The Caspase-3/PKCd/Akt/VEGF-A Signaling Pathway Mediates … · Akt/VEGF-A pathway may be a promising target for blocking tumor repopulation during radiotherapy, and the increased

Translational Cancer Mechanisms and Therapy

The Caspase-3/PKCd/Akt/VEGF-A SignalingPathway Mediates Tumor Repopulation duringRadiotherapyJin Cheng1, Sijia He1, Min Wang2, Ling Zhou3, Zhengxiang Zhang1, Xiao Feng1, Yang Yu1,JingjingMa4,ChenyunDai5, ShengpingZhang5, Lianhui Sun5,YanpingGong1,YiweiWang1,Minghui Zhao1, Yuntao Luo1, Xinjian Liu6, Ling Tian5, Chuanyuan Li6, and Qian Huang1

Abstract

Purpose: Tumor repopulation is known as a major cause oftreatment failure and/or tumor recurrence after radiotherapy.The underlying mechanism remains unclear. Our previousstudy demonstrated that irradiated apoptotic cells mediatedtumor repopulation, in which caspase-3 played an importantrole. Herein, we investigated downstream effectors of caspase-3 involved in this process.

Experimental Design: A dominant-negative protein kinaseCd (DN_PKCd)mutant that could not be cleaved by caspase-3and therefore could not be activated by irradiation-inducedapoptosis was constructed. DN_PKCd stably transducedtumor cells were compared with wild-type tumor cells fortheir growth stimulation effects in in vitro and in vivo tumorrepopulation models. Downstream effectors of caspase-3 andPKCd were investigated. The role of PKCd was further verifiedin human colorectal tumor specimens.

Results: Inactivation of caspase-3 or caspase-7 attenuatedtumor repopulation and weakened PKCd cleavage. BothDN_PKCd and PKCd inhibitors restrained tumor repopula-tion both in vitro and in vivo. Phosphorylated Akt wasattenuated in caspase-3–, caspase-7–, or PKCd-inactivatedtumor cells. Furthermore, expression of vascular endothelialgrowth factor (VEGF)-A but not hypoxia-inducible factor 1a(HIF1a) was decreased in PKCd- or Akt-inactivated tumorcells. In addition, inhibition of p-Akt, HIF1a, VEGF-A, orVEGF-A receptor reduced tumor repopulation significantly.Finally, increased nuclear translocation of PKCd in colorec-tal tumor specimens was associated with worse patientprognosis.

Conclusions: The caspase-3/PKCd/Akt/VEGF-A axis isinvolved in tumor repopulation and could be exploited as apotential target to enhance the efficacy of radiotherapy.

IntroductionRadiotherapy is amajormodality to treatmalignant cancer (1).

The aim of radiotherapy is to eliminate tumor mass efficientlywith only limited damage to normal organs or adjacent tissues. In

order to do so, the total radiation dose for treating tumor is usuallysplit into 20 to 30 fractions and given over a period of 4 to 6weeks,for example, a total of 45 to50Gy in25 to28 fractions according tothe National Comprehensive Cancer Network Clinical PracticeGuidelines for rectal cancer and colon cancer (2, 3). The intervalsbetween each fraction are designed for normal tissue recovery fromsublethal radiation damage.However, there is evidence that tumorcells also get the chance to regrow, in some instances at a remark-ably accelerated pace. This phenomenon is called tumor repopu-lation or accelerated tumor repopulation (4–7). Despite very earlyrecognition of its existence, the molecular mechanism for tumorrepopulation is not well understood. Some of the mechanismsinvestigated include tumor hypoxia and reoxygenation (8), vas-culogenesis (9), and stimulation of tumor stem cells (10).

In our previous studies, we had demonstrated that eitherradiotherapy or chemotherapy treated, dying breast tumor cellswere able to generate potent growth-stimulating signals andpromote the proliferation of living tumor cells nearby. In thatprocess, activated caspase-3, a key executioner in apoptosis,played an important role. More specifically, activated caspase-3could activate ca2þ-independent phospholipase A2 (iPLA2),which catalyzed the release of arachidonic acid (AA), resultingin increased prostaglandin E2 (PGE2) production. PGE2 hasbeen shown to function as a potent mitogen and strongly stim-ulate the proliferation of tumor cells. We named this novelcaspase-3/iPLA2/AA/PGE2 signaling pathway as the "PhoenixRising" pathway (11–13). Our recent studies showed that in

1Molecular Diagnostic Laboratory of Cancer Center, Shanghai General Hospital,Shanghai Jiao Tong University School of Medicine, Shanghai, China. 2Depart-ment of Pathology, Shanghai General Hospital, Shanghai Jiao Tong UniversitySchool of Medicine, Shanghai, China. 3Department of General Surgery, ShanghaiFourth People's Hospital, Shanghai, China. 4Department of Pharmacy, The FirstAffiliated Hospital, School of Medicine, Soochow University, Suzhou, China.5Institute of Translational Medicine, Shanghai General Hospital, Shanghai JiaoTong University School of Medicine, Shanghai, China. 6Department of Derma-tology, Duke University Medical Center, Durham, North Carolina.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

J. Cheng and S. He contributed equally to this article.

Corresponding Authors: Qian Huang, Molecular Diagnostic Laboratory ofCancer Center, Shanghai General Hospital, Shanghai Jiao Tong UniversitySchool of Medicine, 650 Xinsongjiang Road, Songjiang District, Shanghai201620, China. Phone: 86-21-37798906, Fax: 86-21-37798833, E-mail:[email protected]; and Chuanyuan Li, Department of Dermatology,Duke University Medical Center, Box 3135, Durham, NC 27710. Phone: 919-613-8754; Fax: 919-681-0909; E-mail: [email protected]

Clin Cancer Res 2019;25:3732–43

doi: 10.1158/1078-0432.CCR-18-3001

�2019 American Association for Cancer Research.

ClinicalCancerResearch

Clin Cancer Res; 25(12) June 15, 20193732

on February 14, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst March 19, 2019; DOI: 10.1158/1078-0432.CCR-18-3001

http://crossmark.crossref.org/dialog/?doi=10.1158/1078-0432.CCR-18-3001&domain=pdf&date_stamp=2019-5-20

http://clincancerres.aacrjournals.org/

irradiated, dying tumor cells, the Sonic Hedgehog (SHH) pathwaywas activated and positively associated with tumor repopula-tion (14). In contrast, the Wnt pathway was negatively associatedwith tumor repopulation (15). Interestedly, secreted frizzled-related protein 1 (SFRP1), which acts as a secretory Wnt antag-onist (16) that can be induced by SHH activation, increasedsignificantly in irradiated tumor cells and was implicated incross-talk betweenWnt and SHH pathways during tumor repopu-lation. Therefore, tumor repopulation is a very sophisticatedprocess involving many factors and different pathways.

