kinase inhibitors in cells cannot always be extrapolated from in vitro assays. Hypothemycin also blocked tyrosine phosphorylation of PDGFR in PDGF-stimulated NRK cells. Somewhat unex- pectedly, phosphorylation levels of the downstream components did not decrease in proportion to PDGFR inhibition. In the presence of 10 m M hypothemycin, phosphorylation of PDGFR declined to less than 1% of the negative control. However, the phosphorylation level of the downstream effec- tor S6 ribosomal protein was above 60%. This indicates that the PI3K pathway can amplify a weak, almost undetectable residual activation signal and attenuate the inhibition observed at the level of the receptor. The MEK-ERK pathway was clearly more sensitive to inhibition by hypothemycin than the PI3K pathway. This is most likely because hypothemycin inhibits not only PDGFR, but also MEK, ERK and p90RSK. In contrast, none of the major components downstream of PDGFR and upstream of S6 ribosomal proteins in the PI3K pathway (PI3K, PDK1, AKT, mTOR and p70S6K) Fig. 7. Hypothemycin (a) and U0126 (b) Selectively Inhibit Anchorage- are targets of RALs. Independent Growth of HT29 and HCT116 Cells Hypothemycin has been shown to potently inhibit the Cells were grown in polyHEMA-coated (closed circles, anchorage-independent growth of cancer cell lines that harbor MEK-ERK pathway growth) or uncoated (open circles, anchorage-dependent growth) tissue culture plated activating mutations.9,11,13) To further confirm the kinase se- with inhibitors and quantified using the MTT assay as described under Materials and Methods. Each plot represents the mean of triplicate wells; bars, SD. HTM, hypothe- lectivity of hypothemycin in cells, we examined whether mycin. hypothemycin can reverse the transformed phenotypes of human colon cancer cell lines dependent on the MEK-ERK did not show any surface selectivity in any of the cells tested pathway. Hypothemycin displayed enhanced activity against (not shown). The results imply that hypothemycin blocks the HT29 (a BRAF V600E mutant) and HCT116 (a KRAS MEK-ERK pathway with selectivity sufficient to impair the G13D mutant) cells deprived of growth factors or anchorage. transformed properties of HT29 and HCT116 cells. We obtained analogous results with another BRAF V600E mutant Colo205 and another Ki-ras mutant DLD1 (not DISCUSSION shown). Our results suggest that in cells, hypothemycin inhibits the MEK-ERK axis with sufficient selectivity to impair RALs containing a cis-enone moiety have been reported to the transformed growth properties of cancer cells requiring inhibit certain protein kinases.2,5—8,14,15) Because the inhibi- this pathway. tion was time dependent and irreversible, these compounds Hypothemycin interferes with the MEK-ERK pathway on were speculated to bind reversibly to targets and then form multiple levels. This is an important feature that likely con- Michael addition products with protein thiols. Schirmer et al. tributes to its selectivity in cells. As shown in Fig. 5, signal- recognized that a cysteine residue adjacent the catalytically ing pathways can be remarkably resilient. Inhibition up- important DFG-motif is the cis-enone RAL binding site.9) stream does not ensure inhibition downstream because even Although this cysteine is contained in 46 human protein ki- near complete inhibition at the level of the receptor may be nases, hypothemycin has been isolated as a potent and selec- restored in the process of signal transduction. Both MEK- tive inhibitor of MEK. Here, we investigated the effects of ERK and PI3K pathways were activated by PDGF in NRK hypothemycin in cells. cells, but the MEK-ERK pathway was more vulnerable to hy- Hypothemycin