K252a, KT5720, KT5926, and U98017support Paditaxel (Taxol

[CANCERRESEARCH54,5889-5894,November15,1994] K252a,KT5720, KT5926, and U98017Support Paditaxel (Taxol)-dependentCells and Synergize with Paclitaxel

Irene Abraham,' Cindy L. Wo1f@KathleenE. Sampson,AliceL. Laborde,JohnA. Shelly,Paul A. Aristoff, and Harvey I. Skulnick

Cell Biology (I. A., C. L W., K. E. S.], Chemical and Biological Screening (A. L L, J. A. S.], and Medicinal Chemistry (P. A. A., H. I. 5.1, The Upjohn Company, Kalamazoo, Michigan 49@%fl

ABSTRACT The pacitaxel-dependent CHO@mutantcell line Tax 2-4 (7) is com pletely dependent on pacitaxel for survival and growth. We have used We have used padlitaxel-dependent Tax 2-4 cells to screen for corn this cell line to select compounds thatmay act functionally like pacitaxeL pounds that have paclitaxel-like functional activity. The indoiocarba Tax 2-4 cells have an unstable microtubule network due to a mutation in zole serine/threonine kinase Inhibitor K252a and analogues such as @3-tubulin.Themicrotubules containing mutant tubulin are thought to be KT5926, KT5720, and K252b partially support the growth of the stabilized in the presence of pacitaxel and thus can progress normally padiltaxel-dependent cells In the absence of paclitaxel. A novel kinase Inhibitor of similar structure, U98017, supports the growth of the through the cell cycle and mitosis only ifpacitaxel is in the medium. The dependent cells to 48% of that seen with paclitaxei. Used in cornblna cells die in the absence of pacitaxel. We have presumed that molecules don with padiltaxel, these compounds reduce the amount of paclitaxel that can replace paditaxel and allow for the growth of the Tax 2-4 cells required for maximum growth of the dependent cells. Isobologram may do so through a mechanism that causes a functional stabilization of analysis demonstrates that these compounds also act synergistically the microtubules. This mechanism may be the same as that of paclitaxel with padiltaxel to promote toxicity In wild-type Chinese hamster ovary or it may be through a more indirect pathway. cells. These selected Indolocarbazoles may act at sites distinct from that We find that the indolocarbazole K252a, isolated from Nocardio of padiltaxel and may specifically inhibit kinases that contribute to the psis sp. (10), and the synthetic derivatives KT5720 and KT5926 (ii) destabilization of microtubules. Other indolocarbazoles such as stau all support moderate growth of Tax 2-4 cells, up to 26% of that seen rosporine and rebeccamycln do not support padlitaxel-dependent cell with paclitaxel. All of these compounds are inhibitors of serine/ growth. Structurally unrelated serine/threonine kinase Inhibitors such threonine protein kinases. In addition, a new analogue, U980i7, as H-9 and H-7 or tyrosine kinase Inhibitors such as lavendustin do not allows improved growth of the Tax 2-4 cells to 48% of that with support the growth of these cells. These results define a screen for fUnctional padlitaxel analogues and suggest that It may be useful to paclitaxel. Staurosporine, a serine/threonine kinase inhibitor with Investigate the possible synergy of selected indolocarbazoles and specificity toward PKC and a structure similar to that of K252a, does padlltaiel in vivo. not support substantial growth of the Tax 2-4 cells. Combination studies show that the addition of either of the active indolocarbazoles KT5720 or KT5926 to paclitaxel allows close to the maximum sur INTRODUCTION vival rate of the Tax 2-4 cells in 5-fold lower concentration of paciitaxel. In addition, the indolocarbazole compounds that function Pacitaxel is a natural product isolated from the bark of the ally substitute for paclitaxel also synergize with paclitaxel to increase Pacific yew Taxus brevifolia; it was identified as an antitumor toxicity to wild-type CHO AUXB1 cells. These results implicate agent in 1971 by Wani et al. (1). It has been very effective in indolocarbazoles as agents that further the stabilization of microtu cancer chemotherapy against refractory ovarian (2) and metastatic bules in intact cells, possibly through kinase inhibition. In addition, breast cancer (3). In addition, there is reason to hope that paclitaxel this suggests that combined drug therapy might decrease the amount may be an effective chemotherapeutic agent for other types of of paclitaxel required to achieve maximum response in chemotherapy. cancers (4, 5). The efficacy of paclitaxel may lie in its unique mode of action; it has been shown to stabilize microtubules in vivo, and MATERIALS AND METHODS it has been presumed to prevent cell division by inhibiting the normal functions of cellular microtubules and mitotic spindles, Chemicals. Paclitaxei(125973 L/24) and rebeccamycin (NSC 359079 and resulting in cell death (6). In studies with cells in vitro, paclitaxel C38383-W159) were obtainedfrom The NationalCancerInstitute.H-7 and appears to antagonize agents such as Colcemid that cause micro H-9 were purchasedfrom Seikagaku America, Inc. (St. Petersburg,FL). tubule destabilization (7). The development of paclitaxel, however, Lavendustin was purchased from GIBCO BRL. K252a, K252b, KT5923, has faced many obstacles. The poor solubility of paclitaxel has KT5823, and KT5720 were purchased from Kamiya Biomedical Company required that it be administered in vehicles such as Cremophor EL @ThousandOaks, CA). K252a was also isolated and purified from Nocardio (jolyethylated castor oil), which has led to a high incidence of Psis sp. as described (10) and was a gift from V. Marshall, H. Waley, and V. Wiley (The Upjohn Company). U98017 was prepared by adding 600 mg of hypersensitivity reactions. It is not clear whether these reactions anhydrous potassium cyanide to a solution of 600 mg of K252a (1.28 mmol) are caused by the vehicle or the drug, but it was found that longer in 60 ml of 3-heptanol. The resulting reaction mixture was stirred at 80Â°Cfor drug infusions (8) and antiallergic regimens (9) reduce their mci anadditional144 h, whereuponitwas pouredinto600 ml of ethylacetate.The dence Paclitaxel chemotherapy also elicits side effects such as ethyl acetate solution was washed with water, dried over anhydrous sodium peripheral neuropathy and myeiosuppression. Development of al sulfate, filtered, and evaporated under a high vacuum. Precipitation from ternative treatments which act by increasing microtubule stability hexanegave 270 mg (38%)of U98017 as anamberamorphoussolid.Chemical or enhancing the effect of paclitaxel itself to allow the use of lower structure was verified with high resolution mass spectrometry and nuclear concentrations of paclitaxel may alleviate these problems. magnetic resonance analysis. Compounds were dissolved in dimethyl sulfoxide and then diluted into cell growth medium or kinase assay buffer for use in Received 4/6/94; accepted 9/27/94. â€˜l'hecostsof publicationofthisarticleweredefrayedinpartby thepaymentofpage 2 The abbreviations used are: CHO, Chinese hamster ovary; PKC, Ca2@/phospholipid charges. This article must therefore be hereby marked advertisement in accordance with dependent protein kinase; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium 18U.S.C.Section1734solelyto indicatethisfact. bromide;PKA,cyclicAMP-dependentproteinkinase;IC50,50%inhibitoryconcentra I To whom requests for reprints should be addressed. lion; PS, phosphatidylserine. 5889

