discoveryofpotentandselectivemrckinhibitors with ...have beneﬁcial therapeutic effects in numerous...

Translational Science

DiscoveryofPotent andSelectiveMRCK Inhibitorswith Therapeutic Effect on Skin CancerMathieu Unbekandt1, Simone Belshaw2, Justin Bower2, Maeve Clarke2,Jacqueline Cordes2, Diane Crighton2, Daniel R. Croft2, Martin J. Drysdale2,Mathew J. Garnett3, Kathryn Gill2, Christopher Gray2, David A. Greenhalgh4,James A.M. Hall3, Jennifer Konczal2, Sergio Lilla5, Duncan McArthur2,Patricia McConnell2, Laura McDonald2, Lynn McGarry6, Heather McKinnon2,Carol McMenemy4, Mokdad Mezna2, Nicolas A. Morrice5, June Munro1,Gregory Naylor1, Nicola Rath1, Alexander W. Sch€uttelkopf2, Mairi Sime2,and Michael F. Olson1,7

Abstract

The myotonic dystrophy–related Cdc42-binding kinasesMRCKa andMRCKb contribute to the regulation of actin–myosincytoskeleton organization and dynamics, acting in concert withthe Rho-associated coiled-coil kinases ROCK1 and ROCK2. Theabsence of highly potent and selective MRCK inhibitors hasresulted in relatively little knowledge of the potential roles ofthese kinases in cancer. Here, we report the discovery of theazaindole compounds BDP8900 and BDP9066 as potent andselective MRCK inhibitors that reduce substrate phosphorylation,leading to morphologic changes in cancer cells along with inhi-bition of their motility and invasive character. In over 750 humancancer cell lines tested, BDP8900 and BDP9066 displayed con-sistent antiproliferative effects with greatest activity in hemato-logic cancer cells. Mass spectrometry identified MRCKa S1003 as

an autophosphorylation site, enabling development of a phos-phorylation-sensitive antibody tool to report onMRCKa status intumor specimens. In a two-stage chemical carcinogenesis modelof murine squamous cell carcinoma, topical treatments reducedMRCKa S1003 autophosphorylation and skin papilloma out-growth. In parallel work, we validated a phospho-selective anti-body with the capability to monitor drug pharmacodynamics.Taken together, our findings establish an important oncogenicrole forMRCK in cancer, and they offer an initial preclinical proofof concept for MRCK inhibition as a valid therapeutic strategy.

Significance: The development of selective small-moleculeinhibitors of the Cdc42-binding MRCK kinases reveals theiressential roles in cancer cell viability, migration, and invasivecharacter. Cancer Res; 78(8); 2096–114. �2018 AACR.

IntroductionThe actin–myosin cytoskeleton provides the structural frame-

work that determines cell shape, and also is the source of

physical force, which directly powers biological activitiesincluding adhesion, migration, and cell division. In addition,numerous processes are promoted by the actin–myosin cyto-skeleton via less direct routes, such as gene transcription andproliferation, which collectively contribute to cancer (1).Although unlikely to be a primary cancer driver, accumulatingevidence indicates that the actin–myosin cytoskeleton providesa critically important ancillary role in tumor growth and spread,which makes actin–myosin cytoskeleton regulators potentialtargets for cancer chemotherapy (2).

In nonmuscle cells, a key event in promoting actin–myosincontractility is the phosphorylation of class 2 regulatory myosinlight chains (MLC2) on Thr18 and Ser19 residues, whichactivates myosin ATP activity to drive the interaction of myosinheavy- and light-chain complexes with filamentous actin(F-actin; ref. 3). Prominent MLC2-phosphorylating enzymesare the ROCK1 and ROCK2 kinases (4), which act downstreamof the RhoA and RhoC small GTPases to regulate cytoskeletonorganization and dynamics (5). However, ROCK1 and ROCK2are not the only kinases regulated by Rho family GTPases; themyotonic dystrophy-related Cdc42-binding kinases (MRCK)interact with Cdc42 and catalyze phosphorylation of a similarset of substrates, including MLC2 (6, 7). There are three MRCKkinases; the widely expressed and closely related MRCKa andMRCKb, and the more divergent MRCKg , which is considerablymore restricted in its tissue expression.

1Molecular Cell Biology Laboratory, Cancer Research UK Beatson Institute,Glasgow, United Kingdom. 2Drug Discovery Unit, Cancer Research UK BeatsonInstitute, Glasgow, United Kingdom. 3Translational Cancer Genomics,WellcomeTrust Sanger Institute, Hinxton, United Kingdom. 4Section of Dermatology andMolecular Carcinogenesis, College of Medical, Veterinary and Life Sciences,University of Glasgow, Glasgow, United Kingdom. 5Mass Spectrometry Facility,Cancer Research UK Beatson Institute, Glasgow, United Kingdom. 6ScreeningFacility, Cancer Research UK Beatson Institute, Glasgow, United Kingdom.7Institute of Cancer Sciences, College of Medical, Veterinary and Life Sciences,University of Glasgow, Glasgow, United Kingdom.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Current address for M. Drysdale: Center for the Development of Therapeutics,Broad Institute of MIT and Harvard, Cambridge, Massachusetts; and currentaddress for M.F. Olson, Department of Chemistry and Biology, Ryerson Univer-sity, Toronto, Ontario, Canada.

CorrespondingAuthor:Michael F. Olson, Cancer Research UKBeatson Institute,Garscube Estate, Switchback Road, Glasgow, G61 1BD, United Kingdom. Phone:4414-1330-3654; Fax: 4414-1942-6521; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-17-2870

�2018 American Association for Cancer Research.

CancerResearch

Cancer Res; 78(8) April 15, 20182096

on March 20, 2020. © 2018 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 30, 2018; DOI: 10.1158/0008-5472.CAN-17-2870

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-17-2870&domain=pdf&date_stamp=2018-4-2

http://cancerres.aacrjournals.org/

The roles of MRCK signaling in normal cell function andcontributions to cancer are less well characterized than forROCK, largely due to two historical factors: ROCK kinases wereidentified before (4) the MRCK kinases (8, 9), and because ofthe discovery in 1997 of the relatively potent and selectivesmall-molecule ROCK inhibitor Y27632 (10), which hasenabled two decades of research on ROCK biology. The largebody of ROCK knowledge also catalyzed small-molecule inhib-itor discovery efforts, and ROCK inhibitors have been shown tohave beneficial therapeutic effects in numerous preclinicalcancer models (11), which has contributed to their furtherdevelopment for clinical use.

One aspect of cancer with which actin–myosin cytoskeletonregulators, including ROCK and MRCK, are clearly associated istumor cell invasion and metastasis (12). The metastatic spread ofcancer cells is the main cause of cancer mortality, believed tocontribute up to 90% of all cancer-related deaths (13). It hasbecome increasingly appreciated that the same proteins thatenable distantmetastasis also contribute toprimary tumor growth(14); therefore, drugs that restrict processes that contribute tocancer spread (e.g., motility, local invasion) also have beneficialeffects on reducing tumor growth and progression.

It has been demonstrated in several contexts that the concertedinhibition of ROCK and MRCK kinases has greater effects thanblocking either ROCK or MRCK alone (15–17). In addition,MRCK knockdown or inhibition alone was sufficient to reduce3D invasion by squamous cell carcinoma (SCC) cells (18, 19).These results suggest that there are likely to be clinical scenarios inwhich MRCK inhibitors would have therapeutic benefits, eitheralone or when combined with ROCK inhibition (20). However,the absence of potent and highly selective small-molecule inhi-bitors has restricted research on MRCK relative to the advancesmade for other kinases forwhichuseful chemical biology tools arereadily available.

To determine how MRCK contributes to biological proces-ses, including regulation of cell morphology and motility, andto evaluate MRCK as a cancer drug target, selective and potentMRCK inhibitors were developed, starting from a ligand-efficient fragment that was identified in a focused fragmentlibrary screen using an MRCKb biochemical assay. Structure-guided fragment elaboration led to the novel MRCK inhibitorsBDP8900 and BDP9066, which are considerably more potentand selective than the previously described BDP5290 (19),or the mixed ROCK-MRCK inhibitor DJ4 (17) or PKC-MRCKinhibitor chelerythrine (21). Screening of more than 750 hu-man cancer cell lines, from more than 40 different cancertypes, for antiproliferative effects of BDP8900 and BDP9066identified hematologic cancers as the most sensitive malignancytype. Mass spectrometry led to the discovery and validation ofMRCKa S1003 autophosphorylation as a pharmacodynamicbiomarker that enabled evaluation of on-target drug actionin tissues, as well as a biomarker of MRCKa activity in skintumors. BDP9066 treatment of SCC cells induced changes incell morphology, motility, and invasion, while topical treat-ment of mice in a two-stage chemical carcinogenesis modelof SCC significantly reduced papilloma growth and MRCKaS1003 autophosphorylation. These results demonstrate that thepotent and selective inhibitors BDP8900 and BDP9066 arevaluable chemical biology tools to identify MRCK biologicalfunctions and roles in cancer, and that MRCK inhibition hasin vivo therapeutic effects on skin cancer in mice.

Materials and MethodsKinase inhibition assays

MRCKa, MRCKb, ROCK1, and ROCK2 assays were performedas described (19). Recombinant kinase proteins (Life Techno-logies) at 8–12 nmol/L were incubated at room temperature for60minutes with 100 nmol/L FAM-S6-ribosomal protein–derivedpeptide (Alta Biosciences) with 1 mmol/L ATP and 0.5 mmol/LMgCl2 in 20 mmol/L Tris buffer (pH 7.4) containing 0.01% (v/v)Tween 20, 1 mmol/L DTT for MRCKa and b; or 1 mmol/L ATP,10 mmol/L MgCl2 in 20 mmol/L Tris buffer (pH 7.5) containing0.25 mmol/L EGTA, 0.01% (v/v) Triton X-100, 1 mmol/L DTTfor ROCK1 and ROCK2. Reactions were stopped by adding2 volumes 0.25%(v/v) IMAPbinding reagent in1� IMAPbindingbuffer A (Molecular Devices). After 30 minutes to allow bindingreagent to bind phosphorylated peptide, fluorescence polariza-tion was measured on a Tecan Saphire2 plate reader at excitation(470 nm) and emission (530 nm) wavelengths. Inhibition wascalculated using no inhibitor (0%) or no enzyme (100%) con-trols. MRCKa and MRCKb kinases assays by out-sourcedsupplier were carried out in 50 mmol/L HEPES pH 7.5, 0.01%(v/v) BRIJ-35, 10 mmol/L MgCl2, 1 mmol/L EGTA, and 2 mmol/LSer/Thr 13 peptide substrate, with 1-hour incubation beforeaddition of development reagent and quantification by fluores-cence resonance energy transfer (FRET). Kinase selectivity profil-ing was performed by Invitrogen with indicated concentrationsof BDP8900 or BDP9066.

Cell cultureMDA-MB-231 D3H2LN Luc cells (Caliper LifeSciences,

obtained 2006), MDA-MB-231 D3H2LN Luc TetOn MRCKb,ROCK1- and ROCK2-inducible cell lines (19), human cancer-associated fibroblasts (obtained 2008) and SCC12 human squa-mous cell carcinoma cells (gift of Erik Sahai, Crick Institute,London, United Kingdom; obtained 2006; ref. 18) were culturedas described previously (19). SCC12 organotypic invasion assayswere performed as described previously (19).

CRISPR/Cas9–mediated MRCKa, MRCKb, and MRCKabknockouts in MDA-MB-231 D3H2LN Luc cells were generatedusing the plasmid lentiCRISPR (gift of Feng Zhang; Addgeneplasmid # 52961; ref. 22). Target sequences used were: MRCKa1:GATTGGTCGAGGAGCTTTTG, MRCKa2: TGTCTGGAGAAGTG-CGTTTG, MRCKb1: TGTCGCGGCGCAGGGCCGAG, MRCKb2:CGTGGCCGAGTTCCTCGAGT. To generate double MRCKabknockout MDA-MB-231 D3H2LN Luc cells, all 4 targeting se-quences were used together.

Western blotsCell lysates and immunoblot analysis were performed as

described in ref. 19. The following antibodies were used: rabbitanti-MRCKa pS1003 antibody raised by Eurogentec usingthe peptide AcNH-KGCPG-S(PO3H2)-TGFPP-CONH2, rabbit anti-pMLC2 Thr18/Ser19 (#3674; Cell Signaling Technology), mouseanti-MRCL3/MRLC2/MYL9 (sc-28329; Santa Cruz Biotechnolo-gy), mouse anti a-tubulin (T9026; Sigma), rabbit anti-ERK2(Chris Marshall, Institute of Cancer Research, London UK),mouse anti-MRCKa (H00008476-M01; Abnova), mouse anti-MRCKb (H00009578-M03; Abnova), mouse anti-MRCKab(MANDM1 6G8; Glenn Morris, Centre for Inherited Neuromus-cular Disease, Oswestry, United Kingdom; ref. 23), mouse anti-ROCK1 (BD611136; BD Biosciences), mouse anti-ROCK2(BD61062; BD Biosciences), mouse anti-Cas9 (C15200229;

MRCK Inhibitor Discovery to Block Invasion and Tumor Growth

www.aacrjournals.org Cancer Res; 78(8) April 15, 2018 2097




Diagenode), mouse anti-FLAG (F4042; Sigma). Secondary anti-bodies used were: goat anti-mouse IgG Dylight 800 (35521;Thermo Scientific), goat anti-rabbit IgG AlexaFluor 680(A12076; Invitrogen), goat anti-rabbit IgG IR Dye 800 CW(926-32211; LI-COR Biosciences), goat anti-mouse IgG IR Dye680 (926-68020; LI-COR Biosciences).

