Protein Kinase D as a Potential Chemotherapeutic Targetfor Colorectal Cancer
Ning Wei1,2, Edward Chu1,2, Peter Wipf2,3, and John C. Schmitz1,2
AbstractProtein kinase D (PKD) signaling plays a critical role in the regulation of DNA synthesis, proliferation, cell
survival, adhesion, invasion/migration,motility, and angiogenesis. Todate, relatively little is knownabout the
potential role of PKD in the development and/or progression of human colorectal cancer. We evaluated the
expression of different PKD isoforms in colorectal cancer and investigated the antitumor activity of PKD
inhibitors against human colorectal cancer. PKD2was the dominant isoform expressed in human colon cancer
cells. PKD3 expression was also observed but PKD1 expression, at both the RNA and protein levels, was not
detected. Suppression of PKD using the small molecule inhibitors CRT0066101 and kb-NB142-70 resulted in
low micromolar in vitro antiproliferative activity against multiple human colorectal cancer cell lines. Drug
treatment was associated with dose-dependent suppression of PKD2 activation. Incubation with CRT0066101
resulted in G2–M phase arrest and induction of apoptosis in human colorectal cancer cells. Further studies
showed that CRT0066101 treatment gave rise to a dose-dependent increase in expression of cleaved PARP and
activated caspase-3, in addition to inhibition of AKT and ERK signaling, and suppression of NF-kB activity.
Transfection of PKD2-targeted siRNAs resulted in similar effects on downstream pathways as observed with
small molecule inhibitors. Daily administration of CRT0066101 resulted in significant inhibition of tumor
growth in HCT116 xenograft nude mice. Taken together, our studies show that PKD plays a significant role in
mediating growth signaling in colorectal cancer and may represent a novel chemotherapeutic target for the
treatment of colorectal cancer. Mol Cancer Ther; 13(5); 1130–41. �2014 AACR.
IntroductionColorectal cancer is amajor public healthproblem in the
United States and globally. In the United States, it is thesecond leading cause of cancer mortality (1). For 50 years,the main chemotherapeutic treatment was the fluoropyr-imidine 5-fluorouracil (5-FU). From 1996 to 2004, 3 newanticancer agents were approved, which include the oralfluoropyrimidine capecitabine, the topoisomerase I inhib-itor irinotecan, and the platinum analog oxaliplatin.Since 2004, the U.S. Food and Drug Administration hasapproved 5molecular targeting agents, including the anti-EGF receptor antibodies cetuximab and panitumumab,the anti-VEGF inhibitors bevacizumab and ziv-afliber-cept, and a small molecule inhibitor that targets multipletyrosine kinases, regorafenib. Significant advances have
been made in chemotherapy treatment options forpatients with metastatic disease, such that improvementsin 2-year survival are now being reported, with medianoverall survival rates of 21 to 24 months (2–4). Despitethese advances, none of the currently available treatmentoptions have impacted on 5-year overall survival, andcellular drug resistance remains a significant obstacle tosuccessful chemotherapy (5). Thus, identification of novelsignaling pathways and targets that mediate the growthand proliferation of colorectal cancer is critically impor-tant for discovering and developing novel therapeuticagents with enhanced antitumor activity that overcomedrug resistance and improve overall quality of life.
Protein kinase D (PKD) is a subfamily of serine/threo-nine kinases of the calcium/calmodulin-dependent kinasesuperfamily, composed of 3 different isoforms, PKD1,PKD2, and PKD3 (6). This signaling pathway functionsdownstreamof protein kinaseC (PKC),Gprotein–coupledreceptors, and tyrosine kinase receptors. PKD can be acti-vated in a PKC-dependent as well as a PKC-independentmanner. In turn, activated PKD phosphorylates a widerange of downstream targets at specific sites, subsequentlyregulating their activity and/or subcellular localization (7).It was beenwell-documented that PKDplays a critical rolein the regulation of several important cellular processes,such as DNA synthesis, proliferation, cell survival, adhe-sion, invasion/migration, motility, and angiogenesis (8).Moreover, PKD signaling has been implicated in several
Authors' Affiliations: 1Division of Hematology-Oncology, Department ofMedicine, University of Pittsburgh School of Medicine; 2Cancer Therapeu-tics Program, University of PittsburghCancer Institute; and 3Department ofChemistry, University of Pittsburgh, Pittsburgh, Pennsylvania
Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).
Corresponding Authors: Ning Wei, Division of Hematology-Oncology,Department of Medicine, University of Pittsburgh School of Medicine,Pittsburgh, PA 15232. Phone: 412-864-7744; Fax: 412-623-1212; E-mail:[email protected]; and John C. Schmitz, [email protected]
human tumors, including breast, pancreatic, and prostatecancer, and glioblastoma. For example, PKD2 and PKD3seem to be highly expressed in breast cancer (9). PKD1expression level is elevated in human ductal adenocarci-noma of the pancreas compared with normal pancreatictissues (10). PKD2 has been shown to be an importantmediator of induction of various angiogenic factors inhuman pancreatic cancer cells and in the angiogenicresponse of the host vasculature (11). Expression of PKD1and PKD3 has been shown to be elevated in humanprostate carcinoma tissue compared with normal prostateepithelial tissue, and advanced stage tumorswere found tohave increased PKD3 nuclear accumulation (12). PKD2was also highly expressed in both low-grade and high-grade human gliomas (13).Given the importance of PKD in tumor biology, inves-
tigators have focused ondeveloping novel smallmoleculeinhibitors targeting PKD. Several such agents with anti-cancer activity havebeen identified, includingCID755673,kb-NB142-70, and CRT0066101 (14–16). CID755673 is anon-ATP competitive pan-PKD inhibitor that was discov-ered in a high throughput screening campaign. The IC50 ofCID755673 for PKD1, PKD2, and PKD3 was 182, 280, and227 nmol/L, respectively, using an established radiomet-ric kinase assay. This compound blocked phorbol 12-myristate 13-acetate (PMA)–induced activation of PKD1in human prostate cancer LNCaP cells. Moreover,CID755673 displayed inhibitory effects on cell prolifera-tion, migration, and invasion in prostate cancer cells (16).A series of structural analogs of CID755673 was subse-quentlydeveloped, including kb-NB142-70, kb-NB165-09,kb-NB165-31, kb-NB165-92, and kb-NB184-02. Theseagents exhibited at least 2-fold higher potency andimproved kinase selectivity when compared with theparent compound (15). The most potent analog, kb-NB142-70, inhibited PKD1, PKD2, and PKD3 enzymaticactivitywith an IC50 of 28, 59, and53nmol/L, respectively.kb-NB142-70 also significantly inhibited cell proliferation,migration, and invasion. Unfortunately, kb-NB142-70 didnot exhibit in vivo antitumor activity because of rapidmetabolism (17). CRT0066101 is a small molecule PKD-specific inhibitor developed by investigators in theUnitedKingdom, and it exhibited in vitro antitumor activity inhuman pancreatic cancer cells. CRT0066101 significantlysuppressed neurotensin-induced PKD1/2 activation,blocked NF-kB–mediated cellular proliferation and sur-vival, and induced apoptosis. Moreover, CRT0066101inhibited Panc-1 cell growth in in vivo xenograft mousemodels (14). In addition to CID755673, kb-NB142-70, andCRT0066101, several other pan-PKD inhibitors have beenreported in the literature (18, 19).In this study, we investigated PKD isoform expression
in colorectal cancer, evaluated the therapeutic efficacy oftargeting PKD in human colorectal cancer, and deter-mined its potential molecular mechanisms of action. Wepresent both in vitro and in vivo evidence showing thatCRT0066101 has cytotoxic as well as antitumor activityagainst human colorectal cancer model systems. These
findings provide evidence that PKD may represent apotential target for colorectal cancer chemotherapy.