Downloaded from cancerres.aacrjournals.org on October 4, 2021. © 1994 American Association for Cancer Research. INDOLOCARBAZOLESSUPPORT PACLITAXEL-DEPENDENTCELL GROW@ assays. MiT, PKA, and kemptide were obtained from Sigma Chemical Co. RESULTS (St. Louis, MO). Phosphatidylseminewas purchased from Serdary Research Laboratories (London, Ontario, Canada). Histone (lysine-rich subgroup) was Paclitaxel-dependent Tax 2-4 cells were grown in the absence of purchased from ICN Biochemicals (Costa Mesa, CA). PKC purified from rat paclitaxel to test several classes of compounds (Fig. 1) that we brain was supplied by Karen Leach (The Upjohn Company). suspected might have the ability to indirectly or directly stabilize Cells and Cell Growth. The paclitaxel-dependent Tax 2-4 mutant, gener microtubules. Cell growth was quantitated by colony counting in ously given by Fernando Cabral (University of Texas Medical School, Hous many of the early experiments and correlated well with the MU ton, TX), was originally derived from the CHO cell line l000la after single assay. The MU assay was used exclusively in later assays presented step mutagenesis and selection in paclitaxel (7). These cells were grown in a here. Compounds were tested in a range of concentrations since we normal CHO medium (a-minimal essential medium-Earle's balanced salt expected that any response of cells would follow a bell-shaped curve solution with 10% fetal bovine serum-2 mM L-glutamine-100 units/mI peni like that seen with paclitaxel (Fig. 2). Paclitaxel has no effect at low cillin-100 j.tg/ml streptomycin) with the addition of 0.1 or 0.125 @LMpaditaxel. concentrations, has an optimal effect (100% growth) at a concentra Tax 2-4 cells were maintained in paclitaxel which was removed before cells tion of 0.125 p.M,and is toxic at higher concentrations. This optimal were grown with test compounds unless otherwise noted. The CHO line concentration for Tax 2-4 survival with paclitaxel is close to the IC50 AUXB1 (12) is an auxotroph for glycine, adenosine, and thymidine. These for the parental wild-type CHO cells (l000la). Therefore, doses cells were maintained in a normal CHO medium plus 10 @g/mladenosineand chosen for this experiment were in a range surrounding the dose 10 @g/mlthymidine.All cell lines were grown at 37Â°Cin 5% CO2 in a shown to give an IC50 with the wild-type cells (l000la) for each humidified incubator. Cell growth was measured by growing cells in 96-well particular compound. The point of highest absorbance for each growth plates with 1.2 x i0@cells/well with paclitaxel and/or test compounds. Plates curve is defined as the growth peak, and this level is compared to the were incubated at 37Â°Cfor 4 days. Cell viability was scored by a MU assay maximum absorbance achieved by the cell growth in paclitaxel to (13). The mitochondrial dye MIT was added to each well at 0.2 mg/mI, and derive a percentage maximal growth as shown in Table 1. The incubation was continued for 3 h. Media were aspirated, and 0.04 N HC1 in concentrations of the inhibitors with which maximum growth is isopropyl alcohol was added to each well, followed by 5 mm of rapid shaking to dissolve the MiT crystals. Absorbance was read at 570 nm on a Bio-Tek EL obtained varies with different compounds, but these concentrations do 312e Bio-kinetics microplate reader. Background absorbance was subtracted, correspond closely with the dose of each compound required to and replicate wells were averaged. Percentage of maximum growth for Tax 2-4 achieve a IC50 with wild-type l000la cells (data not shown). We find cells was determined by dividing the absorbance reading of each data point by that the indolocarbazole compounds K252a, KT5720, and KT5926 the absorbance reading of growth in 0.125 @Mpachtaxel. Growth tests used to that inhibit protein kinases have the ability to support moderate determine the effect of various compounds on the support of growth of growth (Table 1; Fig. 2). With these three