Protein crystallization, data collection, and structuredetermination

Expression, purification, and crystallization of the kinasedomain of human MRCKb as well as subsequent ligand soakswere performed as described previously (19). After a brief soakin mother liquor supplemented with 20% ethylene glycol,crystals were flash-frozen to 100K and data were collected atbeamlines i24 (BDP8900 complex) and i02 (BDP9066 com-plex) of Diamond Light Source (Didcot). Data were processedwith XDS (24) and scaled using Aimless (25) [via xia2 (26) inthe case of the BDP8900 complex] to resolutions of 1.68Å(BDP8900) and 2.00Å (BDP9066), then phased by manualplacement of the protein component of the MRCKb ADPcomplex structure (PDB ID 4UAK) followed by rigid bodyrefinement with REFMAC (27). Models were improved throughcycles of model building with Coot (28) and refinement withREFMAC. Ligand restraints and starting coordinates were gen-erated using PRODRG (29), model validation was carried outusing Coot and MolProbity (30). Data collection and finalmodel statistics are presented in Supplementary Table S1. Allstructural images were rendered using PyMOL (PyMOL Molec-ular Graphics System, Version 1.8 Schr€odinger, LLC).

Cell viabilityTo determine the effect of BDP9066 on SCC12 cell viability, a

dose curve was established using the CellTiterGlo LuminescentCell Viability Assay (Promega). A total of 1.5 � 104 cells wereplated per well of a 96well plate. After 24 hours, medium wasremoved and cells cultured for 24 hours with medium supple-mented with DMSO vehicle or a dose range of BDP9066. Mea-surements were taken using a Tecan Safire plate reader.

Operetta high content morphology analysisSCC12 cells were plated at 1.5 � 104 per well of a 96well

plate. After 24 hours, medium was removed and cells werecultured for 2 hours with medium supplemented with DMSOvehicle or a dose range of BDP9066. Cells were fixed with 4%paraformaldehyde in PBS for 15 minutes, permeabilized with0.5% (v/v) Triton-X100 in PBS for 15 minutes, and blocked in1% (w/v) BSA in PBS for 30 minutes. Cells were stained withAlexa 568 Phalloidin (A12380; Thermo Fisher Scientific), DAPI(D9564; Sigma) and Deep Red Whole Cell Stain (H32721;Thermo Fisher Scientific). The cells were imaged and analyzedusing the Operetta high content imaging system.

ImmunofluorescenceSCC12 cells were seeded onto glass coverslips and allowed to

settle and grow overnight. Next day, medium was replaced withmedium containing 1 mmol/L BDP9066 or DMSO vehicle. After1 hour of drug treatment, cells were fixed in 4% paraformal-dehyde/PBS for 15 minutes and permeabilized with 0.5%Triton X-100/PBS for 5 minutes. F-actin was visualized byincubation with Alexa Fluor 488 Phalloidin (Thermo FisherScientific). Cells were mounted with ProLong Diamond includ-

ing DAPI (Molecular Probes). Images were taken on a Zeiss 710confocal microscope.

Random migrationSCC12 cells were plated at 2 � 104 per well of an Essen

Bioscience ImageLock 96-well plate. After 24 hours of culture,medium was replaced with medium supplemented with eitherDMSO vehicle or 0.4 mmol/L BDP9066. After 1 hour, plates wereplaced in an Incucyte Zoom apparatus to image cells every 15minutes for 6 hours. Image stacks were analyzed using ImageJwith theManual Tracking plugin, and 120 cells in 6 different wellsper condition were tracked. Data was plotted using ImageJ withthe Chemotaxis and Migration tool.

Plasmids and protein productionThe pEF-FLAG-MRCKa plasmid containing N-terminal

FLAG tagged full-length MRCKa isoform b and the pEF-FLAG-MRCKa-K106M plasmid with a mutation of Lysine106 to Methionine (K106M) were gifts from Simon Wilkinson(University of Edinburgh, Edinburgh, United Kingdom). ThepEF-FLAG-MRCKa-S1003A was generated by site directedmutagenesis of pEF-FLAG-MRCKa using Quickchange II XL(Agilent Technologies) following manufacturer's recommenda-tions. Primers used were: S1003A forward: aaaaaggatgtcctggtg-caactggctttccacct, S1003A reverse: aggtggaaagccagttgcaccagga-catccttttt. The pEF-FLAG-MRCKaDC plasmid was createdusing site-directed mutagenesis of pEF-FLAG-MRCKa by insert-ing a stop codon after nucleotide 4860 corresponding toamino acid 1620. The resulting protein is 100 amino acidsshorter than full-length MRCKa isoform b. The primers usedwere: MRCKaDC forward: gagtgctagcagtggctgattgtcagcaaggtcat,MRCKaDC reverse: atgaccttgctgacaatcagccactgctagcactc.

A pGEX2T-MLC2 plasmid was used for bacterial production ofrecombinant MLC2. BL21 bacteria were transformed withpGEX2T-MLC2 and grown on ampicillin plates, 1 colony waspicked and cultured overnight in 100 mL of LB medium supple-mented with ampicillin. The culture was diluted 1 in 10 (30 mLin 300 mL of LB þ ampicillin) and cultured for 1 hour, then50 mmol/L IPTG was added and incubated for 3 hours. Cultureswere centrifuged at 4,000 rpm for 15 minutes at 4�C and thepellet lysed in Tris-buffered saline (TBS) with 5 mmol/L MgCl2,1 mmol/L DTT, 1 mmol/L PMSF. Lysates were sonicated 3 timesfor 1 minute on ice and then centrifuged at 10,000 rpm for 10minutes at 4�C. The supernatant was collected and used for in vitroMLC2 phosphorylation assays.

MRCKa transfection and immunoprecipitationHEK293 human embryonic kidney cells were plated at 1 �

106 per 10 cm culture dish. The next day, medium was replacedwith OPTIMEM, and after 1 hour, cells were transfected inOPTIMEM with 10 mg of pEF-FLAG-MRCKa, pEF-FLAG-MRCKaDC, pEF-FLAG-MRCKa-K106M or pEF-FLAG-MRCKa-S1003A using Fugene HD (Promega). After 6 hours, mediumwas replaced with DMEM þ 10% FCS þ L-glutamine. Two daysafter the transfection, cells were placed on ice, washed in icecold PBS and lysed in 1 mL of ice-cold lysis buffer (TBS with1 mmol/L EDTA, 1% (v/v) Triton-X100, 1 mmol/L PMSF, 1�Complete protease inhibitor (Roche), 20 mmol/L NaF,20 mmol/L b-glycophosphate, 0.2 mmol/L Na3VO4, 20 mg/mLAprotinin). Lysates were then incubated on a rotating wheelfor 30 minutes at 4�C, then centrifuged at 13,200 rpm for

Unbekandt et al.

Cancer Res; 78(8) April 15, 2018 Cancer Research2098




10 minutes at 4�C and supernatants collected. The lysates werethen either used directly for Western blots or MRCKa wasimmunoprecipitated.

For immunoprecipitation, lysates were incubated with anti-Flag agarose beads (A2220; Sigma) on a rotating wheel for2 hours at 4�C. Beads were washed three times in lysis bufferby successive centrifugations at 3,000 rpm for 1 minute at 4�C.For in vitro MRCKa autophosphorylation, phosphatase assaysand MLC2 phosphorylation assays, beads were used as de-scribed in their respective sections. For mass spectrometry andWestern blots, beads were boiled at 95�C in 1% (w/v) SDS for5 minutes, centrifuged at 3,000 rpm for 2 minutes, and super-natants collected.

Cell treatment with MRCK inhibitorsSCC12 cells were plated at 5 � 105 per well of a 6-well plate.

After 24 hours, medium was replaced with medium supplemen-ted with DMSO or a dose range of BDP9066. After 2 hours, cellswere washed in PBS and lysed in 1% (w/v) SDS, 50 mmol/LTris-HCl, pH 7.4. Lysates were passed through QIAshredder spincolumns (79654; Qiagen) and analyzed by SDS-PAGE.

HEK293 cells were transfected with pEF-FLAG-MRCKa. After48hours,mediumwas replacedwithmediumsupplementedwitheither DMSO, 3 mmol/L BDP5290 or 1 mmol/L BDP9066. After 2hours, cells were washed in PBS and lysed in 1% (w/v) SDS,50 mmol/L Tris-HCL pH 7.4. Lysates were passed throughQIAshredder spin columns and analyzed by SDS-PAGE.

MDA MB 231 cells expressing doxycycline-inducible ROCK1,ROCK2, or MRCKb kinase domains were plated at 1.1 � 105 perwell of a 12-well plate. After 24 hours, cells were treated with1 mg/mL doxycycline for 18 hours to induce kinase domainexpression and then tested with BDP8900 or BDP9066 at theconcentrations indicated for 60 minutes. Cells were then wash-ed with PBS and lysed with Tris-SDS lysis buffer [50 mmol/LTris-HCl pH 7.4, 0.5% (v/v) SDS, 1� PhosStop Inhibitors(04 906 837 001; Roche), and 1� Complete Protease Inhibitors(04 693 124 001; Roche)]. Whole-cell lysates were clarified bypassing through QIAshredder spin columns and analyzed bySDS-PAGE.

Mass spectrometryFull-length MRCKa isoform b and full-length MRCKa iso-

form b K106M were expressed in HEK293 cells and immuno-precipitated with FLAG antibody in triplicate experiments.Samples were run on SDS-PAGE and gels stained using InstantBlue (Expedeon). MRCK bands were excised and digested withtrypsin (Promega) as described previously (31).

Protein tryptic digests were separated by nanoscale C18reverse-phase liquid chromatography using an EASY-nLC II(Thermo Scientific) coupled on-line to a Linear Trap Quadru-pole (LTQ) Orbitrap Velos mass spectrometer (Thermo Scien-tific) via a nanoelectrospray ion source (Sonation). The massspectrometer was operated in data-dependent acquisition. Frag-mentation was performed on the ten most intense ions usingcollision-induced dissociation (CID) with multistage activationenabled and higher energy collision dissociation (HCD) in twoseparated acquisitions.

Raw data were processed with MaxQuant version 1.5.5.1(32). Andromeda (33) peak list files (.apl) were converted toMascot generic files (.mgf) using APL to MGF Converter (www.wehi.edu.au/people/andrew-webb/1298/apl-mgf-converter).MGF

files were searched using Mascot (Matrix Science, version 2.4.1),querying the UniProt Homo sapiens database (09/07/2016;92,939 entries) plus an in-house database containing commonproteomic contaminants and the MRCKa K106M sequence.Mascot was searched assuming trypsin digestion, allowingtwo maximum missed cleavages, and with fragment ion andparent ion mass tolerances of 0.1 kDa and 15 ppm, respectively.The iodoacetamide derivative of cysteine was specified inMascot as a fixed modification. Methionine oxidation andserine, threonine, or tyrosine phosphorylation were specifiedin Mascot as variable modifications. MS/MS phosphopeptidedata were manually curated with Xcalibur Qual Browser ver-sion 2.2 (Thermo Scientific), and the MS-Product utility ofProtein Prospector v 5.12.4 (prospector.ucsf.edu/) was used togenerate theoretical product ions fragmentation series.

Raw msms.txt data files from MaxQuant and Mascot DATfiles were imported into Skyline to build a library of MRCKapeptides. Extracted ions chromatograms (XIC) of the 3 mainisotopic peaks (30 K resolution at 400 m/z) of precursor ionsfrom unmodified and phosphorylated MRCKa peptides thatcarried 2þ and 3þ charges were used for quantification of auto-phosphorylation sites. The peak areas of phosphopeptide 999–1009 that were measured in the three replicate experimentswere normalized to "Global Standard" areas derived from theunmodified MRCKa tryptic peptides 310–315, 1113–1122, and1268–1286.

In vitro MRCKa assaysFor treatments with phosphatase and in vitro autophosphor-

ylation, full-length MRCKa or mutants bound to anti-FLAGbeads were resuspended in 100 mL of phosphatase buffer withlambda protein phosphatase (P0753; New England Biolabs)following manufacturer's instructions and incubated withconstant agitation at 30�C for 1 hour. For samples treated forin vitro autophosphorylation and activity assays, beads werewashed three times in PBS and 1 time in kinase buffer (20mmol/L Tris-HCl pH 7.4, 0.5 mmol/L MgCl2, 0.01% (v/v)Tween20, 1 mmol/L DTT). For autophosphorylation, the beadswere resuspended in 98 mL of kinase buffer supplemented with2 mL of 5 mmol/L ATP. For MLC2 phosphorylation assays,the beads were resuspended in 94 mL of kinase buffer supple-mented with 2 mL of 5 mmol/L ATP and 4 mL of MLC2. ForCDC42 assays, the beads were resuspended in 97 mL of kinasebuffer supplemented with 2 mL of 5 mmol/L ATP and 1 mg ofconstitutively active CDC42 Q61L (Cytoskeleton, Inc. catalogno. C6101). The beads were then incubated with constantagitation at 30�C for 1 hour. To stop the reactions, 100 mL ofboiling 2% (w/v) SDS was added to samples, and reactions wereincubated for 5 minutes at 95�C. Samples were centrifuged at3,000 rpm for 2 minutes and supernatants were collected andanalyzed by Western blot analysis.

In vitro32P incorporationImmunoprecipitated full-length MRCKa and MRCKa K106M

bound to anti-FLAG beads were resuspended in kinase buffersupplemented with g[32 P]-labeled ATP with constant agitation at30�C for 1 hour. Samples were washed three times in kinasebuffer. Beads were resuspended in 50 mL of boiling 1% (w/v) SDSand incubated for 5 minutes at 95�C. Samples were then run onSDS-PAGE, gel was stained with Instant Blue (Expedeon) and anautoradiograph collected.





www.wehi.edu.au/people/andrew-webb/1298/apl-mgf-converter

www.wehi.edu.au/people/andrew-webb/1298/apl-mgf-converter


Dot blots and peptide competitionTwo peptides including MRCKa isoform b S1003 were synthe-

tized by Eurogentec. One peptide contained phosphorylatedserine (P) ¼ AcNH-KGCPG-S(PO3H2)-TGFPP-CONH2, the othercontained nonphosphorylated serine (U) ¼ AcNH-KGCPG-S-TGFPP-CONH2. A dose range of peptides P and U were spottedon Protran nitrocellulose membrane (Sigma) presoaked in TBSþ0.01% (v/v) Tween20 (TBST), and the membrane was dried.When dry, the membrane was blocked in TBST þ 0.5% (w/v)BSA for 30minutes and incubated overnight at 4�CwithMRCKa-pS1003 antibody at 1:500 in TBST þ 0.5% (w/v) BSA. Blots wereimaged using SuperSignal West Femto Maximum SensitivitySubstrate (34095; Thermo Scientific).