Materials and MethodsChemicals and reagents
CRT0066101 was kindly provided by Dr. S. Guha andCancer Research Technology Inc. For in vitro use, the drugwas resuspended in dimethyl sulfoxide (DMSO; Sigma),whereas itwas resuspended in 5% sterile dextrose solutionfor in vivo studies. CID755673 and kb-NB142-70 weresynthesized as previously described (15). The DMSO con-centration never exceeded 0.1% in any experiment. Thisdose had no effect on cell growth nor did it affect proteinexpression. WST-1 was purchased from Roche Diagnos-tics. PMA and other chemicals were obtained from Sigma.The following siRNAs were synthesized by DharmaconResearch (ThermoScientific): siPKD2—50-UGAGACAC-CUUCACUUCAA-30 (#D-004197-05); siPKD3—50-GGGA-GAGUGUUACCAUUGA-30 (#D-005029-06); and siCon—50-GGAUACUGCCAAUCUCUAGG-30.
Tissue culture and human colorectal cancer cell linesNormal human colon epithelial CCD 841 CoN and
FHC cell lines and the human cancer RKO cell linewere obtained from American Type Culture Collection(ATCC). HCT116 p53þ/þ and p53�/� cell lines werekindly provided by Dr. B. Vogelstein (20). H630 andH630R1 cells have been maintained in our laboratoryafter being originally obtained from Dr. A. Gazdar (21).All cell lines, with the exception of FHC, were main-tained in RPMI-1640 (Invitrogen) with 10% (v/v) FBS at37�C in a humidified incubator with 5% CO2. FHC cellswere maintained according to ATCC guidelines. HCT116and RKO cells were authenticated by short tandemrepeat profiling at the University of Pittsburgh CellCulture and Cytogenetics Facility (August 2013). Cellswere tested monthly for mycoplasma by the MycoAlertMycoplasma detection assay (Cambrex BioScience).
Cell viability assayHuman colorectal cancer cells were plated in 96-well
plates at a density of 800 to 1,500 cells/well. On thefollowing day, cells were incubated with various concen-trations of PKD inhibitors for 72 hours. Cell viability wasdetermined by the WST-1 assay. The IC50 value wasdefined as the drug concentration that inhibits 50% cellgrowth compared with the untreated controls and calcu-lated by Graphpad Prism 6.0 software.
Clonogenic assayHCT116 and RKO cells were seeded in 6-well plates at
density of 400 cells/well. On the following day, cells wereexposed to various concentrations of CRT0066101 for24 hours, after which time, the growth medium was thenreplaced. After 10 to 14 days, cell colonies were fixedwithtrypan blue solution (75% methanol/25% acetic acid/0.25% trypan blue) for 15 minutes, washed, and air-driedbefore counting colonies >50 cells.
Protein Kinase D: A Novel Target for Colorectal Cancer Therapy
www.aacrjournals.org Mol Cancer Ther; 13(5) May 2014 1131
siRNA transfectionCells were plated at a density of 2 � 105 cells/well. On
the following day, siRNAs (10 nmol/L) were complexedwith Lipofectamine 2000 (Invitrogen) in serum-freeRPMI-1640 medium and added to the plated cells. After48 hours, cells were processed forWestern blot analysis orfor flow cytometry.
Western blot analysisCell lysate protein concentrations were determined
using the DC Protein Assay (Bio-Rad). Equal amounts ofprotein (30 mg) from each cell lysate were resolved onSDS-PAGE using the method of Laemmli and trans-ferred onto 0.45-mm nitrocellulose membranes (Bio-Rad). Membranes were blocked and incubated over-night with primary antibodies at 4�C. The followingantibodies were used in the experiments: anti-p-PKD2[(Ser876) #07–385; Upstate], anti-PKD2 (#07–488;Upstate), anti-PKD1 (gift from Dr. P. Storz), anti-PKD3(#5655; Cell Signaling), anti-p-ERK (#sc-7383; SantaCruz Biotechnology), anti-ERK (#sc-94; Santa Cruz Bio-technology), anti-p-AKT (#9542; Cell Signaling), anti-AKT (#9272; Cell Signaling), anti-PARP (#9542; CellSignaling), anti-cleaved caspase-3 (#9661; Cell Signal-ing), anti-glyceraldehyde-3-phosphate dehydrogenase(GAPDH; #47724; Santa Cruz Biotechnology), anti-b-actin (#4970; Cell Signaling), and anti-a-tubulin (EMDBiosciences). Detection of b-actin, GAPDH, or a-tubulinwas routinely used as protein loading controls. Aftermultiple TBST washes (1 � TBS, 0.1% Tween-20), mem-branes were incubated with corresponding horseradishperoxidase-conjugated secondary antibodies (Bio-Rad)for 1 hour at room temperature. Proteins were detectedby the enhanced chemiluminescence method (Super-Signal West Pico substrate; Pierce). Quantitation ofsignal intensities was performed by densitometry ona Xerox scanner using ImageJ software.
RNA extracts and real-time qRT-PCR analysisTotal RNA was extracted by the guanidine isothio-
cyanate/phenol/chloroform method (TRizol; Invitro-gen). The integrity and purity of the RNA was deter-mined by UV spectrophotometry at OD260/OD280. Thefirst-strand cDNA was synthesized using 1.0 mg totalRNA and the iScript Reverse Transcription Supermixfor real-time quantitative PCR (qRT-PCR; Bio-Rad).PCR was performed in triplicate using the SsoFastProbes Supermix (Bio-Rad) in a final reaction volumeof 10 mL with gene-specific primer/probe sets, and astandard thermal cycling procedure (40 cycles) on a Bio-Rad CFX96 Real-Time PCR System. PKD1, PKD2, PKD3,and 18S RNA levels were assessed using TaqMan GeneExpression real-time PCR assays (Applied Biosystems;assay ID: Hs00177037_m1, ID: Hs00212828_m1, ID:Hs00178657_m1 and Hs03928990_g1, respectively). AcDNA array of 24 human colon tumors (Colon CancercDNA Array III; Origene) was also analyzed for PKDisoform expression (normalized by b-actin RNA levels;
assay ID: Hs01060665_g1). Results were expressed asthe threshold cycle (Ct). The relative quantification ofthe target transcripts was determined by the compara-tive Ct method (DDCt) according to the manufacturer’sprotocol. The 2�DDCt method was used to analyze therelative changes in gene expression. Control experi-ments were conducted without reverse transcription toconfirm that the total RNA was not contaminated withgenomic DNA.