structurally related com paclitaxel-dependent cells by paclitaxel were performed in 96-well plates using pounds, around 25% of optimal growth is obtained. Other serine/ the MTT assay described above. Cells were treated with the experimental threonine protein kinase inhibitors of dissimilar structure, such as H-9 compound and in varying concentrations of paclitaxel. or H-7, or the tyrosine kinase inhibitor, lavendustin, have little effect Isobolograrn Analysis. Assays to test the effects of various compounds on (Table 1). Staurosporine, a kinase inhibitor, which has a structure the toxicity of paclitaxel were set up in duplicate sets of two Corning 96-well similar to that of K252a, KT5720, and KT5926, has very little effect plates. AUXB1 cells were used at a fmal concentration of 1.2 X 10@cells/well. on growth of the paclitaxel-dependent cells. Rebeccamycin, an N- The cells were treated with 132â€”168different dose combinations of the test glycoside indolocarbazole that produces single-strand DNA breaks compound plus paclitaxel or with test compound or paclitaxel alone. Corn (16), also allows little growth. We also testedvarious other com pounds and paclitaxel concentrations were also tested alone. IC50 for drug combinations was defmed to be the various concentrations of drugs in com pounds including puromycin, Adriamycin, actinomycin D, ethidium bination which caused a 50% inhibition in cell growth. Synergy of the two bromide, vinblastine, bleomycin, ouabain, herbimycin A, dimethyl compounds on AUXB1 cells was determined by graphing the concentrations of sulfoxide, and glycerol. None of these compounds supports the the combinations which gave IC50in the form of an isobologram (14, 15). growth of the paclitaxel-dependent cells (data not shown). The syn Kinase Assays. PKA and PKC activities were determinedby measuring thesized indolocarbazole analogue U98017 (Table 1; Fig. 2) supports the transfer of 32P from [â€˜y-32P]ATPtoa PKA-specific peptide substrate, 48% of the growth of these cells compared to the optimal growth kemptide, or to a histone, respectively. PKA reaction mixtures contained 50 achieved with paclitaxel. Because the K252a, KT5720, and KT5926 mM 2-(N-morpholineo)ethanesulfonic acid (pH 7.0), 10 msi magnesium ace tate, 5 mM NaF, 1 mM ethylene g1ycol-bis(@3-uninoethyl ether)-N,N,N',N tetraacetic acid, 0.3 mg/mi Kemptide, 0.25 mg/ml bovine serum albumin, 5 @M [-y-32P]ATP(500â€”2000cpm/prnol),10 @.tMcyclicAMP (not included in negative controls), and 200 ng PKA holoenzyme in 100 @.dvolume. PKC reaction mixtures contained 50 mM Tris-HCI (jH 7.5), 7.5 mM magnesium acetate, 0.75 mg/ml histone (lysine-rich subgroup), 2 mM dithiothmeitol, 0.8mM @y@o H CaCl2, 5 @LM[-y-32P]ATP(1000â€”2000 cpm/pmol), 100 @.tg/mlphosphatidyi serine (phosphatidylserine, not included in negative controls), and 200 ng PKC cH3 enzyme in 50 @.dvolume. KT5720 and U98017 were tested at 0.1â€”10p@M 1@axol K252a, R- Cu,1 concentrations in the kinase reactions. Triplicate samples were incubated at 1cr5720.A- 30Â°Cfor10 mm; then they were placed on ice, and 25 pi were spotted onto phosphocellulose paper (Whatman P81). The papers were washed four times in 0.5% phosphoric acid for 2 min at room temperature and then dried and counted in scintillation fluid. No enzyme blanks were subtracted from total incorporation, and values were corrected for activity in the absence of cyclic AMP (for PKA) or phosphatidyisenne (for PKC). Kinase activity units were defined as pmol of 32P incorporated per mg of enzyme per min at 30Â°C. Inhibitors were tested in a range of concentrations against PKA and PKC, and cHrc@-cH@cH@rcH, the percentage of inhibition was determined by comparison with control kinase U-98017 Stauros@n. R.b.ccam@dn activity. The IC50 was defined for each compound as the concentration which Fig. 1. Molecular structures of paclitaxel (taxol), K252a, KT5720, U98017, KT5926, inhibited kinase activity by 50%. staurosporine, and rebeccamycin. 5890