For peptide competition, 2 mL of MRCKa-pS1003 antibodyin 1 mL of TBST þ 0.5% (w/v) BSA was incubated with 4 ngpeptide P or 4 ng peptide U, and incubated for 2 hours.Solutions were centrifuged at 13,200 rpm for 10 minutes,supernatants collected, and reincubated with 4 ng peptide P or4 ng peptide U for 2 hours. The solutions were centrifuged at13,200 rpm for 10 minutes. Antibodies were then applied toWestern blots of immunoprecipitated MRCKa.

Cancer cell line screenThe Translational Cancer Genomics team at the Wellcome

Trust Sanger Institute screened 757 cancer cell lines withBDP8900 and BDP9066. Cells were seeded in 1,536-wellplates and compounds were screened using a 7-point dose–response curve and half-log compound dilution series. Cellviability was measured using CellTitre Glo (Promega).Detailed information about the screening process can beobtained at www.cancerrxgene.org.

IHCFormalin-fixed, paraffin-embedded sections were dewaxed and

rehydrated. Antigen retrieval was performed in a boiling waterbath with 10 mmol/L citric acid buffer at pH 6 for 20 minutes.After cooling, endogenous peroxidase was blocked with 3% (v/v)H2O2 for 15 minutes. Sections were subsequently blocked inPBSþ 0.5% (v/v) goat serum for 30 minutes. Primary antibodieswere incubated overnight at 4�C in PBSþ 0.5% (v/v) goat serum:rabbit anti-MRCKa (ab96659; Abcam) was used at 1:100, rabbitanti-MRCKb (HPA022821; Sigma) was used at 1:1,000, rabbitanti-MRCKa pS1003 was used at 1:50. Envision þ SystemHRPlabeled Polymer (Dako) and LiquidDABþ Substrate Chromogen(Dako) were used for visualization. For staining quantifications,sections were imaged using a Leica SCN 400f scanner. Intensitythresholds were defined for absence of staining, low intensitystaining,medium intensity staining, andhigh intensity staining ofMRCKa pS1003 IHC pictures using Leica Slidepath image anal-ysis software. Areas of normal skin and of papillomawere selectedfor each sample and analyzed. Automatic cell detection wasperformed and the cytoplasmic intensities of the MRCKa pS1003staining were measured. Histoscores were measured by multiply-ing the percentage of low intensity staining in the area by 100 andadding it to the percentage of medium intensity staining in thearea multiplied by 200 and adding it to the percentage of highintensity staining in the area multiplied by 300.

Mouse modelsParental MDA MB 231 D3H2LN Luc cells or MRCKab KO

cells (3.5 � 106) were subcutaneously injected in the right flank

of CD-1 immunocompromised nude mice. Mice were culledand sampled 2 weeks postinjection before tumors reached12-mm diameters.

Mouse genetic skin cancer models expressing activatedHras alone, or with c-Fos plus conditional deletion of Pten, inepidermal keratinocytes form papillomas that convert to car-cinoma as described previously (34–36).

To study the effect of BDP9066 on MRCKa pS1003 stain-ing in mouse skin, 5 FVB mice were treated topically with50 mL of 80% (v/v) DMSO (vehicle) and 5 FVB mice weretreated topically with of 25 mg of BDP9066 in 50 mL 80% (v/v)DMSO. Treatments were applied 4 times per mouse over2 days. Mice were culled 2 hours after final treatment. Skinand blood BDP9066 concentrations were measured byPharmidex. To study BDP9066 pharmacokinetics on mouseskin and blood, FVB mice were treated topically once with10 mg or 25 mg BDP9066 in 80% (v/v) DMSO, 4 times dailywith 25 mg BDP9066 in 80% (v/v) DMSO, or 8 times daily(4 successive days on, 2 days off, then 4 days on) with 80%(v/v) DMSO or 25 mg BDP9066 in 80% (v/v) DMSO. Micewere culled 2 hours, 4 hours, 8 hours, or 24 hours after finaltreatment. Skin and blood BDP9066 concentrations weremeasured by Pharmidex.

For DMBA/TPA experiments, 40 FVB mice were treated topi-cally with 25 mg DMBA in acetone on day 1. From day 5, 4.7 mgTPA in acetone was applied 3 times weekly. From day 5, 20 micewere treated with 25 mg BDP9066 in 50 mL 80% (v/v) DMSO and20 mice were treated with 50 mL 80% (v/v) DMSO 5 times perweek. Experimenters were fully blinded to the identity of treat-ment groups. Mice weights and general conditions were moni-tored at least 2 times per week, tumor sizes and numbers wererecorded weekly. Mice were culled when papillomas reached12 mm in diameter. Analysis of tumor volume and papillomanumbers were performed blinded to the identity of treatmentgroups. Skin and blood BDP9066 concentrations were measuredby Pharmidex.

All mouse experiments have been conducted in accordancewith UK HomeOffice regulations, and with the approvals of theCancer Research UK Beatson Institute Animal Care and UseCommittee (IACUC) and University of Glasgow Ethical Review.

Statistical analysisData analysis was performed using GraphPad Prism. Statistical

tests used for each data set are indicated in respective figurelegends.

ResultsDiscovery and characterization of MRCK inhibitorsBDP8900 and BDP9066

A focused fragment library was screened using an MRCKbbiochemical assay (16, 19), and structure-guided elaborationof a ligand-efficient 7-azaindole-3-carbonitrile fragment hit(Fig. 1A), prioritizing the identification of MRCK-selectiveinhibitors, led to the discovery of BDP8900 and BDP9066(Fig. 1A). Synthesis routes and methods for BDP8900 andBDP9066 will be described in a subsequent manuscript (inpreparation). When the enzyme activities of MRCKa, MRCKb,and the closely related AGC family kinases ROCK1 and ROCK2,were assayed in vitro at 1 mmol/L ATP, which approached ATPKm values for these kinases under identical assay conditions

Unbekandt et al.




www.cancerrxgene.org


(Supplementary Table S2), both compounds induced dose-dependent inhibition with greater selectivity for the MRCKkinases relative to ROCK kinases (Fig. 1B and C). Using exper-imentally determined ATP Km values and the Cheng–Prusoffequation (37), calculated Ki values revealed affinities forMRCKa and MRCKb that were 318- to 2,371-fold greater for

both compounds than for ROCK1 or ROCK2 (Supplemen-tary Table S2). Inhibition of myosin light-chain phosphoryla-tion (pMLC2) in MDA-MB-231 human breast cancer cells,engineered to express tetracycline-inducible MRCKb, ROCK1,or ROCK2 kinase domains (19), by BDP8900 was >562 timesmore selective for MRCKb relative to ROCK1 or ROCK2

pMLC2

pMLC2

pMLC2

MR

CK

βR

OC

K1

RO

CK

20 .01.03 .1 .3 1 0 .01 .03 .1 .3 1

0 0.3 1 3 10 30 0 0.3 1 3 10

0 0.3 1 3 10 30 0 0.3 1 3 10 30

30

MR

CK

βR

OC

K1

RO

CK

2

BDP8900 (μmol/L) BDP9066 (μmol/L)

0 0.3 1 3 10 30 0 0.3 1 3 10

0 0.3 1 3 10 30 0 0.3 1 3 10

0 .01.03 .1 .3 1 0 .01 .03 .1 .3

30

30

1

αTubulin αTubulin

αTubulin

αTubulin

αTubulin

αTubulin

pMLC2

pMLC2

pMLC2

BDP8900 BDP9066

A

B C

ED

-10 -9 -8 -7 -6 -5 -40

20

40

60

80

100

BDP8900 (Log10 mol/L)

Inhi

bitio

n (%

)

MRCKMRCK

ROCK1ROCK2

-10 -9 -8 -7 -6 -5 -40

20

40

60

80

100


Inhi

bitio

n (%

)

MRCKMRCK

ROCK1ROCK2

-8 -7 -6 -50

25

50

75

100

125

BDP8900 (log10 mol/L)

Rel

ativ

e pM

LC (%

)

Rel

ativ

e pM

LC (%

)

MRCK EC50 = 43 nmol/LROCK1 EC50 = 33 mol/LROCK2 EC50 = 24 mol/L

-8 -7 -6 -50

25

50

75

100

125


MRCK EC50 = 20 nmol/LROCK1 EC50 = 6.0 mol/LROCK2 EC50 = 2.0 mol/L

N

NHN

NH N SN

NHN

N

NNH N

NHN

Fragmenthit

Figure 1.

Inhibition ofMRCK activity in vitro and incells by BDP8900 and BDP9066. A,Structures of 7-azaindole-3-carbonitrilehit fragment, BD8900 and BDP9066.B, BDP8900 dose–response curves forinhibition of MRCKa, MRCKb, ROCK1,and ROCK2 kinase activity in vitro at 1mmol/L ATP. C, BDP9066 dose–response curves for inhibition ofMRCKa, MRCKb, ROCK1, and ROCK2kinase activity in vitro at 1 mmol/L ATP.Results shown are mean � SD ofduplicate independent replicates. D,Cells expressing doxycycline-inducedMRCKb, ROCK1, or ROCK2 kinasedomains were treated with BDP8900 atindicated concentrations for 60minutesprior to lysis and quantitative Westernblotting. Inhibition of MLC2phosphorylation by BDP8900 for eachinduced kinase domain. Results shownare mean � SEM of 3–4 independentreplicates. E, Cells expressingdoxycycline-induced MRCKb, ROCK1, orROCK2 kinase domains were treatedwith BDP9066 at indicatedconcentrations for 60 minutes prior tolysis and quantitative Western blotting.Inhibition of MLC2 phosphorylation byBDP8900 for each induced kinasedomain. Results shown aremean� SEMof 3–4 independent replicates.






(Fig. 1D), and by BDP9066 was >100 times more selective forMRCKb relative to ROCK1 or ROCK2 (Fig. 1E).

To extend selectivity profiling, a larger panel of 115 kinaseswas screened with 1 mmol/L BDP8900 or BDP9066 eitherfor inhibition of kinase activity at the apparent Km for ATP ofeach kinase, or for competitive inhibition of binding to theATP site (Supplementary Table S3). Kinase inhibition resultswere mapped onto an annotated human kinase phylogenetictree (38) using KinMap (Supplementary Fig. S1; ref. 39), anddisplayed as a heatmap (Fig. 2A). The majority of kinaseswere unaffected by BDP8900 or BDP9066 [indicated by greencircles with labels, with >75% inhibition (red circles) of theAGC kinases PKCg and DMPK by BDP8900 (SupplementaryFig. S1A; Fig. 2A and B)], and ROCK1, ROCK2, CDK2/Cyclin A,CDK9/(Cyclin K or Cyclin T1), PKCa, TSSK1 (STK22D) andDMPK by BDP9066 (Supplementary Fig. S1B; Fig. 2A and B). Tomore precisely characterize the selectivity of BDP8900 andBDP9066, dose–response assays were performed against kinasesidentified by the screen as comparatively sensitive (Fig. 2B, reddots). Under the assay conditions, BDP8900 and BDP9066 weremarkedly selective against all kinases tested (Fig. 2C and D).Using ATP Km values and the Cheng–Prusoff equation (37),calculated Ki values from this screen revealed minimum fold-selectivity for MRCKb that was 43 times greater for BDP8900(Supplementary Table S4) and 27 times greater for BDP9066against all kinases tested (Supplementary Table S5).

The myotonin–protein kinase, also known as DMPK, has highhomology to MRCKa and MRCKb with over 61% amino acididentity in its kinase domain compared with both their kinasedomains (80.8% identity between MRCKa and MRCKb). A com-petitive LanthaScreen Eu time-resolved fluorescence resonanceenergy transfer (TR-FRET) kinase binding assay in whichBDP8900 and BDP9066 were assayed in vitro for displacementof a fluorescent tracer that binds to the DMPK ATP-binding sitewith 43 nmol/L Kd (Fig. 2E) yielded 0.80 nmol/L (BDP8900) and0.98 nmol/L (BDP9066) Kd values using the Cheng–Prusoffequation (37).

Given the >1.0 Hill slopes (�1.9) for the MRCKb inhibitioncurves for BDP8900 (Fig. 2C) and for BDP9066 (Fig. 2D), andthe lower apparent potencies of both MRCK inhibitors in thisassay format relative to the potencies observed assay usingoptimized conditions established in-house (Fig. 1B and C),the selectivities of BDP8900 and BDP9066 were likely under-estimated. Plotting the natural log Ki values determined in-house (Fig. 2F, red dots) against out-sourced values (Fig. 2F,black dots) indicated that the differences between in-house andoutsourced Ki values for ROCK1 inhibition were only 4.4-foldfor BDP8900 and 2.9-fold for BDP9066, but for MRCKb dif-fered by 41-fold for BDP8900 and 59-fold for BDP9066,consistent with on-target compound potencies being under-valued due to insufficient sensitivity of the out-sourced assays.While numerous parameters were rigorously optimized tomaximize the sensitivity of the in-house kinase assays, includ-ing detection method, pH, MgCl2 and substrate concentrations,the out-sourced kinase assay conditions were not necessarilyoptimized for each kinase, but were generic for wide use againstmany different kinases. When compared against MRCKb Ki

values determined in-house (Fig. 2F, red dots), the affinity ofBDP8900 was more than 2,083 times greater, and of BDP9066was more than 1,635 times greater, relative to all other kinasestested for inhibition of activity (Fig. 2F, black dots), and 33-fold

higher than the DMPK Kd determined for BDP8900 and42-fold higher than for BDP9066 (Fig. 2F, purple dot). Takentogether, these results reveal that BDP8900 and BDP9066are potent and highly selective inhibitors of MRCK with robuston-target actions in vitro and in whole cells.

