triptolide induces cell killing in multidrug- resistant ... · reduced p-gp and mdr1 mrna in...

Small Molecule Therapeutics

Triptolide Induces Cell Killing in Multidrug-Resistant Tumor Cells via CDK7/RPB1 Ratherthan XPB or p44Jun-Mei Yi1, Xia-Juan Huan1, Shan-Shan Song1, Hu Zhou2, Ying-Qing Wang1,and Ze-Hong Miao1

Abstract

Multidrug resistance (MDR) is a major cause of tumor treat-ment failure; therefore, drugs that can avoid this outcome areurgently needed. We studied triptolide, which directly kills MDRtumor cells with a high potency and a broad spectrum of celldeath. Triptolide did not inhibit P-glycoprotein (P-gp) drug effluxand reduced P-gp andMDR1mRNA resulting from transcriptioninhibition. Transcription factors including c-MYC, SOX-2, OCT-4, and NANOG were not correlated with triptolide-induced cellkilling, but RPB1, the largest subunit of RNA polymerase II, wascritical in mediating triptolide's inhibition of MDR cells. Tripto-lide elicited antitumor and anti-MDR activity through a universalmechanism: by activating CDK7 by phosphorylating Thr170 inboth parental and MDR cell lines and in SK-OV-3 cells. TheCDK7-selective inhibitor BS-181 partially rescued cell killing

induced by 72-hour treatment of triptolide, which may be dueto partial rescue of RPB1 degradation. We suggest that a precisephosphorylation site on RPB1 (Ser1878) was phosphorylated byCDK7 in response to triptolide. In addition, XPB and p44, twotranscription factor TFIIH subunits, did not contribute to tripto-lide-driven RPB1 degradation and cell killing, although XPB wasreported to covalently bind to triptolide. Several clinical trials areunderway to test triptolide and its analogues for treating cancerand other diseases, so our datamay help expand potential clinicaluses of triptolide, as well as offer a compound that overcomestumor MDR. Future investigations into the primary moleculartarget(s) of triptolide responsible for RPB1 degradation maysuggest novel anti-MDR target(s) for therapeutic development.Mol Cancer Ther; 15(7); 1495–503. �2016 AACR.

IntroductionDrug resistance, especially multidrug resistance (MDR), is a

major cause of tumor treatment failure (1). Clinically, few drugsare available that do not produce resistance, and resistance-reversal agents are not available because most tested compoundsreverse MDR by inhibiting drug transporter function, such asP-glycoprotein (P-gp), which mediates tumor MDR. However,drug transporters are critical for physiologic processes at theblood brain barrier, intestine, kidney, and liver (2).

We reported that specific compounds can directly kill MDRtumor cells without affecting the function of P-gp, such as thenatural products salvicine (3, 4), pseudolaric acid B (5, 6), methylspongoate (7), tanshinone I (2, 8), and synthetic small moleculesYCH337 (9) and MT series (10–12). Among them, the MDR-

overcoming activities of natural products were associated withthe regulation of certain transcription factors. However, thesecompounds do not hold promise for chemotherapeutic use, asthey have relatively poor anticancer activity either in vitro(methyl spongoate and tanshinone I), in vivo (pseudolaric acidB, YCH337, and MT series), or in clinical trials (salvicine).

Triptolide (Fig. 1A), a principle ingredient of Tripterygiumwilfordii Hook F, is a unique transcription inhibitor (13) withpotent anticancer activity. Its analogue minnelide is in clinicaltrials for cancer therapy, and several others are also inclinical development, such as LLDT8 for rheumatoid arthritis andPG490-88 and WilGraf for graft rejection after organ transplan-tation (14). Triptolide has been reported to covalently bind to theCys342 residue of XPB, one subunit of TFIIH, a general transcrip-tion factor that regulates RNA polymerase I and II (RNAPII;refs. 15–17). Our previous work indicates that CDK7 and p44,two subunits of TFIIH, may contribute to the degradation ofRPB1 (the largest subunit of RNAPII) and therefore to tumorcell killing (13), but these relationships require clarification.

Here, we report that triptolide directly kills various tumorMDRcells without inhibiting P-gp drug-efflux function. Reduced P-gpby triptolide was due to transcription inhibition. Transcriptionfactors including c-MYC, SOX-2, OCT-4, and NANOG do notcontribute to proliferative inhibition of triptolide, but RPB1does facilitate this in MDR sublines and in parental tumorcell lines. We report that triptolide leads to the phosphorylationof CDK7 at its Thr170 and RPB1 at its Ser1878. XPB and p44appear not to be correlated with triptolide-driven RPB1 degrada-tion and cell killing.

1Division of Antitumor Pharmacology, State Key Laboratory of DrugResearch, Shanghai Institute of Materia Medica, Chinese Academy ofSciences, Shanghai, P.R. China. 2CAS Key Laboratory of ReceptorResearch, Shanghai Institute of Materia Medica, Chinese Academy ofSciences, Shanghai, P.R. China.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Authors: Ze-Hong Miao, Shanghai Institute of Materia Medica,ChineseAcademyof Sciences, 555 ZuChongZhi Road, ZhangjiangHi-Tech Park,Shanghai 201203, China. Phone: 8621-5080-6820; Fax: 8621-5080-6820; E-mail:[email protected]; and Ying-Qing Wang, [email protected]

doi: 10.1158/1535-7163.MCT-15-0753

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 1495

on November 5, 2020. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst March 29, 2016; DOI: 10.1158/1535-7163.MCT-15-0753

http://mct.aacrjournals.org/

Materials and MethodsDrugs, chemicals, and reagents

Triptolide was purchased from Sigma-Aldrich, and BS-181 wasfromSelleck. Verapamil, doxorubicin (DX), and vincristine (VCR)were obtained from Melonepharma. RIPA lysis buffer, ProteinAþG beads, and Rhodamine 123 (Rh123) were from the Beyo-time Institute of Biotechnology (Haimen, China). All antibodieswere commercially available: RPB1 [carboxy-terminal domain(CTD) repeat], p-S5-RPB1, XPB, OCT-4, SOX-2, NANOG, ubiqui-tin, and p-Thr170-CDK7were fromAbcam,GAPDHand IgGwerefrom Beyotime Institute of Biotechnology (Haimen, China),CDK7 and p44 were from Santa Cruz Biotechnology, and c-MYCwas from BD Biosciences.