Flow cytometryHCT116 and RKO cells were seeded in 6-well plates at a
densityof 4� 105 cells/well.After exposure toCRT0066101for 24 hours, cells were harvested, washed twice with PBS,resuspended with 1� binding buffer, stained with FITCAnnexin V Apoptosis Detection Kit (BD Biosciences), andanalyzed on the BDAccuri C6 FlowCytometer (BDAccuriCytometers Inc.) at the UPCI Cytometry Facility. For cellcycle analysis, cells were washed with PBS and fixed with70% ethanol overnight. Cells were washed with PBS, andthen treatedwith 1mg/mLRNaseA for 30minutesat 37�C.Cells were incubated with propidium iodide (1 mg/mL)for 45 minutes before detection by flow cytometry.
NF-kB activity assayHCT116 and RKO cells were seeded in 24-well plates at
a density of 1 � 105 cells/well. On the following day,pGL3-Luc-NF-kB or pGL3-LucDNA (0.5 mg)was cotrans-fected with 0.1 mg of Renilla luciferase plasmid DNA intocells using Lipofectamine 2000 according to the manu-facturer’s protocol (Invitrogen). The Renilla luciferaseplasmid DNA was used as an internal control for trans-fection efficiency. After 6 hours, transfectionmediumwaschanged, and on the following day, cells were incubatedfor 2 hours with various concentrations of PKD inhibitorsfollowedbyadditionof 50ng/mLTNF-a for anadditional5 hours. Firefly luciferase values were normalized withrenilla activity and the reporter assays were performed intriplicate. To generate RKO cells that stably express NF-kB–driven luciferase, RKO cells were transfected withpGL3-Luc-NF-kB.After 48hours, 400mg/mLhygromycinB was added to the cells. After 2 weeks of antibioticselection, the heterogeneous RKO cells were incubatedfor 2 hours with various concentrations of PKD inhibitorsfollowed by the presence or absence of 100 nmol/L PMAfor an additional 5 hours. Firefly luciferase activity wasthen determined as described above and was normalizedto total soluble protein.
Xenograft mouse model experimentsAll animal experiments were approved by the Institu-
tional Animal Care and Use Committee of the Universityof Pittsburgh. HCT116 cells (<70% confluent) were har-vested, and 5 � 106 cells in 0.1 mL of medium wereimplanted subcutaneously on the back of athymic nudefemalemice.When the tumor size reached approximately100 mm3, mice were randomized into the followinggroups (5 mice per group): (A) control (vehicle; 5%
Wei et al.
Mol Cancer Ther; 13(5) May 2014 Molecular Cancer Therapeutics1132
dextrose); (B) 40mg/kg, (C) 80mg/kg, and (D) 120mg/kgCRT0066101 (dissolved in 5% dextrose) administeredorally once daily. Therapy was administered for 3 weeks,andanimalswere sacrificedonday21 after treatmentwithCRT0066101. Tumorvolumewasmeasured asV¼ 1/2ab2,in which "a" and "b" represents length andwidth of tumor(22). Tumor volumesweremonitored 3 timesperweek.Atthe time the animals were euthanized, half of the tumortissuewas fixedwith formalin andparaffin-embedded forimmunohistochemistry. Theother halfwas snap-frozen inliquid nitrogen and stored at �80�C. Tissue slides wereprocessed by the Department of Pathology DevelopmentLaboratory and the Tissue and Research Pathology Ser-vices at the University of Pittsburgh for Ki67 (#9027; CellSignaling), in situ terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) staining(APOPTAG Peroxidase Kit; Chemicon), p-ERK (#4370;Cell Signaling), and M30 (#12140322001; Roche).
Statistical analysisData were expressed as the mean � SD. Statistical
analysis of the data was performed using one-way
ANOVA (SPSS software). P < 0.05 was considered statis-tically significant.
ResultsPKD expression in colorectal cancer
We analyzed the expression of the 3 different PKDisoforms in 2 normal epithelial colon cell lines (FHC and841) and 3 human colorectal cancer cell lines. As shownin Fig. 1A and B, PKD1 was only expressed in normalcolon cells and not in colorectal cancer cells. In contrast,PKD2 and PKD3 were expressed, at the protein andmRNA level, in all cell lines. To provide further supportfor the differential level of expression of the respectivePKD RNA isoforms in human colorectal cancer, we usedthe CellMiner web-based program to interrogate RNAtranscript patterns in the NCI-60 cell panel (23). The7 human colorectal cancer cell lines in the panel areCOLO205, HCC2998, HCT116, HCT15, HT29, KM12, andSW620. The level of PKD1 RNA expression was signifi-cantly reduced below the median transcript expression(Z-score ¼ 0) in all cell lines, whereas PKD3 RNA
A B
C D
E
RK
O
HC
T11
6
− − 0.3 1 3 10 CRT (μmol/L)
p-PKD2
PKD2
α-Tubulin
p-PKD2
PKD2
α-Tubulin
− + + + + + PMA
PKD1 PKD2 PKD3
mR
NA
exp
ress
ion
(PK
D/β
-act
in)
p-P
KD
2 ex
pres
sion
− + + + + + PMA− − 0.3 1 3 10 CRT (μmol/L)
mR
NA
exp
ress
ion
(PK
D/1
8S)
PKD1
PKD2
PKD3
α-Tubulin
FH
C
841
RK
O
HC
T11
6 p5
3+/+
HC
T11
6 p5
3-/-
H63
0
H63
0R1
FH
C
841
RK
O
HC
T11
6 p5
3+/+
HC
T11
6 p5
3-/-
H63
0
H63
0R1
3.0
2.5
2.0
1.5
1.0
0.5
0.0
400
300
200
100
0
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
HCT116RKO
###
###
###
***
***
PKD1PKD2PKD3
Figure 1. Expression of PKD inhuman colorectal cancer cell lines.Normal colon epithelial cells andcolorectal cancer cells wereharvested and processed forWestern blot analysis (A) and qRT-PCR analysis (B). C, qRT-PCRanalysis of cDNAs from 24 humancolon cancers. D, cells wereincubated with CRT0066101 for24 hours, followed by stimulationwith 100 nmol/L PMA for 30minutes. Cells were immediatelyprocessed for Western blotanalysis. E, expression valuesrepresent the mean � SD from3 individual experiments.Densitometry measurements ofp-PKD2 after treatment withPMA alone were normalized to 1.###, P < 0.001; ���, P < 0.001.