Structures of BDP8900 and BDP9066 associated withMRCKb kinase domain

To investigate the binding mode of these azaindole com-pounds, crystal structures of MRCKb in complex with BDP8900and BDP9066 were determined to 1.68Å and 2.00Å resolutions,respectively (Supplementary Table S1). The overall proteinconformations in these complexes were essentially identical tothose seen previously, with RMSD values after superpositiononto all PDB-deposited MRCKb structures (16, 19) rangingfrom 0.5Å to 1.5Å for �395 Ca atoms. As noted previously(19), the glycine-rich loop (specifically residues 84–88) showedmost conformational diversity. While in the MRCKb�BDP9066complex structure it was in the closed state observed for otherinhibitor complexes (16, 19), the MRCKb�BDP8900 complexshowed an open conformation similar to that described forthe ADP complex structure (19). This was surprising given thesimilarity between the two compounds in structure (Fig. 1A)and binding mode (see below). It should be noted that in bothcomplexes there were hints of electron density supporting the"other" loop conformation, suggesting that the glycine-richloop is somewhat flexible and able to switch between closedand open states in the presence of azaindole inhibitors. It ispossible that crystal-to-crystal variation, rather than ligandfeatures, determine the apparently preferred state.

A second region of conformational divergence was formedby the C-terminal end of the activation loop (around Val235)and the aEF/aF loop (around Met255), though again boththese loops were relatively flexible in all MRCKb ligand com-plexes and there is no clear connection between the contentsof the ATP-binding site and the major loop conformations.

BDP8900 and BDP9066 adopt highly similar poses in the ATP-binding site (Fig. 3A andB): the all-atom coordinate RMSD for thetwo ligands after superposition of the two complex structures onthe protein residues highlighted in Fig. 3A was 0.105Å for 21comparable atoms (i.e., excluding the thiazole/pyrimidine). Theazaindole acts as the hinge binder, forming hydrogen bonds withthe backbone carbonyl of Asp154 and the backbone amine ofTyr156. Both ligands preserved the two "pocketwaters" seen in theBDP5290 complex (19) and satisfy the hydrogen bonding poten-tial of the outer (left in Fig. 3A and B) water by accepting an H-bond to the thiazole (BDP8900) or pyrimidine (BDP9066)nitrogen. In addition, the second pyrimidine nitrogen ofBDP9066 accepts a hydrogen bond of relatively poor geometryfrom Lys105.

The spiro moiety of the azaindole ligands provides a `plug'for the ATP-binding site opening, interacting with residuesaround the glycine-rich loop (Ile82, Gly83, Arg84, Val90), thecatalytic loop (Asp204, Asn205, Leu207) as well as Phe370(Fig. 3C). In the MRCKb�BDP8900 complex, the open confor-mation of the glycine-rich loop results in a loss of the inter-action with Arg84. In addition to the mostly hydrophobic directprotein-ligand contacts, the ring nitrogen of the distal piperi-dine formed hydrogen bonds with (and orders) two watermolecules, one of which (on the right in Fig. 3C) providedindirect interactions with the backbone of Asp204 and the side

Unbekandt et al.





BA

C

Inhi

bitio

n at

1 μ

mol

/L (%

)

BDP8900 BDP9066AKT1

AMPK A1/B1/G1AXLBTK

CAMK1DCAMK4

FESGRK5JAK2

MAP3K9MLCK2MYLK2

PLK3SRPK1

STK17ASYKZAK

MLCKPDK1IKKIKKß

IRAK4LIMK1DDR1

EEF2KERK1HIPK3

MAPK8PIM1

FGFR3TYK1AKT2

MAPK9mTOR

RONDAPK3ErbB1ERK2MKK6

PHKG2Aurora A

cRAF (DD)PIK3CA/PIK3R1

STK23TBK1ABL1

DYRK1AINSRRPLK1

TAOK3CLK1

RIPK2ALK4ALK5ALK2

BRAF V600ENEK2

AMPK A2/B1/G1MAP3K7

FYNBMX

CHK1PDGFRA

IGF1RZAP70

FAKcMET

CSNK1A1EPHA1

KITCHK2

LYN APDGFRß

LCKFLT1

MAPK14SRC

DAPK1SGK1

PRKD2STK3MEK1PAK2WEE1

RPS6KA5GSK3

PKN1PAK1

GSK3ßSLK

CDK8/cyclin CCDK7/cyclinH/MNAT1

CaMKIIßFGFR1

PKACSF1RPRKX

CDK5/p25CDK5/p35

FLT3RPS6KB1

CDK1/cyclin BMARK1

CDK2/cyclin ARPS6KA1

STK22DCDK9/cyclin K

PKCROCK2

PKCCDK9/cyclin T1

ROCK1DMPK

MRCKMRCKß

0

10

20

30

40

50

60

70

80

90

100

D-10 -9 -8 -7 -6 -5

0

20

40

60

80

100


Inhi

bitio

n (%

)In

hibi

tion

(%)

CDK1/cyclinBCDK2/cyclinA

PKCPKCROCK1

STK22D

MRCK

CDK9/cyclin T1

-10 -9 -8 -7 -6 -50

20

40

60

80

100

BDP9066 (Log10mol/L)

MRCK

CDK1/cyclin BCDK2/cyclin APKC

PKC

ROCK1STK22D

CDK9/cyclin T1

E

0 25 50 75 1000

25

50

75

100

Inhibition by 1 mmol/L BDP8900 (%)

Inhi

bitio

n by

1 m

mol

/LB

DP9

066

(%) CDK1/cyclin B

CDK2/cyclin ASTK22D

PKC

ROCK2PKC

CDK9/cyclin T1ROCK1

MRCKMRCK /DMPK

BDP8900 Ki(log nmol/L)e

F

BD

P906

6 K

i(lo

gnm

ol/L

)e

In-house assayOut-sourced assay

-10 -9 -8 -7 -6 -50

20

40

60

80

100

Compound (Log10mol/L)

DM

PK b

indi

ng in

hibi

tion

(%)

BDP8900 Kd = 0.80 nmol/LBDP9066 Kd = 0.98 nmol/L

Competitivebinding assay

-4 -2 2 4 6 8

-4

-2

2

4

6

8

MRCKMRCK

ROCK1ROCK2

MRCK

CDK9/cyclin T1

PKCPKC

ROCK1STK22D

DMPK

Figure 2.

Detailed selectivity profiles for BDP8900 and BDP9066. A, Percentage kinase inhibition by 1 mmol/L BDP8900 and BDP9066 were ranked anddisplayed by heatmap. B, Plot of inhibition by BDP8900 and BDP9066. Red dot, kinases selected for detailed dose–response analysis. ROCK1 wasselected to additionally represent ROCK2 (black dot). Dotted line, Deming regression (slope deviation from zero, P < 0.0001). C, BDP8900 dose–responsecurves for inhibition of indicated kinases in vitro. Results shown are mean � SD of duplicate independent replicates. D, BDP9066 dose–responsecurves for inhibition of indicated kinases in vitro. Results shown are mean� SD of duplicate independent replicates. E, BDP8900 and BDP9066 dose–responsecurves for competitive inhibition of tracer binding to DMPK in vitro, with calculated Kd values. Results shown are mean � SD of duplicate independentreplicates. F, Natural log Ki or Kd (nmol/L) values of BDP8099 and BDP9066 determined using in-house optimized assay conditions (red dots) byoutsourced assays (black dots) or competition binding assay (purple dot). Dashed line, Deming regression (slope deviation from zero, P < 0.0001).






chain of Asp160. The other water was more exposed to bulksolvent and accordingly appears less well bound and lessconsistent in its interactions, which can involve Asp204 and(via additional ordered waters) Arg84, Asn205, and Asp218.

Comparing the binding modes of the azaindole ligands andthe previously described pyrazole amide BDP5290 (19) revealsthat most binding interactions are maintained across compoundseries despite their unrelated chemical structures and bindingmodes (Fig. 3D). The azaindole and pyrazole cores form equiv-alent interactions with the hinge. Pocket water hydrogen bond-

ing duties fulfilled by the amide linker in BDP5290 are takenover by the heterocyclic 3-substituent (thiazole or pyrimidine),which also partly fills the space occupied by the chloropyrazoleof BDP5290. The diazaspiro[5.5]undecane substituent of theazaindole compounds provided an approximate replacementfor the piperidine of BDP5290, although it appears that the shiftin location and nitrogen position of the terminal piperidinemoiety allowed BDP8900/9066 to form more extensive inter-actions with the protein, which may contribute to the higheraffinity of these compounds for MRCKb. Another notable

AD154

Y155

Y156

T137 F219

E124

D218

K105

D204D160

D154

Y155

Y156

T137 F219

E124

D218

K105

D204D160

BD154

Y155

Y156

T137 F219

E124

D218

K105

D204D160

D154

Y155

Y156

T137 F219

E124

D218

K105

D204D160

D

BDP5290

BDP9066

D154

Y155

Y156T137

F219

E124

D218

F370

C

R84

G83

I82

V90

F370

D160

L207D204

N205

Figure 3.

Structure of MRCKb in complex withBDP8900 and BDP9066. BDP8900(A) or BDP9066 (B) bound to theMRCKb ATP-binding site in stereoviews. Protein residues (gray) arelabeled with the single-letter aminoacid code and residue number;selected water molecules areindicated by red spheres. Ligands areshown in purple with sA-weighted|Fo|-|Fc|,fc electron density mapscalculated prior to the initial inclusionof the ligand in refinement contouredat 3.0s (dark blue). Potentialhydrogen bonds are highlighted bydotted black lines. C, Detailed viewof the BDP9066 spiro moiety. Theligand is shown in cyan with asemitransparent surface. D,Comparison of the binding modes ofBDP9066 (cyan) and BDP5290 (blue,PDB ID 4UAL). Protein residues,pocket waters, and potentialhydrogen bonds are shown as beforeand are colored to match the ligand.

Unbekandt et al.





difference between the two series is that, unlike the 2-pyridyl-pyrazole of BDP5290, the smaller azaindole of BDP8900/9066did not displace the Phe370 side chain on binding (Fig. 3D),avoiding energetic penalties associated with this conformationalchange and preserving a more compact ATP-binding site con-formation for the spiro moiety to interact with, which couldcontribute to the improved affinity of the azaindoles comparedto BDP5290.

We previously proposed the ligand interaction with thepocket waters as a determinant of selectivity for MRCK overROCK, and this appears to hold true when comparingBDP8900 and BDP9066. The hydrogen-bonding geometrybetween the thiazole of BDP8900 and the outer pocket wateris closer to ideal than that seen for the pyrimidine of BDP9066,and correspondingly the former compound exhibit greaterselectivity.

Effect of MRCK inhibitors on cancer cell line viabilityTo identify cancer cell lines with significant sensitivity to the

antiproliferative effects of MRCK inhibitors as single agents, 757cancer cell lines were screened using a 7-point dose range upto 10 mmol/L of either BDP8900 or BDP9066, and viabilitymonitored after 72 hours as described in ref. 40. By comparingcell line EC50 values within a given cancer type against all cancercell lines (Supplementary Table S6), significantly (P < 0.05)sensitive (green) and resistant (red) cancer types were identified.Figure 4A shows natural log EC50 values (mmol/L) of BDP8900for each cell line within each cancer type, with purple linesindicating mean EC50, arranged from low to high mean EC50.Relative to the effect of BDP8900 on all cancer cell lines together(blue), there were 10 cancer types significantly sensitive and6 cancer types significantly resistant. Similarly, 8 cancer typeswere significantly sensitive and 4 were significantly resistant toBDP9066 relative to all cancer cells (Fig. 4B). When the meanEC50 values of BDP8900 and BDP9066 for each cancer typewere plotted, Deming regression analysis revealed a strongcorrelation (P < 0.001) and a line slope that approached 1.0,with common sensitive cancer types (green dots) and resistantcancer types (red dots) relative to all cancers (blue dots)indicated (Fig. 4C). Plotting the EC50 values of BDP8900 andBDP9066 for each cell line also revealed a strong correlation(P < 0.0001) by Deming regression and a line slope that appro-ached 1.0 (Supplementary Fig. S2). These results strongly sup-port the conclusion that the effect of both inhibitors on cancercell proliferation is due to a common mechanism of actionthat is highly likely to be on-target inhibition of MRCK activ-ity. Interestingly, all 8 cancer types sensitive to both BDP8900and BDP9066 were hematologic cancers, with 2 additionalblood cancers being significantly sensitive to one compoundand others clustering, albeit not reaching statistical significance,to the sensitive end of the rankings.

MRCKa S1003 autophosphorylation as a biomarkerof activity

An additional strategy to identify cancer types that poten-tially would benefit from MRCK inhibitor treatment is tosearch for indications of increased MRCK signaling. By iden-tifying phosphorylation sites that could be used as surrogatemarkers of activity, it was previously shown that ROCK1S1333 phosphorylation (41) and ROCK2 S1366 phosphory-lation (42) were associated with kinase activity, and that

phosphorylation-state sensitive antibodies could be used toassess ROCK activity in clinical cancer samples (43). Whenwild-type MRCKa, or a kinase-dead version with a K106Mmutation that prevents ATP binding, were expressed andimmunoprecipitated from HEK293 cells (Fig. 5A, left), in vitrokinase assays with [g-32P] ATP revealed considerable32P labeling of wild-type MRCKa, but not MRCKa K106M(Fig. 5A, right). To identify autophosphorylation sites,wild-type MRCKa and MRCKa K106M were expressed andimmunoprecipitated from HEK293 cells (Fig. 5B, left), andphosphorylations measured by mass spectrometry (Fig. 5B,right). A total of 22 phosphorylations were identified onMRCKa (Supplementary Table S7), of which 8 were not detec-ted in MRCKa K106M, including S1003 (Fig. 5B, y7 m/z peaksin right; Supplementary Table S8), consistent with thesephosphorylations being autocatalyzed. S1003 is locatedbetween the coiled-coil (CC) and phorbol ester/diacylgly-cerol-binding C1 domains (Fig. 5B, right inset). A comparisonwith MRCKb revealed no conservation of MRCKa auto-phosphorylated residues (Supplementary Fig. S3).

A polyclonal phosphospecific antibody against pS1003 was rais-ed using the phospho-peptide (P) AcNH-KGCPGpSTGFPP-CONH2.Dot blots with varying quantities (10–500 pg) of phosphor-ylated (P) or unphosphorylated peptides (U) confirmed thephosphorylation-selective binding of this antibody (Fig. 5C).Detection of immunoprecipitated FLAG-tagged MRCKa byWestern blotting was blocked by preincubation of the anti-body with phosphopeptide P, but not peptide U (Fig. 5D).Incubation of immunoprecipitated FLAG-tagged MRCKawith lambda phosphatase also led to loss of pS1003 antibodyreactivity (Fig. 5E). Immunoreactivity with the pS1003 anti-body was detected on immunoprecipitated FLAG-taggedMRCKa, but not kinase-dead K106M or nonphosphory-latable S1003A mutants, consistent with this site being auto-phosphorylated (Fig. 5F).