Cell cultureHuman cancer KB, IM-9, and MES/SA cell lines and doxoru-

bicin-selected resistant MES-SA/DX5 cell line were purchasedfrom the ATCC. Human cancer K562 cells and adriamycin-select-ed resistant K562/A02 cells were purchased from the Institute ofHematology, Chinese Academy of Medical Sciences (Tianjin,China). Human cancer SK-OV-3 cells were obtained from theJapanese Foundation of Cancer Research (Tokyo, Japan).GM21071 and GM02252 cells were purchased from the CoriellInstitute (Camden, NJ). A vincristine-selected resistant KB/VCRsubline was from the Sun Yat-Sen University of Medical Sciences(Guangzhou, China). During this study, all cell lineswere authen-ticated using the short tandem repeat (STR) profiling at ShanghaiGenesky Bio-Tech CO., LTD (KB and KB/VCR,May 2013;MES/SAand MES-SA/DX5, June 2013; SK-OV-3, August 2013; K562,March 2013; K562/A02 and IM-9, February 2014). Cells werealso periodically authenticated with morphologic inspection andtested for mycoplasma contamination. Cell lines were culturedaccording to the manufacturer's instructions.

Proliferative inhibition assaysIC50 values of different agents in adherent and suspended

cells were measured using a sulforhodamine B (SRB; Sigma)assay and the Cell Counting Kit-8 (Dojindo Laboratories)assay, respectively. Cells were seeded into 96-well plates, cul-tured overnight, and treated with gradient concentrations of thetested agents for 72 hours. Optical density for both assays wasread with a SpectraMax 190 (Molecular Devices). Averaged IC50

values were calculated using logit method from three indepen-dent experiments (9).

Colony formation assaysKB and KB/VCR cells were plated into 6-well plates (200 cells/

well). After overnight incubation, cellswere treatedwith triptolideat the indicated concentrations for 72 hours and then fixed with10% trichloroacetic acid (TCA; Sangon), stained with 0.4% SRB,washed with 1% acetic acid, dried, and photographed.

Western blot analysisWestern blot analysis was performed as described previously in

the published literature (10).

Rh123 efflux assaysCells were treated with 5 mmol/L verapamil or 1 mmol/L

triptolide for 90 minutes, followed by incubation with Rh123(1 mg/mL) for 30 minutes. Then, cells were collected and sus-

pended in PBS for 90 minutes and assessed with flow cytometrywith a FACSCalibur cytometer (BD Biosciences; ref. 2).

Plasmid transfectionPlasmid-expressing pMSCV-c-MYC (#18775) was obtained

fromAddgene. Transfectionwas conducted as reported previouslyin the published literature (18).

RNA interferenceXPB gene expression was reduced with specific siRNA duplexes

from Santa Cruz Biotechnology. siRNA transfection was per-formedwith RNAiMAX Transfection Reagent (Invitrogen) accord-ing to the manufacturer's instructions.

qRT-PCRTotal RNA was prepared with the TRIzol reagent (Invitrogen)

and reverse transcribed into cDNA with a PrimeScript RT ReagentKit (TaKaRa). cDNAwas amplifiedwith the SYBR Premix EX TaqIIKit (TaKaRa) in a 7500 Fast Real-Time PCR System (AppliedBiosystems). The PCR program was as follows: 95�C, 30 seconds;40 cycles (for each cycle 95�C, 5 seconds; 64�C, 20 seconds; 72�C,15 seconds); 72�C, 10 minutes. All primers were synthesized bySangon as follows: 50-GTATTCAACTATCCCACCC-30 (forward)and 50- GCTTTATTTCTTTGCCATC-30 (reverse) for MDR1; 50-TCTACAATGAGCT GCGTGTG-30 (forward) and 50-GGTGAG-GATCTTCAT-GAGGT-30 (reverse) for b-actin; 50-CGTCTCCACA-CATCAGCACAA-30 (forward) and 50-TGTTGGCAGC AGGA-TAGTCCTT-30 (reverse) for c-MYC.

Eliminating p44 expression with transcription activator–likeeffector nuclease technique

The transcription activator–like effector nuclease technique(TALEN) technique is a new method for genome editing andgenetic modifications by inducing DNA double-strand breaksthat stimulate error-prone nonhomologous end joining orhomology-directed repair at specific genomic locations (19).To eliminate p44 expression, TALEN was performed with aFASTALE TALEN Kit (SiDanSai Biotechnology) specifically tar-geting p44. Transfected positive TALEN plasmids into SK-OV-3cells and screened with 1.5 mg/mL puromycin. After puromycinscreening, surviving cells were cultured and selected for p44-deficient monoclonal cells.

ImmunoprecipitationSK-OV-3 cells were treated with 1 mmol/L triptolide for the indi-

cated times, and cells were treated as described previously (10).

Determination of the RPB1 phosphorylation site induced bytriptolide

An RPB1-interacting protein mixture from immunoprecipi-tation was separated with SDS-PAGE separation, and in-geldigestion was performed as reported previously (20). Then,protein bands were excised, dehydrated with acetonitrile, anddigested with trypsin at 37�C overnight. The resulting trypticpeptides were dissolved with 0.1% formic acid and centrifugedat 12,000 rpm for 15 minutes. Supernatant was analyzed byLC/MS-MS using an Orbitrap Elite high-resolution mass spec-trometer. MS-MS spectra were searched against the humandatabase using pFind software (21).

Yi et al.

Mol Cancer Ther; 15(7) July 2016 Molecular Cancer Therapeutics1496




Statistical analysesAll data, if applicable, were expressed as mean � SD from at

least three independent experiments. Comparisons between twogroups were performed using Student t test. P < 0.05 was con-sidered statistically significant.