In the present study, we attempted to identify other factorsdownstream of caspase-3 in the "Phoenix Rising" pathway. Pro-tein kinase Cd (PKCd) drew our attention because it contains acaspase-3 cutting site. More importantly, its serine/threonineprotein kinase is activated after caspase-3 cleavage (17, 18).According to previous studies, it participates in the regulation ofmany cellular biological behaviors such as proliferation, differ-entiation, metabolism, apoptosis, cellular survival, and carcino-genesis (17–22). PKCd consists of an N-terminal regulatoryfragment and a C-terminal catalytic fragment. The N-terminalregulatory fragment has a pseudosubstrate domain (PSD) andtwo cysteine-rich zinc-finger-like sequences (C1A and C1B) in theC1 region. The C-terminal catalytic fragment has an ATP-bindingregion (C3), a catalytically active/substrate-binding region (C4)and a nuclear localization signal (NLS). In between the N- andC-terminal fragments, there is a hinge region with a caspase-3–specific cleavage site. Under normal circumstances, the PSDat the N-terminus binds to the C-terminal substrate-bindingregion and keeps the kinase inactive (17, 23, 24). In response tostress exposures such as ionizing radiation, activated caspase-3proteolytically cleaves PKCd and separates the N-fragment fromthe C-fragment, resulting in the generation of a constitutivelyactivated 40-kDa catalytic fragment (24). The activated catalyticfragment of PKCd translocates from the cytoplasm intothe nucleus, where it phosphorylates downstream proteins(23–25). On the basis of this information, we hypothesized that

caspase-3–mediated PKCd cleavage and activation might beinvolved in tumor repopulation during radiotherapy.

In the present study, we created in vitro and in vivo tumorrepopulation models to study the involvement of caspase-3–mediated PKCd activation and possible downstream factors. Weprovided evidence that the caspase-3/PKCd/Akt/VEGF-A axisplayed an important role in tumor repopulation during radio-therapy. Furthermore, in human colorectal cancer patients, theincreased nuclear translocation of PKCd was significantly associ-ated with worse prognosis. Taken together, the caspase-3/PKCd/Akt/VEGF-A pathway may be a promising target for blockingtumor repopulation during radiotherapy, and the increasednuclear translocation of PKCd might be a novel prognosis bio-marker for patients with colorectal cancer.

Materials and MethodsCell culture and irradiation

Human colorectal cancer cell lines, HT-29 and HCT116, werepurchased from the Cell Bank of the Chinese Academy ofSciences and cultured in Dulbecco's Modified Eagle Medium(DMEM; Thermo Fisher Scientific Inc.) supplemented with10% fetal bovine serum (FBS; Thermo Fisher Scientific Inc.)and 1% penicillin–streptomycin (Thermo Fisher Scientific Inc.)at 37�C under 5% CO2. Each cell line was authenticated andregularly tested for Mycoplasma contamination. Irradiation ofcells with X-ray was performed at a dose rate of approximately3.6 Gy/min using an Oncor linear accelerator (Siemens).

Clonogenic cell survival assayClonogenic cell survival assay was performed following

published instructions (26). In brief, different numbers oftumor cells (200, 400, 800, 1,600, 1 � 104, and 1 � 105 cells)were seeded into 6-cm dishes in triplicates. Twenty-four hourslater, cells were irradiated with different dose of X-ray (0 Gy,2 Gy, 4 Gy, 6 Gy, 8 Gy, and 10 Gy, respectively) and thenincubated for 14 days with medium change once at day 7. Thecolonies were stained with crystal violet, and the numbers ofcolonies with 50 or more cells were counted. The survivingfraction was then calculated.

Constructs and gene transductionThe pLEX lentiviral vector system (Thermo Fisher Scientific

Inc.) was used to deliver an exogenous gene into target cells. Thefirefly luciferase (Fluc) and green fluorescent protein (GFP)fusion gene, caspase-3 and caspase-7 dominant-negativemutantswith a key amino acid change in the catalytic fragment (C163A incaspase-3 and C186A in caspase-7) were generated in our labs.PKCd dominant–negative mutants with a key amino acid changein the caspase-3–specific cleavage site (D329A) at the hingeregion and in the ATP-binding site (K378A) at the C-terminalcatalytic fragment were kindly provided by Dr. Hui Xiong(Shanghai-MOST Key Laboratory of Health and Disease Geno-mics, Chinese National Human Genome Center, Shanghai,China). Lentiviral vectors were packaged in 293T cells followingthe manufacturer's instructions. Tumor cells that stably express-ed the Fluc-GFP fusion protein, DN_caspase-3 (DN_C3),DN_caspase-7 (DN_C7), DN_PKCd (D329A), and DN_PKCd(K378A) were obtained through lentivirus infection and selec-tion with 2 mg/mL puromycin for 14 days and confirmed byWestern blotting.

Translational Relevance

Tumor repopulation is a major cause of treatment failureafter radiotherapy. Tumor hypoxia and reoxygenation, vascu-logenesis, or tumor stem cell proliferation have beenproposedas potential underlying mechanisms. Our previous studysuggested that activated caspase-3 in irradiated or chemo-treated apoptotic tumor cells played an important role. Toidentify potential effectors downstream of caspase-3, we cre-ated a dominant-negative protein kinase Cd (DN_PKCd)mutant that could not be cleaved by caspase-3. We found thatirradiated tumor cells expressing DN_PKCd or treated withPKCd inhibitors exhibited attenuated stimulation effect ontumor repopulationwhen comparedwith irradiatedwild-typetumor cells. Furthermore, we showed that the caspase-3/PKCd/Akt/vascular endothelial growth factor (VEGF)-A axiswas positively associated with tumor repopulation, and thatincreased nuclear translocation of PKCd was significantlycorrelated with worse prognosis in colorectal cancer patients.Our study thus suggests that the caspase-3/PKCd/Akt/VEGF-Asignaling pathway might be a suitable target for blockingtumor repopulation.

The Caspase-3/PKCd/Akt/VEGF-A Pathway in Tumor Repopulation

www.aacrjournals.org Clin Cancer Res; 25(12) June 15, 2019 3733




Establishment of in vitro and in vivo tumor repopulation modeland bioluminescence imaging

The in vitro tumor repopulationwas simulated by coculture of alarge number of dying tumor cells (2.5 � 105 cells/well) with asmall number of Fluc-GFP–labeled living tumor cells (500 or1,000 cells/well) in 24-well plates or 24-well transwell plates inDMEM supplemented with 2% FBS. The dying tumor cells (alsocalled feeder cells) were irradiated or pretreated with 15 mmol/Lcisplatin for 24hours prior touse and thenplated in 24-well platesor in the lower chamber of 24-well transwell plates. The Fluc-GFP–labeled living tumor cells (also called reporter cells) werenontreated andplated in24-well plates or in theupper chamber of24-well transwell plates. The culture medium was replaced everyother day until time to be imaged. For chemical inhibitor assay,the inhibitors were added into the fresh medium at indicatedconcentrations while medium changed. The coculture continuedfor 10 to 14 days, and then D-Luciferin (Promega) in phosphatebuffer saline (PBS) was added to the final concentration of0.15 mg/mL for each well; bioluminescence imaging was per-formed 5 minutes later.

BALB/c nude mice (4–6 weeks old) were purchased fromShanghai SIPPR/BK Laboratory Animal Co. Ltd. The in vivo tumorrepopulation was performed by a subcutaneous injection of100 mL cell suspension containing 1 � 106 10 Gy irradiatedfeeder cells with 1 � 105 Fluc-GFP–labeled reporter cells intothe right hind legs of nude mice, whereas the 1 � 105 Fluc-GFP–labeled reporter cells alone were injected into contralateralhind legs as controls. Bioluminescence imaging was conductedright after tumor cell injection as baseline of bioluminescenceintensity, and further imaging was performed once or twice aweek. Before imaging, the mice were injected intraperitoneallywith 150 mg/kg D-luciferin dissolved in deionized water and10minutes later anesthetized with continuous flow of isoflurane.