Protein Kinase D: A Novel Target for Colorectal Cancer Therapy
www.aacrjournals.org Mol Cancer Ther; 13(5) May 2014 1133
expression was somewhat more variable (SupplementaryFig. S1). In sharp contrast, expression of PKD2 RNA waselevated in 5 of 7 cell lines. A search of the Broad-NovartisCancer Cell Line Encyclopedia database (24), which con-tains genetic information on 1,000 cell lines, revealed thathuman colorectal cancer cell lines express the lowestPKD1 mRNA levels among all the cell lines (Supplemen-tary Fig. S2). PKD2 expression in the 61 colorectal cancercell lines was slightly higher than the mean mRNA levels(8.36 vs. 8.26; RMA, log2) whereas PKD3 expression wassomewhat lower than themean (8.07 vs. 8.42; RMA, log2).To provide further support for the potential clinical rel-evance of PKD expression, we performed qPCR analysison cDNAs obtained from 24 human colon tumors (Ori-gene cDNA array). After estimation of all PKD isoforms,we observed a nearly identical RNA expression patternwith PKD2 being the dominantly expressed isoform (Fig.1C). Taken together, these data suggest that PKD2 is themost abundant isoform in colorectal cancer.
We next investigated the effect of pan-PKD inhibitorson PKD activation in the human colorectal cancer celllines. Given the high level of PKD2 expression in HCT116and RKO and the availability of a specific p-PKD2 anti-body, we monitored phosphorylation of the PKD2isoform by Western blot analysis. Cells were incubatedwith PKD inhibitors followed by stimulationwith PMA, aknown activator of the PKC/PKD pathway. As shownin Fig. 1D and E, treatment with CRT0066101 significantlyinhibited PKD2 phosphorylation in a dose-dependentmanner. The highest concentration of CRT0066101 (10mmol/L) almost completely blocked PKD2 activation inHCT116 and RKO cells. The concentration that inhibited50% of PKD2 activationwas 2 and 3 mmol/L, respectively.Treatment of RKO cellswith a different PKD inhibitor, kb-NB142-70, also decreased p-PKD2 levels, but to a muchlesser extent than CRT0066101 (Supplementary Fig. S3A;IC50, 38 mmol/L).
Effect of PKD inhibition on growth of humancolorectal cancer cells
To determine the potential impact of PKD suppressionon cell growth, a panel of colorectal cancer cell lines wastreated with different PKD inhibitors. Cells were exposed
to CRT0066101, CID755673, and kb-NB142-70 for 72hours, and cell proliferation was determined by theWST-1 assay. CRT0066101 exhibited low mmol/L IC50
values against all colorectal cancer cell lines in the panel(Table 1). kb-NB142-70 had similar inhibitory effects oncell growth albeit with slighter higher IC50 values. Incontrast, CID755673 was significantly less potent, as hadbeen shown previously observed with different humancancer model systems (16). Growth inhibition was similarin both p53þ/þ and p53�/� HCT116 cells suggesting thatthe cytotoxic effects of these molecules are mediatedthrough p53-independent pathways. These compoundswere also able to maintain their inhibitory activity in cellsthat are resistant to both chemotherapy and radiationtherapy. Of note, the 5-FU–resistant H630R1 cells wereas sensitive to the PKD inhibitors as parentalH630 cells. Inaddition to the WST-1 assay, we utilized the clonogenicassay to determine the effect of CRT0066101 on HCT116and RKO clonogenic growth. CRT0066101 effectivelydecreased colony number in a dose-dependent manner(Supplementary Fig. S4).
Effect of PKD inhibition on cell-cycle distributionand apoptosis
Given the growth inhibitory activity of CRT0066101 onhuman colorectal cancer cells, we evaluated the potentialmechanisms of action of PKD inhibition in these cells. Todetermine the effect of CRT0066101 on distribution of thecell cycle, HCT116 and RKO cells were treated withCRT0066101 for 24 hours, stainedwith propidium iodide,and analyzed by flow cytometry. As shown in Fig. 2A,CRT0066101 blocked cell-cycle progression at the G2–Mphase in a dose-dependent manner. Similar results wereobtained with kb-NB142-70 (Supplementary Fig. S3B andS3C). This block at the G2–M checkpoint coincided with adecrease in the fraction of cells in the G1 population. Wealso observed a significant increase in the sub-G1 phase inboth cell lines suggesting that CRT0066101 may induceapoptosis of human colorectal cancer cells. To furtherinvestigate the potential impact of this agent on inducingapoptosis, cells were treated with CRT0066101 for 24hours, stained with FITC-Annexin V and propidiumiodide and analyzed by flow cytometry. As shown
Table 1. Effect of PKD inhibitors on cell proliferation
Figure 2. Effect of PKD inhibition on cell-cycle distribution and apoptosis. HCT116 and RKO cells were incubatedwith various concentrations of CRT0066101(0.3–10 mmol/L) for 24 hours. A, cell-cycle distribution was detected by flow cytometry. Cycle percentages represent the mean � SD from 3 individualexperiments using HCT116 and RKO cells. �, P < 0.05; ��, P < 0.01; ���, P < 0.001. B, apoptotic cells were detected by flow cytometry. Apoptotic cellpercentages represent the mean � SD from 3 individual experiments. ##, P < 0.01, ###, P < 0.001 early apoptosis versus control; ���, P < 0.001 lateapoptosis versus control. C, expression of cleaved-PARP and cleaved caspase-3 was determined by Western blot analysis after 24 hours exposure toCRT0066101. D, cells were transfected with 10 nmol/L siRNAs for 48 hours, and expression of cleaved PARP and caspase-3 was analyzed by Western blotanalysis. E, cells were transfected with 10 nmol/L siRNAs for 48 hours and processed for apoptotic cell detection by flow cytometry.
Protein Kinase D: A Novel Target for Colorectal Cancer Therapy
www.aacrjournals.org Mol Cancer Ther; 13(5) May 2014 1135
in Fig. 2B, CRT0066101 significantly induced apoptosis inboth cell lines in a dose-dependent manner. A differentmolecule kb-NB142-70 induced slightly lower levels ofapoptosis (Supplementary Fig. S3D and S3E). We theninvestigated the effect of drug treatment on the expressionof other markers of apoptosis. As determined byWesternblot analysis, CRT0066101 treatment (1–3 mmol/L)resulted in cleavage of both PARP and caspase-3, respec-tively (Fig. 2C). Because the PKD inhibitors suppress allPKD isoforms, we transfected isoform-specific siRNAsinto colorectal cancer cells to identifywhich isoformmightbe responsible for the growth inhibitory activity of thesmall molecule inhibitors. Each siRNA specifically andpotently suppressed only the intended targeted isoform(Fig. 2D). Knockdown of PKD2, but not PKD3, resulted incleavage of PARP and caspase-3. Furthermore, transfec-tion of PKD2 siRNA induced apoptosis as determined byflowcytometry,whereasPKD3siRNAtransfectionhadnoeffect on Annexin V/propidium iodide staining (Fig. 2E).