ResultsTriptolide kills MDR cells effectively

We evaluated triptolide potency in P-gp–expressing cellvariants, including doxorubicin-selected variants, MES-SA/DX5(22), and K562/A02 (2), and vincristine-selected KB/VCR(9). These sublines were resistant to agents used for theirestablishment, respectively as depicted in Fig. 1B. Triptolidepotently killed MDR sublines and was 2-fold more potent inMDR MES-SA/DX5 and KB/VCR sublines but approxi-mately equipotent in K562 and K562/A02 cell lines and theresistance factor is in Fig. 1C. Triptolide reduced RPB1 in time-

and concentration-dependent manners in parental cell linesas reported previously (13) and in respective MDR sublines(Supplementary Fig. S1A and S1B and Fig. 1D). However, short-time treatments (within 24 hours) with triptolide did notreduce P-gp or MDR1 mRNA in KB/VCR cells (SupplementaryFig. S1A–S1C).

In contrast, persistent treatments (36 hours or longer)reduced P-gp and MDR1 mRNA in KB/VCR and MES-SA/DX5sublines (Fig. 1D and E), and greater exposure (�72 hours) alsoreduced P-gp and MDR1 mRNA in K562/A02 cells (Supple-mentary Fig. S1D and S1E). Therefore, less P-gp and MDR1mRNA might be due to transcription inhibition caused bydegradation of RPB1. After rescuing RPB1 degradation usinga CDK7-selective inhibitor, BS-181, reduced P-gp was rescuedin KB/VCR cells (Supplementary Fig. S1F). In addition, shorttreatments did not change accumulation of the P-gp substrateRh123 (Fig. 1F). Thus, triptolide is potent and broad spectrumfor overcoming tumor drug resistance, and this activity is

C

E

D

A B

F

200

RF

0.55

1.23

0.54 Triptolide

10

KB/VCR

RF

36.67

163.86

40

0 36 48 60 0 36 48 60

KB/VCR MES-SA/DX5

Triptolide (50 nmol/L) (h)0 36 48 60

K562/A02

P-gp

GAPDH

RPB1

Rel

ativ

e m

RN

A le

vel o

f MD

R1

0 36 48 60 (h)Triptolide (50 nmol/L)

KB Control Rh123 Rh123 + Verapamil Rh123 + Triptolide

KB/VCR1234

5678

OO

H

O

OOH

O

40 KB

KB/VCRK562

K562/A02MES/SA

MES-SA/DX5

IC50 (nmol/L)30

0 0.5 1KB

0 5 10 15 20 25 30 K562

K562/A02MES/SA

MES-SA/DX5

Adriam

ycin Vincristine

IC50 (μmol/L)

Cou

nts

Fluorescence

0.00.20.40.60.81.01.21.4

KB/VCRMES-SA/DX5K562/A02

1

040

80

100 101 102 103

23

45

6

78

Figure 1.Triptolide kills MDR tumor cells. A, thechemical structure of triptolide. B,MDR sublines were resistant to drugsthat were used for their establishment.C, triptolide elicited potent cell killingin MDR sublines and respectiveparental cell lines. Data from threeindependent experiments areexpressed as mean � SD. Theresistance factor (RF) was calculatedas the ratio of IC50 value of MDR cellsto that of corresponding parental cells.D and E, three MDR cell lines weretreatedwith triptolide at 50 nmol/L forindicated time. Cells were lysed andimmunoblotted for RPB1 and P-gp (D),or RNA was extracted and MDR1mRNA was measured (E). F, triptolidedid not affect the efflux of Rh123. Cellswere treated as indicated in Materialsand Methods and assessed with flowcytometry.

Triptolide Kills MDR Tumor Cells

www.aacrjournals.org Mol Cancer Ther; 15(7) July 2016 1497




independent of drug transporters, such as P-gp. However,differences in reducing P-gp expression in different MDR sub-lines might explain different sensitivities to triptolide, possiblyby removing the inhibition of P-gp on caspases (23).

Triptolide inhibits colony formation and downregulatesc-MYC in KB and MDR KB/VCR cells

In addition to overexpression of drug transporters, cancer stemcells have been proposed to contribute to tumor MDR to che-

motherapeutics (24, 25). Cancer stem cells within some specifictumors may be able to self-renew (26), and colony formationassays could be used to reflect this. We observed that MDRKB/VCR cells had enhanced colony formation (�10%) comparedwith parental KB cells. However, triptolide inhibited colonyformation of KB/VCR more potently than that of KB cells, (IC50,0.68 and 0.41 nmol/L respectively; Fig. 2A and SupplementaryFig. S2A). Therefore, MDR KB/VCR cells are more sensitive totriptolide than parental KB cells, similar to that shown in Fig. 1C.

Triptolide (nmol/L, 72 h)

Triptolide (50 nmol/L)

Triptolide (50 nmol/L, h)

Triptolide IC50 (nmol/L)

c-MYC

c-MYC

c-MYC

pMSCV-c-MYC

pMSCV-c-MYC

c-MYC

Rel

ativ

e le

vels

of

c-M

YC

mR

NA

Figure 2.Triptolide differentially inhibitscolony formationofMDRKB/VCRandparental KB cells and the expressionof c-MYC, OCT-4, SOX-2, andNANOG.A, KB andKB/VCR cellswerecultured as indicated in Materials andMethods, and colonies were fixed,stained, and photographed withImageQuant LAS 4000. B, KB andKB/VCR cells were as treated asindicated and immunoblotted forc-MYC, OCT-4, SOX-2, NANOG, andGAPDH. C and D, comparisons ofc-MYC protein (C) and c-MYCmRNA (D) in MDR sublines andrespective parental cell lines. E, KBand KB/VCR cells were transfected asindicated in Materials and Methodsand immunoblotted for c-MYC andGAPDH. F, KB and KB/VCR cells weretransfected as indicated in Materialsand Methods and 72-hour IC50s weremeasured using SRB assay. Data fromthree independent experiments areexpressed as mean � SD.

Yi et al.