Bioluminescence imaging was performed using the NC100instrument (Berthold Technologies GmbH&Co. KG), SPECTRALAmi X (Spectral Instruments Imaging), and IVIS Lumina Series III(PerkinElmer). Bioluminescent signals were quantitativelyanalyzed using the software supplied by the correspondingmanufacturer.

Western blottingWestern blotting was performed as we previously described

(14). For the separation of cytoplasmic and nuclear proteins, theNuclear/Cytosol Fractionation Kit (BioVision) was used accord-ing to the manufacturer's instructions.

ELISAThe VEGF-A concentration in a cell culture supernatant was

evaluated using the Human VEGF Valukine ELISA Kit (R&DSystem). Procedureswere performed strictly following the instruc-tions supplied by the manufacturer.

Immunofluorescence stainingImmunofluorescence staining was performed as we previously

described (27).

Xenograft tumor growthApproximately 1 � 106 HT-29 and HT-29DN_PKCd (D329A)

cells as well as 2 � 106 HT-29 and HT-29DN_C3 cells in 100 mLPBS were injected subcutaneously into the hind legs or flanks ofnude mice, respectively. Tumor size was measured using vernier

caliper once or twice a week until 5 to 6 weeks, then themice wereeuthanized. Tumor volume (V) was calculated using the formula:V ¼ 0.5 � length � width2. Tumor tissue sections were used forsubsequent IHC staining.

IHC stainingIHC staining of xenograft tumor sections was performed as we

previously described (28). Images were taken by an automatedupright microscope system (Leica Microsystems Inc.).

Tumor tissue microarray staining and analysisA tissue microarray including 68 colonic and rectal carcinoma

(colorectal cancer) tissue sections was built from formalin-fixedand paraffin-embedded tissue blocks as we previously described(28). The detailed clinical information about the 68 colorectalcancer samples is described in Supplementary Table S1.

Antibodies and chemicals used in this studyAntibodies against GAPDH, b-Actin, Lamin B1, hemagglutinin

(HA)-tag, Ki67, pan-Akt, phospho-Akt (Ser473), pan-p38, phos-pho-p38, pan-JNK1/2, phospho-JNK1/2, caspase-3 (recognizingboth full-length and cleaved fragment of caspase-3), and caspase-7(recognizing both full-length and cleaved fragment of caspase-7)were purchased from Cell Signaling Technology. Antibodiesagainst HIF1a, VEGF-A, VEGFR2, and PKCd (recognizing bothfull-length and cleaved fragment of PKCd) were purchased fromSanta Cruz Biotechnology. Cisplatin was purchased from Sigma-Aldrich Corporation. GF109203X and Rottlerin were purchasedfrom Tocris Bioscience. LY294002, MK-2206 2HCl, GSK690693,PX-478 2HCl, and Ki8751 were purchased from Selleck Chemi-cals. Bevacizumab was purchased from Roche.

EthicsAnimal studies and protocol for human tumor tissue micro-

array were approved by the Animal Ethics Committee and EthicalReview Board of Shanghai General Hospital, Shanghai Jiao TongUniversity School of Medicine, China (No. 2014DW107 and No.2014KY107, respectively).

Statistical analysisStatistical analysis was carried out using the Statistical Package

for the Social Sciences 24.0 (SPSS, IBM). Normally distributeddata were presented as mean � standard error of the mean.Differences were analyzed by two-tailed Student t test or one-wayanalysis of variance. Survival analysis was carried out using theKaplan–Meier analysis and log-rank test. DifferenceswithP<0.05were determined as statistically significant.

ResultsIrradiation-induced dying colorectal tumor cells potentlystimulate tumor repopulation both in vitro and in vivo

We initially carried out a clonogenic assay to select an optimalX-ray dose that could be used to make dying tumor cells (feedercells). As shown in Supplementary Fig. S1A and S1B, the survivalfractions of 8Gy irradiatedHT-29 andHCT116 cells were approx-imately 0.39%� 0.10% and 0.17%� 0.07%, respectively, where-as 10 Gy irradiated HT-29 or HCT116 cells showed no formationof colonies. As to reporter cells, bioluminescence imaging dem-onstrated that the intensity of bioluminescence signals waslinearly correlated with Fluc-GFP–labeled cell numbers

Cheng et al.

Clin Cancer Res; 25(12) June 15, 2019 Clinical Cancer Research3734




(Supplementary Fig. S2A and S2B; the R2 value of HT-29Fluc andHCT116Fluc cells was 0.9979 and 0.9978, respectively).

Subsequently, we conducted in vitro and in vivo tumor repop-ulation assay to determine whether irradiated dying cells couldpromote living tumor cell proliferation. The results were shownin Fig. 1. In vitro, HT-29Fluc reporter cells growing on irradiatedHT-29 feeder cells exhibited a remarkable increase in luciferaseactivity when compared withHT-29Fluc reporter cells growing onnonirradiated HT-29 feeder cells and HT-29Fluc reporter cellsalone (Fig. 1A). Similar growth-stimulating results were observedin irradiated HCT116 feeder cells (Fig. 1B) as well as cisplatin-treated HT-29 feeder cells (Supplementary Fig. S3). In addition,similar results were also observed when reporter cells were placedin the upper chamber and irradiated feeder cellswere placed in thelower chamber of transwell plates (Fig. 1C and D). More impor-tantly, the presence of irradiated HT-29 feeder cells significantlystimulated the growth ofHT-29Fluc reporter cells (right hind legs)in vivo when compared with HT-29Fluc reporter cells alone (lefthind legs; Fig. 1E). In the same experiment, the tumor volumemeasured by a caliper also showed similar results (Fig. 1F).

Caspase-mediated PKCd cleavage and activation promotetumor repopulation both in vitro and in vivo

We first attempted to confirm that caspase-3 and caspase-7 areinvolved in the tumor repopulation of irradiated colorectal tumorcells. Western blotting showed that the expression of cleaved

caspase-3 and caspase-7 in X-ray (10 Gy) irradiated HT-29 cellsincreased significantly (Fig. 2A). After transduction of DN_C3 orDN_C7 (Fig. 2B), growth stimulation from irradiated HT-29DN_C3 or DN_C7 cells attenuated significantly in comparisonwith irradiated wild-type HT-29 cells (Fig. 2C). These results werecoincident with our previous study in which shRNA of caspase-3was used to reduce caspase-3 expression (11, 12).

Next, we examined PKCd expression levels and its cleavage andits potential involvement in tumor repopulation. Western blot-ting showed an obvious increase of cleaved PKCd in 10 Gyirradiated HT-29 cells (Fig. 2D). As expected, PKCd cleavage wassignificantly decreased in both irradiated HT-29DN_C3 andHT-29DN_C7 cells (Fig. 2E).

We further confirmed the role of PKCd by conducting tumorrepopulation assay and found that 0.4 mmol/L and 1.0 mmol/LGF109203X (GF, a pan-PKC inhibitor) as well as 0.2 mmol/L and0.4 mmol/L Rottlerin (Rot, a PKCd-specific inhibitor) remarkablysuppressed the dying cell-mediated tumor repopulation whencompared with vehicle control. It was noteworthy that the sameconcentration of GF or Rot had no influence on the proliferationof reporter cells alone (Fig. 2F–I). Because PKCd possesses acaspase-3–specific cleavage site (DMQD329/N) at the hinge regionthat is critical to caspase-3 and caspase-7–mediated cleavage andactivation (24), we transduced the DN_PKCd (D329A) mutantinto the HT-29 and HCT116 cells (Fig. 2J and K; ref. 29). Ourresults showed that DN_PKCd (D329A) stably transduced HT-29

Figure 1.