compounds originally tested are all protein kinase inhibitors, we tested A U98017 to determine if it also is effective as a protein kinase inhibitor 5@ KT5720 (Table 2). U98017 is similar in its ability to inhibit PKA and PKC in vitro as compared to KT5720, although it appears to be somewhat more potent as an inhibitor of PKC. Cells were grown for 4 days in the presence of various concentra tions of pacitaxel with or without the presence of various concentra tions of KT5720 or KT5926. Dose combinations of these inhibitors with pacitaxel give far more than additive effectiveness in supporting the growth of the Tax 2-4 cells (Fig. 3). The presence of the inhibitor shifts the growth peak so that lower doses of paclitaxel are required to 0 0.01 0.10 1.@Xi achieve maximum cell growth. At levels of paclitaxel too low for cells B to grow, the additionof thekinaseinhibitordramaticallyimprovescell I viability. On the other hand, maximal growth of the cells in the 3 combination of indolocarbaxole and lower doses of paclitaxel does not fully reach the growth obtained with higher doses of paclitaxel

0.7 . ThxoI U KT5720 A 1196011 Thxol(@tM) 0.5 Fig. 3. Potentiation of Tax 2-4 cell growth in paclitaxel in the presence of KT5720 or L KT5926.Tax 2-4 cells weregrown,as in Fig.2, in combinationsof paclitaxelandKT5720 (A) or paclitaxel and KT5926 (B). 0.3

alone. This suggests that while KT5720 and KT5926 can potentiate 0.1 the effect of paclitaxel, they cannot fully replace it; optimal growth of the Tax 2-4 cells is reached only with a full dose of paclitaxel. The cell morphology after treatment reflects the effect on cell viability (Fig. 4). @ OMCOfl@fiOfl The Tax 2-4 cells grown without paclitaxel (Fig. 4A) show a lack of Fig.2. Tax 2-4 cell growthin paclitaxel,KT5720,or U98017.Tax2-4 cellswere growth as seen by the low density, and many cells appear to be dead. grownin variousconcentrationsofeitherpacitaxel,U98017,or KT5720for4 days. A low amount of paciitaxei, 001 gM (Fig. 4B), is not sufficient for Viability was determined by incorporation of MTT and detected by absorbance at 570 nm significantly improved growth or morphology of the cells. Growth in asdescribedinâ€œMaterialsandMethods.â€• 5 @LMKT5720 results in a higher cell density and clusters of growing cells (Fig. 4C). Growth in a combination oflow amounts of paclitaxel, analoguesTaxTable 1 Growth of Tax 2-4 cells in protein kinase inhibitors and 0.01 @LM,andof KT5720, 5 @LM(Fig.4D), results in cells that appear induplicate2.4 cells were grown in serial dilutions of paclitaxel or test compounds very similar in morphology and density to the same cells grown in the ofMTF wellsof 96-wellplates.After4days,theviabilitywasdeterminedbyuptake subtractedandandquantitatedbyabsorbanceat570nm.Backgroundabsorbancewas standard amount of paclitaxel, 0.1 p.M (Fig. 4E). togrowthduplicatewellswereaveraged.OrowthisexpressedaspercentageÂ±SEcompared We have also examined the effectiveness of combinations of var in padlitaxel.Optimal ious kinase inhibitors with paclitaxei on the killing of wild-type, ofCompound % non-paclitaxel-dependent cells by means of isobolograms (Fig. 5). growthControl dose Previous work has shown that staurosporine and other protein kinase 0.4Paclitaxel 4.0 Â± inhibitors can act as substrates for the P-glycoprotein efflux pump 100.01398017 0.125 @si 0.2KT5720 2.5 psi 48.6 Â± associated with muitidrug resistance (17, 18) and that they can cause 2.3K252a 5.0 @M 26.6Â± increased drug sensitivity to hydrophobic natural product-derived 1.41CF5926 0.25 @st 25.0Â± drugs such as vincristine and vinblastine (19, 20). Tax 2-4 and its 2.5K252b 0.25 @@si 24.0Â± 1.8KT5823 1.6@.&M 12.5Â± wild-type parent cell, 10001, both express low but measurable levels 1.1Lavendustin 6.0MM 9.5Â± of P-glycoprotein (21).@The inhibitors used here might act to sensitize 0.0H-9 6 @g/od 5.0 Â± 0.7Staurospomine 25.0psi 4.0Â± cells that express P-giycoprotein and to generate misleading isobolo 0.7H-7 2.5 ma 3.8 Â± gram results; therefore, we used CHO AUXB1 cells that are not 3.0Â±0.7Rebeccamycin 12.Sgi.sm multidrug resistant (22), are not paclitaxel dependent, and do not 0.3 @u@t 1.3Â±0.2 express P-giycoprotein (Ref. 23; data not shown). The AUXB1 cells were treated with combined doses of test compound and of paclitaxel Table2PKCCompoundIC@ Kinase-inhibitoryactivityofKf5720 andU98017 towardPKA and for a total of 168 different doses. After subtracting background ab sorbance and averaging duplicates, the combined doses that resulted (a@)aPKAPKCKT57201.455.8U980171.02.4 in 50% mortality (IC50) were determined. The IC50s were graphed as isobolograms (14, 15) of test compound versus paclitaxel concentra tion. The dashed line drawn between the IC50 for paclitaxel alone and the IC50for test compound alone indicates the alignment of theoretical @ a of compoundwhichinhibitskinaseactivityby 50% measuredas describedinâ€œMaterialsandMethods.â€•Resultsarebasedon meanvaluesof samples measured in triplicate. 3 K. E. Sampson and I. Abraham, unpublished data. 5891