Transcription factors such as c-MYC, SOX-2, OCT-4, andNANOG are critical for cancer stem cells (27). Both KB andKB/VCR cells highly express c-MYC, SOX-2, and OCT-4 proteinsbut express low NANOG protein (Fig. 2B). Triptolide reducedc-MYC in a time-dependent manner but only slightly loweredSOX-2 and did not change OCT-4 or NANOG in these cells(Fig. 2B). c-MYC mRNA consistently decreased in triptolide-treated cells (Supplementary Fig. S2B). Similarly, triptolidereduced c-MYC in other cells (Supplementary Fig. S2C andS2D). Notably, all MDR cells expressed more c-MYC proteinand c-MYC mRNA than their corresponding parental cells(Fig. 2C and D and Supplementary Fig. S2B). Thus, tripto-lide-driven reduction of c-MYC expression might contribute tocell killing or enhanced sensitivity of MDR KB/VCR cells.However, ectopic expression of c-MYC did not confirm thisbecause increased c-MYC did not significantly reverse cellkilling caused by triptolide in KB and KB/VCR cells (Fig. 2Eand F). Therefore, exogenous c-MYC might not rescue the cellkilling of triptolide, either in MDR cells or parental cells.

Triptolide activates CDK7 by phosphorylating Thr170, whichleads to phosphorylation of RPB1 at Ser1878

Cell-killing activity of triptolide is correlated with CDK7-mediated RPB1 degradation (13), and this may be true fordrug-resistant tumor cells as well, because triptolide induceddegradation of RPB1 in drug-resistant MES-SA/DX5 andKB/VCR cells in a manner similar to respective parentalMES/SA and KB cells (Fig. 3A). Moreover, pretreatments withBS-181 (28) reversed triptolide-induced RPB1 degradationsimilarly in both MDR KB/VCR and parental KB cells and inovarian cancer SK-OV-3 cells (Fig. 3B).

Phosphorylation of CDK7 at Thr170 has been revealed toaugment phosphorylation of RPB1 at its CTD (29). Weobserved that triptolide increased phosphorylation of CDK7at Thr170 in SK-OV-3 cells (Fig. 3C), as well as in KB and KB/VCR cells (Fig. 3D), which may contribute to activation drivenby triptolide. BS-181 rescued the degradation of RPB1 inducedby triptolide, so we studied whether BS-181 could rescue cellkilling induced by triptolide. Data showed that 10 mmol/L BS-181 could partially rescue cell killing after 72 hours of treat-ment with triptolide (Fig. 3E) in K562, K562/A02, KB, KB/VCR,and SK-OV-3 cells. RPB1 measurements after combining 10mmol/L BS-181 with 30 nmol/L triptolide for 72 hours in bothKB and KB/VCR cells confirmed that BS-181 could partiallyrescue RPB1 degradation (Fig. 3F). Triptolide elicited antitumorand anti-MDR activity through a universal mechanism.

Our previous work indicated that triptolide could phosphor-ylate Ser5 residue(s) in the CTD repeats of RPB1; however,which Ser5 residue(s) can be phosphorylated by triptolide-activated CDK7 is unclear. Thus, using immunoprecipitationto enrich RPB1 in control and triptolide-treated SK-OV-3 cells,we analyzed the phosphorylated site(s) of RPB1 with high-resolution mass spectrometry. RPB1 protein sequence (Sup-plementary Fig. S3) coverage was obtained from controland triptolide-treated SK-OV-3 cells, and changes in molec-ular mass revealed a mass shift of þ80.65 Da at Ser1878within the peptide spanning the residues from 1874 to 1887(1874YSPTSPTYSPTTPK1887) in triptolide-treated cells (Sup-plementary Fig. S4), but not in untreated cells (SupplementaryFig. S5). This mass shift corresponded to the molecular mass ofa phosphate group (Fig. 3G). Moreover, the 1878th serine

residue is exactly the serine residue at the 5th site of thestandard CTD repeat located between the 1874th and 1880thamino acid residues of the RPB1 protein (13). Therefore, here,we defined a precise Ser5 site in CTD repeats, that is, Ser1878,which can be phosphorylated by triptolide-activated CDK7.

Rickert and colleagues reported that CDK7 could phosphor-ylate the Serine 5 at the consensus heptapeptide YSPTSPXin vitro by synthesizing different types of peptides, includingYSPTSPT (30). As the sequence coverage in the mass spectrom-etry analysis is only 33% (Supplementary Fig. S3) and lackof lysines in the CTD resulted in poor trypsin digestions, wefailed to identify other phosphorylated Ser5 site in the CTDrepeats.

XPB does not contribute to RPB1 degradation or cell killinginduced by triptolide

Triptolide can bind to XPB at its Cys342 residue due to acovalent modification by the 12,13-epoxide group of triptolide(15, 16). However, a Cys342 mutation of XPB to threonine, butnot to alanine or to serine, could rescue cell killing of triptolide(16, 17), challenging the notion that XPB contributes to trip-tolide-driven RPB1 degradation and cell killing. Smurnyy andcolleagues studied the relationship between XPBmutations andcell killing induced by triptolide using CRISPR/Cas9 geneediting technology (31); however, various XPB mutations dif-ferent from the Cys342 residue could lead to triptolide resis-tance. Silencing XPB with specific siRNA duplexes (siXPB) didnot reverse the effects of triptolide on Rpb1 (Fig. 4A) or cellproliferation (Fig. 4B) in SK-OV-3 cells. Consistently, triptolidecaused similar RPB1 degradation (Fig. 4C) and proliferationinhibition (Fig. 4D) in both human XPB-deficient (GM02252)and XPB-proficient (IM-9) lymphocyte cells. Triptolide alsoinduced degradation of RPB1 in GM21071, a human fibroblastcell line with XPB deficiency (Fig. 4E). Thus, XPB does notcontribute to RPB1 degradation or cell killing induced bytriptolide, although it covalently binds to XPB.