Dying tumor cells stimulated living tumor cell growth both in vitro and in vivo. A, In vitro tumor cell growth on irradiated feeder cells as represented by luciferaseactivities (top). Bottom, representative bioluminescence images of HT-29Fluc cells when seeded among irradiated or nonirradiated HT-29 cells or alone in24-well plates (N¼ 3). B,Quantitative estimates of HCT116Fluc cell growth when seeded with irradiated or nonirradiated HCT116 cells or alone in 24-well platesby bioluminescence imaging (N¼ 4). C,Quantitative estimates of HT-29Fluc cell growth when seeded in the upper chamber with irradiated or nonirradiatedHT-29 cells in the lower chamber or alone in 24-well transwell plates (N¼ 4). D,Quantitative estimates of HCT116Fluc cell growth when seeded with irradiated ornonirradiated HCT116 cells or alone in 24-well transwell plates as mentioned in C (N¼ 4). E, Growth of 1� 105 HT-29Fluc cells injected subcutaneously alone (left)or together with unlabeled but irradiated (10 Gy) 1� 106 HT-29 cells (right) in nudemice (N¼ 5 for each group). Tumor growth represented by luciferase levels.F, Volume of HT-29 tumors from the experiment described in E as measured by use of a caliper 35 days after tumor cell injection. The right plot shows a typicalmouse bearing tumor derived from HT-29Fluc reporter cell alone on its left hind leg and HT-29Fluc reporter cell with irradiated (10 Gy) 1� 106 HT-29 cells on theright hind leg.��, P < 0.01; �� , P < 0.001.






Figure 2.

Caspase-mediated PKCd cleavage and activation in dying tumor cells stimulated living tumor cell growth both in vitro and in vivo. A,Western blotting (top) anddensitometric analysis (bottom) of full-length and cleaved caspase-3 and caspase-7 as well as b-actin in 10 Gy irradiated HT-29 cells on days 0, 1, 2, 3, 4, 5, and 6.C3FL, full-length caspase-3; C3CF, caspase-3 catalytic fragment; C7FL, full-length caspase-7; C7CF, caspase-7 catalytic fragment. B,Western blotting (left) anddensitometric analysis (right) of HA-tag (labeling DN_C3 or DN_C7) and b-actin in wild-type HT-29, HT-29DN_C3, and HT-29DN_C7 cells. C,Quantitativeestimates of HT-29Fluc cell growthwhen seeded alone or with nonirradiated wild-type HT-29 or with irradiated wild-type HT-29 or HT-29DN_C3 andHT-29DN_C7 cells in 24-well plates (N¼ 3). D,Western blotting (top) and densitometric analysis (bottom) of full-length and cleaved PKCd as well as b-actin in10 Gy irradiated HT-29 cells on days 0, 1, 2, 3, 4, 5, and 6. PKCdFL, full-length PKCd; PKCdCF, PKCd catalytic fragment. E,Western blotting (top) anddensitometric analysis (bottom) of full-length and cleaved PKCd as well as GAPDH in nonirradiated or irradiated wild-type HT-29, HT-29DN_C3, andHT-29DN_C7 cells 3 days after radiation. (Continued on the following page.)

Cheng et al.





or HCT116 cells had significantly reduced capacity to promotereporter cell proliferation after irradiation (Fig. 2L and M).Another DN_PKCd mutant that had a key kinase-inactivatinglysinemutation (K378A) in theATP-binding site at theC-terminalcatalytic fragment also reduced the proliferation of reporter cells(Fig. 2J; Supplementary Fig. S4A and S4B; ref. 18). Furthermore,the presence of irradiated HT-29DN_PKCd (D329A) cells dra-matically restrained the growth of reporter cells (right hind legs)in vivo when compared with irradiated wild-type HT-29 cells(left hind legs; Fig. 2N).

Inactivation of PKCd or caspase-3 attenuates tumorigenesisTo further demonstrate the proliferation promoting effect of

PKCd and caspase-3 in tumor repopulation, we performed

tumor xenograft assay. Both HT-29DN_PKCd (D329A) andHT-29DN_C3 cells exhibited remarkably reduced tumorigenicitywhen compared with wild-type HT-29 cells (Fig. 3A and B). Inaddition, IHC staining of the xenograft tumor sections alsoshowed significantly reduced expression of Ki67 in HT-29DN_PKCd (D329A) andHT-29DN_C3derived tumor sections,which indicated attenuated tumor cell proliferation in vivo(Fig. 3C–F).

Akt is activated by the caspase-3/PKCd signaling in dying tumorcells and associated with tumor repopulation

The activation of PKCd after ionizing radiation has been shownto affect the function of several signal transduction proteins. Akt,p38 kinase, and c-Jun N-terminal kinase/stress-activated protein

(Continued.) F–I, Bioluminescence measurement of HT-29Fluc cells or HCT116Fluc cells alone or seeded with irradiated feeder cells with vehicle or GF109203X(a pan-PKC inhibitor) or Rottlerin (a PKCd-specific inhibitor) in 24-well plates, respectively (F, N¼ 3; G–I, N¼ 4). J, A schematic representation of PKCd showingthe location of the D329Amutation in the caspase-3 cleavage site at the hinge region as well as the K378A mutation in the ATP-binding site at the C-terminalcatalytic fragment. NLS, nuclear localization signal; PSD, pseudosubstrate domain. K,Western blotting (left) and densitometric analysis (right) of HA-tag–labeledDN_PKCd (D329A), full-length PKCd, and GAPDH in wild-type and DN_PKCd (D329A) transduced HT-29 and HCT116 cells. L andM, Bioluminescence images ofHT-29Fluc cells and HCT116Fluc cells alone or seeded with nonirradiated wild-type or irradiated wild-type and DN_PKCd (D329A) transduced tumor cells in24-well plates (N¼ 3). N, Growth kinetics of subcutaneously injected 1� 105 HT-29Fluc cells together with 1� 106 irradiated wild-type (left leg) or DN_PKCd(D329A) transduced HT-29 cells (right leg) (N¼ 5). A representative image (right) showed bioluminescence signal at days 0 and 35, which suggested differentialgrowth mediated by irradiated wild-type (left leg) and DN_PKCd (D329A) transduced HT-29 cells (right leg). � , P < 0.05; �� , P < 0.01; �� , P < 0.001; n.s., nosignificance.

Figure 3.

Inactivation of PKCd or caspase-3 attenuated tumor growth in vivo. A, Tumor growth kinetics (left) and a representative image of tumor size (right, on day 35)from nudemice injected with wild-type HT-29 cells (left) or DN_PKCd (D329A) transduced HT-29 cells (right; N¼ 3). B, Tumor growth kinetics (left) and arepresentative image of tumor size (right, on day 42) from nudemice injected wild-type HT-29 cells (left) or HT-29DN_C3 cells (right; N¼ 3). C, RepresentativeIHC images of HA-tag and Ki67 staining from tumor sections derived fromwild-type HT-29 and HT-29DN_PKCd (D329A) xenografts. Scale bar, 25 mm. D,Representative IHC images of HA-tag and Ki67 from tumor sections derived fromwild-type HT-29 and HT-29DN_C3 xenografts. Scale bar, 25 mm. E, Percentageof Ki67-positive cells in wild-type HT-29 and HT-29DN_PKCd (D329A) xenografts. F, Percentage of Ki67-positive cells in wild-type HT-29 and HT-29DN_C3xenografts. � , P < 0.05; �� , P < 0.01; �� , P < 0.001.