To establish whether MRCKa S1003 autophosphorylationcould be used as a MRCKa activity biomarker, recombinantconstitutively-active CDC42 Q61L was incubated with immu-noprecipitated FLAG-tagged MRCKa, which resulted in a sig-nificant approximately 30% increase in S1003 autophosphor-ylation (Fig. 5G). Treatment of HEK293 cells expressing FLAG-tagged MRCKa with DMSO vehicle, 3 mmol/L BDP5290(19) or 1 mmol/L BDP9066 MRCK inhibitors blocked pS1003immunoreactivity (Fig. 5H), indicating that the phospho-selective antibody could be used as a surrogate marker ofMRCKa activity.

To determine the mechanism of MRCKa S1003 autopho-sphorylation, a FLAG-tagged MRCKa mutant lacking the last100 amino acids (MRCKaDC), or the kinase-dead K106Mmutant, were expressed and purified from HEK293 cells. Fol-lowing FLAG-immunoprecipitation and lambda phosphatasetreatment to reverse S1003 phosphorylation, purified proteinswere incubated with Mg2þ-ATP to allow autophosphorylation.While the faster-migrating MRCKaDC was able to autopho-sphorylate S1003 in cis when alone or with K106M, there wasno evidence of trans phosphorylation of the larger K106Mmutant by active MRCKaDC (Fig. 5I). Therefore, MRCKa uti-lizes an intramolecular mechanism for S1003 autophosphor-ylation. When dephosphorylated MRCKa (Fig. 5J, top) wasallowed to phosphorylate itself and recombinant MLC2 in thesame in vitro reaction, both autophosphorylation (Fig. 5J,






bottom left) and MLC2 phosphorylation (Fig. 5J, bottom right)were blocked by BDP9066, supporting the relationshipbetween MRCKa S1003 autophosphorylation and kinaseactivity.

To characterize whether S1003 autophosphorylation isrequired for substrate phosphorylation, FLAG-tagged MRCKa,kinase-dead K106M and nonphosphorylatable S1003A mutants

were immunoprecipitated from HEK293 cells and assayedfor MLC2-phosphorylating activity in vitro. While kinase-deadK106M was significantly less active than wild-type MRCKa orthe S1003A mutant, there was no significant difference in activitybetween wild-typeMRCKa or the S1003Amutant (Fig. 5K). Theseresults indicate that S1003 autophosphorylation is not requir-ed for MRCKa activation and substrate phosphorylation.

AB

DP8

900

EC(lo

gmm

ol/L

)50

eB

DP9

066

EC(lo

gmm

ol/L

)50

e

BD

P906

6 EC

(log

mmol

/L)

50e

BDP8900 EC (log mmol/L)50 e

B

C

Tce

llle

ukem

iaAc

ute

mye

loid

leuk

emia

Lym

phoi

dne

opla

smot

her

Lym

phob

last

icle

ukem

iaB

cell

lym

phom

aH

emat

opoi

etic

neop

lasm

othe

rLe

ukem

iaB

urki

tt's

lym

phom

aH

airy

cell

leuk

emia

Lym

phob

last

icT

cell

leuk

emia

Bce

llle

ukem

iaC

hron

icm

yelo

idle

ukem

iaM

yelo

ma

Fibr

osar

com

aSo

fttis

sue

othe

rH

odgk

in's

lym

phom

aM

edul

lobl

asto

ma

Cho

ndro

sarc

oma

Stom

ach

Ost

eosa

rcom

aN

euro

blas

tom

aLu

ngN

SCLC

SCC

Rha

bdom

yosa

rcom

aLu

ngN

SCLC

larg

ece

llH

ead

and

neck

Ewin

gsa

rcom

aK

idne

yAl

lcan

cers

Glio

ma

Bla

dder

Cer

vix

Mel

anom

aLu

ngsm

allc

ellc

arci

nom

aEn

dom

etriu

mEs

opha

gus

Lung

NSC

LCca

rcin

oid

Mes

othe

liom

aPr

osta

teLa

rge

inte

stin

eLi

ver

Ova

ryLu

ngN

SCLC

aden

ocar

cino

ma

Thyr

oid

Bre

ast

Panc

reas

Bili

ary

trac

t-2

0

2

4

6

8

10

Bce

llly

mph

oma

Tce

llle

ukem

iaLe

ukem

iaLy

mph

oid

neop

lasm

othe

rB

urki

tt's

lym

phom

aH

emat

opoi

etic

neop

lasm

othe

rLy

mph

obla

stic

leuk

emia

Acut

em

yelo

idle

ukem

iaM

yelo

ma

Hai

ryce

llle

ukem

iaC

hron

icm

yelo

idle

ukem

iaH

odgk

in's

lym

phom

aLy

mph

obla

stic

Tce

llle

ukem

iaB

cell

leuk

emia

Med

ullo

blas

tom

aEw

ing

sarc

oma

Ost

eosa

rcom

aR

habd

omyo

sarc

oma

Stom

ach

Soft

tissu

eot

her

Fibr

osar

com

aC

hond

rosa

rcom

aN

euro

blas

tom

aEs

opha

gus

Hea

dan

dne

ckLu

ngN

SCLC

larg

ece

llLu

ngN

SCLC

carc

inoi

dAl

lcan

cers

Lung

NSC

LCSC

CG

liom

aTh

yroi

dC

ervi

xLa

rge

inte

stin

eLu

ngsm

allc

ellc

arci

nom

aEn

dom

etriu

mB

ladd

erK

idne

yPr

osta

teM

elan

oma

Mes

othe

liom

aLi

ver

Ova

ryPa

ncre

asLu

ngN

SCLC

aden

ocar

cino

ma

Bre

ast

Bili

ary

trac

t-2

0

2

4

6

8

10

654 71

2

3

4

5

HN Eso

AML

B cell leukemia

B cell lymphoma

Burkitt's lymphoma

CML

HNO

HLeu Hodgkin's

LeuLLeu

LTLeu

LNO

Myeloma

T cell leukemia

Cho

EwingsOsteo

Breast

Biliary

Large intestineLiver

Sto

Kidney

Lung NSCLC Adenocarcinoma

LungCarLungLCLungSCC LungSmCCL

Meso

Glioma

Medullo

Neuro

PancreasMela

FibroRhabSoft

ThyroidBladder

CervixEndo

OvaryPro

All

No significant effectSignificant resistance to one compoundSignificant resistance to both compounds

Significant sensitivity to both compoundsSignificant sensitivity to one compound

Figure 4.

Evaluation of cancer cell line sensitivities to MRCK inhibitors.A, Cancer cell lines (n ¼ 757), divided into 45 cancer types,were treated with 7 concentrations of a half-log dilutionseries of BDP8900 (top concentration ¼ 10 mmol/L). After72 hours, cell density was measured, EC50 values calculated,and MANOVA performed to identify cancer typessignificantly (P < 0.05) sensitive (green triangles) orresistant (red triangles) to BDP8900 relative to allcancers (blue triangles) considered together. Purple lines,mean EC50 for each cancer type. B, Sensitivity or resistanceto BDP9066 was determined as described above. Cancertypes significantly (P < 0.05) sensitive (red triangles) orresistant (red triangles) from all cancers (blue triangles) areindicated. NSCLC, non–small cell lung cancer. C, Mean EC50

values for each cancer type to BDP8900 and BDP9066were plotted. Dark green dots, cancer types significantlysensitive to both inhibitors; light green dots, significantsensitivity to one inhibitor; dark red dots,significant resistance to both inhibitors; pink dots,significant resistance to one inhibitor (all P < 0.05), relativeto all cancers (blue dot). Deming regression indicatesthat the there is a significant (P < 0.0001) relationshipbetween the responses of cancer types to both drugs.Sensitive cancer types: AML, acute myeloid leukemia; CML,chronic myeloid leukemia; HNO, hematopoietic neoplasmother; LLeu, lymphoblastic leukemia; LNO, lymphoidneoplasm other.

Unbekandt et al.





MRCK expression and activation in mouse models of skincancer

Analysis of publicly available human microarray data usingOncomine (44) revealed significantly elevated levels of bothMRCKa and MRCKb in human epidermal SCC relative tonormal skin (Fig. 5L; ref. 45). Previous studies revealed a rolefor MRCKa in contributing to SCC in genetically modified(GM) mice expressing an oncogenic Hras transgene combinedwith Notch1 inhibition (46). To determine whether MRCKaand MRCKb expression or activation were associated with SCCin additional GM mouse models, antibodies against MRCKa,MRCKa pS1003 and MRCKb were first validated for IHC usingMDA MB 231 D3H2LN human breast cancer cells (47)knocked out for MRCKa and MRCKb by CRISPR/Cas9 (Sup-plementary Fig. S4A). Parental and MRCKab knockout (KO)cells were grown subcutaneously in immunocompromisednude mice for 2 weeks. Sections of parental cell xenograftsstained with the 3 antibodies were positive, but staining wasabrogated in sections of KO cell xenografts (SupplementaryFig. S4B). Antibody phosphoselectivity was validated whenMRCKa pS1003 staining of MDA MB 231 subcutaneoustumors was blocked by antibody preincubation with phos-phopeptide P, but not unphosphorylated peptide U (Supple-mentary Fig. S4C).

MRCKa, MRCKa pS1003, and MRCKb immunoreactivitywere examined in GM mouse SCC models expressing epider-mis-targeted activated Hras alone (34), or in combination withepidermis-targeted transgenic c-Fos and conditional epidermaldeletion of Pten (35), which give rise to stable papillomas thatmay stochastically convert to malignancy via additional eventssuch as p53 loss (36). MRCKa was expressed at moderatelyhigh levels in benign Hras and Hras/c-Fos/Pten�/� tumors,which increased further following malignant conversion tobecome strong and uniform in invasive SCC layers (Fig.5M). MRCKa pS1003 and MRCKb were expressed at low levelsin benign tumors with slight increases in early stage carcinomas(Fig. 5M). MRCKa pS1003 expression was mainly confined topost-mitotic, differentiating SCC layers, while MRCKb dis-played weak expression maintained in proliferative layers andstrong expression in occasional invasive cells. The observationswere similar in the 7,12-Dimethylbenz[a]anthracene (DMBA)/12-O-tetradecanoylphorbol-13-acetate (TPA) (DMBA/TPA)two-stage chemical carcinogenesis SCC model, with moderateto high levels of MRCKa and MRCKb, and a restricted patternof MRCKa pS1003 staining in post-mitotic differentiating SCCcells (Fig. 5M). These results indicate that MRCKa and MRCKbexpression, and MRCKa activation are associated with mouseskin tumors induced by genetic modifications and chemicalcarcinogenesis.

BDP9066 inhibits MLC phosphorylation and blocks SCC12squamous cell carcinoma motility and invasion

It was previously determined that the phosphorylation ofMLC and invasion of human SCC12 cells in an organotypic3D collagen invasionmodel was sensitive to combinedMRCKa/MRCKb knockdown (18). Furthermore, the MRCK inhibitorBDP5290 also reduced MLC phosphorylation and inhibitedSCC12 organotypic invasion (19). To investigate the effect ofBDP9066 on pMLC2 in SCC12 cells, which express MRCK butno detectable DMPK (Supplementary Fig. S5), varying BDP9066concentrations incubated with cells for 2 hours led to dose-

dependent inhibition of MLC2 phosphorylation (Fig. 6A, left)with an EC50 ¼ 64 nmol/L (Fig. 6A, right), corresponding to a5-fold increase in cellular potency relative to BDP5290 (19).SCC12 cell viability following 24-hour treatment was unaffect-ed by all BDP9066 concentrations up to 0.5 mmol/L (Fig. 6B),despitemaximal pMLC2 inhibition at this concentration (Fig. 6A),with a 25% decrease in viability at 1 mmol/L (Fig. 6B). Theseresults indicate that BDP9066 is relatively nontoxic at concentra-tions that profoundly inhibit substrate phosphorylation.

Treatment of SCC12 cells with non-toxic 0.5 mmol/L BDP9066led to changes in cell morphology, with reduced proportions ofregularly (rounded, few protrusions) shaped cells (Fig. 6C, greencells), and increased irregularly (greater spreading, increasedprotrusions) shaped cells (Fig. 6C, blue cells) as determined byhigh content imaging. By varying BDP9066 concentrations, itwas determined that there was a parallel redistribution fromregular to irregular cell morphologies with similar EC50 values(Fig. 6D). The proportion of elongated cells (Fig. 6D, red cells) didnot vary with BDP9066 concentrations (Fig. 6D). When individ-ual morphologic parameters were examined, there were dose-dependent increases in cell length and width (Fig. 6E, left), withdose-responsive changes in length exceeding width (Fig. 6E,middle). The collective effect of these BDP9066-induced mor-phologic alterations resulted in increased 2D cell area anddecreased roundness (Fig. 6E, right).

Consistent with our previous observations using the MRCKinhibitor BDP5290 (19), BDP9066 similarly reduced the bun-dling of filamentous actin at cortical and cytoplasmic regions(Fig. 6F). To determine how MRCK inhibition affected cellmotility, SCC12 cells were treated with DMSO vehicle controlor nontoxic 0.4 mmol/L BDP9066, and individual cells weretracked over 6 hours to quantify random migration. Scatter-plots of cell paths showed marked reduction of cell motilityinduced by BDP9066 (Fig. 6G). Mean cell velocity and accu-mulated distance travelled were significantly reduced by 21%,while mean Euclidean distance travelled was significantlyreduced by 18% (Fig. 6H). No differences were observed fordirectionality of cell motility.