The p44 subunit of TFIIH is not correlated with RPB1degradation and proliferative inhibition induced by triptolide

p44, another subunit of TFIIH, possesses E3 ubiquitin ligaseactivity in yeast (32). Possibly, triptolide causes degradation ofRPB1 via p44-mediated ubiquitination. To confirm this, theTALEN technique (Supplementary Fig. S6) was used to eliminatethe expression of p44 in SK-OV-3 cells, and different p44-deficientclonal cells were generated (no. 26-4, 26-5, 26-10, 26-12, and26-23; Fig. 5A). However, p44 protein elimination could notprevent RPB1 degradation induced by triptolide (Fig. 5B). At bothlow (10 nmol/L) and high (100 nmol/L) concentrations, tripto-lide also caused similar proliferative inhibition in parentalSK-OV-3 cells and the p44-deficient 26-10 or 26-12 clonal cells(Fig. 5C). Although p44-deficient 26-5 clonal cells had enhancedsurvival at 72 hours with 10 nmol/L triptolide, similar enhance-ments did not occur under other conditions (Fig. 5C). In responseto triptolide treatment, RPB1 was ubiquitinated (and phos-phorylated at Ser5) in p44-deficient 26-5 clonal cells, just as inp44-proficient SK-OV-3 cells (Fig. 5D). These data indicate thatp44 does not mediate ubiquitinated degradation of RPB1 causedby triptolide.

In addition, we show that WWP2, a reported E3 ubiquitinligase that can mediate ubiquitinated degradation of RPB1(33), was not responsible for RPB1 degradation in triptolide-






B

A

BS-181(10 μmol/L)Triptolide (200 nmol/L, 2 h) −

−−

−+

+ ++

β-A

ctin

RP

B1

SK-OV-3

SK-OV-3

C

0 15 30 60 120


45

CDK7

p-T170-CDK7

GAPDH

p-S5-RPB10 30 60 0 30 60 (min)

KB KB/VCRSK-OV-3

Tubulin

Cell lines

K562

K562/A

02 KB

KB/VCR

SK-OV-3

0

50

100

150

IC50

(nm

ol/L

)

Triptolide Triptolide + BS-181

BS-181(10 μmol/L)Triptolide (30 nmol/L, 72 h) −

−−

−+

+ ++ −

−−

−+

+ ++

KB KB/VCR

Tubulin

D

E F

CDK7

p-T170-CDK7

p-S5-RPB1

(min)


200 300 400 500 600 700 800 900 1,000 1,100100

1.7e+003

010

2030

4050

6070

8090

100

y1+

147.

11y3

++ 1

73.4

6

y4++

224

.10

y2+

244.

16b2

+ 25

1.10

b3+

348.

10

y9-P

hosp

ho[S

]++

447.

10

y9++

496

.26

y10-

Phos

pho[

S]++

530

.77

y5+5

43.3

1y1

0++

580.

10

b11+

+ 63

1.35

y6+

630

.1 y

11++

630

.35

y7+

793.

42

y8+

894.

49 y9+

991.

50

y10-

Phos

pho[

S]+

1060

.56

G

Y S P T S P T Y S P T T P K2+

y1y2y3y4y5y6y7y8y9y10y11

b2 b3 b11

MES-SA/Dx5

GAPDH

Triptolide (1 μmol/L, min ) 0 10 20 30 40 50 60 90 120 240

RPB1

MES/SA

KB/VCR

KB

0 10 20 30 40 50 60 90 120 240

KB/VCR

KB

KB/VCR

KB

RPB1

Rel

ativ

e in

tens

ity (%

)

m/z

Heptapeptide consensus sequence in CTD

1878

Figure 3.Triptolide induces RPB1 degradation in both parental and MDR cell lines and RPB1 phosphorylation at Ser1878. A, MES/SA, MES-SA/MX5, KB, and KB/VCR cellswere treated as shown in Materials and Methods and immunoblotted for RPB1. B, SK-OV-3, KB, and KB/VCR cells were pretreated with a CDK7-specific inhibitor and then assayed for RPB1. C, SK-OV-3 cells were treated as indicated and immunoblotted for phosphorylated RPB1 at Ser5 andphosphorylated CDK7 at Thr170. D, KB and KB/VCR cells were treated as indicated and immunoblotted for phosphorylated RPB1 at Ser5 andphosphorylated CDK7 at Thr170. E, cells were cultured as depicted in Materials and Methods, and IC50 values were measured using a Cell CountingKit-8 (CCK-8) assay. F, KB and KB/VCR cells were cultured as indicated and immunoblotted for RPB1. G, SK-OV-3 cells were treated as indicated andimmunoprecipitated RPB1 and separated with SDS-PAGE. Gel bands were excised, digested, and assessed using LC/MS-MS to identify thephosphorylation site of RPB1. m/z, mass-to-charge ratio.

Yi et al.





treated WWP2-deficient F9 cells (Supplementary Fig. S7A).Using BRCA1-reexpressed UWB1.289 cells, we confirmed ourprevious conclusion that triptolide-induced RPB1 degradationwas independent of BRCA1 (Supplementary Fig. S7B; ref. 13).The E3 ubiquitin ligase(s) correlated with ubiquitinated deg-radation of RPB1 responding to triptolide, which remains tobe clarified.

DiscussionWe report that triptolide had potent direct cell killing on three

MDR cell variants (averaged resistance factor of 0.77). P-gp–expressing sublines were derived from the corresponding non-P-gp–expressing parental tumor cell lines by selection with dif-ferent anticancer drugs (2, 9, 22).We also observed that triptolidenearly equipotently killed human pancreatic cancer Capan-1 andEwing sarcoma SK-ES-1 cells and PARP inhibitor simmiparib-selected resistant variants Capan-1/SP (RF: 0.84) and SK-ES-1/SP(RF: 1.36), and these variants did not express detectable drugtransporters, including P-gp, MRP1, or BRCP (data not shown).Thus, triptolide offers broad-spectrum resolution of drug resis-tance in tumor cells, regardless of their tissue source, drug selec-tion, or drug transporter status.

Triptolide hadunique effects onMDR1 gene expression in threeMDR cell variants: with persistent treatments (�36 hours)reduced P-gp and MDR1 mRNA in both KB/VCR and MES-SA/DX5 sublines, but longer exposure (�72 hours) reduced P-gp andMDR1 mRNA in K562/A02 cells for unclear reasons.