Figure 4.

Akt activation correlated to the caspase-3/PKCd signaling, and Akt inactivation attenuated tumor repopulation. A,Western blotting (left) and densitometricanalysis (right) of phosphorylated Akt (Ser473), pan-Akt, phosphorylated p38, pan-p38, phosphorylated JNK1/2, pan-JNK1/2, and GAPDH in irradiated (10 Gy)HT-29 cells at different time points after radiation exposure. B,Western blotting (left) and densitometric analysis (right) of phosphorylated Akt (Ser473), pan-Akt, and GAPDH in nonirradiated or irradiated wild-type HT-29 and HT-29DN_PKCd (D329A) cells 4 hours after radiation. C,Western blotting (left) anddensitometric analysis (right) of phosphorylated p38, pan-p38, phosphorylated JNK1/2, pan-JNK1/2, and GAPDH in nonirradiated or irradiated wild-type HT-29and HT-29DN_PKCd (D329A) cells 48 hours after radiation. D,Western blotting (left) and densitometric analysis (right) of phosphorylated Akt (Ser473), pan-Akt, and GAPDH in nonirradiated or irradiated wild-type HT-29, HT-29DN_C3, and HT-29DN_C7 cells 4 hours after radiation. E and F,Western blotting (left) anddensitometric analysis (right) of phosphorylated Akt (Ser473), pan-Akt, and GAPDH in nonirradiated or irradiated HT-29 cells (4 hours after radiation) pretreatedwith vehicle or 20 mmol/L LY294002 (a PI3K/Akt inhibitor) or 1 mmol/L MK-2206 2HCl (a specific allosteric Akt inhibitor). G–L, Bioluminescence measurement ofHT-29Fluc cells or HCT116Fluc cells alone or seeded with irradiated feeder cells with vehicle or 10 mmol/L LY or 0.2 mmol/L MK or 0.8 mmol/L GSK690693 (aspecific ATP-competitive Akt inhibitor) in 24-well plates, respectively (N¼ 4). ��, P < 0.01; �� , P < 0.001; n.s., no significance.

Cheng et al.





kinase (JNK/SAPK or JNK1/2) were shown to act downstream ofprotein kinases and be involved in regulating the cell cycle, cellproliferation, cell survival, and carcinogenesis (30–38). There-fore, we decided to investigate the activities of those proteins inirradiated dying cells to see if they were involved in tumorrepopulation during radiotherapy.

Western blotting showed remarkably enhanced expression ofphospho-Akt (ser473; 4 hours after radiation), phospho-p38, andphospho-JNK1/2 (48 hours after radiation) in 10 Gy irradiatedHT-29 cells (Fig. 4A). Interestingly, the expression of phospho-Akt(ser473) but not phospho-p38 and phospho-JNK1/2 was dra-matically lessened in irradiated HT-29DN_PKCd (D329A) cellswhen compared with irradiated wild-type HT-29 cells (Fig. 4Band C). Moreover, the expression of phospho-Akt (ser473) wasalso sharply weakened in both irradiated HT-29DN_C3 andHT-29DN_C7 cells (Fig. 4D).

The role of Akt during tumor repopulation was further testedusing a PI3K/Akt inhibitor, LY294002 (LY), and a specific allo-steric Akt inhibitor, MK-2206 2HCl (MK). In comparison withvehicle, 20 mmol/L LY as well as 1 mmol/L MK significantlydecreased the phospho-Akt (ser473) expression in irradiatedHT-29 cells (Fig. 4E and F). Consistent with the Western blottingresults, 10 mmol/L LY as well as 0.2 mmol/L MK remarkablysuppressed the reporter cell growth stimulatedby irradiated feedercells. Meanwhile, the same concentration of LY or MK had noinfluence on controls (Fig. 4G–J). Similar results were alsoobserved when using a specific ATP-competitive Akt inhibitor,GSK690693 (GSK; Fig. 4K and L).

VEGF-A secreted from dying tumor cells mediates the effect ofcaspase-3/PKCd/Akt signaling in an HIF1a-independent wayduring tumor repopulation

In Fig. 1C and D, we had demonstrated that dying cellsstimulated tumor cell repopulation through paracrine signals.Other studies showed that PKCd activation was linked toincreased expression of hypoxia-inducible factor 1a (HIF1a) andsubsequently elevated secretion of vascular endothelial growthfactor (VEGF), an archetypical angiogenic factor involved intumor angiogenesis and tumor growth (30, 39). Recent studiesshowed that VEGF could also directly stimulate tumor cell pro-liferation (40–43). Therefore, we also examinedHIF1a and VEGFinduction as well as their relationship with caspase-3/PKCd/Aktsignaling in tumor repopulation during radiotherapy. Westernblotting showed significantly elevated expression of HIF1a,VEGF-A, and VEGFR2 in 10 Gy irradiated HT-29 cells (Fig. 5A).The increased VEGF-A secretion was also confirmed by ELISA(Fig. 5C). Interestingly, the expression and secretion of VEGF-Abut not HIF1a were significantly attenuated in irradiatedHT-29DN_PKCd (D329A) cells (Fig. 5B and C). In addition,IHC staining of the xenograft tumor sections indeed showedthat VEGF-A expression was significantly lessened in bothHT-29DN_PKCd (D329A) and HT-29DN_C3 xenografts(Fig. 5D). Furthermore, 20 mmol/L LY as well as 1 mmol/L MKsharply suppressed the expression of VEGF-A but not HIF1a inirradiated HT-29 cells (Fig. 5E and F).

To further confirm the role of VEGF-A during tumor repop-ulation, a specific VEGFR2 inhibitor, Ki8751 (Ki), as well as ahumanized monoclonal antibody against VEGF-A (bevacizu-mab, Bev), was used. Bev can bind to and neutralize all humanVEGF-A isoforms and their bioactive proteolytic fragments. Asshown in Fig. 5G– J, Ki or Bev dramatically lessened the reporter

cell growth stimulated by irradiated feeder cells. In contrast,the same concentration of Ki or Bev had no influence oncontrols. In addition, similar growth-inhibition effects werealso observed in transwell plates when Bev was added into thelower chamber with feeder cells (Fig. 5K and L). However, aspecific inhibitor of HIF1a (PX-478 2HCl, PX) could alsoinhibit the stimulation effect of dying feeder cells on livingreporter cells (Supplementary Fig. S5).

Elevated nuclear translocation of PKCd in colorectal tumortissues correlates with worse clinical prognosis

Previous literature has shown that the biological function ofPKCd depends on its catalytic activity and spatial localization.Because PKCd contains an NLS at the C-terminal catalytic frag-ment (Fig. 2J), in response to stress exposures such as ionizingradiation, the catalytic fragment is liberated by caspase-3 andbecomes active and translocates into the nucleus (23, 24). Toconfirm if this is true in our experiments, HT-29 cells wereirradiated with 10 Gy X-ray, and the cytoplasmic and nuclearproteins were separated and extracted, respectively. Western blot-ting showed that the expression of cleaved PKCd significantlyincreased in the nucleus after irradiation when comparedwith thecytoplasm (Fig. 6A). Similar results were observed in irradiatedHT-29 cells when using immunofluorescence staining (Fig. 6Band C). More importantly, IHC staining of the xenograft tumorsections showed that the expression of nuclear PKCd was signif-icantly less in HT-29DN_C3 xenografts (Fig. 6D and E). Todetermine the clinical relevance of PKCd, we evaluated PKCdexpression and its location in 68 human colorectal cancer tumortissue specimens with their clinical data and found that highexpression of PKCd in the nucleus (P ¼ 0.0453) but not in thecytoplasm was significantly associated with worse prognosis incolorectal cancer patients (Fig. 6F and G).