Effect of CRT0066101 on key cellular survivalpathways
The PI3K-AKT, MAPK, and NF-kB signaling pathwaysare well-established survival pathways in human colo-rectal cancer. Constitutive activation of each of thesepathways, as a result of point mutations, has been shownto increase cancer cell proliferation and drug resistance(25). It has also been demonstrated that each of thesepathways are downstream of PKD signaling (26).We nextdetermined whether any of these pathways might repre-sent active downstream components of PKD signaling incolorectal cancer cells. We treated human HCT116 andRKO cells with CRT0066101 for 24 hours, and then deter-mined expression of p-AKT, p-ERK, total AKT, and totalERK by Western blot analysis. CRT0066101 treatmentresulted in a significant reduction in expression of p-AKTand p-ERK, with no effect on total AKT and ERK proteinlevels (Fig. 3A–C). Transfection of PKD2 siRNA resultedin a similar decrease in p-AKT expression in RKO cellsand a decrease in p-ERK expression in both cell lines (Fig.3D). In contrast, treatment with the PKD3 siRNA did notalter these pathways. We next detected the effect of PKDinhibition on NF-kB activity. HCT116 and RKO cellswere transiently transfected with a luciferase plasmidunder the control of the NF-kB response element. Afterincubation of PKD inhibitors for 2 hours, cells werestimulated by TNF-a (50 ng/mL) for an additional 5hours. NF-kB activity was determined by the dual lucif-erase assay. As shown in Fig. 3E, CRT0066101 effectivelysuppressed, in a dose-dependent manner, the activationof NF-kB. Significant inhibition was observed after treat-ment with 1 mmol/L CRT0066101. In contrast, higherdoses of kb-NB142-70 were required to obtain similarsuppression of NF-kB activity (Supplementary Fig. S3F).To determine whether PKD suppression can alter basalexpression of NF-kB, RKO cells were stably transfectedwith the NF-kB–driven luciferase plasmid. Addition ofCRT0066101 to these cells in the absence of an inducer
resulted in a dose-dependent inhibition of basal NF-kBactivity (Fig. 3F). Although TNF-a activates the NF-kBpathway, it does not directly induce the PKC/PKDpathway (data not shown). Thus, cells preincubated withCRT0066101 were stimulated with PMA, a known PKDstimulator. As shown in Fig. 3F, PMA-induced NF-kBactivation was suppressed with the PKD inhibitor. Takentogether, our findings suggest that PKD inhibition and inparticular PKD2, results in suppression of key signalingpathways in human colorectal cancer, including AKT,ERK, and NF-kB.
Antitumor activity of PKD small molecule inhibitorsOur in vitro results suggest that PKDplays an active role
in colorectal cancer growth and proliferation. AsCRT0066101 is an orally bioavailable PKD inhibitor, wefurther evaluated the in vivo antitumor activity ofCRT0066101 using HCT116 tumor-bearing athymic nudemice. Mice were administered different oral doses ofCRT0066101 (40, 80, and 120 mg/kg) once daily for3 weeks. At the end of the 21-day treatment period, tumorvolume was decreased 55.6%, 65.2%, and 69.5%, respec-tively, as compared with control, vehicle-treated mice(Fig. 4A and C). Although the reduction in mean tumorvolume seemed to be dose dependent, this effect was notstatistically significant. As compared with the vehicle-treated group, significant antitumor activity for each dosewas achieved on day 16, day 12, and day 9, respectively.No gross toxicities were evident as determined by bodyweight, even at the highest dose (Fig. 4B). We also deter-mined the potential effect of drug treatment on PKD2activationwithin the tumor xenografts. As seen in Fig. 4D,treatment with CRT066101 resulted in a dose-dependentsuppression of p-PKD2 in the xenograft tumors. Tumorxenografts were examined for additional markers ofgrowth and apoptosis. For these studies, Ki67 expressionand the level of TUNEL and M30 staining was investi-gated. Ki67 is a well-established marker for cell prolifer-ation. Treatment with the lowest CRT0066101 dose usedin this experiment, 40 mg/kg, resulted in a significantreduction in Ki67 expression (Fig. 5). In addition, expres-sion levels of p-ERK were investigated. We observed adose-dependent inhibition of p-ERK expression, whichcorrelated with our in vitro results. The TUNEL assaydetects DNA fragmentation as a result of apoptotic sig-naling cascades. CRT0066101 treatment significantlyincreased TUNEL staining in a dose-dependent manner(Fig. 5). To verify that the increased TUNEL stainingsignified apoptosis and not necrosis, staining for M30, acaspase-cleaved cytokeratin 18, was performed. Asshown in Fig. 5, CRT0066101 administration significantlyincreased M30 staining suggesting that the PKD inhibitorinduced apoptosis in our in vivo colorectal cancer model.
DiscussionIt is now well-established that PKD signaling mediates
several key cellular processes, including DNA synthesis,
Wei et al.
Mol Cancer Ther; 13(5) May 2014 Molecular Cancer Therapeutics1136
Figure 3. Effect of CRT0066101 on downstream cell survival pathways. A, cells were exposed to CRT0066101 for 24 hours, and protein expressionwas determined by Western blot analysis. B, semiquantitative analysis of Western blots for p-ERK/total ERK. C, semiquantitative analysis of Westernblots for p-AKT/total AKT. Values represent the mean � SD from at least 3 individual blots. E, HCT116 and RKO cells were transiently transfected with aluciferase plasmid under the control of a NF-kB response element. On the following day, cells were treated with CRT0066101 for 2 hours, and thenstimulated with TNF-a for an additional 5 hours. NF-kB activity was determined by the dual-luciferase assay. F, RKO cells stably expressing NF-kB luciferasewere treated with CRT0066101 for 2 hours, and then stimulated with PMA (100 nmol/L) for an additional 5 hours. Luciferase values represent themean�SD from3 to 5 separate experiments. �,P <0.05; ��,P <0.01; ���,P< 0.001 versus untreated cells (B, C), TNF-a–stimulated cells (E), unstimulated cells(-PMA; F), or PMA-stimulated cells (F). &, P < 0.001 versus untreated cells (E).