Data show that KB/VCR and MES-SA/DX5 were more sensi-tive to triptolide compared with K562/A02. Tainton and col-leagues reported that P-gp could suppress the activation ofcaspases and reduce apoptosis caused by many chemothera-peutic drugs (23). Therefore, we speculated more but slower

reduction of P-gp in K562/A02 cells led to the distinct sensi-tivities of triptolide compared with KB/VCR and MES-SA/DX5. We also identified that the CDK7-selective inhibitor,BS-181, could partially rescue degradation of RPB1 in KB/VCRcells, which would in turn rescue reduction of P-gp. In addition,triptolide did not significantly change SOX-2, OCT-4, andNANOG, but it did reduce c-MYC. This did not appear tocontribute to its ability to reverse drug resistance. In contrast,triptolide led to RPB1 degradation in all three tested MDRsublines. Thus, considering our previous results (13), it isreasonable to conclude that triptolide kills MDR tumor cellsby driving RPB1 degradation just as it kills their correspondingparental cells.

Previous work from our laboratory and others' suggested apossible coordination between XPB, CDK7, and p44, the threesubunits of TFIIH in triptolide-mediated RPB1 degradation(13, 15–17). We report that triptolide activated CDK7 bystimulating its phosphorylation at Thr170 and then led to thephosphorylation of RPB1 at Ser1878, which is located withinthe CTD repeats. However, that either XPB or p44 did notseem to contribute to triptolide-induced RPB1 degradationand cell killing was intriguing. Other known E3 ubiquitinligases, such as BRCA1, VHL (13), and WWP2, were not cor-related with RPB1 degradation induced by triptolide either.Notably, data to confirm whether XPB mediates RPB1 degra-dation and proliferative inhibition induced by triptolide areinconsistent. Titov and colleagues (15), He and colleagues(16), and Titov (17) revealed that triptolide covalently boundto XPB at its Cys342 and thereby inhibited ATPase activityand led to proliferative inhibition. The triptolide analoguelacking the C12,13-epoxide or mutation of Cys342 of XPB tothreonine dramatically reduced triptolide-induced inhibitionof cell proliferation and ATPase activity of XPB (16). However,

0

20

40

60

80

100

1 10 100 1,000

Inhi

bitio

n ra

te (%

)

NC siRNA XPB

15.316.0

RPB1

XPB

GAPDH

Triptolide (45 min) 0 100 200 500 1,000 (nmol/L)0 100 200 500 1,000siNC siXPB

RPB1

XPB

GAPDH

0 0.5 1 2 3 4GM02252 IM-9

Triptolide (1 μmol/L) (h)0 0.5 1 2 3 4

AB

C

D

E

IC50 (nmol/L)

Concentration of triptolide (nmol/L)

IM-9GM02252

Inhi

bitio

n ra

te (%

)

13.9 13.9

0

20

40

60

80

100

1 10 100 1,000Concentration of triptolide (nmol/L)

IC50 (nmol/L)GM21071

0 0.5 1 2 3 4Triptolide (1 μmol/L, h) SK-OV-3

RPB1XPB

β-Actin

Figure 4.XPB does not contribute to triptolide-driven RPB1 degradation and cell killing. A and B, XPB was silenced with specific siRNA in SK-OV-3 cells, whichwere treated with triptolide at the indicated concentrations for 45 minutes and then immunoblotted for XPB and RPB1 (A). Other cells were treatedwith triptolide at the indicated concentrations for 72 hours and then assessed by SRB. The inhibition rate was calculated from three independentexperiments (B). C, IM-9 and GM02252 cells were treated with 1 mmol/L triptolide for the indicated time, and cells were immunoblotted for XPB and RPB1.D, IM-9 and GM02252 cells were treated with triptolide at the indicated concentrations for 72 hours and then assessed by SRB. The inhibition ratewas calculated from three independent experiments. E, GM21071 cells were treated with 1 mmol/L triptolide for the indicated time, and cells wereimmunoblotted for XPB and RPB1. SK-OV-3 cells were used as positive controls.






the studies by the authors cited above (15–17) also found thatmutation of Cys342 of XPB to alanine or serine did not rescuetriptolide-induced proliferative inhibition, although thesemutants were resistant to binding and ATPase inhibition bytriptolide (17). They did not study the effects of triptolidebinding to XPB and RPB1 degradation (15–17). Smurnyy andcolleagues reported that different mutations of XPB could resultin drug resistance to triptolide, but these mutations were alldifferent from the Cys342 residue that the above-mentionedauthors reported (15–17). Smurnyy and colleagues also did notexamine the effects of triptolide binding to XPB and RPB1degradation. In addition, overexpression of mutants of XPB inWT HCT-116 cells did not induce drug resistance to triptolide(31). We found that both XPB knockdown by siRNA and XPBdeficiency (GM02252 and GM21071, 2 cell lines with differenttissue origins, commercially available) did not ease RPB1degradation and proliferative inhibition induced by triptolide.Thus, binding of triptolide to XPB is not a critical factor forinducing RPB1 degradation and subsequent cell killing.

In contrast, all current evidence reveals that triptolide drivesRPB1 degradation via CDK7-mediated Ser5 phosphorylationand subsequent ubiquitination, which is responsible for theproliferative inhibition induced by triptolide. Therefore, find-ing kinase(s) and E3 ubiquitin ligase(s) that are respectivelyresponsible for triptolide-driven CDK7 activation via phos-phorylation at Thr170 and RPB1 ubiquitination will be thefocus of future investigations.

In addition to triptolide, many other natural products, in-cluding salvicine (3, 4), pseudolaric acid B (5, 6), methyl

spongoate, (7) and tanshinone I (2, 8), can induce nearlyequipotent tumor cell killing in MDR sublines and their respec-tive parental cell lines. A common feature of these agents is thatthey do not inhibit P-gp drug-efflux function. Another similar-ity is that their regulation effects on transcription factors,including RPB1, c-Jun, HIF1a, and Stat3, are correlated withthe killing of MDR tumor cells. However, among these com-pounds, triptolide is the most potent for overcoming MDR,possibly because RPB1 as a general transcription factor is farmore critical in regulating transcription of precursor mRNAthan all the other transcription factors. Also, possibly, triptolideapparently reduces P-gp in triptolide-sensitive MDR cells (i.e.,MES-SA/DX5 and KB/VCR cells). Because triptolide and itsanalogues are undergoing clinical testing, overcoming MDRmay assist with its clinical development. Our data offer amolecular mechanism of triptolide-driven RPB1 degradation,which may inform future studies to offer new molecular target(s) for overcoming MDR.