DiscussionThe primary events driving tumor repopulation remain

incompletely defined. In this study, we demonstrated that irra-diation-induced dying colorectal tumor cells could generatepotent growth-stimulating signals to promote living tumor cellgrowth both in vitro and in vivo. Importantly, our results suggestedthat the caspase-3/PKCd/Akt/VEGF-A signaling pathway in dyingtumor cells mediated tumor repopulation during cancer radio-therapy (Fig. 6H).

The finding of PKCd's involvement in tumor repopulation issurprising given contradicting reports in the literature. Someprevious studies have reported that the caspase-3–mediated PKCdcleavagewas akey event facilitating theapoptotic cascade (17, 18).The catalytic fragment of PKCd liberated by caspase-3 cleavagebecame constitutively activated and translocated into the nucleus,where it phosphorylated several proapoptotic proteins, includingp53, p73b, and Rad9 (24, 25). However, PKCd also functions asan antiapoptotic protein. It has been reported that PKCd couldpromote cell survival in several cancers, such as non–small celllung cancer (NSCLC), breast cancer, pancreatic cancer, and livercancer (44), which was achieved via several prosurvival pathways,such as Akt and nuclear factor kB (NF-kB; refs. 31, 44, 45).Moreover, PKCd activation has also been reported to be relatedto therapy resistance. For instance, a study indicated that frac-tionated radiation induced an increase in the glioma-initiatingcell population and less sensitivity to cancer treatment and






Figure 5.

VEGF-A expression and secretion correlated to caspase-3/PKCd/Akt signaling and inhibition of VEGF-A restrained tumor repopulation. A,Western blotting(top) and densitometric analysis (bottom) of HIF1a, VEGF-A, VEGFR2, and GAPDH in 10 Gy irradiated wild-type HT-29 cells at 0, 4, 12, 24, and 48 hours. B,Western blotting (top) and densitometric analysis (bottom) of HIF1a, VEGF-A, and GAPDH in nonirradiated or irradiated wild-type HT-29 and HT-29DN_PKCd(D329A) cells 48 hours after radiation. C, VEGF-A concentration in a cell culture supernatant was measured by ELISA. The supernatant was collected 48 hourslater fromwild-type HT-29 or HT-29DN_PKCd (D329A) cells that were nonirradiated or irradiated with 10 Gy of X-ray (N¼ 3). D, Representative IHC images ofVEGF-A from tumor sections derived fromwild-type HT-29, HT-29DN_PKCd (D329A), and HT-29DN_C3 xenografts. Scale bar, 25 mm. E and F,Western blotting(top) and densitometric analysis (bottom) of HIF1a, VEGF-A, and GAPDH in nonirradiated or irradiated HT-29 cells (48 hours after radiation) pretreated withvehicle or 20 mmol/L LY or 1 mmol/L MK. G and H, Bioluminescence measurement of HT-29Fluc cells or HCT116Fluc cells alone or seeded with irradiated feedercells with vehicle or Ki8751 (a specific VEGFR2 inhibitor) in 24-well plates (G, N¼ 3; H, N¼ 4). I–L, Bioluminescence measurement of HT-29Fluc cells orHCT116Fluc cells alone or seeded with irradiated feeder cells with vehicle or 100 mg/mL bevacizumab (a monoclonal antibody against VEGF-A) in 24-well platesor 24-well transwell plates (N¼ 4). �, P < 0.05; �� , P < 0.01; �� , P < 0.001; n.s., no significance.

Cheng et al.





both implicated in activation of Abl/PKCd signaling (46). Ourresults demonstrating that caspase-3–mediated PKCd cleavageplayed a positive role in tumor repopulation are therefore con-sistent with its protumor survival roles, providing new insights

into the sophisticated interplay between cell death and tumorrepopulation.

New questions do arise with regard to roles of downstreamfactors of PKCd. Although Akt, p38, and JNK1/2 had all been

Figure 6.

Elevated nuclear expression of PKCd in colorectal tumor tissues correlated with worse clinical prognosis. A,Western blotting (top) and densitometric analysis(bottom) of full-length and cleaved PKCd as well as nuclear marker Lamin B1 and cytoplasmic marker GAPDH in irradiated HT-29 cells 0, 12, 24, and 48 hours afterradiation. B, Representative images of immunofluorescence staining of PKCd in HT-29 cells 48 hours after 10 Gy X-ray irradiation (nonirradiated HT-29 ascontrol). Scale bar, 25 mm. C, Fraction of PKCd-positive nuclei from the experiment described in B. D, Representative image of IHC staining of PKCd in tumorsections derived fromwild-type HT-29 and HT-29DN_C3 cell xenografts. Scale bar, 25 mm. E, Fraction of PKCd-positive nuclei from the experiment described inD. F, Representative images of IHC staining of PKCd in human colorectal cancer tumor tissues. Scale bar, 25 mm. G, Kaplan–Meier survival analysis in 68 cases ofpatients with colorectal cancer (PKCd-negative nucleus versus PKCd-positive nucleus, P¼ 0.0453).H, A schematic representation showing the dying tumorcell–mediated tumor repopulation during cancer radiotherapy through the novel caspase-3/PKCd/Akt/VEGF-A signaling pathway. �� , P < 0.001.






reported to be phosphorylated and activated by PKCd (30, 31,36, 37), our data showed that only Akt but neither p38 norJNK1/2 was involved in tumor repopulation in a PKCd-depen-dent way. Further studies will be needed to understand thepreferential use of Akt as the downstream effector of PKCd overthe other two.

Another interesting observation from our study is the differ-ential dependence of HIF1a and VEGF-A on PKCd/Akt activationin tumor cell repopulation. Previously, PKCd activation waslinked to increased expression of HIF1a and subsequently ele-vated secretion of VEGF (30, 39). Moreover, activated Akt couldpromote VEGF expression in an HIF1a-dependent or -indepen-dentway (33, 47). In thepresent study,we showed that theHIF1a,VEGF-A and its receptor VEGFR2 were all significantly increasedin tumor cells after radiation. Furthermore, VEGF-A but notHIF1a was dramatically decreased when PKCd or Akt was inhib-ited. In addition, tumor repopulation was also remarkably atten-uated when blocking VEGF-A by use of a neutralizing antibody orinhibiting VEGFR2 activity by use of a chemical inhibitor. Despiteits independence from PKCd/Akt activation, suppression ofHIF1a expression could also prevent tumor repopulation. Theseresults suggested that both HIF1a and VEGF-A were necessaryin tumor repopulation but the caspase-3/PKCd/Akt signalingpathway mediated VEGF-A secretion in dying cells via anHIF1a-independent way.

The discovery of VEGF-A's involvement in tumor repopulationalso has significant translational implications. Numerous reportshave demonstrated the importance of the VEGF-A/VEGFR2signaling axis in human cancer patients, and inhibitors of bothhave found success in treating human cancer patients (48–50).Our finding of their involvement in radiotherapy-induced tumorcell repopulation suggests that combing Bev with radiotherapycould be an effective strategy to combat tumor repopulation insome patients.