Protein Kinase D: A Novel Target for Colorectal Cancer Therapy
www.aacrjournals.org Mol Cancer Ther; 13(5) May 2014 1137
proliferation, survival, adhesion, invasion/migration,motility, and angiogenesis (8). This pathway has also beenimplicated in a broad range of human tumors. Moreover,there is evidence suggesting that different PKD isoformsmay be associated with specific cancers (27). In the liter-ature, relatively little is known about the precise role of
PKD in colorectal cancer. For this reason, we initiallydetermined the respective mRNA and protein expressionlevels of each PKD isoform in normal human colon andcolorectal cancer cell lines. PKD1 expression was detect-able only in the normal colon cell lines. However, signif-icant expression of PKD2 and PKD3 was observed in all
Control 40 80 120 mg/kg
p-PKD2
PKD2
β-Actin
Ctrl
40
80
120
A B
C D
Figure 4. Effect of CRT0066101 onin vivo tumor growth. Athymic nudemice bearing HCT116 tumorxenografts were administered dailyvehicle control (dextrose) orCRT0066101 at 40, 80, and 120mg/kgorally for 3weeks (5micepergroup). Tumor volume (A) andbody weight (B) were measured 3times per week. C, mouse andexcised tumor images aftertreatment with CRT0066101. D,protein expression in xenograftswas determined after 3 weeks ofCRT0066101 treatment. Tumorswere harvested 2 hours after thelast dose.
Control 40 80 120 mg/kg
Ki67
p-ERK
TUNEL
M30
100μm100μm
Figure 5. In vivo biologic activity ofCRT0066101. Tumors werefixed in formalin and paraffin-embedded. Slides were stainedfor Ki67, p-ERK, TUNEL, and M30expression (�200) as described inthe Materials and Methods.
Wei et al.
Mol Cancer Ther; 13(5) May 2014 Molecular Cancer Therapeutics1138
cell lines. These findings were validated in a search of theCellMiner and CCLE databases and subsequently con-firmed with expression analysis in human colon tumorsamples. The absence of PKD1 in colorectal cancer wassomewhat puzzling as it had been previously reportedthat PKD1 is abundantly expressed in mouse intestinalcells and that expression of this particular isoform isnecessary for DNA synthesis and cellular migration(28). It is conceivable that the loss of PKD1 expressionmay be the result of epigenetic silencing as normal intes-tinal cells transform into cancer cells. In support of thispossibility, investigators have also identified loss of PKD1expression in gastric and breast cancer resulting fromepigenetic regulatory mechanisms (9, 29). With this inmind, we attempted to reexpress PKD1 with the use ofDNA methyltransferase inhibitors (decitabine andRG108) but were unsuccessful in doing so (data notshown). TheCCLEdatabase states thatHCT116 cells havea point mutation in PKD1 (S625N) and RKO cells have 2pointmutations (G779D; C853R). However, at this time, itis unclear as to the potential impact of these discrete pointmutations on protein expression and/or stability.The development and evaluation of small molecule
inhibitors directed against novel targets, such as PKD,would seem to be critically important for identifying newtherapeutic options that can ultimately improve patientresponse and outcome. To date, several PKD inhibitorshave been developed with activity against a wide varietyof tumor types (8). Our studies reveal that 2 of theseinhibitors, kb-NB142-70 and CRT0066101, exhibited sig-nificant activity against a wide range of human colorectalcancer cell lines. Interestingly, these PKD inhibitors equal-ly suppressed growth of multidrug-resistant cells as wellas p53-deficient cells, suggesting that PKD inhibitors maybe able to overcome chemoresistance and radioresistancemechanisms.With respect to the potentialmode of action, our studies
show that PKD inhibition resulted in cell-cycle arrest atthe G2–M phase, a finding that was previously demon-strated in human prostate cancer cells (15). Similarly,Kienzle and colleagues demonstrated that PKD1 andPKD2 are necessary in the G2 phase of the cell cycle inHeLa cells (30). Interestingly, siRNAknockdown of PKD3resulted in accumulation of prostate cancer cells in G0–G1
phase (12). Studies byAzoitei and colleagues showed thatglioblastoma cells accumulate in G0–G1 phase after PKD2suppression (13). Thus, in addition to cell-specific expres-sion, the different PKD isoforms may play different rolesin mediating cell division and proliferation.Using our human colorectal cancer model systems, we
observed that PKD inhibition, through the use of eithersmall molecule inhibitors or siRNAs, resulted in signifi-cant induction of apoptosis. This effect was detected byboth Annexin V/propidium iodide staining andWesternblot analysis of cleaved PARP and caspase-3. In additionto induction of apoptotic pathways, PKD inhibition sup-pressedAkt andERKsignaling. Previously, PKDhas beenshown to mediate the MEK/ERK/RSK pathway and
promote cell proliferation through a stimulatory effect onGPCR (31). The pro-proliferative effects of PKD in cancercells are associated with activation of both ERK1/2 andAKT. In human prostate cancer, PKD3 modulated boththe extent and duration of ERK1/2 activation (12). In thisstudy, we showed that the small molecule inhibitorCRT0066101 strongly suppressed activation of ERK inhuman colorectal cancer. This agent also significantlyblocked AKT activation. Transfection of PKD2 siRNAresulted in similar effects on these 2 downstream path-ways. Other key regulators of cell survival, proliferation,andmotility are the NF-kB transcription factors. Previousstudies have shown that PKD is a mediator of NF-kBinduction in a variety of cells exposed toGPCRagonists oroxidative stress (32). PKD2 gene silencing dramaticallyblocked LPA-stimulated NF-kB promoter activity in non-transformed human colonic epithelial NCM460 cells (33).However, they observed no decrease in ERK signaling.PKD2 has also been implicated in mediating NF-kB acti-vation by Bcr-Abl in myeloid leukemia cells (34). Inkeratinocytes, in which phorbol esters are major tumorpromoters, PKDs stimulate proliferation and prevent dif-ferentiation. These findings suggest that PKDs and, inparticular, PKD2 may play an important role in phorbolester-sensitive tumors, such as skin tumors and coloncancer. PMA induces NF-kB activation, and PKDs hasbeen shown to promote cell survival through activatingNF-kB signaling pathway in response to oxidative stress(8). Thus, it is conceivable that the prosurvival effects ofPKD2 in colorectal cancer cells, in response to PMAtreatment, may be partly attributable to NF-kB activation.Herein, we confirmed that suppression of PKDwith smallmolecule inhibitors leads to significant inhibition of NF-kBactivity in colorectal cancer. Taken together, these 3 keysurvival pathways, AKT, ERK, and NF-kB, each of whichare located downstream of PKD signaling, seem to becritical for the cytotoxic activity of PKD inhibitors inhuman colorectal cancer.