normalized morphology of Ki-ras/NRK pothemycin. Thus, targeting of more than one component in cells and selectively inhibited their anchorage-independent the same pathway by inhibitors such as hypothemycin or a growth on polyHEMA over anchorage-dependent growth on combination of inhibitors may be an efficient way to shut plastic. We have tested various compounds for growth inhibi- down signaling. tion of Ki-ras-transformed cells on polyHEMA, and only Several cis-enone RALs such as hypothemycin, LL- MEK-inhibitors have shown noteworthy selectivity. It thus Z1640-2, L-783,277, Ro 09-2210, radicicol A, and 4-O- seems reasonable to suppose that hypothemycin mainly tar- demethylhypothemycin have been reported to inhibit protein geted the MEK-ERK pathway in Ki-ras/NRK cells. As ex- kinases.2,5—8,14,15) Whether these compounds differ in selec- pected, hypothemycin potently inhibited ERK phosphoryla- tivity has not been thoroughly investigated. The Ki of hy- tion in Ki-ras/NRK and PMA-stimulated NRK cells, but did pothemycin for reversible binding to protein kinases varied not inhibit the autophosphorylation of PKD1, another kinase as much as 10000-fold.9) It is possible that slight differences containing the target cysteine. Autophosphorylation of PKD1 in structure could cause significant differences in the inhibitory in PDGF-stimulated NRK cells was also resistant to hypothe- spectrum. Further modifications may lead to enhanced selec- mycin. PKD1 was more susceptible to hypothemycin than tivity and broaden the application range of this new type of February 2010 173 kinase inhibitor that binds covalently to kinases. M., Mihara M., Tsuchiya M., Matsumoto K., J. Biol. Chem., 278, 18485—18490 (2003). 7) Williams D. H., Wilkinson S. E., Purton T., Lamont A., Flotow H., Acknowledgments We thank the members of our lab for Murray E. J., Biochemistry, 37, 9579—9585 (1998). helpful discussions. This work was supported by Grants-in- 8) Zhao A., Lee S. H., Mojena M., Jenkins R. G., Patrick D. R., Huber H. Aid for Scientific Research on Priority Areas “Cancer” from E., Goetz M. A., Hensens O. D., Zink D. L., Vilella D., Dombrowski The Ministry of Education, Culture, Sports, Science and A. W., Lingham R. B., Huang L., J. Antibiot. (Tokyo), 52, 1086—1094 Technology, Japan. (1999). 9) Schirmer A., Kennedy J., Murli S., Reid R., Santi D. V., Proc. Natl. Acad. Sci. U.S.A., 103, 4234—4239 (2006). REFERENCES 10) Fukazawa H., Uehara Y., Cancer Res., 60, 2104—2107 (2000). 11) Solit D. B., Garraway L. A., Pratilas C. A., Sawai A., Getz G., Basso 1) Winssinger N., Barluenga S., Chem. Commun. (Camb.), 2007, 22—36 A., Ye Q., Lobo J. M., She Y., Osman I., Golub T. R., Sebolt-Leopold (2007). J., Sellers W. R., Rosen N., Nature (London), 439, 358—362 (2006). 2) Camacho R., Staruch M. J., DaSilva C., Koprak S., Sewell T., Salituro 12) Fukazawa H., Mizuno S., Uehara Y., Anal. Biochem., 228, 83—90 G., Dumont F. J., Immunopharmacology, 44, 255—265 (1999). (1995). 3) Sonoda H., Omi K., Hojo K., Nishida K., Omura S., Sugita K., Life 13) Wee J. L., Sundermann K., Licari P., Galazzo J., J. Nat. Prod., 69, Sci., 65, 381—394 (1999). 1456—1459 (2006). 4) Tanaka H., Nishida K., Sugita K., Yoshioka T., Jpn. J. Cancer Res., 90, 14) Rawlins P., Mander T., Sadeghi R., Hill S., Gammon G., Foxwell B., 1139—1145 (1999). Wrigley S., Moore M., Int. J. Immunopharmacol., 21, 799—814 5) Dakas P. Y., Barluenga S., Totzke F., Zirrgiebel U., Winssinger N., (1999). Angew. Chem. Int. Ed. Engl., 46, 6899—6902 (2007). 15) Jensen B. M., Beaven M. A., Iwaki S., Metcalfe D. D., Gilfillan A. M., 6) Ninomiya-Tsuji J., Kajino T., Ono K., Ohtomo T., Matsumoto M., Shiina J. Pharmacol. Exp. Ther., 324, 128—138 (2008).