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

Authors' ContributionsConception and design: J.-M. Yi, Y.-Q. Wang, Z.-H. MiaoDevelopment of methodology: J.-M. YiAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): H. Zhou, Z.-H. MiaoAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): J.-M. Yi, Z.-H. MiaoWriting, review,and/or revisionof themanuscript: J.-M.Yi, Y.-Q.Wang,Z.-H.Miao

A B

C D

p4426

-4 26

-526

-1026

-1226

-23

β-Actin

SK-OV-3

0 0.1 1 0 0.1 1 0 0.1 1 0 0.1 1SK-OV-3 26-5 26-10 26-12

RPB1

Triptolide (2 h, μmol/L)

GAPDH

β-ActinRPB1

IP

Triptolide(1 μmol/L, 25 min)

+

_

++

+

_

++

SK-OV-3 26-5

Ubiquitin

p-S5-RPB1

250 kD

250 kD

Cel

l via

bilit

y (%

)

Triptolide (10 nmol/L, h) 0 24 48 72

0

40

80

100

60

20

Triptolide (100 nmol/L, h)0 24 48 72

Cel

l via

bilit

y (%

)

0

40

80

100

60

20

26-526-10SK-OV-3

26-12

Figure 5.p44 does not mediate RPB1degradation and proliferativeinhibition induced by triptolide. A,expression of p44 protein waseliminated by TALEN in SK-OV-3 cellsand five p44-deficient cell cloneswere obtained. SK-OV-3 and p44-deficient cells were collected, lysed,and immunoblotted for p44. B, SK-OV-3 cells and p44-deficient 26-5,26-10, and 26-12 clonal cells weretreated with triptolide at theindicated concentrations for 2 hoursand immunoblotted for RPB1. C,cells were treated with triptolide at10 or 100 nmol/L for the indicatedtime, and then the cell viability wasassessed using SRB assay andcalculated from three independentexperiments. D, SK-OV-3 and p44-deficient cells were treated with1 mmol/L triptolide for 25 minutesand then immunoprecipitated (IP).Western blot analysis was used tomeasure ubiquitination andphosphorylation of RPB1.

Yi et al.





Administrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): X.-J. Huan, S.-S. Song, Z.-H. MiaoStudy supervision: Y.-Q. Wang, Z.-H. Miao

Grant SupportThis work was supported by the National Basic Research Program of

China (no. 2012CB932502; to Z.H. Miao), the National Natural Science

Foundation of China (no. 81321092; to Z.H. Miao), and the State Key Labo-ratory of Drug Research.

The costs of publication of this article were defrayed in part by the pay-ment of page charges. This article must therefore be hereby marked advertise-ment in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received September 11, 2015; revised March 16, 2016; accepted March 19,2016; published OnlineFirst March 29, 2016.

References1. Pluchino KM, Hall MD, Goldsborough AS, Callaghan R, Gottesman MM.

Collateral sensitivity as a strategy against cancer multidrug resistance.Drug Resist Updat 2012;15:98–105.

2. Xu L, Feng JM, Li JX, Zhu JM, Song SS, Tong LJ, et al. Tanshinone-1 inducestumor cell killing, enhanced by inhibition of secondary activation ofsignaling networks. Cell Death Dis 2013;4:e905.

3. Miao ZH,Ding J. Transcription factor c-Jun activation repressesmdr-1 geneexpression. Cancer Res 2003;63:4527–32.

4. Miao ZH, Tang T, Zhang YX, Zhang JS, Ding J. Cytotoxicity, apoptosisinduction and downregulation of mdr-1 expression by the anti-topoisom-erase II agent, salvicine, in multidrug-resistant cells. Int J Cancer 2003;106:108–115

5. Yu B, Li MH,WangW,Wang YQ, Jiang Y, Yang SP, et al. Pseudolaric acid B-driven phosphorylation of c-Jun impairs its role in stabilizing HIF-1alpha:a novel function-converter model. J Mol Med 2012;90:971–81.

6. Yu B,Miao ZH, Jiang Y, LiMH, YangN, Li T, et al. c-Jun protects HIF-1 alphafrom degradation via its Oxygen-Dependent-Degradation domain in anon-transcriptional manner. Cancer Res 2009;69:7704–12.

7. Jiang Y, Miao ZH, Xu L, Yu B, Gong JX, Tong LJ, et al. Drug-transporter-independent liver cancer cell killing by a marine steroid methyl spongoatevia apoptosis induction. J Biol Chem 2011;286:26461–69.

8. Wang Y, Li JX, Wang YQ, Miao ZH. Tanshinone I Inhibits Tumor Angio-genesis by Reducing Stat3 Phosphorylation at Tyr705 and Hypoxia-induced HIF-1a Accumulation in both Endothelial and Tumor Cells.Oncotarget 2015;6:16031–42.

9. Yi JM, Zhang XF, Huan XJ, Song SS, Wang W, Tian QT, et al. DualTargeting of Microtubule and Topoisomerase II by a-CarbolineDerivative YCH337 for Tumor Proliferation and Growth Inhibition.Oncotarget 2015;6:8960–73.

10. Wang W, Wang YQ, Meng T, Yi JM, Huan XJ, Ma LP, et al. MCL-1Degradation Mediated by JNK Activation via MEKK1/TAK1-MKK4 Con-tributes to Anticancer Activity of New Tubulin Inhibitor MT189.Mol Cancer Ther 2014;13:1480–91

11. Zhang Z, Meng T, Yang N, WangW, Xiong B, Chen Y, et al. MT119, a newplanar-structured compound, targets the colchicine site of tubulinarresting mitosis and inhibiting tumor cell proliferation. Int J Cancer2011;129:214–24.