The effect of nontoxic 0.4 mmol/L BDP9066 was tested onorganotypic invasion, in which the movement of SCC12 cellswas measured from the surface downwards into a dense 3D rattail collagen matrix that had been conditioned by cancer-associated fibroblasts for 7 days prior to their removal (Fig.6I; ref. 18). A significant 50% decrease in the percentage ofinvading cells was observed for cells treated with 0.4 mmol/LBDP9066 relative to DMSO vehicle (Fig. 6J). Taken together,the results indicate that in SCC12 cells BDP9066 potentlyinhibits MRCK at submicromolar levels, leading to inhibitionof MLC phosphorylation, changes in cell morphology, andreduced cell motility and 3D collagen invasion, without sig-nificantly affecting cell viability. These findings are consistentwith the previously reported inhibitory effects of MRCKa þMRCKb knockdown by siRNA (18), and the MRCK inhibitorBDP5290 (19) on SCC12 organotypic invasion.

In vivo action of BDP9066 on mouse skin cancerHaving established that BDP9066 affects SCC tumor cell

motility and invasion at submicromolar concentrations andviability at 1 mmol/L in vitro (Fig. 6), and that MRCKa, MRCKapS1003, and MRCKb levels were elevated in GM and chemically-induced mouse models of skin cancer (Fig. 5L), BDP9066 was






MR

CK

α

K10

6M

MR

CK

α

K10

6M

MR

CK

α

K10

6M

Con

trol

500

pg

250

pg

125

pg

50 p

g

25 p

g

10 p

g

P

U

DM

SO

BD

P52

90(3

μm

ol/L

)

(1 μ

mol

/L)

DM

SO

BD

P90

66

Phosphatase ++ +++--AutoPhos +++-- --

K106M ++ ++-- -MRCK CΔ ++ ++- - -

ControlPhosph-

ataseCon

trol

Pep

tide

P

Pep

tide

U

ladd

er

ladd

er

ladd

er

Rel

ativ

e M

LC2

phos

phor

ylat

ion

(%)

MR

CK

α

K10

6M

S10

03A

*ns

MR

CK

αK

106M

S10

03A

MLC2/pMLC2

MLC2

pMLC2

BA

DC E

H

I K

KD CC C1 PH CH

CR

IB

S1003

Stained gel Autoradiograph

pS1003

FLAG/pS1003

FLAG/pS1003

FLAG-MRCK

pS1003

FLAG-MRCK

pS1003

FLAG/pS1003

FLAG-MRCK

FLAG/pS1003

FLAG-MRCK

pS10030

50

100

150

FLAG-MRCKα

Hras

MR

CK

αpS

1003

MR

CK

αM

RC

Kβ

DMBA/TPA

Rel

ativ

e M

RC

Kα

expr

essi

on

SCC

**

Rel

ativ

e M

RC

Kβ

expr

essi

on

L M Hras/c-Fos/Pten -/-

**

Normal

Normal SCC0

1

2

3

0

1

2

3

170130100

70

5540

3525

MW

15

17013010070

55

40

3525

MW

15100 200 300 400 500 600 700 800 900 1,000 1,100

m/z

05

101520253035404550556065707580859095

100

Rel

ativ

e ab

unda

nce

341.22

869.45

244.17

218.06529.24

488.29147.11

484.22z=2435.23

z=2129.10

772.40

543.76z=2

363.13

715.38813.35

967.43323.21627.22

592.34z=2

887.47

851.45

1084.581038.43

b2y1

y2

y3

y4

y6

y7

y9

b10

b7-H PO3 4MH+

-H PO3 4 599.22

746.42

b8-H PO3 4

y7-H PO3 4

y8+2y9+2

y9-H PO3 4

y8-H PO3 4

646.36

999 1009G C P G G F P P KS Tb2 b7 b8 b10

y9 y7 y6 y4 y3 y2 y1

J

FLAG/pS1003FLAG-MRCK

pS1003Phosphatase +-

FLAG/pS1003

FLAG-MRCKpS1003

MLC2pMLC2

MLC2/pMLC2

BDP9066(1 μmol/L) (1 μmol/L)

+- +-BDP9066

F

MR

CK

αK

106M

S10

03A

pS1003

FLAG/pS1003

FLAG-MRCK

GpS1003

FLAG/pS1003

FLAG-MRCK

CDC42 Q61L+-

CDC42 Q61L

CDC42+-0

20406080

100120140

Rel

ativ

e S

1003

phos

phor

ylat

ion

(%) ***

Figure 5.

MRCKa S1003 autophosphorylation as a biomarker of activity. A, Left, stained SDS-PAGE gel of FLAG immunoprecipitated FLAG-MRCKa andFLAG-MRCKa K106M after in vitro32P incorporation assay. Arrow, MRCKa. Right, gel autoradiograph shows 32P labeling of MRCKa. MW, molecular weight.B, Left, stained SDS-PAGE gel of FLAG immunoprecipitates from control HEK293 cells, or cells transfected with pEF-FLAG-MRCKa or pEF-FLAG-MRCKa K106M. (Continued on the following page.)

Unbekandt et al.





evaluated for in vivo pharmacological proof-of-concept as anSCC chemotherapeutic agent. Topical application of 25 mgBDP9066 on FVB mouse skin twice per day for 2 days led to26 mmol/L mean BDP9066 concentration in skin, but only0.04 mmol/L in blood (Fig. 7A). BDP9066 application led tosignificantly reduced epidermal MRCKa pS1003–positive stain-ing (Fig. 7B). To determine how repeated dosing would affectBDP9066 accumulation and distribution, skin and blood con-centrations were determined after 10 mg was administered once,or 25 mg was repeated over 4 days (Fig. 7C). Although relative tothe single 10 mg dose, repeated 25 mg doses did result in 2.8-foldhigher concentrations in skin (Fig. 7C, left), and 4-fold higherconcentrations in blood (Fig. 7C, right), these differences wereless than the 10-fold difference in total BDP9066 administered,indicating that compound accumulation was less than additive.To characterize clearance, BDP9066 was either given once ata 25 mg dose, or 8 times on a 4 days on, 2 days off, 4 days onschedule, and then compound concentrations were determinedin skin and blood at 2, 4, 8, and 24 hours following the finaldose. In either dosing regimen, over 16 mmol/L BDP9066 wasdetected in skin 24 hours after final dosing (Fig. 7D, left),while the low levels detected in blood were undetectable after24 hours (Fig. 7D, right). These results indicated that it waspossible to achieve sustainable BDP9066 levels in mouse skinby repeated topical application, which were sufficient to inducephenotypic responses in squamous cell carcinoma cells in vitro,without significant compound accumulation following sequen-tial administration.

Given the MRCKa, MRCKa pS1003, and MRCKb levelsobserved in skin tumors induced by the DMBA/TPA two-stagechemical carcinogenesis protocol (Fig. 5L), this model was select-ed to evaluate the in vivo efficacy of BDP9066. FVB mice weretopically treated with 25 mg DMBA (Fig. 7E, green arrow), thenwith 4.7 mg TPA 3 times per week (Fig. 7E, purple arrow). Inaddition, mice were topically treated with 25 mg BDP9066, or anequal volume of DMSO vehicle control (Fig. 7E, red arrow),5 times per week (5 days on, 2 days off) for 14 weeks. At studyend, skin papillomas in the BDP9066-treated group appeared

visually smaller than in the DMSO-treated group (Fig. 7F).Although total papilloma numbers per mouse were not differentbetween the DMSO and BDP9066 treatment groups (Fig. 7G),both the total tumor volume (Fig. 7H, left) and average papil-loma volume (Fig. 7H, right) per mouse were significantlyreduced in the BDP9066 treatment group. Topical BDP9066application resulted in undetectable compound in blood, and>1 mmol/L mean BDP9066 concentration in skin at experi-mental endpoint (Fig. 7I), which was associated with significantdecreases in MRCKa pS1003 IHC staining (Fig. 7J) and reducedhistoscores both in the treated skin area (Fig. 7K, left) and inpapillomas (Fig. 7K, right). These results provide pharmacologicproof-of-concept evidence indicating in vivo therapeutic actionsof BDP9066.

DiscussionThe small-molecule inhibitors BDP8900 and BDP9066 are

the most potent and selective MRCK inhibitors discovered todate. Structure-guided medicinal chemistry facilitated com-pound optimization for potency and selectivity over closelyrelated kinases including ROCK1 and ROCK2 (Figs. 1 and 2;Supplementary Fig. S1; Supplementary Tables S1–S4). Consis-tent with the previously characterized cell biological roles ofthese proteins, MRCK inhibition led to dose-dependentdecreases in MLC2 phosphorylation, morphologic changes, andreduced cell motility and invasion into 3D matrices at concen-trations below cytotoxic levels (Fig. 5). Inhibition of prolifer-ation (e.g., Figs. 4 and 6B) was observed at BDP9066 concen-trations higher than required for significant substrate phosphor-ylation, likely indicating a need for near complete inhibition ofactivity to provoke antiproliferative responses. Furthermore,topical application of BDP9066 to mice undergoing a DMBA/TPA chemical carcinogenesis protocol resulted in the therapeuticeffect of significantly reduced papilloma volumes associatedwith micromolar compound levels in skin and reduced MRCKaS1003 autophosphorylation (Fig. 7). The discovery of thesecompounds will enable further exploration of MRCK functions

(Continued.) Arrow, MRCKa. MW, molecular weight. Right, higher energy collision dissociation (HCD) MS/MS fragmentation spectra of MRCKa trypticpeptide 999–1,009 containing S1003 (red lettering) in MRCKa or MRCKa K106M (phosphorylation inferred from y7, y7-H3PO4, and y6 signals isindicated with green arrows). See Supplementary Table S8 for theoretical y and b fragmentation ion series masses. Diagram of MRCKa protein domains (inset)illustrating location of S1003. KD, kinase domain; CC, coiled-coiled domain; C1, C1 diacylglycerol binding domain, PH, Pleckstrin homology domain; CH, Citronhomology domain; CRIB, Cdc42/Rac interactive binding motif. C, Dot blot of MRCKa peptides (10 to 500 pg) containing phosphorylated S1003 (peptide P) orunphosphorylated S1003 (peptide U), stained with pS1003 antibody. D, Western blots of FLAG-immunoprecipitated FLAG-MRCKa stained withuntreated pS1003 antibody, or pS1003 antibody preincubated with phosphorylated S1003 peptide P or unphosphorylated peptide U. E, Western blotsof FLAG-immunoprecipitated FLAG-MRCKa, incubated in phosphatase buffer without (control) or with lambda phosphatase (phosphatase). F,Western blot analysis of FLAG-immunoprecipitated FLAG-MRCKa, FLAG-MRCKa K106M, or FLAG-MRCKa S1003A. G, Western blot analysis (left) ofFLAG-immunoprecipitated FLAG-MRCKa incubated in kinase buffer in the absence or presence of recombinant CDC42 Q61L protein. Relative S1003phosphorylation revealed that CDC42 significantly increased autophosphorylation. Results shown are means � SEM from three experiments. Student t test( , P < 0.001). H, Western blots of HEK293 cells expressing FLAG-MRCKa treated with DMSO, 3 mmol/L BDP5290, or 1 mmol/L BDP9066. I, Western blotsof FLAG immunoprecipitated FLAG-MRCKaDC or FLAG-MRCKa K106M untreated or treated with lambda phosphatase, or treated with lambdaphosphatase and then incubated in kinase buffer to allow autophosphorylation (AutoPhos). J, Top, Western blots of FLAG-immunoprecipitated FLAG-MRCKa,untreated or treated with lambda phosphatase. Incubation with kinase buffer resulted in autophosphorylation (pS1003; bottom left) and recombinant MLC2substrate phosphorylation (pMLC2; bottom right). K, Western blots of FLAG-immunoprecipitated FLAG-MRCKa, FLAG-MRCKa K106M, or FLAG-MRCKaS1003A incubated in vitro with recombinant MLC2 in a kinase buffer (left). Relative MLC2 phosphorylation by each kinase revealed that K106Msignificantly reduced activity, while S1003A was not different from wild-type. Results shown are means � SEM from five independent experiments. One-wayKruskal–Wallis ANOVA followed by post hoc Dunn multiple comparison ( , P < 0.05). ns, nonsignificant. L, Relative MRCKa (top) and MRCKb (bottom)expression was significantly higher in cutaneous SCC relative to normal human skin determined by microarray analysis (45). Results shown are means � SDfrom 6 normal skin and 5 SCC patients. Two-tailed Mann–Whitney test of significance ( , P < 0.01). M, IHC of MRCKa, MRCKa pS1003, or MRCKb inbenign papillomas from mice expressing oncogenic Hras in epidermal keratinocytes, or in well-differentiated squamous cell carcinomas resulting fromHras/c-Fos/Pten�/� genetic modifications or DMBA/TPA chemical carcinogenesis. Scale bars, 100 mm.






DMSO BDP9066

**

% o

f Inv

adin

g ce

lls

αTubulin

pMLC2MLC2

MLC2pMLC2

DM

SO

0.02

μm

ol/L

0.05

μm

ol/L

0.1

μmol

/L

0.2

μmol

/L

0.5

μmol

/L

1 μm

ol/L

BDP9066

DMSO BDP9066 (0.5 μmol/L)

BA

C

H

I

Acc

umul

ated

Eucl

idea

ndi

stan

ce (m

m)

dist

ance

(mm

)

*

Velo

city

(mm

/min

) ***D

irect

iona

lity

ns

J

Rel

ativ

e pM

LC

BDP9066 (log mol/L)10



Viab

ility

(%)

D

-8.0 -7.5 -7.0 -6.5 -6.00

25

50

75

100

IC = 64 nmol/L50

-8.5 -8.0 -7.5 -7.0 -6.5 -6.00

25

50

75

100

Cel

l mor

phol

ogy

0 200-200

-200

200

0

DMSO

200-200 0

200

-200

0

x-axis (μm)

x-axis (μm)

y-ax

is (μ

m)

y-ax

is (μ

m)

BDP9066 (0.4 μmol/L)

0

100

200

300

400

500

0

50

100

150

200

250

0.0

0.5

1.0

1.5

0.0

0.2

0.4

0.6

0.8

1.0

DMSO BDP9066 DMSO BDP9066

DMSO BDP9066 DMSO BDP9066

***

DMSO BDP9066 (0.4 μmol/L)

0

50

100

150

200

E

EC = 0.39 μmol/L50

EC = 0.30 μmol/L50

EC = 0.28 μmol/L50

EC = 0.19 μmol/L50EC = 0.28 μmol/L50

-8.5 -8.0 -7.5 -7.0 -6.5 -6.050

55

60

65

70

75 *

-8.5 -8.0 -7.5 -7.0 -6.5 -6.026.25

27.50

28.75

30.00

31.25

Cel

l wid

th (m

m)

Cel

l len

gth

(mm

)

*

-8.5 -8.0 -7.5 -7.0 -6.5 -6.00.425

0.445

0.465

0.485

0.505

0.525

Wid

th/L

engt

h ra

tio

*

-8.5 -8.0 -7.5 -7.0 -6.5 -6.01,250

1,500

1,750

2,000

BDP9066 (log10 mol/L)BDP9066 (log10 mol/L)BDP9066 (log10 mol/L)BDP9066 (log10 mol/L)

Cel

l are

a (m

m2 )

* *

-8.5 -8.0 -7.5 -7.0 -6.5 -6.00.65

0.70

0.75

0.80

0.85


Cel

l rou

ndne

ss

*

-8.5 -8.0 -7.5 -7.0 -6.5 -6.00

20

40

60

80Regular (%)EC50 = 0.54 μmol/L

Irregular (%)EC50 0.6= 6 μmol/L

Elongated (%)

GF

BDP9066 (1 μmol/L)

DMSO

Figure 6.