In summary, our study demonstrated that radiotherapyinduced caspase-3–mediated cleavage of PKCd, which caused its

activation and translocation from the cytoplasm into the nucleusand in turn triggered the activation of the PKCd/Akt/VEGF-Asignaling cascade that strongly promotes the repopulation ofthe irradiated tumor. Our findings suggest PKCd/Akt/VEGF-A asnovel therapeutic targets for preventing radiotherapy-inducedtumor repopulation. They also suggest that nuclear translocationof PKCdmaybe used in conjuntionwith cleaved caspase-3 (28) aspotential biomarkers to predict clinical outcomes in colorectalcancer patients.

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

Authors' ContributionsConception and design: Q. Huang, J. Cheng, L. Tian, C. LiDevelopment of methodology: Q. Huang, J. Cheng, S. He, C. LiAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): J. Cheng, S. He, Z. Zhang, X. Feng, Y. Yu, J. Ma,C. Dai, Y. Gong, Y. Wang, M. Zhao, Y. LuoAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Q. Huang, J. Cheng, S. He, Z. Zhang, C. LiWriting, review, and/or revision of the manuscript: Q. Huang, J. Cheng, C. LiAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases):M.Wang, L. Zhou, S. Zhang, L. Sun, L. Tian, C. LiStudy supervision: X. Liu, L. Tian

AcknowledgmentsWe thank the staff at the Department of Radiation Oncology in Shanghai

General Hospital for their continuous help in carrying out cell-irradiationexperiments. This study was supported by the National Natural Science Foun-dation of China (grant 81502648 to J. Cheng, grants 81120108017 and81572951 to Q. Huang, and grant 81572788 to X. Liu).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received September 12, 2018; revised January 12, 2019; accepted March 11,2019; published first March 19, 2019.

References1. Rodriguez-Wallberg KA. Principles of cancer treatment: impact on repro-

duction. Adv Exp Med Biol 2012;732:1–8.2. Benson AB, Venook AP, Al-Hawary MM, Cederquist L, Chen YJ, Ciombor

KK, et al. NCCN Rectal Cancer Guidelines (Version 2. 2018) [Internet].Fort Washington, PA: NCCN. Available from: https://www.nccn.org/professionals/physician_gls/pdf/rectal.pdf.

3. Benson AB, Venook AP, Al-Hawary MM, Cederquist L, Chen YJ, CiomborKK, et al. NCCN Colon Cancer Guidelines (Version 2. 2018) [Internet].Fort Washington, PA: NCCN. Available from: https://www.nccn.org/professionals/physician_gls/pdf/colon.pdf.

4. Hermens AF, Barendsen GW.Changes of cell proliferation characteristicsin a rat rhabdomyosarcoma before and after x-irradiation. Eur J Cancer1969;5:173–89.

5. Stephens TC, Currie GA, Peacock JH. Repopulation of gamma-irradiatedLewis lung carcinomabymalignant cells andhostmacrophage progenitors.Br J Cancer 1978;38:573–82.

6. Withers HR, Taylor JM, Maciejewski B. The hazard of accelerated tumorclonogen repopulation during radiotherapy. Acta Oncol 1988;27:131–46.

7. Kim JJ, Tannock IF.Repopulation of cancer cells during therapy:an important cause of treatment failure. Nat Rev Cancer 2005;5:516–25.

8. Saggar JK, Tannock IF. Chemotherapy rescues hypoxic tumor cells andinduces their reoxygenation and repopulation-an effect that is inhibitedby the hypoxia-activated prodrug TH-302. Clin Cancer Res 2015;21:2107–14.

9. Moeller BJ, Cao Y, Li CY, Dewhirst MW. Radiation activates HIF-1 toregulate vascular radiosensitivity in tumors: role of reoxygenation, freeradicals, and stress granules. Cancer Cell 2004;5:429–41.

10. Duru N, Candas D, Jiang G, Li JJ. Breast cancer adaptive resistance: HER2and cancer stem cell repopulation in a heterogeneous tumor society.J Cancer Res Clin Oncol 2014;140:1–14.

11. Huang Q, Li F, Liu X, Li W, Shi W, Liu FF, et al. Caspase 3-mediatedstimulation of tumor cell repopulation during cancer radiotherapy.Nat Med 2011;17:860–6.

12. Donato AL, Huang Q, Liu X, Li F, Zimmerman MA, Li CY. Caspase 3promotes surviving melanoma tumor cell growth after cytotoxic therapy.J Invest Dermatol 2014;134:1686–92.

13. Li F, Huang Q, Chen J, Peng Y, Roop DR, Bedford JS, et al. Apoptotic cellsactivate the "phoenix rising" pathway to promote wound healing andtissue regeneration. Sci Signal 2010;3:ra13.

14. Ma J, Tian L, Cheng J, Chen Z, Xu B,Wang L, et al. Sonic hedgehog signalingpathway supports cancer cell growth during cancer radiotherapy. PLoSOne2013;8:e65032.

15. Ma J, Cheng J, Gong Y, Tian L, Huang Q. Downregulation of Wnt signalingby sonic hedgehog activation promotes repopulation of human tumor celllines. Dis Models Mech 2015;8:385–91.

16. Surana R, Sikka S, Cai W, Shin EM, Warrier SR, Tan HJ, et al. Secretedfrizzled related proteins: Implications in cancers. Biochim Biophys Acta2014;1845:53–65.

17. Zhao M, Xia L, Chen GQ. Protein kinase cdelta in apoptosis: a briefoverview. Arch Immunol Ther Exp (Warsz) 2012;60:361–72.


Cheng et al.



https://www.nccn.org/professionals/physician_gls/pdf/rectal.pdf

https://www.nccn.org/professionals/physician_gls/pdf/rectal.pdf

https://www.nccn.org/professionals/physician_gls/pdf/colon.pdf

https://www.nccn.org/professionals/physician_gls/pdf/colon.pdf


18. Ghayur T,HuguninM, Talanian RV, Ratnofsky S,QuinlanC, EmotoY, et al.Proteolytic activation of protein kinase C delta by an ICE/CED 3-likeprotease induces characteristics of apoptosis. J Exp Med 1996;184:2399–404.

19. Allen-Petersen BL, Carter CJ, Ohm AM, Reyland ME. Protein kinase Cdeltais required for ErbB2-drivenmammary gland tumorigenesis and negativelycorrelates with prognosis in human breast cancer. Oncogene 2014;33:1306–15.

20. Sassano A, Altman JK, Gordon LI, Platanias LC. Statin-dependent activa-tion of protein kinase Cdelta in acute promyelocytic leukemia cells andinduction of leukemic cell differentiation. Leuk Lymphoma 2012;53:1779–84.

21. Mayr M, Siow R, Chung YL, Mayr U, Griffiths JR, Xu Q. Proteomic andmetabolomic analysis of vascular smooth muscle cells: role of PKCdelta.Circ Res 2004;94:e87–96.

22. Chen Z, Forman LW, Miller KA, English B, Takashima A, Bohacek RA, et al.Protein kinase Cdelta inactivation inhibits cellular proliferation anddecreases survival in human neuroendocrine tumors. Endocr Relat Cancer2011;18:759–71.

23. DeVries TA, Neville MC, Reyland ME. Nuclear import of PKCdelta isrequired for apoptosis: identification of a novel nuclear import sequence.EMBO J 2002;21:6050–60.