12. ZhangZ,MengT,He J, LiM, Tong LJ, XiongB, et al.MT7, a novel compoundfrom a combinatorial library, arrests mitosis via inhibiting the polymer-ization of microtubules. Invest New Drug 2010;28:715–28.

13. Manzo SG, Zhou ZL, Wang YQ, Marinello J, He JX, Li YC, et al. NaturalProduct Triptolide Mediates Cancer Cell Death by Triggering CDK7-Dependent Degradation of RNA Polymerase II. Cancer Res 2012;72:5363–73.

14. Zhou ZL, Yang YX, Ding J, Li YC, Miao ZH. Triptolide: structural modifica-tions, structure-activity relationships, bioactivities, clinical developmentand mechanisms. Nat Prod Rep 2012;29:457–75.

15. TitovDV,GilmanB,HeQL, Bhat S, LowWK,DangY, et al. XPB, a subunit ofTFIIH, is a target of the natural product triptolide. Nat Chem Biol2011;7:182–8.

16. He QL, Titov DV, Li J, Tan M, Ye Z, Zhao Y, et al. Covalent modificationof a cysteine residue in the XPB subunit of the general transcription

factor TFIIH through single epoxide cleavage of the transcription inhib-itor triptolide. Angew Chem Int Ed Engl 2015;54:1859–63.

17. Titov DV. Mechanism of action of the antiproliferative natural pro-duct triptolide [dissertation]. Baltimore, MD: The Johns HopkinsUniversity; 2012.

18. Hou LL, Gao C, Chen L, Hu GQ, Xie SQ. Essential role of autophagy infucoxanthin-induced cytotoxicity to human epithelial cervical cancerHeLa cells. Acta Pharmacol Sin 2013;34:1403–10.

19. Gaj T, Gersbach CA, Barbas CF 3rd. ZFN, TALEN, and CRISPR/Cas-basedmethods for genome engineering. Trends Biotechnol 2013;31:397–405.

20. Ji S,Qin Y, Shi S, Liu X,HuH, ZhouH, et al. ERK kinase phosphorylates anddestabilizes the tumor suppressor FBW7 in pancreatic cancer. Cell Res2015;25:561–73.

21. Fu Y, Yang Q, Sun R, Li D, Zeng R, Ling CX, et al. Exploiting the kerneltrick to correlate fragment ions for peptide identification via tandemmass spectrometry. Bioinformatics 2004;20:1948–54.

22. Duran GE, Wang YC, Francisco EB, Rose JC, Martinez FJ, Coller J, et al.Mechanisms of resistance to cabazitaxel. Mol Cancer Ther 2015;14:193–201.

23. Tainton KM, Smyth MJ, Jackson JT, Tanner JE, Cerruti L, Jane SM, et al.Mutational analysis of P-glycoprotein: suppression of caspase activation inthe absence of ATP-dependent drug efflux. Cell Death Differ 2004;11:1028–37.

24. Dean M, Fojo T, Bates S. Tumour stem cells and drug resistance. Nat RevCancer 2005;5:275–84.

25. Cojoc M, Mabert K, Muders MH, Dubrovska A. A role for cancer stem cellsin therapy resistance: Cellular and molecular mechanisms. Semin CancerBiol 2015;31C:16–27.

26. ClarkeMF,Dick JE,Dirks PB, Eaves CJ, JamiesonCH, JonesDL, et al. Cancerstem cells–perspectives on current status and future directions: AACRWorkshop on cancer stem cells. Cancer Res 2006;66:9339–44.

27. Gonzalez-Moles MA, Scully C, Ruiz-Avila I, Plaza-Campillo JJ. Thecancer stem cell hypothesis applied to oral carcinoma. Oral Oncol2013;49:738–46.

28. Ali S, Heathcote DA, Kroll SH, Jogalekar AS, Scheiper B, Patel H, et al. Thedevelopment of a selective cyclin-dependent kinase inhibitor that showsantitumor activity. Cancer Res 2009;69:6208–15.

29. Larochelle S, Chen J, Knights R, Pandur J, Morcillo P, Erdjument-BromageH, et al. T-loop phosphorylation stabilizes the CDK7-cyclin H-MAT1complex in vivo and regulates its CTD kinase activity. EMBO J 2001;20:3749–59.

30. Rickert P, Corden JL, Lees E. Cyclin C/CDK8 and cyclin H/CDK7/p36 arebiochemically distinct CTD kinases. Oncogene 1999;18:1093–102.

31. Smurnyy Y, Cai M, Wu H, McWhinnie E, Tallarico JA, Yang Y, et al. DNAsequencing and CRISPR-Cas9 gene editing for target validation in mam-malian cells. Nat Chem Biol 2014;10:623–5.

32. Takagi Y, Masuda CA, Chang WH, Komori H, Wang D, Hunter T, et al.Ubiquitin ligase activity of TFIIH and the transcriptional response to DNAdamage. Mol Cell 2005;18:237–43.

33. Li H, Zhang Z, Wang B, Zhang J, Zhao Y, Jin Y. Wwp2-mediated ubiqui-tination of the RNA polymerase II large subunit in mouse embryonicpluripotent stem cells. Mol Cell Biol 2007;27:5296–305.






2016;15:1495-1503. Published OnlineFirst March 29, 2016.Mol Cancer Ther Jun-Mei Yi, Xia-Juan Huan, Shan-Shan Song, et al. via CDK7/RPB1 Rather than XPB or p44Triptolide Induces Cell Killing in Multidrug-Resistant Tumor Cells

Updated version

10.1158/1535-7163.MCT-15-0753doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/03/29/1535-7163.MCT-15-0753.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/15/7/1495.full#ref-list-1

This article cites 32 articles, 9 of which you can access for free 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://mct.aacrjournals.org/content/15/7/1495To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-15-0753

http://mct.aacrjournals.org/content/suppl/2016/03/29/1535-7163.MCT-15-0753.DC1

http://mct.aacrjournals.org/content/15/7/1495.full#ref-list-1

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

mailto:[email protected]

http://mct.aacrjournals.org/content/15/7/1495


triptolide induces cell killing in multidrug- resistant ... · reduced p-gp and mdr1 mrna in...

Documents