INDOLOCARBAZOLESSUPPORT PACLITAXEL.DEPENDENTCELL GROWrH

@ â€¢@r' â€¢@1. @ .. .@ @. @@ ...@. ,: :@ @ -â€˜ .@ l@'

@ â€˜1

â€¢â€˜.@: @.

A B

@ . Â° .

@ ..@4.

.@Â°.

@ @.::/ l@. @... @@@ . @a.'@4â€œ@ â€˜ , @. I' @T ..

â€˜@. 1@'?.. [email protected] @.

., .â€˜,@

@, @t#.k' . @@ . j' .&@c.. -a, @p,, @, E Fig. 4. Morphology of paclitaxel-dependent cells in paclitaxel, KT5720, and combination. Tax 2-4 cells were grown for 3 days in the presence of unsupplemented media, paclitaxel, @ KT5720, and a combination of KT5720 and paclitaxel. A, unsupplemented media; B, 0.01 @Mpaclitaxel; C, 5 @LMKT5720;D, 0.01 @Mpaclitaxel and 5.0 @MKT5720;E, 0.1 paclitaxel. Bar, 20 p@.

IC50 points for additive interactions between the paclitaxel and tested DISCUSSION compound. The true IC50 points were plotted and compared to the The mutant CHO cell line Tax 2-4 requires paclitaxel for growth additive line; IC5@sin the area below the line on the graph indicate a due to a mutation in its @3-tubulinwhichdestabilizes its microtubules. synergistic or potentiating effect, while those above the line indicate an antagonistic interaction. The data shown in Fig. 5, A-C, clearly These cells can survive in an amount of paclitaxel that stabilizes the demonstrate a potentiating effect of K252a, KT5720, and U98017 on microtubules sufficiently for them to proceed with normal cellular paclitaxel toxicity toward AUXB1 cells. In contrast, rebeccamycin, functions. By using paclitaxel-dependent cells we have been able to which is another indolocarbazole analogue, does not support growth select compounds that allow for moderate growth of these cells. of the paclitaxel-dependent cells and shows no synergy or potentiation Survival of these cells indicates the presence of a compound which with paclitaxel. Staurosporine also does not demonstrate synergy with has activity similar to that of paclitaxel. Such a compound presumably paclitaxel (data not shown) in these cells. acts by directly or indirectly increasing microtubule stability. Because 5892

A B

50 40 30 20 Fig. 5. IsobologmamofAUXB1 cells with pacli tazel and K252a, KT5720, U98017, or mebeccamy cAn.TheIC50ofdrugcombinationsgeneratedfrom growthcurveswere plotted. ----, theoretical val -_ 50 160 150 3 0 0.25 0.50 0.75 10 1.25 1.50 ins of an additiveinteractionbetweenthe two K25@ D KT5720(IIM) compounds.Valuesabovethisdashedlineindicate ,@ C an antagonisticinteraction,andvaluesbelowindi cats potentiationoftoxicity.A,K252aversuspa 50 dlitaxel; B, 1CF5720versus paclitaxel; C, U98017 versus paclitaxel; D, rcbeccamycin versus pad taxcL