BDP9066 affects MLC2 phosphorylation, morphology, migration, and invasion of human SCC12 squamous cell carcinoma. A, Western blots (left) showingdose-dependent inhibition (right) of MLC2 phosphorylation by BDP9066 in SCC12 cells. Results shown are means � SEM from three independentreplicates. B, SCC12 viability as a function of BDP9066 concentration after 24-hour treatment. Results shown are means � SEM from three independentreplicates. C, High content analysis of SCC12 morphology following treatment with DMSO vehicle or 0.5 mmol/L BDP9066 for 2 hours. Multiparametricanalysis was used to define regular (green), elongated (red), and irregular (blue) SCC12 cell morphologies. Scale bar, 100 mm. D, Percentages of eachSCC12 cell shape as functions of BDP9066 concentration after 2-hour treatment. Results shown are means � SEM from three independent replicates.E, Individual SCC12 cell morphology parameters (cell length, cell width, width/length ratio, cell area, cell roundness) as functions of BDP9066concentration after 2-hour treatment. Results shown are means � SEM from three independent replicates. F, Representative images of phalloidin-stainedfilamentous actin structures in SCC12 cells treated with DMSO or 1 mmol/L BDP9066 for�18 hours. Scale bar, 10 mm. G, Tracks of SCC12 cell migration (120 cells)treated with DMSO (left) or 0.4 mmol/L BDP9066 (right) over 6 hours. Each line represents the path of a single cell from the origin, with final position indicatedby a dot. H, Accumulated distance, Euclidean distance, velocity, and directionality of SCC12 cells treated with DMSO or 0.4 mmol/L BDP9066 for each trackedcell over 6 hours. Results shown are means� SEM from 120 individual cells. Two tailed unpaired t-tests were used to determine significance. , P < 0.05; , P <0.001; ns, nonsignificant). I, Representative images of organotypic invasion by SCC12 cells cultured with medium supplemented with DMSO (left) or 0.4 mmol/L BDP9066 (right). Scale bars, 100 mm. J, Percentage of invading SCC12 cells in organotypic invasion assays following treatment with DMSO or 0.4 mmol/LBDP9066. Results shown are means � SEM from 6 independent replicates. Two-tailed Mann–Whitney test of significance (, P < 0.01).

Unbekandt et al.





DMSO

pS1003

BDP9066

Blood SkinB

DP9

066

(mm

ol/L

) B

DP9

066

(mm

ol/L

)

BD

P906

6 ( m

mol

/L)

DMSO pS1003 BDP9066

BDP9066DMSO

DMSO BDP9066Tota

l pap

illom

a nu

mbe

r

Ave

rage

pap

illom

avo

lum

e (m

m3 )

Tota

l tum

or v

olum

e (m

m3 )

BDP9066DMSO DMSO BDP9066

pS10

03-p

ositi

veep

ider

mis

(%)

DMSO BDP9066

*

DMSO BDP9066

pS10

03 H

isto

scor

es

Skin

A B

C D

J K

pS1003

10 μg 25 μg(1 dose) (4 doses)

10 μg 25 μg(1 dose) (4 doses)

BloodSkin Skin

Time (hr) Time (hr)42 24

25 μg X 1 dose25 μg X 8 doses

8 42 248

Blood

**

**

BD

P906

6 (m

mol

/L)

**

Blood Skin

*

I

pS10

03 H

isto

scor

es

DMSO BDP9066

Papilloma*

pS1003

HG

X14DMBA TPA

DMSO/BDP9066

E

F

0

10

20

30

40

50

0.00

0.05

0.10

0.15

0.00

0.05

0.10

20

30

0

10

20

30

40

50

0

20

40

60

0.0

0.5

1.0

1.5

2.0

0102030405060708090

0

500

1,000

1,500

2,000

2,500

0

50

100

150

200

0

25

50

75

100

125

0

10

20

30

0.00

0.05

0.10

Figure 7.

In vivo application of BDP9066 in a DMBA/TPA mouse model of SCC. A, Topical application of 4� 25 mg BDP9066 over 2 days to dorsal skin led to measurabledrug levels in the skin that were significantly higher than in blood. Results shown are means � SD from 5 independent mice, each indicated by a datapoint. Two-tailed Mann–Whitney test of significance ( , P < 0.01). B, Left, representative MRCKa pS1003 immunohistochemical staining of mouse skinsections topically treated with DMSO or 4 � 25 mg BDP9066 over 2 days. Scale bars, 100 mm. Right, topical administration of BDP9066 (red dots) led tosignificant reduction in positive epidermal staining for MRCKa pS1003. Results shown are means � SD from 5 independent mice per condition, each indicatedby a data point. Two-tailed Mann-Whitney test of significance ( , P < 0.05). C, Skin (left) or blood (right) BDP9066 concentrations following topicaladministration of a single 10 mg dose or 4 � 25 mg doses of BDP9066 over 2 days. Results shown are means � SD from three independent mice per condition,each indicated by a data point. D, Skin (left) or blood (right) BDP9066 concentrations following topical administration of a single 25 mg dose (pink dots) or8 � 25 mg doses of BDP9066 (red dots; 4 days on, 2 days off, 4 days on) at indicated times after final administration. Results shown are means � SDfrom three independent mice per condition, each indicated by a data point. E, Timeline of DMBA (green arrow), TPA (purple arrow), DMSO (black arrow),or BDP9066 (red arrow) administration, with each day represented by a rectangle. F, Representative images of cagemate DMBA/TPA-treated micewith topical application of DMSO or BDP9066 as indicated. G, Total endpoint papilloma numbers per mouse. Results shown are means � SD from20 independent mice per condition, each indicated by a data point. H, Total tumor volume (left) and average papilloma volume (right) per mouse. Resultsshown are means � SD from 20 independent mice per condition, each indicated by a data point. Two tailed unpaired t-tests were used to determinesignificance ( , P < 0.05). I, Blood and skin BDP9066 concentrations at endpoint. Results shown are means � SD from three independent mice for blood and10 mice for skin, each indicated by a data point. Two-tailed Mann-Whitney test of significance ( , P < 0.001). J, Representative MRCKa pS1003 stainingof DMBA/TPA skin sections treated topically with DMSO or BDP9066. Scale bars, 100 mm. K, Cytoplasmic histoscores of MRCKa pS1003 staining of DMBA/TPA-treated skin (left) or papilloma (right) sections that had been administered DMSO (black dots) or BDP9066 (red dots). Results shown are means � SDfrom 19 independent mice per condition, each indicated by a data point. Two-tailed unpaired t tests were used to determine significance (, P < 0.05).






and roles in cancer, which will be important for determininghow MRCK inhibition would be clinically applicable.

The Rho GTPase family effectors ROCK1, ROCK2, MRCKa,and MRKCb phosphorylate common substrates includingMLC2, the myosin-binding subunit (MYPT1) of the MLC2phosphatase complex, and the LIM kinases 1 and 2 (6, 7,48). However, the ROCK and MRCK kinases appear to operatein different subcellular locales, with ROCK actions being moredispersed and MRCK acting primarily close to the plasmamembrane. These differing sites of action likely explain whythe combined inhibition of ROCK and MRCK together is moreeffective at blocking tumor cell invasion (15, 16). It has beenproposed that compounds that inhibit both ROCK and MRCKkinases would be more effective as antimetastatic therapeuticsthan agents that were selective for one or the other (20). Thishypothesis may be correct, but less selective inhibitors wouldnot be useful as chemical biology tools to explore kinase-specific biological roles. In addition, in situations where onlyinhibition of MRCK was necessary for therapeutic effect, theadditional ROCK inhibitory actions might lead to undesirableresponses, such as dose-limiting hypotension. The highlypotent and selective inhibitors BDP8900 and BDP9066 willenable studies focused on discovery of MRCK actions and rolesin cancer.

Compounds that block tumor cell invasion are likely to bebeneficial through a combination of effects, including reducedprimary tumor growth and local invasion, as well as decreaseddissemination and metastasis. For example, there were multiplebeneficial effects on pancreatic tumor growth and spread whenmice were treated with ROCK inhibitors, including sensitiza-tion of tumors to standard-of-care therapeutics (49, 50). Giventhat compounds that target processes that contribute to tumorcell invasion do not directly lead to cytotoxicity, they areunlikely to induce easily monitored responses such as tumorshrinkage in patients with advanced cancers, making the clin-ical development process more challenging than for "classical"cytotoxic type chemotherapeutics. Instead, clinical benefit fromanti-invasive drug administration would more likely beobserved in alternative scenarios, such as long-term adjuvanttreatment as part of post-therapy management, or in situationswhere local tissue invasion rather than distant metastasis is thecause of morbidity and mortality, as is the case for invasivegliobastoma (51). Additional preclinical studies will help todefine the contexts in which anti-invasive compounds, such asthe MRCK inhibitors BDP8900 and 9066, are most likely to bebeneficial for cancer patient therapy.

Indications consistent with a role of MRCKa or MRCKb incancer include elevated expression and gene copy number varia-tions. A "poor prognosis gene" expression signature, associatedwith increased incidence of distant metastasis in fewer than 5years, included MRCKb (indicated as PK428; ref. 52), while theMRCKa gene was found to be amplified in 491/2,051 (24%) ofbreast cancer samples (53). However, an additional possibility isthat increased MRCK activity is a common mechanism contrib-uting to cancer, which would not necessarily be associated withchanges in MRCK expression. Increased MRCK activity might betrigged from upstream signals via CDC42, which could be acti-vated through elevated expression, loss of inhibitory GTPase-activating proteins (GAP), increased expression or activatingmutation of guanine-nucleotide exchange factors (GEFs), oraberrant activity of regulators further upstream, such as receptor

tyrosine kinases (54). Furthermore, mechanical stimulation ofCDC42 by the extracellular microenvironment (55) may beanother pathway leading to MRCK activation. A direct way toassess MRCK activation state is through the identification ofautophosphorylation sites that report on activity status. A totalof 8 phosphorylation sites were identified in wild-type MRCKa,but not the kinase dead K106M mutant, consistent with theirbeing autophosphorylation events (Fig. 5; Supplementary TablesS7 and S8). A phospho-sensitive pS1003 confirmed that phos-phorylation of this site was mediated by an intramolecularmechanism, and was dependent on MRCK activity, supportingthe conclusion that S1003 was autophosphorylated (Fig. 5).Furthermore, the antibody reagent could be used in formalin-fixed paraffin embedded mouse tissue samples to interrogateMRCKa activity in tumors, or following BDP9066 treatment(Figs. 5 and 7). Similarly, ROCK1 S1333 phosphorylation (41)and ROCK2 S1366 phosphorylation (42) were shown to beassociated with kinase activity, and that phosphorylation-statesensitive antibodies could be used to assess ROCK activity inclinical cancer samples (43). Therefore, MRCKa and MRCKbphospho-selective antibodies raised against autophosphorylationsites are likely to be very useful tools to identify cancer typesthat would benefit from MRCK inhibitor treatment. AlthoughMRCKa S1003 autophosphorylation was validated and thephospho-sensitive antibody revealed increased staining inmouse skin tumors (Fig. 5), other sites that have been identifiedin broad-scale phosphoproteomics studies might be better candi-dates for further evaluation, such as the putative autophos-phorylation site S1629.

A role for MRCK in squamous cell carcinoma was previouslyindicated by microarray expression analysis of normal humanskin versus epidermal SCC (Fig. 5K; ref. 44) and by a GMmousemodel of epidermal SCC induced by oncogenic Hras expressioncombined with Notch1 inhibition (46). Using validated anti-bodies for MRCKa, phosphorylated MRCKa S1003 andMRCKb, there was increased staining of each in two GM skincancer models, and especially increased staining in the DMBA/TPA chemical carcinogenesis model (Fig. 5L). Furthermore,BDP9066 topical application significantly decreased phosphor-ylated MRCKa S1003 staining and tumor volumes (Fig. 7),consistent with antibody staining for MRCK autophosphoryla-tion being both an indicator of an association between MRCKactivity with cancer, and a predictor of therapeutic response toinhibitor treatment. The approach described, in which differ-ences between the phosphorylation of wild-type versus kinase-dead versions were used to identify autophosphorylationevents that could be used to monitor activation states, has thepotential to be applied to additional kinases to evaluate theassociations between activity and cancer, and to predict ther-apeutic responses. Monitoring kinase activity by IHCapproaches would be a complementary or alternative approachthat could be applied to signaling pathways in which activatingmutations are not major factors.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: M. Unbekandt, J. Bower, D. Crighton, M.J. Drysdale,D. McArthur, H. McKinnon, J. Munro, G. Naylor, M. Sime, M.F. Olson

Unbekandt et al.