24. KurokawaM, Kornbluth S. Caspases and kinases in a death grip. Cell 2009;138:838–54.

25. Griner EM, Kazanietz MG. Protein kinase C and other diacylglyceroleffectors in cancer. Nat Rev Cancer 2007;7:281–94.

26. Munshi A, Hobbs M, Meyn RE. Clonogenic cell survival assay.Methods Mol Med 2005;110:21–8.

27. He S, Cheng J, Sun L, Wang Y, Wang C, Liu X, et al. HMGB1 released byirradiated tumor cells promotes living tumor cell proliferation via para-crine effect. Cell Death Dis 2018;9:648.

28. Zhang Z, Wang M, Zhou L, Feng X, Cheng J, Yu Y, et al. Increased HMGB1and cleaved caspase-3 stimulate the proliferation of tumor cells and arecorrelated with the poor prognosis in colorectal cancer. J Exp Clin CancerRes 2015;34:51.

29. D'Costa AM,DenningMF.A caspase-resistantmutant of PKC-delta protectskeratinocytes from UV-induced apoptosis. Cell Death Differ 2005;12:224–32.

30. Choi YH, Jin GY, Li LC, Yan GH. Inhibition of protein kinase Cdelta attenuates allergic airway inflammation through suppressionof PI3K/Akt/mTOR/HIF-1 alpha/VEGF pathway. PLoS One 2013;8:e81773.

31. Diaz Bessone MI, Berardi DE, Campodonico PB, Todaro LB, Lothstein L,Bal de Kier Joffe ED, et al. Involvement of PKC delta (PKCdelta) in theresistance against different doxorubicin analogs. Breast Cancer Res Treat2011;126:577–87.

32. Fang Y, Xue JL, Shen Q, Chen J, Tian L. MicroRNA-7 inhibits tumor growthandmetastasis by targeting the phosphoinositide 3-kinase/Akt pathway inhepatocellular carcinoma. Hepatology 2012;55:1852–62.

33. Schuurbiers OC, Kaanders JH, van der Heijden HF, Dekhuijzen RP, OyenWJ, Bussink J. The PI3-K/AKT-pathway and radiation resistance mechan-isms in non-small cell lung cancer. J Thorac Oncol 2009;4:761–7.

34. Vivanco I, Sawyers CL.The phosphatidylinositol 3-kinase AKT pathwayin human cancer. Nat Rev Cancer 2002;2:489–501.

35. Seok JS, Jeong CH, Petriello MC, Seo HG, Yoo H, Hong K, et al. Piper-longumine decreases cell proliferation and the expression of cell cycle-associated proteins by inhibiting Akt pathway in human lung cancer cells.Food Chem Toxicol 2018;111:9–18.

36. Cardoso VG, Goncalves GL, Costa-Pessoa JM, Thieme K, Lins BB, CasareFAM, et al. Angiotensin II-induced podocyte apoptosis is mediated byendoplasmic reticulum stress/PKC-delta/p38 MAPK pathway activationand trough increased Na(þ)/H(þ) exchanger isoform 1 activity.BMC Nephrol 2018;19:179.

37. Yu J, Ok SH, Kim WH, Cho H, Park J, Shin IW, et al. Dexmedetomidine-induced contraction in the isolated endothelium-denuded rat aortainvolves PKC-delta-mediated JNK phosphorylation. Int J Med Sci 2015;12:727–36.

38. Wagner EF, Nebreda AR. Signal integration by JNK and p38 MAPK path-ways in cancer development. Nat Rev Cancer 2009;9:537–49.

39. Kim J, Koyanagi T, Mochly-Rosen D. PKCdelta activation mediates angio-genesis via NADPH oxidase activity in PC-3 prostate cancer cells. Prostate2011;71:946–54.

40. Devery AM,Wadekar R, Bokobza SM,Weber AM, Jiang Y, Ryan AJ. Vascularendothelial growth factor directly stimulates tumour cell proliferation innon-small cell lung cancer. Int J Oncol 2015;47:849–56.

41. KodamaM, Kitadai Y, Tanaka M, Kuwai T, Tanaka S, Oue N, et al. Vascularendothelial growth factor C stimulates progression of human gastric cancervia both autocrine and paracrine mechanisms. Clin Cancer Res 2008;14:7205–14.

42. Liang Y, Brekken RA,Hyder SM. Vascular endothelial growth factor inducesproliferation of breast cancer cells and inhibits the anti-proliferativeactivity of anti-hormones. Endocr Relat Cancer 2006;13:905–19.

43. Yu X, Wang W. Tumor suppressor microRNA613 inhibits glioma cellproliferation, invasion and angiogenesis by targeting vascular endothelialgrowth factor A. Mol Med Rep 2017;16:6729–35.

44. Basu A, Pal D. Two faces of protein kinase Cdelta: the contrasting rolesof PKCdelta in cell survival and cell death. Sci World J 2010;10:2272–84.

45. Castilla C, Chinchon D, Medina R, Torrubia FJ, Japon MA, Saez C. PTPL1and PKCdelta contribute to proapoptotic signalling in prostate cancer cells.Cell Death Dis 2013;4:e576.

46. Kim MJ, Kim RK, Yoon CH, An S, Hwang SG, Suh Y, et al. Importance ofPKCdelta signaling in fractionated-radiation-induced expansion of glio-ma-initiating cells and resistance to cancer treatment. J Cell Sci 2011;124:3084–94.

47. Choi SB, Park JB, Song TJ, Choi SY. Molecular mechanism of HIF-1-independent VEGF expression in a hepatocellular carcinoma cell line.Int J Mol Med 2011;28:449–54.

48. Li J, Qin S, Xu RH, Shen L, Xu J, Bai Y, et al. Effect of fruquintinib vs placeboon overall survival in patients with previously treated metastatic colorectalcancer: the FRESCO randomized clinical trial. JAMA 2018;319:2486–96.

49. Goel HL,Mercurio AM.VEGF targets the tumour cell. Nat Rev Cancer 2013;13:871–82.

50. Ciombor KK, Berlin J, Chan E. Aflibercept. Clin Cancer Res 2013;19:1920–5.






2019;25:3732-3743. Published OnlineFirst March 19, 2019.Clin Cancer Res Jin Cheng, Sijia He, Min Wang, et al. Tumor Repopulation during Radiotherapy

/Akt/VEGF-A Signaling Pathway MediatesδThe Caspase-3/PKC

Updated version

10.1158/1078-0432.CCR-18-3001doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2019/04/19/1078-0432.CCR-18-3001.DC2

http://clincancerres.aacrjournals.org/content/suppl/2019/03/19/1078-0432.CCR-18-3001.DC1Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/25/12/3732.full#ref-list-1

This article cites 48 articles, 10 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/25/12/3732.full#related-urls

This article has been cited by 1 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://clincancerres.aacrjournals.org/content/25/12/3732To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-18-3001

http://clincancerres.aacrjournals.org/content/suppl/2019/03/19/1078-0432.CCR-18-3001.DC1 http://clincancerres.aacrjournals.org/content/suppl/2019/04/19/1078-0432.CCR-18-3001.DC2

http://clincancerres.aacrjournals.org/content/suppl/2019/03/19/1078-0432.CCR-18-3001.DC1 http://clincancerres.aacrjournals.org/content/suppl/2019/04/19/1078-0432.CCR-18-3001.DC2

http://clincancerres.aacrjournals.org/content/25/12/3732.full#ref-list-1

http://clincancerres.aacrjournals.org/content/25/12/3732.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/25/12/3732


The Caspase-3/PKCd/Akt/VEGF-A Signaling Pathway Mediates … · Akt/VEGF-A pathway may be a promising target for blocking tumor repopulation during radiotherapy, and the increased

Documents