0 50 100 150 200 250 1198017(usM) R.bs@(nM)

one is selecting for the positive end pointof cell survival,this type of the amount of growth of Tax 2-4 cells with the active analogues; none screen has the potential to be highly sensitive and to select few false are able to support growth to the same level as paclitaxel. We positives. The implementation and utility of this screen will be de examined the effects of short-term (4 h) treatment of wild-type cells sen@bedinmore detail elsewhere.4 In addition to its use in discovering with paclitaxel, KT5720 alone, and KT5720 in combination in wild agents that help stabilize microtubules directly or indirectly, this type CHO cells on microtubules by immunofluorescence to determine technique would also be useful in screening for new, effective com whether KT5720 alone or in combination with paclitaxel caused the pounds that are structural analogues of paclitaxel. induction of microtubule bundles (6). We used a low concentration of We show here that several known indolocarbazoles, K252a, paclitaxel (0.05 ELM)that when used by itself does not cause the 1cr5720, and KT5926, and the novel serine/threonine protein kinase formation of bundles. We did not see a dramatic increase in the inhi@bitor,U98017, support the growth of the dependent cells. These number of microtubule bundles per field with the combination treat compounds represent the only currently defined molecules which will ment as compared to the treatments with paclitaxel alone. At higher substitute for pacitaxel in the growth support of the Tax 2-4 cells. Our concentrations KT5720 by itself appears to cause a change in cellular results also show that the indolocarbazoles KT5720 and KT5926 shape, but it does not induce bundles (data not shown). The nature of potentiate the effects of paclitaxel in pacitaxel-dependent cells. The the induced shape change is under further investigation. The fact that argument might be raised that these compounds are supporting the we did not see a dramatic increase in microtubule bundling with growth of the Tax 2-4 cells by allowing cells to use residual paclitaxel KT5720 alone or in combination with low concentrations of paclitaxel that is left on the cells or in the tissue culture plates after the media reinforces the supposition that these compounds act differently from were changed from paclitaxel-containing to paclitaxel-free media. If paclitaxel. this were the case, agents such as verapamil, which allow multidrug The concentrations at which the indolocarbazoles are maximally cells to increase drug accumulation in multidrug resistant cells, would effective for support of the paclitaxel-dependent cells are close to their be expected to support the growth of the Tax 2-4 cells by allowing IC50s for PKA and PKC inhibition. Therefore the kinase inhibition increased accumulation of the residual paclitaxel in the cell. However, may be critical for their activity. Kinase inhibition might stimulate verapainil has no such effect; it does not support the growth of the microtubule stability through various mechanisms, such as inhibition pacitaxel-dependent cells@'.In addition, the analogues that support the of phosphorylation of transcription factors, phosphatases, or other growth of the paclitaxel-dependent cells synergistically increase the kinases. Alternatively, they might affect the phosphorylation state of toxicity ofpacitaxel on the AUXB1 cells. The compounds that do not tubulin or of microtubule-associated proteins. Interestingly, stauros potentiate the effect of paclitaxel on AUXB1 cells also do not support porine, a potent PKC inhibitor, does not support the growth ofTax 2-4 the growth of the Tax 2-4 cells, e.g., rebeccamycin and staurosporine cells. This suggests that if inhibition of a specific kinase is involved (data not shown). If the mode of action of the active indolocarbazoles is similar to that of paclitaxel, one would expect the compounds to in the support of Tax 2-4 cell growth and paclitaxel synergy it is work in an additive manner. The potentiating effect suggests that the unlikely to be a PKC. compounds are working on a distinct site and perhaps in a mechanistic If the potentiation is indeed due to kinase inhibition, the effective fashion dissimilar to that of pacitaxel. The active compounds do not compounds may inhibit neither PKA nor PKC but a distinct protein promote polymerization of bovine brain tubulin in vitro (data not kinase(s) which affects the stability of microtubules at some level. shown), which also suggests that they are acting in an indirect manner Numerous reports have implicated phosphorylation in controlling or distinct from that of paclitaxel. This conclusion is also supported by modulating microtubules and other cytoskeletal structures (24â€”27). An effect of K252a and a related indolocarbazole, RK-286C, on 4N. Lopes,B. ICBhuyan,C.L Wolf andI. Abraham,unpublisheddata. overriding endoreduplication in rat 3Y1 cells has been reported, but 5 A. L Saborde and J. A. Shelly, unpublished data. the putative kinase that is inhibited has not been definitively identified 5893

(28). Several kinases including PKA (29, 30), mitogen-activated pro to proliferationandcytotoxicityassays. J. Immunol.Methods,65: 55-63, 1983. tein kinase (31, 32), CaM kinase II (33, 34), mos (35), and p34 14. Steel, 0. G., and Peckham, M. J. Exploitable mechanisms in combined radiotherapy chemotherapy: the concept of additivity. J. Rad. BioL Phys., 5: 85â€”91,1979. cdc2-like kinase (36) have been shown to phosphorylate microtubule 15. Brunden,M. N., Vidmar,T. J., andMcKean,J. W. Chapter1:Earlymethodsapplied associated proteins or tubulin or colocalize with microtubules (for to drugconcentration.In: Drug InteractionandLethalityAnalysis, pp. 1â€”11.Boca review see Ref. 37). On the other hand, the mechanism involved in Raton, FL: CRC Press, Inc., 1988. 16. Bush,J.A.,Long,B.H.,Catino,J.I., Bradner,W.T.,andTomita,K.Productionand growth support may not involve kinase inhibition at all. In support of biological activity ofrebeccamycin, a novel antitumor agent. J. Antibiot. (Tokyo), 40: this alternative, we have found no kinase inhibitors that are structur 668â€”678, 1987. ally distinct from K252a that will substantially support growth of the 17. Bradley, 0., Juranka P. F., and Ling, V. Mechanism ofmultidrug resistance. Biochim. Biophys. Acts, 948: 87â€”128,1988. Tax 2-4 cells. 2-Aminopurine, a protein kinase inhibitor which has 18. Pastas, I., and Gottesman, M. M. Multidrugresistance. Annu. Rev. Met, 42: been shown previously to cause baby hamster kidney cells to override 277â€”286,1991. cell cycle check points (38), has no effect on Tax 2-4 cells (data not 19. Sampson, K. E., Wolf, C. L, and Abraham,1. Proteinkinase inhibitortreatment reduces P-glycoprotein expression and modulates multidrug resistance. Cancer LetL, shown). We are continuing studies to distinguish between these 68: 7â€”14,1993. mechanistic alternatives. 20. Sato,W.,Yusa,K., Naito,M.,andTsuruo,T. Staurosporine,apotentinhibitorof Further study of these compounds promises to shed light on the C-kinasc, enhances drug accumulation in multidrug-resistant cells. Biochem. Bio phys.Res.Commun.,173:1252-1257,1990. mechanism of microtubule stabilization. In addition, they may warrant 21. Abraham, L, Chin, K-V., Gottesman, M. M., Mayo, J. K., and Sampson, K. E. investigation as chemotherapeutic candidates to be used in combina Transfection of a mutant regulatory subunit gene of cAMP.depcndcnt protein kinase tion with paclitaxel to allow administration of reduced concentrations causes increased drug sensitivity and decreased expression of P-glycoprotein. Exp. Cell Res., 189: 133â€”141,1990. of paclitaxel while maintaining efficacy. 22. Ling,V.,andThompson,LH. Reducedpermeabifityin010 cellsas a mechanism of resistancetocolchicinc.J.Cell.PhysioL,83:103â€”116,1988. 23. Kartner,N.,Riordan,J.R.,andLing,V.CellsurfaceP-glycoproteinassociatedwith ACKNOWLEDGMENTS multidrug resistance in mammalian cell lines. Science (Washington DC)@221:1285- 1288, 1983. We thankKarenLeachforprovidingpurifiedPKCandFernandoCabralfor 24. Jameson, L, Frey, 1., Zeeberg, B., Dalldorf, F., and Caplow, M. Inhibition of generously supplying the Tax 2-4 cells. We also thank Vincent Marshall, microtubule assembly by phosphorylation of microtubule-associated protein& Howard Whaley, and Veronica Wiley for supplying us with purified K252a, Biochemistry,19: 2472-2479, 1980. 25. Jameson, L, and Caplow, M. Modification of microtubule steady-state dynamics by Bijoy Bhuyan and Narima Lopes for help in setting up the MTI' assay, and phosphorylation of the microtubule-associatcd protiens. Proc. Natl. Acad. Sd. USA, Jacque Hercik for secretarial assistance. 78: 3413â€”3417,1981. 26. Nishida, E., Kuwaki,T., and Sakai, H. Phosphorylationof imrmtubule-associated proteins (MAPs) and pH of the medium control interaction between MAPs and acm REFERENCES filaments. J. Biochem. (Tokyo), 90: 575â€”578,1981. 1. Wani, M. C., Taylor, H. L, Wall, M. E., Coggon, P., and McPhail,A. T. Plant 27. Selden,S.C.,andPollard,T.D.Phosphorylationofmicrotubule-associatedproteins regulates their interaction with actin flisments@ J. Biol Chem., 258: 7064-7071, antitumor agents. VI. The isolation and structure of taxol, a novel antileukemic and antitumor agent from Texas brevifolia. J. Am. Chem. Soc., 93: 2325â€”2327,1971. 1983. 2. McOuire,W.P., Rowinsky,E.K., Rosenshein,N.B., Grumbine,F.C., Ettinger, 28. Usui,T., Yoshida,M.,Abe,K.,Osada,H.,Isono,K.,andBeppu,T.Uncoupledcell D. S., Armstrong, D. K., and Donehower, R. C. Taxol: a unique antineoplastic agent cyclewithoutmitosisinducedbyaproteinkianseinhibiter,K-252a.J.CellBiol.,115: with significant activity in advanced ovarian epitheial neoplasms. Ann. intern. Med., 1275â€”1282,1991. 111:273â€”279,1989. 29. Sloboda, R. D., Rudolph, S. A., Rosenbaum, J. L, and Greengard, P. Cyclic AMP 3. Holmes, F. A., Walters, R. S., TherisuIt, R. L, Forman, A. D., Newton, L K., Raber, dependent endogenous phosphorylation of a microtuble-associated protein. Proc. M. N., Buzdar,A. U., Frye, D. IC, and Hortobagyi,0. N. PhaseII trialof taxol, an NatL Mad. Sd. USA, 72: 177-181,1975. active drug in the treatment of metastatic breast cancer. J. Nat! Cancer Inst., 83: 30. Theurkauf,W. E., and Vallee, R. B. Molecular characterizationof the cAMP 1797â€”1805,1991. dependent protein kinase bound to mirmtubule-asmciated protein 2. J. Biol @1em., 4. Ettinger, D. S. Overview of paclitaxel (taxol) in advanced lung cancer. Semin. Oncol., 257: 3284â€”3290,1982. 20: 46â€”49,1993. 31. Theurkauf, W. E., and Vance, R. B. Extensive cAMP-dependent and cAMP-inde 5. Forastiere, A. A. Use of paditaxel (taxol) in squamous cell carcinoma of the head and pendentphosphorylationof microtubule-associatedprotein2. J. Biol. Chem., 258: neck. Semin. Oncol., 20: 56â€”60,1993. 7883â€”7886,1983. 6. Schiff, P. B., and Horwitz@ S. B. Taxol stabilizes microtubules mouse flbroblast 32. Sturgill, T. W., and Ray, L B. Muscle proteins related to microtubule associated cells.Proc.NaIl.Acad.Sd. USA,77: 1561â€”1565,1980. protein-2are substratesfor an insulin-stimulatablekinase.Biochem. Biophys. Res. 7. Schibler,M.J., andCabral,F. Taxol-dependentmutantsof Chinesehamsterovary Commun.,134:565â€”571,1986. cells with alterations in a- and @-tubulin.J.Cell Biol., 102: 1522â€”1531,1986. 33. Goldenring, J. R., Gonzalez, B., McGuire, J. S., Jr., and DeLorenzo, R. J. Purification 8. Weiss, R. B., Donehower, R. C., Wiernik, P. H., Ohnuma, T., Gralia, R. J., Trump, and characterizationofcalmodulin-dependentkinasefrom ratbraincytosol able to D. L, Baker, J. R. Jr., VanEcho, D. A., VonHoff, D. D., and Leyland-Jones, B. phosphorylate tubulin and microtubule-associated proteins. J. Biol. Chem., 258: Hypersensitivity reactions from taxoL J. Clin. OncoL, 8: 1263-1268, 1990. 12632â€”12640,1983. 9. Rowinsky,E. K., Cazenave,L A, and Donehower,R. C. Taxol: a novel investiga 34. Schulman,H.Phosphorylationofmicrotubule-associatedproteinsbya Ca2@/calmod tional antimicrotubule agent. J. Natl. Cancer Inst., 82: 1247â€”1259,1990. ulin-dependent protein kinase. J. Cell Biol., 99: 11â€”19,1984. 10. Kase, H., lwahashi, K, Nakanishi,S., Matsuda,Y., Yamada,K., Takahashi,M., 35. Thou, R., Oskarsson, M., Panics, R. S., Schulz, N., aeveland, D., and Vande Woude, Murakata,C.,Sato, A., and Kaneko,M. K-252 compounds,novel andpotentinhib 0. F. Ability of the c-ems product to associate with and phosphorylate tubulin. itots of protein kinase C and cyclic nucleotide-dependent protein kinases. Biochem. Science(WashingtonDC),251:671â€”675,1991. Biophys. Res. Common., 142: 436-440, 1987. 36. Tombes,R. M., Peloquin,J. 0., andBorisy,0. 0. Specific associationofan M-phase 11. Kase, H. New inhibitorsof proteinkinases from microbialsource. In: 0. Yoshim kinasewith isolatedmitoticspindlesand identificationoftwo of its substratesas (ed), Biologyof Actinomycetes,pp.159â€”164.Tokyo:JapanScientificSocieties MAP4 and MAP1B. Cell Regul., 2: 861-874, 1991. Press, 1988. 37. Wiche, 0., Oberkanins, C., and Himmler, A. Molecular structure and function of 12. McBurney, M. W., and Whitmore, 0. F. Isolation and biochemical characterization of microtubule-associated proteins. bit. Rev. CytoL, 124: 217â€”273,1991. folate deficient mutants of Chinese hamster cells. Cell, 2: 173â€”182,1974. 38. Andreassen,P. R., and Margolis,R. L 2-Aminopurineoverridesmultiplecell cycle 13. Mosmann, T. Rapid colorimetric assay for cellular growth and survival: application checkpoints in BIlK cells. Proc. Natl. Acad. Sal. USA, 89: 2272-2276, 1992.

5894

Downloaded from cancerres.aacrjournals.org on October 4, 2021. © 1994 American Association for Cancer Research. K252a, KT5720, KT5926, and U98017 Support Paclitaxel (Taxol)-dependent Cells and Synergize with Paclitaxel

Irene Abraham, Cindy L. Wolf, Kathleen E. Sampson, et al.

Cancer Res 1994;54:5889-5894.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/54/22/5889

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/54/22/5889. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.