Development of methodology: M. Unbekandt, J. Bower, D. Crighton,D.R. Croft, C. Gray, D.A. Greenhalgh, S. Lilla, P. McConnell, H. McKinnon,M. Mezna, M.F. OlsonAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M. Unbekandt, M. Clarke, D. Crighton, D.R. Croft,M.J. Garnett, K. Gill, C. Gray, D.A. Greenhalgh, J.A.M. Hall, J. Konczal,P. McConnell, L. McDonald, L. McGarry, H. McKinnon, C. McMenemy,N.A. Morrice, J. Munro, G. Naylor, N. Rath, A.W. Sch€uttelkopfAnalysis and interpretation of data (e.g., statistical analysis, biostati-stics, computational analysis):M. Unbekandt, J. Bower, J. Cordes, D.R. Croft,D.A. Greenhalgh, L. McGarry, H. McKinnon, M. Mezna, N.A. Morrice,J. Munro, G. Naylor, A.W. Sch€uttelkopf, M.F. OlsonWriting, review, and/or revision of the manuscript: M. Unbekandt, J. Bower,D.A. Greenhalgh, H. McKinnon, A.W. Sch€uttelkopf, M.F. OlsonAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): M. Unbekandt, S. Belshaw, D. Crighton,P. McConnell, M.F. OlsonStudy supervision: J. Bower, D. Crighton, M.J. Drysdale, H. McKinnon,M.F. OlsonOther (working as laboratory assistant for M. Unbekandt): J. CordesOther (synthesis of azaindole chemical compound): K. Gill

Acknowledgments

This work was funded by Cancer Research UK (A18276), MedicalResearch Council (MR/J005126/1), and Worldwide Cancer Research(14-0223 to M.F. Olson). The work in the M.J. Garnett lab is supported byfunding from CRUK (C44943/A22536), SU2C (SU2C-AACR-DT1213), andthe Wellcome Trust (102696). The authors would like to thank DiamondLight Source for beamtime (proposal mx8659) and the staff of beamlines i02and i24 for their assistance. The authors thank Sara Zanivan and CatherineWinchester (Beatson Institute) for critical reading of the manuscript.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received September 19, 2017; revised November 20, 2017; accepted January24, 2018; published first January 30, 2018.

References1. Hall A. The cytoskeleton and cancer. Cancer Metastasis Rev 2009;28:5–14.2. Mardilovich K, Olson MF, Baugh M. Targeting Rho GTPase signaling for

cancer therapy. Future Oncol 2012;8:165–77.3. Heissler SM, Sellers JR. Various themes of myosin regulation. J Mol Biol

2016;428:1927–46.4. Julian L,OlsonMF. Rho-associated coiled-coil containing kinases (ROCK).

Small GTPases 2014;5:e29846.5. Olson MF. Rho GTPases, their post-translational modifications, disease-

associated mutations and pharmacological inhibitors. Small GTPases.2016 Aug 12. [Epub ahead of print].

6. Unbekandt M, Olson MF. The actin-myosin regulatory MRCK kinases:regulation, biological functions and associationswith human cancer. JMolMed 2014;92:217–25.

7. Zhao Z, Manser E. Myotonic dystrophy kinase-related Cdc42-bindingkinases (MRCK), the ROCK-like effectors of Cdc42 and Rac1. SmallGTPases 2015;6:81–8.

8. Leung T, Chen XQ, Tan I, Manser E, Lim L. Myotonic dystrophy kinase-related Cdc42-binding kinase acts as a Cdc42 effector in promotingcytoskeletal reorganization. Mol Cell Biol 1998;18:130–40.

9. Ng Y, Tan I, Lim L, Leung T. Expression of the human myotonicdystrophy kinase-related Cdc42-binding kinase gamma is regulatedby promoter DNA methylation and Sp1 binding. J Biol Chem 2004;279:34156–64.

10. Uehata M, Ishizaki T, Satoh H, Ono T, Kawahara T, Morishita T, et al.Calcium sensitization of smooth muscle mediated by a Rho-associatedprotein kinase in hypertension. Nature 1997;389:990–4.

11. Rath N, Olson MF. Rho-associated kinases in tumorigenesis: re-considering ROCK inhibition for cancer therapy. EMBO Rep 2012;13:900–8.

12. Olson MF, Sahai E. The actin cytoskeleton in cancer cell motility. ClinExp Metastasis 2009;26:273–87.

13. Chaffer CL, Weinberg RA. A perspective on cancer cell metastasis. Science2011;331:1559–64.

14. Waclaw B, Bozic I, Pittman ME, Hruban RH, Vogelstein B, Nowak MA. Aspatial model predicts that dispersal and cell turnover limit intratumourheterogeneity. Nature 2015;525:261–4.

15. Wilkinson S, Paterson HF, Marshall CJ. Cdc42-MRCK and Rho-ROCKsignalling cooperate in myosin phosphorylation and cell invasion.Nat Cell Biol 2005;7:255–61.

16. Heikkila T, Wheatley E, Crighton D, Schroder E, Boakes A, Kaye SJ, et al.Co-crystal structures of inhibitors with MRCKbeta, a key regulator oftumor cell invasion. PLoS One 2011;6:e24825.

17. Kale VP,Hengst JA,DesaiDH,Dick TE, ChoeKN,Colledge AL, et al. Anovelselective multikinase inhibitor of ROCK and MRCK effectively blockscancer cell migration and invasion. Cancer Lett 2014;354:299–310.

18. Gaggioli C, Hooper S, Hidalgo-Carcedo C, Grosse R, Marshall JF, Harring-ton K, et al. Fibroblast-led collective invasion of carcinoma cells withdiffering roles for RhoGTPases in leading and following cells. Nat Cell Biol2007;9:1392–400.

19. UnbekandtM, Croft DR, CrightonD,MeznaM,McArthur D,McConnell P,et al. A novel small-molecule MRCK inhibitor blocks cancer cell invasion.Cell Commun Signal 2014;12:54.

20. Kale VP, Hengst JA, Desai DH, Amin SG, Yun JK. The regulatory roles ofROCK and MRCK kinases in the plasticity of cancer cell migration. CancerLett 2015;361:185–96.

21. Tan I, Lai J, Yong J, Li SF, Leung T. Chelerythrine perturbs lamellaractomyosin filaments by selective inhibition of myotonic dystro-phy kinase-related Cdc42-binding kinase. FEBS Lett 2011;585:1260–8.

22. Sanjana NE, Shalem O, Zhang F. Improved vectors and genome-widelibraries for CRISPR screening. Nat Methods 2014;11:783–4.

23. Pham YC,Man N, Lam LT, Morris GE. Localization of myotonic dystrophyprotein kinase in human and rabbit tissues using a new panel of mono-clonal antibodies. Hum Mol Genet 1998;7:1957–65.

24. Kabsch W. Xds. Acta Crystallogr D Biol Crystallogr 2010;66:125–32.25. Evans PR, Murshudov GN. How good are my data and what is the

resolution? Acta Crystallogr D Biol Crystallogr 2013;69:1204–14.26. Winter G. xia2: an expert system for macromolecular crystallography data

reduction. J Appl Crystallogr 2010;43:186–90.27. Murshudov GN, Vagin AA, Dodson EJ. Refinement of macromolecular

structures by the maximum-likelihood method. Acta Crystallogr D BiolCrystallogr 1997;53:240–55.

28. Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development ofCoot. Acta Crystallogr D Biol Crystallogr 2010;66:486–501.

29. Sch€uttelkopf AW, van Aalten DM. PRODRG: a tool for high-throughputcrystallography of protein-ligand complexes. Acta Crystallogr D BiolCrystallogr 2004;60:1355–63.

30. Chen VB, Arendall WB III, Headd JJ, Keedy DA, Immormino RM,Kapral GJ, et al. MolProbity: all-atom structure validation for macro-molecular crystallography. Acta Crystallogr D Biol Crystallogr 2010;66:12–21.

31. McGarry DJ, Shchepinova MM, Lilla S, Hartley RC, Olson MF. A cell-permeable biscyclooctyne as a novel probe for the identification of proteinsulfenic acids. ACS Chem Biol 2016;11:3300–4.

32. Cox J, Mann M. MaxQuant enables high peptide identification rates,individualized p.p.b.-range mass accuracies and proteome-wide proteinquantification. Nat Biotechnol 2008;26:1367–72.

33. Cox J, Neuhauser N, Michalski A, Scheltema RA, Olsen JV, Mann M.Andromeda: a peptide search engine integrated into the MaxQuant envi-ronment. J Proteome Res 2011;10:1794–805.






34. Greenhalgh DA, Rothnagel JA, Quintanilla MI, Orengo CC, Gagne TA,Bundman DS, et al. Induction of epidermal hyperplasia, hyperkeratosis,and papillomas in transgenic mice by a targeted v-Ha-ras oncogene. MolCarcinog 1993;7:99–110.

35. Yao D, Alexander CL, Quinn JA, Chan WC, Wu H, Greenhalgh DA.Fos cooperation with PTEN loss elicits keratoacanthoma not carci-noma, owing to p53/p21 WAF-induced differentiation triggered byGSK3beta inactivation and reduced AKT activity. J Cell Sci 2008;121:1758–69.

36. Macdonald FH, Yao D, Quinn JA, Greenhalgh DA. PTEN ablation in Ras(Ha)/Fos skin carcinogenesis invokes p53-dependent p21 to delay con-version while p53-independent p21 limits progression via cyclin D1/E2inhibition. Oncogene 2014;33:4132–43.

37. Yung-Chi C, Prusoff WH. Relationship between the inhibition constant(KI) and the concentration of inhibitor which causes 50 per centinhibition (I50) of an enzymatic reaction. Biochem Pharmacol 1973;22:3099–108.

38. Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. Theprotein kinase complement of the human genome. Science 2002;298:1912–34.

39. Eid S, Turk S, Volkamer A, Rippmann F, Fulle S. KinMap: a web-based toolfor interactive navigation through human kinome data. BMC Bioinfor-matics 2017;18:16.

40. Iorio F, Knijnenburg TA, Vis DJ, Bignell GR, MendenMP, Schubert M, et al.A landscape of pharmacogenomic interactions in cancer. Cell 2016;166:740–54.

41. Chuang HH, Liang SW, Chang ZF, Lee HH. Ser1333 phosphorylationindicates ROCKI activation. J Biomed Sci 2013;20:83.

42. ChuangHH,YangCH, Tsay YG,HsuCY, Tseng LM,Chang ZF, et al. ROCKIISer1366 phosphorylation reflects the activation status. Biochem J 2012;443:145–51.

43. Hsu C-Y, Chang Z-F, Lee H-H. Immunohistochemical evaluation of ROCKactivation in invasive breast cancer. BMC Cancer 2015;15:943.

44. Rhodes DR, Yu J, Shanker K, Deshpande N, Varambally R, Ghosh D, et al.ONCOMINE: a cancer microarray database and integrated data-miningplatform. Neoplasia 2004;6:1–6.

45. Nindl I, Dang C, Forschner T, Kuban RJ, Meyer T, Sterry W, et al.Identification of differentially expressed genes in cutaneous squa-mous cell carcinoma by microarray expression profiling. Mol Cancer2006;5:30.

46. Lefort K, Mandinova A, Ostano P, Kolev V, Calpini V, Kolfschoten I, et al.Notch1 is a p53 target gene involved in human keratinocyte tumorsuppression through negative regulation of ROCK1/2 andMRCKa kinases.Genes Dev 2007;21:562–77.

47. Jenkins DE, Hornig YS, Oei Y, Dusich J, Purchio T. Bioluminescent humanbreast cancer cell lines that permit rapid and sensitive in vivo detection ofmammary tumors and multiple metastases in immune deficient mice.Breast Cancer Res 2005;7:R444–54.

48. Scott RW, Olson MF. LIM kinases: function, regulation and associationwith human disease. J Mol Med 2007;85:555–68.

49. Rath N, Morton JP, Julian L, Helbig L, Kadir S, McGhee EJ, et al. ROCKsignaling promotes collagen remodeling to facilitate invasive pan-creatic ductal adenocarcinoma tumor cell growth. EMBO Mol Med2017;9:198–218.

50. Vennin C, Chin VT, Warren SC, Lucas MC, Herrmann D, Magenau A, et al.Transient tissue priming via ROCK inhibition uncouples pancreatic cancerprogression, sensitivity to chemotherapy, and metastasis. Sci Transl Med2017;9:eaai8504.

51. Paw I, Carpenter RC, Watabe K, Debinski W, Lo HW. Mechanisms regu-lating glioma invasion. Cancer Lett 2015;362:1–7.

52. van 't Veer LJ, DaiH, van de VijverMJ,He YD,Hart AAM,MaoM, et al. Geneexpression profiling predicts clinical outcome of breast cancer. Nature2002;415:530–6.

53. Pereira B, Chin S-F, Rueda OM, Vollan H-KM, Provenzano E, BardwellHA, et al. The somatic mutation profiles of 2,433 breast cancers refinetheir genomic and transcriptomic landscapes. Nat Commun 2016;7:11479.

54. Arias-Romero LE, Chernoff J. Targeting Cdc42 in cancer. Expert Opin TherTargets 2013;17:1263–73.

55. Liu Z, WuH, Jiang K, Wang Y, ZhangW, ChuQ, et al. MAPK-mediated YAPactivation controls mechanical-tension-induced pulmonary alveolarregeneration. Cell Rep 2016;16:1810–9.


Unbekandt et al.




2018;78:2096-2114. Published OnlineFirst January 30, 2018.Cancer Res Mathieu Unbekandt, Simone Belshaw, Justin Bower, et al. Therapeutic Effect on Skin CancerDiscovery of Potent and Selective MRCK Inhibitors with

Updated version

10.1158/0008-5472.CAN-17-2870doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2018/01/30/0008-5472.CAN-17-2870.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/78/8/2096.full#ref-list-1

This article cites 54 articles, 10 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/78/8/2096.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

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

.http://cancerres.aacrjournals.org/content/78/8/2096To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-17-2870

http://cancerres.aacrjournals.org/content/suppl/2018/01/30/0008-5472.CAN-17-2870.DC1

http://cancerres.aacrjournals.org/content/78/8/2096.full#ref-list-1

http://cancerres.aacrjournals.org/content/78/8/2096.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/78/8/2096


discoveryofpotentandselectivemrckinhibitors with ...have beneﬁcial therapeutic effects in numerous...

Documents