To date, several PKD inhibitors have been studied in invivo animal models. As one example, kb-NB142-70, dem-onstrated significantly improved in vitro activity com-pared with the structurally related CID755673, but exhib-ited no antitumor activity in vivo, which is most likelybecause of rapid metabolism (17). Alternatively, anaphthyridine-based PKD inhibitor showed PKD inhibi-tion and suppression of PKD-dependent downstreampathways in an in vivo rat model (35). The PKD inhibitorskb-NB142-70 andCRT0066101displayed similar activitiesin our in vitromodels, andCRT0066101 also demonstratedsignificant antitumor activity in 2 pancreatic animal mod-els (14). Our data confirm the antitumor activity ofCRT0066101 and further demonstrate that colon cancerxenografts are responsive to in vivo inhibition of the PKDpathway. With the colorectal cancer models that wereused in this study, we observed greater tumor suppres-sion at lower doses than previously reported, suggestingthat colon cancer may be more susceptible to blockage ofthis growth-mediating pathway. However, given the
Protein Kinase D: A Novel Target for Colorectal Cancer Therapy
www.aacrjournals.org Mol Cancer Ther; 13(5) May 2014 1139
critical roles of PKDs in both endothelial and epithelialcells, amajor concern is that the use of pan-PKD inhibitorsmay lead to off-target effects. Several PKD inhibitors havedemonstrated such off-target effects against other kinases(19, 35). It is conceivable that the orally bioavailableinhibitor CRT0066101 may bind to and inhibit otherkinases. Notably, the 2-(4-aminopyrimidin-2-yl)phenolmoiety in CRT0066101 is also present in small moleculekinase inhibitors targeting the proto-oncogene serine/threonine-protein kinase (Pim-1), checkpoint kinase 2,inhibitor of NF-kB kinase subunit b, tropomyosin-relatedkinases and others (36–39). However, specific PKD iso-form targeting with either siRNAs or antisense moleculescan suppress in vivo tumor growth (11, 40). Thus, thedevelopment of small molecule inhibitors and/or nucleicacid–based molecules, such as siRNAs, targeting specificPKD isoforms is warranted.
In conclusion, we report that small molecule inhibi-tors directed against PKD signaling exhibited potent invitro cytotoxic and in vivo antitumor activity. The bio-logic effects of these agents seem to be mediated byinhibition of PKD2 activation, G2–M phase arrest,induction of apoptosis, and inhibition of AKT, ERK,and NF-kB signaling pathways. These studies areimportant as they provide support for the potential roleof PKD as a novel target for cancer chemotherapy.Moreover, they provide a rational basis for the designand development of novel agents that may act eitheralone or in combination with presently available anti-
cancer agents to enhance clinical activity and/or over-come cellular drug resistance.
Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.
Authors' ContributionsConception and design: N. Wei, E. Chu, P. Wipf, J.C. SchmitzDevelopment of methodology: N. Wei, E. Chu, J.C. SchmitzAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): N. Wei, J.C. SchmitzAnalysis and interpretation of data (e.g., statistical analysis, biostatis-tics, computational analysis): N. Wei, E. Chu, J.C. SchmitzWriting, review, and/or revision of the manuscript: N. Wei, E. Chu, P.Wipf, J.C. SchmitzAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): J.C. SchmitzStudy supervision: N. Wei, J.C. Schmitz
AcknowledgmentsThe authors thank Drs. S. Wu and W. Yang for excellent technical
assistance and K.G. Rosenker for the preparation of kb-NB142-70.This project used the UPCI Tissue and Research Pathology Services,the Animal Facility, and the Cytometry Facility that are supported inpart by award P30CA047904.
Grant SupportResearch funds provided by a grant from theNIH/NCI (P30CA047904,
to N.E. Davidson).The costs of publication of this article were defrayed in part by the
payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.
Received October 15, 2013; revised February 21, 2014; accepted March10, 2014; published OnlineFirst March 14, 2014.
References1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer
J Clin 2013;63:11–30.2. Chu E. An update on the current and emerging targeted agents in
metastatic colorectal cancer. Clin Colorectal Cancer 2012;11:1–13.3. Kohne CH, Lenz HJ. Chemotherapy with targeted agents for the
treatment of metastatic colorectal cancer. Oncologist 2009;14:478–88.
4. Lucas AS,O'Neil BH,Goldberg RM. A decade of advances in cytotoxicchemotherapy formetastatic colorectal cancer. ClinColorectal Cancer2011;10:238–44.
5. Diaz LA Jr., Williams RT, Wu J, Kinde I, Hecht JR, Berlin J, et al. Themolecular evolution of acquired resistance to targeted EGFR blockadein colorectal cancers. Nature 2012;486:537–40.
6. Rozengurt E, Rey O, Waldron RT. Protein kinase D signaling. J BiolChem 2005;280:13205–8.
7. Rozengurt E. Protein kinaseDsignaling:multiple biological functions inhealth and disease. Physiology 2011;26:23–33.
8. LaValle CR, George KM, Sharlow ER, Lazo JS, Wipf P, Wang QJ.Protein kinase D as a potential new target for cancer therapy. BBARevCancer 2010;1806:183–92.
9. Eiseler T, Doppler H, Yan IK, Goodison S, Storz P. Protein kinase D1regulates matrix metalloproteinase expression and inhibits breastcancer cell invasion. Breast Cancer Res 2009;11:R13.
10. Trauzold A, Schmiedel S, Sipos B, Wermann H, Westphal S, RoderC, et al. PKCmu prevents CD95-mediated apoptosis and enhancesproliferation in pancreatic tumour cells. Oncogene 2003;22:8939–47.
11. Azoitei N, Pusapati GV, Kleger A, Moller P, Kufer R, Genze F, et al.Protein kinase D2 is a crucial regulator of tumour cell-endothelialcell communication in gastrointestinal tumours. Gut 2010;59:1316–30.
12. Chen J, Deng F, Singh SV, Wang QMJ. Protein kinase D3 (PKD3)contributes to prostate cancer cell growth and survival through a PKCepsilon/PKD3 pathway downstream of Akt and ERK 1/2. Cancer Res2008;68:3844–53.
13. Azoitei N, Kleger A, Schoo N, Thal DR, Brunner C, Pusapati GV, et al.Protein kinase D2 is a novel regulator of glioblastoma growth andtumor formation. Neuro Oncol 2011;13:710–24.
14. Harikumar KB, Kunnumakkara AB, Ochi N, Tong Z, Deorukhkar A,Sung B, et al. A novel small-molecule inhibitor of protein kinase Dblocks pancreatic cancer growth in vitro and in vivo. Mol Cancer Ther2010;9:1136–46.
15. Lavalle CR, Bravo-Altamirano K, Giridhar KV, Chen J, Sharlow E, LazoJS, et al. Novel protein kinase D inhibitors cause potent arrest inprostate cancer cell growth and motility. BMC Chem Biol 2010;10:5.
16. Sharlow ER, Giridhar KV, LaValle CR, Chen J, Leimgruber S, Barrett R,et al. Potent and selective disruption of protein kinase D functionalityby a benzoxoloazepinolone. J Biol Chem 2008;283:33516–26.
17. Guo J, Clausen DM, Beumer JH, Parise RA, Egorin MJ, Bravo-Alta-mirano K, et al. In vitro cytotoxicity, pharmacokinetics, tissue distri-bution, and metabolism of small-molecule protein kinase D inhibitors,kb-NB142-70 and kb-NB165-09, inmice bearing human cancer xeno-grafts. Cancer Chemother Pharmacol 2013;71:331–44.
18. Gamber GG, Meredith E, Zhu QM, Yan WL, Rao C, Capparelli M, et al.3,5-Diarylazoles as novel and selective inhibitors of protein kinase D.Bioorg Med Chem Lett 2011;21:1447–51.
19. Sharlow ER, Wilson GM, Close D, Leimgruber S, Tandon M, Reed RB,et al. Discovery of diverse small molecule chemotypes with cell-basedPKD1 inhibitory activity. PLoS ONE 2011;6:e25134.
20. Bunz F, Dutriaux A, Lengauer C, Waldman T, Zhou S, Brown JP, et al.Requirement for p53 and p21 to sustain G2 arrest after DNA damage.Science 1998;282:1497–501.
Mol Cancer Ther; 13(5) May 2014 Molecular Cancer Therapeutics1140
21. Park JG, Oie HK, Sugarbaker PH, Henslee JG, Chen TR, Johnson BE,et al. Characteristics of cell lines established from human colorectalcarcinoma. Cancer Res 1987;47:6710–8.
22. Wei N, Liu GT, Chen XG, Liu Q, Wang FP, Sun H. H1, a derivative oftetrandrine, exerts anti-MDR activity by initiating intrinsic apoptosispathway and inhibiting the activation of Erk1/2 and Akt1/2. BiochemPharmacol 2011;82:1593–603.
23. Reinhold WC, Sunshine M, Liu H, Varma S, Kohn KW, Morris J, et al.CellMiner: a web-based suite of genomic and pharmacologic tools toexplore transcript and drug patterns in the NCI-60 cell line set. CancerRes 2012;72:3499–511.
24. Barretina J, Caponigro G, StranskyN, Venkatesan K,Margolin AA, KimS, et al. The cancer cell line encyclopedia enables predictive modellingof anticancer drug sensitivity. Nature 2012;483:603–7.
25. Markowitz SD, Bertagnolli MM. Molecular origins of cancer: molecularbasis of colorectal cancer. N Engl J Med 2009;361:2449–60.
26. Guha S, Tanasanvimon S, Sinnett-Smith J, Rozengurt E. Role ofprotein kinase D signaling in pancreatic cancer. Biochem Pharmacol2010;80:1946–54.
27. Borges S, Storz P. Protein kinase D isoforms: new targets fortherapy in invasive breast cancers? Expert Rev Anticancer Ther2013;13:895–8.
28. Sinnett-Smith J, Rozengurt N, Kui R, Huang C, Rozengurt E. Proteinkinase D1 mediates stimulation of DNA synthesis and proliferation inintestinal epithelial IEC-18 cells and in mouse intestinal crypts. J BiolChem 2011;286:511–20.
29. Kim M, Jang HR, Kim JH, Noh SM, Song KS, Cho JS, et al.Epigenetic inactivation of protein kinase D1 in gastric cancer andits role in gastric cancer cell migration and invasion. Carcinogenesis2008;29:629–37.
30. Kienzle C, Eisler SA, Villeneuve J, Brummer T,OlayioyeMA,Hausser A.PKD controls mitotic Golgi complex fragmentation through a Raf-MEK1 pathway. Mol Biol Cell 2013;24:222–33.
31. Sinnett-Smith J, Zhukova E, Hsieh N, Jiang X, Rozengurt E. Proteinkinase D potentiates DNA synthesis induced byGq-coupled receptors
by increasing the duration of ERK signaling in Swiss 3T3 cells. J BiolChem 2004;279:16883–93.
32. Storz P, Doppler H, Toker A. Protein kinaseDmediatesmitochondrion-to-nucleus signaling and detoxification from mitochondrial reactiveoxygen species. Mol Cell Biol 2005;25:8520–30.
33. Chiu TT, Leung WY, Moyer MP, Strieter RM, Rozengurt E. Proteinkinase D2 mediates lysophosphatidic acid-induced interleukin 8 pro-duction in nontransformed human colonic epithelial cells through NF-kB. Am J Physiol Cell Physiol 2007;292:C767–77.
34. Mihailovic T, Marx M, Auer A, Van Lint J, Schmid M, Weber C, et al.Protein kinaseD2mediates activation of nuclear factor kBbyBcr-Abl inBcr-Ablþ human myeloid leukemia cells. Cancer Res 2004;64:8939–44.
35. Meredith EL, Ardayfio O, Beattie K, Dobler MR, Enyedy I, Gaul C, et al.Identification of orally available naphthyridine protein kinase D inhibi-tors. J Biol Chem 2010;53:5400–21.
36. Caldwell JJ, Welsh EJ, Matijssen C, Anderson VE, Antoni L, Boxall K,et al. Structure-based design of potent and selective 2-(quinazolin-2-yl)phenol inhibitors of checkpoint kinase 2. J Med Chem 2011;54:580–90.
37. EkambaramR, Enkvist E, Vaasa A, Kasari M, Raidaru G, KnappS, et al.Selective bisubstrate inhibitors with sub-nanomolar affinity for proteinkinase Pim-1. ChemMedChem 2013;8:909–13.
38. Nagarajan S, Doddareddy M, Choo H, Cho YS, Oh KS, Lee BH,et al. IKKb inhibitors identification part I: homology model assistedstructure based virtual screening. Bioorg Med Chem 2009;17:2759–66.
39. Park H, Chi O, Kim J, Hong S. Identification of novel inhibitors oftropomyosin-related kinase A through the structure-based virtualscreening with homology-modeled protein structure. J Chem InfModel 2011;51:2986–93.
40. LaValle CR, Zhang LY, Xu SP, Eiseman JL, Wang QJ. Induciblesilencing of protein kinase D3 inhibits secretion of tumor-promot-ing factors in prostate cancer. Mol Cancer Ther 2012;11:1389–99.
www.aacrjournals.org Mol Cancer Ther; 13(5) May 2014 1141
Protein Kinase D: A Novel Target for Colorectal Cancer Therapy
2014;13:1130-1141. Published OnlineFirst March 14, 2014.Mol Cancer Ther Ning Wei, Edward Chu, Peter Wipf, et al. Colorectal CancerProtein Kinase D as a Potential Chemotherapeutic Target for
Updated version
10.1158/1535-7163.MCT-13-0880doi:
Access the most recent version of this article at:
