1

CPX Targeting DJ-1 Triggers ROS-induced Cell Death and 1

Protective Autophagy in Colorectal Cancer 2

3

Jing Zhou1,2†, Lu Zhang2†, Meng Wang2†, Li Zhou2, Xuping Feng2, Linli Yu1, Jiang 4

Lan2, Wei Gao2, Chundong Zhang1, Youquan Bu1, Canhua Huang2, Haiyuan Zhang3 , 5

Yunlong Lei1 6

1 Department of Biochemistry and Molecular Biology, Molecular Medicine and Cancer 7

Research Center, Chongqing Medical University, Chongqing, 400016, P.R.China 8

2 State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, and West 9

China School of Basic Medical Sciences & Forensic Medicine, Sichuan University, and 10

Collaborative Innovation Center for Biotherapy, Chengdu, 610041, P.R. China 11

3 Key Laboratory of Tropical Diseases and Translational Medicine of Ministry of Education 12

& Department of Neurology, the First Affiliated Hospital of Hainan Medical University, 13

Haikou, P. R. China 14

† These authors contributed equally to this work. 15

To whom requests for reprints should be addressed: 16

Prof. Yunlong Lei, Department of Biochemistry and Molecular Biology, and 17

Molecular Medicine and Cancer Research Center, Chongqing Medical University, 18

Chongqing 400016, P. R. China. 19

Tel: +86 13648323625, Email: leiyunlong@cqmu.edu.cn, leiyunlong12345@163.com 20

Prof. Haiyuan Zhang, Key Laboratory of Tropical Diseases and Translational 21

Medicine of Ministry of Education & Department of Neurology, the First Affiliated 22

Hospital of Hainan Medical University, Haikou, P. R. China 23

Tel: + 86 15027037562, Email: hyzhang_88@163.com. 24

2

Short Title: CPX Targeting DJ-1 to treat CRC 25

Abstract 26

Rationale: Colorectal cancer (CRC) is one of the most common cancers worldwide. 27

Ciclopirox olamine (CPX) has recently been identified to be a promising anticancer 28

candidate, however, novel activities and detailed mechanisms remain to be uncovered. 29

Methods: The cytotoxic potential of CPX towards CRC cells was examined in vitro 30

and in vivo. The global gene expression pattern, ROS levels, mitochondrial function, 31

autophagy, apoptosis, etc. were determined between control and CPX-treated CRC 32

cells. 33

Results: We found that CPX inhibited CRC growth by inhibiting proliferation and 34

inducing apoptosis both in vitro and in vivo. The anti-cancer effects of CPX involved 35

the downregulation of DJ-1, and overexpression of DJ-1 could reverse the cytotoxic 36

effect of CPX on CRC cells. The loss of DJ-1 resulted in mitochondrial dysfunction 37

and ROS accumulation, thus leading to CRC growth inhibition. The cytoprotective 38

autophagy was provoked simultaneously, and blocking autophagy pharmacologically 39

or genetically could further enhance the anti-cancer efficacy of CPX. 40

Conclusion: Our study demonstrated that DJ-1 loss-induced ROS accumulation play 41

a pivotal role in CPX against CRC, providing a further understanding for CRC 42

treatment via modulating compensatory protective autophagy. 43

Keywords: Ciclopirox olamine, DJ-1, mitochondria, ROS, Autophagy 44

45

3

Graphical Abstract 46

CPX transcriptionally inhibits DJ-1 to suppress CRC by inducing mitochondrial 47

dysfunction and the accumulation of ROS. Meanwhile, ROS can also activate 48

cytoprotective autophagy through activating AMPK in CRC cells. 49

50

51

52

53

54

55

56

57

58

59

60

61

62

4

Introduction 63

Colorectal cancer (CRC) is one of the most commonly diagnosed cancers among both 64

men and women worldwide, with 8.1% of all new cancer cases and 8.3% of all cancer 65

deaths in 2018. Although over the past several decades historical changes in risk 66

factors (eg. smoking, increased use of aspirin, red and processed meat consumption, 67

obesity, and diabetes) and the dissemination of colonoscopy with polypectomy have 68

decreased the incidence and mortality rates [1], almost 50% of the patients after 69

surgical resection have a risk of recurrence and metastasis leading to death within 5 70

years of diagnosis [2]. Randomized trials have shown that the death and recurrence in 71

patients treated with adjuvant chemotherapy (for patients with stage III/IV and 72

high-risk stage II colon cancer) have been reduced [3, 4]. While the lack of 73

informative markers for the identification of patients at high risk of relapse who might 74

benefit from adjuvant therapy, the effectiveness of chemotherapy has often been 75

limited [5, 6]. Hence at present, searching for novel target-based therapeutic agents 76

for colorectal cancer is an urgent need. 77

Ciclopirox olamine (CPX), a synthetic antifungal agent and iron chelator used to treat 78

mycoses of the skin and nails for more than 20 years, has recently been identified as a 79

promising antitumor candidate [7, 8]. Most recent studies have revealed that CPX 80

induced cell death in many types of tumor, including acute myeloid leukemia (AML), 81

breast cancer, colorectal cancer, neuroblastoma, and inhibited tumor growth in 82

primary AML cells xenografts as well as human breast cancer MDA-MB-231 83

5

xenografts without gross organ toxicity or loss of body weight [9-13]. CPX was 84

capable to inhibit proliferation and induce apoptosis through regulating expression of 85

cell cycle and apoptosis-regulating proteins in breast cancer [10, 11, 14]. Moreover, 86

CPX could induce autophagy by ROS-mediated JNK activation[15]. Notably, previous 87

human phase I study of oral ciclopirox olamine has evaluated its safety, dose 88

tolerance, pharmacokinetics, and pharmacodynamics in patients with refractory 89

hematologic malignancies, demonstrated that ciclopirox was rapidly absorbed and 90

revealed a favorable therapeutic index of CPX, but affected by a short half-life [16]. 91

However, the detailed molecular mechanisms underlying ciclopirox-mediated 92

suppression of tumor growth remain to be further elucidated. 93

DJ-1 (RS/PARK7/CAP1) is a highly conserved homodimeric protein which was 94

initially cloned as a putative oncogene capable of transforming NIH-3T3 cells in 95

cooperation with H-Ras [17]. The Leu166Pro (L166P) mutation and degradation of 96

DJ-1 is known to be linked with early-onset, autosomal recessive Parkinson's disease 97

(PD) and thus DJ-1 is thought to play a role in neuroprotection [18, 19]. In addition, 98

DJ-1 functions primarily as an endogenous antioxidant and protects cells from 99

oxidative injury through modulating many signal transduction, such as extracellular 100

signal-regulated kinase (ERK) 1/2, apoptosis signal-regulating kinase (ASK) 1, 101

Akt/mTOR, and HIF1α signaling pathways [17, 20, 21]. Several lines of evidence 102

subsequently suggest that DJ-1 is over-expressed in multiple types of tumor and is 103

positively correlated with tumor progression, tumor recurrence, and chemotherapeutic 104

6

drug resistance [22]. In our previous study, we demonstrated that DJ-1 promotes CRC 105

proliferation and metastasis and is negatively correlated with patient survival [23]. 106

Therefore, DJ-1 expression level may be utilized for identifying CRC patients that are 107

more likely to relapse and those who may benefit from more aggressive adjuvant 108

chemotherapeutic treatment. 109

Autophagy, a lysosome-dependent catabolic pathway by which damaged or senescent 110

organelles are removed, plays an important role in the regulation of cancer 111

progression and in determining the response of tumor cells to anti-cancer therapy. 112

Four functional forms of autophagy have been described to date: the cytostatic, 113

nonprotective, cytotoxic and cytoprotective autophagy [24], of which cytoprotective 114

autophagy is more frequent in response to chemotherapy. A growing body of evidence 115

implicates that autophagy can enhance the acquired resistance of some cancer cells 116

during chemotherapy resulting in limited effectiveness of anti-cancer drugs [25, 26], 117

suggesting autophagy as a potential target for cancer treatment. 118

In this work, we find that CPX induces apoptosis and growth inhibition in CRC cells 119

via massive accumulation of ROS. CPX promotes transcriptional-mediated 120

downregulation of DJ-1, a potential biomarker for CRC diagnosis and prognosis, 121

which results in the mitochondrial dysfunction, and thereby increases the levels of 122

ROS. Concomitantly, the accumulation of ROS induces cytoprotective autophagy in 123

response to CPX treatment, thus impairs the antitumor effect of CPX. CPX treatment 124

accompanied with autophagy inhibition can effectively inhibit CRC growth. 125

7

Collectively, our results demonstrate an experimental foundation for downregulating 126

DJ-1 as an important mechanism underlying CPX anticancer activity and provide 127

evidence for targeting cytoprotective autophagy as a potential chemotherapeutic 128

strategy. 129

130

Material and methods 131

Cell culture 132

The HIEC, SW480, HT29, DLD-1 and SW620 cell lines were purchased from the 133

American Type Culture Collection (ATCC), HCT116 and RKO cell lines were 134

purchased from Shanghai cell bank, NCM460 was purchased from In Cell. All cell 135

lines were cultured according to the guidelines and were maintained in DMEM(Gibco) 136

supplemented with 100 U/mL penicillin (Sigma), 10 mg/L streptomycin (Sigma), and 137

10% serum (Hyclone) in a humidified incubator at 37 ℃ under 5% CO2 atmosphere. 138

Reagents and antibodies 139

The following primary antibodies were used: DJ-1 (Santa Cruz Biotechnology), 140

RUNX1 (Santa Cruz Biotechnology), Lamp1 (Santa Cruz Biotechnology), LC3 141

(MBL International Corporation), p62 (MBL International Corporation), PARP (Cell 142

Signaling Technology), PRKAA/AMPK (Cell Signaling Technology), 143

phosphor-PRKAA/AMPK (Cell Signaling Technology), Beclin1 (Cell Signaling 144

Technology), ATG5 (Cell Signaling Technology), PCNA (abcam), NFIL3 (Santa Cruz 145

Biotechnology), ETS1 (Santa Cruz Biotechnology), FOXA1 (Santa Cruz 146

8

Biotechnology), C-JUN (Santa Cruz Biotechnology), EGR3 (Santa Cruz 147

Biotechnology), ATF3 (Santa Cruz Biotechnology), MAFF3 (Santa Cruz 148

Biotechnology), BHLHE40 (Santa Cruz Biotechnology), NR4A2 (Santa Cruz 149

Biotechnology). Ciclopirox olamine, Rotenone, APO, NDGA, NAC, Glucose, 150

Oligomycin, 2-DG, CQ, Bafilomycin A1, E64D and PepA were purchased from 151

Sigma. 152

Animal models 153

All studies were approved by the Institutional Animal Care and Treatment Committee 154

of Sichuan University. For establishment of tumor xenografts, DJ-1-overexpressing 155

and vector HCT116 cells were stably transfected using pReceiver-Lv103 lentivirus 156

vector containing puromycin for screening. 1×107 cells were suspended in PBS and 157

injected subcutaneously into 5-week-old female Balb/c mice. When the tumor 158

volumes reached 100 mm3, mice were randomized into two groups, respectively, 159

treated with CPX (25 mg/kg) prepared in a solution (4% ethanol) or vehicle control 160

via oral gavage once daily. Mice were euthanized for analysis after three weeks. 161

Tumor tissues were isolated and frozen in liquid nitrogen or fixed in 10% formalin 162

immediately. 163

Quantitative RT-PCR (qRT-PCR) 164

Total RNA was extracted using Trizol reagent (Invitrogen) and reverse transcribed 165

using Reverse Transcription PrimeScript 1st Stand cDNA Synthesis kit (TaKaRa, 166

Otsu, Japan). qRT-PCR were performed using quantitative PCR reagents SYBR 167

9

PremixEx TaqTM (TaKaRa) following the manufacturer’s instructions. Levels of 168

GAPDH quantified with target genes acts as an internal control and fold-changes were 169

analyzed using the 2−ΔΔCt method. The qRT-PCR primer sequences were shown in 170

Supplementary Table 2. 171

Luciferase reporter constructs and reporter assays 172

The DJ-1 promoter region (from 1 kb upstream of transcription start site to 1 kb 173

downstream) were cloned into pGL3-luciferase reporter vector (Promega). The PCR 174

primers for DJ-1 promoter were as follows: Forward primer: 175

GCGTGCTAGCCCGGGCTCGAGGGATCCTTCTAAGCTCATTCAA, Reverse 176

primer: CAGTACCGGAATGCCAAGCTTGAGCTCTTTTGGAAGCCAT. 177

For luciferase reporter analysis, HCT116 cells were cotransfected with pRL-TK 178

vector (Promega) encoding Renilla luciferase, RUNX1 expression vector, and 179

pGL3-Basic or DJ-1-reporter construct, respectively. After 24 hours, cells were 180

treated with or without ciclopirox at the indicated concentrations for 24 hours, then 181

cells were lysed and luciferase activity was measured using a Dual Luciferase Assay 182

System (Promega) according to the manufacturer's instructions. All transfection 183

experiments were performed 3 to 4 times in triplicate. 184

Mitochondrial and glycolysis stress tests 185

Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) were 186

measured using the Seahorse XF24 analyzer. Briefly, HCT116 and SW480 cells were 187

seeded in plates after 24-hour ciclopirox treatment. After 12 hours, the cells washed 188

10

three times using seahorse assay medium and then loaded into the machine. Repeat 189

measurement cycles (3 minutes Mix, 2 minutes Wait, 3 minutes Measure) three times 190

for Basal and after injection of 10 mM glucose, 1.0 μM oligomycin and 50 mM 191

2-deoxyglucose. 192

Immunofluorescence 193

Cells grown on coverslips that treated with ciclopirox for 24 hours were fixed with 4% 194

paraformaldehyde (Sigma) for 30 minutes and then washed three times with PBS. 195

Fixed cells were permeabilized with 0.4% Triton 100 and 5% BSA for 1 hour at 37 ℃. 196

For staining, cells were incubated with primary antibodies for 12 hours at 4 ℃, 197

followed by incubation with secondary antibodies (Thermo Scientific) for 1 hour at 198

37 ℃. Finally, Nuclei were stained with DAPI for 10 minutes. For autophagy flux 199

studies, cells were transfected with GFP-RFP-LC3 for 24 hours and then treated with 200

ciclopirox for another 24 hours. Images were captured using a confocal microscopy 201

(Carl Zess Micromaging). 202

Immunohistochemistry 203

Immunohistochemical (IHC) staining was performed as described previously [27]. 204

After blocking with goat serum for 1 hour, the sections were then probed with the 205

desired primary antibody, including DJ-1, p-AMPK, LC3, and PCNA. The 206

immunostaining intensity (A) was divided into four grades (0, negative; 1, weakly 207

positive; 2, positive; 3, strongly positive), and the proportion of staining-positive cells 208

(B) was indicated by five grades (0, <5%; 1, 6-25%; 2, 26-50%; 3, 51-75%; 4, >75%). 209

11

The final score for each slide was calculated as ΣA*B. 210

Cell viability and proliferation assays 211

CCK8 assay was performed to determine cell viability. Briefly, cells were seeded in 212

96-well plates at a density of 5000 cells. After treatment, CCK8 (Dojindo, 213

Kamimashiki-gun, Kumamoto, Japan; CK04) were added and incubated for 40 214

minutes. The absorbance value was then determined at 450 nm. 215

The long-term effects of ciclopirox on CRC cell proliferation were analyzed with a 216

colony formation assay. After treatment, 1500-2000 cells were seeded in six-well 217

plates and the medium was changed every 3 days. After two weeks, cells were fixed 218

with 4% paraformaldehyde (Sigma) for 30 minutes and stained with Crystal Violet for 219

another 30 minutes, then the colonies were washed three times and taken photos. 220

Cell proliferation were determined using EDU assay. Cells grown on coverslips were 221

treated with ciclopirox for 24 hours and stained with EDU (RiBoBio) for 12 hours. 222

Then, operating according to the instructions of the manufacturer. Finally, Nuclei were 223

stained with DAPI for 10 minutes and observed using fluorescence microscopy 224

(Leica). 225

Immunoblot and immunoprecipitation 226

Cells were lysed with RIPA buffer (50 mM Tris base, 1.0 mM EDTA, 150 mM NaCl, 227

0.1% SDS, 1% Triton X-100, 1% sodium deoxycholate, 1 mM PMSF) supplemented 228

with protease inhibitor cocktail (Sigma) and then protein lysates were centrifuged and 229

boiled with loading buffer. For immunoprecipitation, Whole cell lysates were prepared 230

12

in RIPA buffer (40 mM TRIS-HCl, pH 7.5, 150 mM NaCl, 0.5% Nonidet P-40, cocktail, 231

5% glycerol, 10 mM NaF) and incubated with 1 μg of antibody overnight at 4 °C. Next 232

day, Sepharose protein A/protein G beads were added for 2 hours. The 233

immune-complexes were then centrifuged and washed 3 times using RIPA buffer. All 234

lysates were quantified by the BCA Protein Assay (Thermo Fisher Scientific) and 235

analyzed by SDS–PAGE. 236

Acridine orange staining 237

To detect the formation of acidic vesicular organelles (AVO), cells were seeded in 238

24-well plates and treated with or without ciclopirox at indicated concentration for 24 239

hours, and then stained with 1 µmol/L acridine orange (Signa-Aldrich) for 15 minutes. 240

Finally, Cells were washed five times and then observed using fluorescence 241

microscopy (Leica). 242

Transfection 243

All siRNAs were designed using BLOCK-iT™ RNAi Designer (Invitrogen) and 244

synthesized by GenePharma (Shanghai, China). The sequences of the siRNAs used 245

are listed in Supplementary Table 3. Cells were transfected with siRNAs using 246

Lipofectamine RNAiMax (Invitrogen) according to the manufacturer’s instruction. 247

pCMV-myc-DJ-1 and Dominant-negative PRKAA1 (DN-PRKAA1/DN-AMPKα1) 248

plasmids were obtained as described previously [23, 28]. The plasmids were 249

transfected into the indicated cells using Lipofectamine 2000 (Invitrogen) according 250

to the manufacturer’s instruction. 251

13

Transmission electron microscopy 252

Transmission electron microscopy assay was used to visualize autophagic vesicles 253

and mitochondrial morphology. After 24-hour ciclopirox treatment, HCT116 cells 254

were fixed in glutaraldehyde (Sigma) and ultrathin sections were prepared using a 255

sorvall MT5000 microtome. Then, the sections were stained by lead citrate and /or 1% 256

uranyl acetate and visualized by Philips EM420 electron microscopy. 257

Flow cytometry 258

The indicated cells treated with ciclopirox for 24 hours were harvested and washed 259

three times with PBS, then resuspended and incubated with PI-ANXA5 solution 260

(KeyGEN Biotech). Apoptosis was analyzed with a FACSCalibur flow cytometer 261

(Becton Dickinson, San Jose, CA USA). 262

Chromatin immunoprecipitation (ChIP) 263

ChIP was performed with the EZ ChIP™ Chromatin Immunoprecipitation kit (catalog 264

# 17-371) according to the manufacturer's instructions. Anti-RUNX1 (Santa Cruz 265

Biotechnology) was used for ChIP. The qRT-PCR primer sequences were shown in 266

Supplementary Table 4. ChIP qPCR data are calculated relative to IgG using the 2−ΔΔCt 267

method. where ΔCt = (Ct [ChIP]−(Ct [Input]−Log2100)). The experiments were 268

repeated at least three times. 269

Data analysis and statistics 270

Data were expressed as means ± s.d. All experiments were performed at least three 271

times. Statistical analysis was performed with GraphPad Prism 6.0 software. 272

Statistical differences between groups were determined using two-tailed Student’s 273

14

t-test. Significance was designated as follows: *, P < 0.05, **, P < 0.01, ***, P < 274

0.001. 275

276

Results 277

CPX inhibits proliferation and induces apoptosis in CRC cell 278

To validate whether CPX exhibited antitumor effect against CRC, CCK8 assay was 279

performed to assess the cell viability in response to CPX treatment in different human 280

CRC cell lines. As shown in Fig. 1A and Fig S1A, CPX treatment for 24 hours 281

markedly decreased the cell viability of various CRC cell lines (HCT116, DLD-1, 282

RKO, HT29, and SW480) with relatively low IC50 values, whereas normal human 283

small intestine or colon mucosal epithelial cell lines HIEC and NCM460 cells 284

demonstrated higher tolerance to CPX. Consistently, the proliferation of CRC cells 285

was significantly inhibited under CPX treatment, as evidenced by reduced colony 286

formation and EdU incorporation (Fig. 1B, 1C, and Fig S1B). We then performed 287

LDH (lactate dehydrogenase) release assay and found that CPX treatment exhibited 288

marked cytotoxicity in HCT116 and SW480 cells (Fig. 1D). Previous studies 289

suggested that CPX exerts antitumor activity by inducing apoptosis [10]. To evaluate 290

whether apoptosis is involved in CPX-induced cytotoxicity in CRC cells, we first 291

detected morphology of CRC cells and found obvious cytoplasmic vacuole formation 292

and morphological changes that are typical of apoptosis in CPX-treated CRC cells 293

(Fig. 1E and Fig. S1C). The pro-apoptotic effect of CPX in HCT116, SW480 and 294

15

HIEC cells was further evidenced by Annexin V/PI staining measured with flow 295

cytometry (Fig. 1F and Fig. S1D). In addition, CPX treatment resulted in increased 296

levels of cleaved PARP, a specific and sensitive marker of apoptosis, in CRC cells 297

(Fig. 1G). Collectively, these results demonstrate that CPX exhibits a considerable 298

antitumor effect in CRC cells in vitro. 299

CPX-induced downregulation of DJ-1 is involved in anti-CRC effects of CPX 300

To demonstrate the molecular mechanisms of CPX-induced CRC suppression, we 301

determined the global gene expression pattern for CPX-treated cells and compared it 302

with that for the controls using RNA-seq. Using a ≥ 2-fold change (FC) and < 0.05 303

P-value as a cut-off to define overexpression or downregulation, we identified a series 304

of differential expressed genes (DEGs) when compared between CPX-treated and 305

control groups. Sequentially, we divided these DEGs into 843 upregulated genes and 306

454 downregulated genes (Fig. 2A and Supplementary Table 1). Notably, the 307

expression of DJ-1 was markedly decreased in CPX-treated CRC cells. We then 308

performed the real-time PCR to ensure the expression levels of DJ-1 under CPX 309

treatment. As shown in Fig. 2B, CPX treatment decreased the levels of DJ-1 mRNA in 310

a dose-dependent manner. This was also confirmed by immunoblot analysis of DJ-1 311

protein levels in CPX-treated CRC cells (Fig. 2C). 312

To investigate whether CPX-induced downregulation of DJ-1 was involved in CRC 313

suppression, we evaluated cell viability and proliferation in CRC cells stably 314

overexpressing DJ-1 under CPX treatment. As shown in Fig. 2D and 2E, DJ-1 315

16

overexpression markedly restored CPX-induced suppression of cell proliferation in 316

HCT116 and SW480 cells. Consistently, the cell viability of DJ-1 overexpression 317

cells was also higher than that of control cells when treated with CPX (Fig. 2F). We 318

then generated a mouse xenograft model by subcutaneously inoculating human CRC 319

HCT116 cells into nude mice. Consistent with in vitro study, exogenous expression of 320

DJ-1 could obviously impair the anti-CRC effects of CPX in vivo, as evidence by 321

faster growth rate and larger size of tumors (Fig. 2G-2I). This was further supported 322

by stronger PCNA intensity in xenografts of DJ-1 overexpression group than those of 323

control group when treated with CPX (Fig. 2J). Taken all, these data suggest that 324

CPX-induced downregulation of DJ-1 is involved in anti-CRC effects of CPX. 325

ROS induced by DJ-1 downregulation are responsible for anti-CRC effects of CPX 326

Previous studies have demonstrated that decreased DJ-1 leads to impaired antioxidant 327

response [29]. Therefore, we determined to detect whether CPX-induced DJ-1 328

downregulation could increase cellular ROS levels. Treatment of CPX dramatically 329

increased cellular ROS levels (Fig. 3A and 3B), whereas exogenous expression of 330

DJ-1 could reverse this phenotype in HCT116 and SW480 cells (Fig. 3C). Next, we 331

would like to identify the role of ROS in CPX-induced growth suppression. We found 332

that combinatorial treatment of CPX with ROS scavenger N-Acetyl-cysteine (NAC) 333

restored CPX-induced growth suppression, as evidence by increased colony formation, 334

EdU incorporation and cell viability in combinatorial treatment cells (Fig. 3D-3F). In 335

addition, treatment of NAC also reduced the pro-apoptotic effect of CPX, as 336

17

evidenced by Annexin V/PI staining and levels of cleaved PARP (Fig. 3G and 3H). 337

Together, these findings reveal that CPX downregulates DJ-1 and induces cellular 338

ROS accumulation, thereby promoting apoptotic cell death. 339

CPX induces mitochondrial dysfunction to promote ROS accumulation in CRC 340

cells 341

In the above study, we have found that CPX induces ROS accumulation and as shown 342

in Fig. 3B, mitochondria, a major source of ROS production, produces more 343

superoxide in CPX-treated CRC cells. To evaluate whether CPX induces 344

mitochondrial alterations in CRC cells, we examined the morphology, quantity, and 345

function of mitochondria. Compared with the normal tubular mitochondria in control 346

group, CPX treatment led to apparent alteration in mitochondrial morphology or even 347

damage (Fig. 4A). After CPX treatment, a significant reduction in mitochondria mass 348

was observed (Fig. 4B). As the relative mitochondrial number can be predicted by 349

quantifying the mitochondrial DNA (mtDNA) copy number, we measured the levels 350

of mtDNA content and found that CPX treatment resulted in decreased mtDNA copy 351

number (Fig. 4C). Next, we tested whether CPX was involved in regulating 352

mitochondria function and measured the oxygen consumption rate (OCR) and 353

extracellular acidification rate (ECAR) in real-time. The results revealed that CPX 354

inhibited mitochondrial oxidative phosphorylation and activated glycolysis (Fig. 355

4D-4E and Fig. S2F). In addition, we performed RNA-seq to analyze the gene 356

expression pattern of OXPHOS and glycolysis pathway. As shown in Fig. 4F and 4G, 357

18

a panel of OXPHOS-related genes were downregulated upon CPX treatment, whereas 358

several key genes involved in glycolysis were upregulated, which was confirmed by 359

Quantitative RT-PCR assay (Fig. 4H and 4I). Collectively, these data suggest that 360

CPX induces mitochondrial dysfunction to promote ROS accumulation in CRC cells. 361

We next investigated the role of DJ-1 in ROS production and mitochondrial function. 362

In the absence of CPX, knockdown of DJ-1 with siRNA increased ROS levels, 363

decreased mtDNA copy number and OXPHOS-related gene expression (Fig. S2A-D), 364

indicating a mitochondrial protective function of DJ-1. Upon the treatment of CPX, 365

overexpression of DJ-1 partially restored the expression of OXPHOS-related genes 366

(Fig. S2E) and the CPX-induced OCR inhibition (Fig. 4D) and attenuated 367

CPX-induced ECAR elevation (Fig. 4E). These results suggest that DJ-1 plays an 368

important role in CPX-induced mitochondrial dysfunction. 369

CPX induces autophagy in CRC cells 370

Increasing evidence has highlighted the potential application of pharmacologically 371

modulating autophagy for cancer treatment [24]. We thus investigated whether 372

autophagy is regulated by CPX in CRC cells. To ascertain this hypothesis, we first 373

examined the protein levels of autophagy-related genes in CPX-treated CRC cells. As 374

results, CPX treatment promoted the turnover of LC3-I to lipidated LC3-II, which is 375

required for autophagosome formation, in a dose- and time-dependent manner in 376

various CRC cells and other types of cancer cells (Fig. 5A and Fig. S3). The 377

autophagic phenotype was further supported by the accumulation of autophagic 378

19

vesicles (LC3 puncta) in CPX-treated cells compared with control cells (Fig. 5B). To 379

further corroborate CPX-induced autophagy, the appearance of double-membraned 380

autophagosomes was investigated by transmission electronic microscopy. As shown 381

in Fig. 5C, there was a significant accumulation of autophagosomes/autolysosomes in 382

CPX-treated cells but not in control cells. Cells were also stained with acridine orange 383

to detect the formation of acidic vesicular organelles (AVO). As shown in Fig. 5D, 384

abundant cytoplasmic AVO formation was readily observed in CPX-treated cells. In 385

addition, mouse xenografts were stained with LC3 to clarify whether CPX could 386

induce autophagy in vivo. As shown in Fig. 5E, CPX-treated xenografts displayed 387

stronger LC3 staining compared with the control group. Consistently, a similar 388

tendency was observed in LC3-II conversion in CPX-treated tumors (Fig. 5F). 389

Also, we examined the expression levels of Atg5 and Beclin 1, two autophagy-related 390

proteins, to clarify whether CPX promoted the formation of autophagic vesicles. As 391

shown in Fig. 5A, CPX enhanced the expression of Atg5 and Beclin 1 in a 392

dose-dependent manner. In addition, it has been reported that the diminished 393

interaction of Beclin 1 with Bcl-2 is a key event in the initiation process of autophagy 394

[30]. We found that CPX treatment decreased the binding of Beclin 1 with Bcl-2 (Fig. 395

5G). Moreover, silencing the expression of either Beclin 1 or Atg5 using siRNA 396

prominently inhibited elevation of LC3 lipidation and endogenous LC3 puncta 397

accumulation in CPX-treated cells (Fig. S4A-D). Co-administration of wortmannin, a 398

PI3K inhibitor, with CPX failed to induce autophagy (Fig. S4E). Taken all, these data 399

20

suggest that CPX induces autophagy in CRC cells both in vitro and in vivo. 400

CPX promotes autophagy flux in CRC cells 401

To determine whether CPX induces complete autophagic flux, we examined the 402

protein levels of p62, a well-known autophagic substrate. As shown in Fig. 5A, we 403

observed decreased p62 levels in CPX-treated cells along with the increased LC3-II 404

levels, implying the induction of complete autophagic flux. In addition, combinatorial 405

treatment of CPX with autolysosome inhibitors (CQ, Baf A1, or E64D and Pepstatin 406

A) resulted in enhanced LC3-II turnover and accumulation of endogenous LC3 puncta 407

(Fig. 6A-D and Fig. S5). Using a tandem monomeric RFP-GFP-tagged LC3, we 408

found that CPX treatment resulted in dramatic formation of red fluorescent 409

autophagosomes, which transformed into yellow under co-treatment with CQ (Fig. 410

6E). To further confirm the fusion of autophagosome with lysosome in CPX-treated 411

CRC cells, we examined the colocalization of LC3 with LAMP1 (lysosome marker). 412

CPX treatment induced obvious colocalization of LC3 with LAMP1, suggesting the 413

fusion of autophagosome with lysosome (Fig. 6F). Together, these findings reveal that 414

CPX induces complete autophagic flux in CRC cells. 415

DJ-1/ROS/AMPK axis contributes to CPX-induced autophagy initiation 416

AMPK pathway is the main autophagy-regulated signaling pathway and tightly linked 417

to the status of mitochondria. We then tend to identify whether AMPK pathway was 418

involved in CPX-induced autophagy in CRC cells. As shown in Fig. S6A and B, CPX 419

treatment resulted in activation of AMPK and inhibition of mTOR pathway, as 420

21

evidence by increased phosphorylation levels of AMPK both in CPX-treated cells and 421

CPX-treated xenografts. To determine whether the activation of AMPK was involved 422

in CPX-induced autophagy, we transfected AMPK siRNA or domain negative AMPK 423

(DN-AMPK) to inhibit AMPK activation. As shown in Fig. S6C-F, inhibition of 424

AMPK prominently inhibited elevation of LC3 lipidation, endogenous LC3 puncta 425

accumulation, and cytoplasmic AVO formation in CPX-treated cells. Thus, these 426

results suggest that CPX-induced autophagy is dependent on AMPK activation in 427

CRC cells. 428

In the above study, we have found that CPX could downregulate DJ-1 in CRC cells. 429

We wonder whether CPX-induced downregulation of DJ-1 was involved in 430

CPX-induced autophagy. To ascertain this hypothesis, we transfected a Myc-DJ-1 to 431

enforce the exogenous expression of DJ-1 in CRC cells. As shown in Fig. S7A and B, 432

exogenous expression of DJ-1 inhibited activation of AMPK and elevation of LC3 433

lipidation both in CPX-treated cells and CPX-treated xenografts. This was further 434

confirmed by decreased formation of endogenous LC3 puncta, reduced AVO 435

formation, and diminished colocalization between LC3 and LAMP1 (Fig. S7C-E). In 436

contrast, knockdown DJ-1 could induce elevation of LC3 lipidation (Fig. S7F). 437

Therefore, these data demonstrate that CPX-induced autophagy through 438

downregulation of DJ-1 in CRC cells. 439

Next, we also elucidated the role of ROS accumulation in CPX-induced autophagy. 440

As shown in Fig. S8A-C, combinatorial treatment of CPX with ROS scavenger (NAC) 441

22

resulted in AMPK inhibition, reduced LC3-II turnover, decreased accumulation of 442

endogenous LC3 puncta and AVO formation. Moreover, we examined the protein 443

levels of LC3 in CRC cells treated with CPX or/and apocynin, rotenone and 444

nordihydroguaiaretic acid (NDGA) used to block NADPH oxidase, mitochondrial and 445

5-LOX-driven ROS release, respectively. The results shown that only rotenone can 446

reduce LC3-II turnover (Fig. S8D and E). Collectively, these findings reveal that 447

CPX-induced autophagy is dependent on increased mitochondria ROS levels in CRC 448

cells. 449

Inhibition of autophagy augments the anti-CRC effects of CPX 450

To evaluate whether autophagy was involved in the anti-CRC effect of CPX, CRC 451

cells were treated with CPX combined with autophagy inhibitor CQ. As shown in Fig. 452

7A and 7B, combinational use of CQ with CPX boost CPX-induced growth inhibition. 453

Consistently, similar results were obtained by inhibition of autophagy using 454

siRNA-targeted ATG5 knockdown (Fig. 7C). A similar decrease in cell proliferation 455

was also observed in CPX-treated CRC cells in combination with CQ as evidenced by 456

colony formation analysis and EdU labeling (Fig. 7D-7F). We also observed that 457

ATG5 knockdown could augment CPX-induced apoptotic cell death, as evidence by 458

Annexin V/PI staining measured with flow cytometry (Fig. 7G). In conclusion, these 459

results suggest that inhibition of autophagy augments the anti-CRC effects of CPX. 460

Transcriptional inhibition of DJ-1 by RUNX1 461

The mRNA expression of DJ-1 was significantly inhibited by CPX treatment (Fig. 462

23

2B). Herein, to explore whether DJ-1 was downregulated at a transcriptional level, we 463

analyzed the expression of transcription factors under CPX treatment using 464

transcriptome sequencing in HCT116 and SW480 cells. As shown in Fig. 8A and 8B, 465

the expression of various transcription factors was regulated by CPX. Quantitative 466

PCR further confirmed the mRNA levels of several transcription factors (Fig. 8C). To 467

validate which of these transcription factors mediated the suppression of DJ-1, we 468

interfered them using specific siRNAs, and found that silence of RUNX1 restored, at 469

least partially, CPX-induced DJ-1 downregulation (Fig. 8D, E and S9A, B). 470

Consistently, overexpression of RUNX1 inhibited the expression of DJ-1, which 471

could not be further inhibited by CPX treatment (Fig. 8F), while overexpression of 472

NFIL3, FOXA1, ETS1 showed no observed effect on DJ-1 expression (Fig. S9C). 473

Promoter prediction revealed that nine RUNX1 binding sites exist in DJ-1 promoter 474

region (Fig. S10). We then cloned DJ-1 promoter sequence into pGL3-luciferase 475

reporter vector and conducted luciferase reporter analysis. As shown in Fig. 8G and 476

S9D, compared to other transcription factors regulated by CPX, only expression of 477

RUNX1 markedly suppressed the luciferase activity, while CPX treatment failed to 478

show additional effect on DJ-1 promoter activity, indicating that RUNX1 bound to 479

DJ-1 promoter and inhibit DJ-1 transcription. To investigate whether RUNX1 directly 480

binds to DJ-1 promoter, we used chromatin immunoprecipitation (ChIP) assay. Six 481

sets of PCR primers which contain nine predicted RUNX1-binding sites were 482

constructed (Table S4). As shown in Fig. 8H, regions 1, 2 and 4, which contain 483

24

RUNX1 binding sites 1, 2, 3 and 7 in DJ-1 promoter, were found to bind with 484

RUNX1. Taken together, these results suggest that CPX-induced DJ-1 inhibition is 485

largely due to the upregulation of RUNX1, which transcriptionally inhibits DJ-1 by 486

directly binding to its promoter. 487

488

Discussion 489

CPX is an off-patent antifungal agent, the effectiveness and safety of which have been 490

well demonstrated [31-34]. Recently, CPX has been found to have considerable 491

potential to inhibit various types of cancer [7, 9, 35-37]. It is of great interest to reveal 492

activities and mechanisms of CPX in cancer treatment. In the present study, we 493

demonstrated that CPX exhibited potent antitumor effect on CRC both in vitro and in 494

vivo. CPX inhibited the growth of CRC via inhibition of cell proliferation and 495

induction of apoptosis, which were involved in the downregulation of DJ-1. Loss of 496

DJ-1 protein resulted in the mitochondrial dysfunction, which led to the accumulation 497

of ROS, thus inhibiting cell proliferation and inducing apoptosis. Interestingly, 498

downregulation of DJ-1 simultaneously provoked autophagy, which played a 499

cytoprotective role under CPX treatment. Blockage of autophagy using CQ or ATG5 500

siRNA markedly enhanced the antitumor effect of CPX. Our results provide 501

supporting evidence for DJ-1 acting as a modulating autophagy to benefit anti-cancer 502

efficacy during anti-cancer agent treatment. 503

25

The therapeutic potential of CPX has attracted increasing interest. Numerous studies 504

have suggested the antitumor activity of CPX. Previous pharmacology and toxicology 505

study have confirmed that the LD50 value of CPX in mice, rats, and rabbits ranged 506

from 1700 to 3290 mg/kg after oral administration [38, 39]. In mammals, the CPX 507

serum concentration of 10 μM were achievable without toxicity when the compound 508

is administered orally in rats and dogs [38, 39]. Recently, several groups have 509

demonstrated that CPX, given to nude mice at 20-25 mg/kg/day, potently inhibit 510

tumor growth in myeloid leukemia, breast cancer and pancreatic cancer xenografts, 511

but does not display obvious toxicity [33, 40]. Additionally, after oral administration 512

to humans, 96% of the administered CPX could be recovered from urine and the drug 513

concentrations of CPX was pharmacologically achievable, although it’s half-life is 514

short and requires dosing repeatedly [39]. Notably, short half-life may represent a 515

limitation for the use of CPX, and the use of CPX prodrugs or new drug delivery 516

systems may rewire a way to improve pharmacodynamic profile. For example, 517

Andrew M. Davidoff et al. delivered CPX via a subcutaneously implanted, 518

continuous release pump to treat neuroblastomas [11]. More importantly, John A. 519

Taylor III et al. designed and synthesized the phosphoryl-oxymethyl ester of CPX, 520

fosciclopirox (CPX-POM) which has higher oral bioavailability and is able to be 521

metabolized rapidly and completely to the active CPX when intravenously injected 522

[41]. The safety, dose tolerance, pharmacokinetics and pharmacodynamics of 523

intravenous CPX-POM are currently being characterized in a United States 524

26

multi-center first-in-human Phase 1 clinical trial in patients with advanced solid 525

tumors (NCT03348514). In addition, CPX has obvious activity to regulate 526

inflammation, thus, it may exert anti-cancer role by regulating tumor 527

microenvironment. A recent report showed that CPX could stimulate immunogenic 528

cell death in pancreatic cancer [42].These studies suggested that CPX can be benefit 529

for treating solid tumors. 530

CPX treatment resulted in a remarkable decrease of DJ-1 expression, which played 531

pivotal roles in CPX-induced CRC inhibition, suggesting that CPX exerted the 532

anti-CRC activity, at least partially, by targeting DJ-1. In cancer cells, DJ-1 has been 533

identified to be a negative regulator of the tumor suppressor PTEN in primary breast 534

cancer and non-small cell lung carcinoma (NSCLC) [43]. Moreover, DJ-1 is 535

upregulated and correlated with patient prognosis in various types of cancer, 536

including NSCLC [43], ovarian carcinoma [44], and CRC [23], indicating that DJ-1 537

acts as an oncogene and can be a potential therapeutic target for cancer treatment [17]. 538

DJ-1 protein expression has been revealed to be downregulated by CPX treatment in 539

HeLa cells through 2D gel electrophoresis [45]. However, the underlying mechanism 540

remains unclear. In this study, we found that CPX decreased the mRNA levels of 541

DJ-1. Further investigations revealed that CPX treatment resulted in upregulation of 542

RUNX1 to bind to the promoter region of DJ-1 and subsequently inhibited its 543

transcription, implying a transcriptional regulation of DJ-1 in CRC cells. 544

27

DJ-1 is widely suspected as a cytoprotective protein in cancer cells. DJ-1 acts as an 545

endogenous antioxidant to eliminate ROS by provoking antioxidant system, including 546

nuclear factor erythroid 2-related factor (NRF2) [46] and glutathione [47]. DJ-1 can 547

also prevent oxidative stress-induced apoptosis [48]. Mechanistic study revealed that 548

DJ-1 blocks ASK1 activation by sequestering the death-associated protein Daxx in the 549

nucleus [49], or by preventing the dissociation of ASK1 from Trx1 during oxidative 550

stress [50], which negatively regulates apoptosis by inhibiting ASK1 [51]. 551

Intriguingly, a recent study revealed that in dihydroartemisinin-resistant HeLa cells, 552

DJ-1 is overexpressed and translocated to the mitochondria [52]. Consistent with 553

these observations, our results showed that CPX-induced ROS accumulation was 554

largely attributed to the downregulation of DJ-1, as restoring DJ-1 expression 555

dramatically decreased ROS level. Moreover, CPX resulted in mitochondrial 556

dysfunction, which was responsible for ROS generation. This may be caused, at least 557

partially, by the downregulation of DJ-1, supported by the observations that DJ-1 acts 558

in parallel to the PINK1/parkin pathway to keep mitochondrial function, and its 559

deficiency results in mitochondrial dysfunction and ROS accumulation [53, 54]. 560

Therefore, the elevated generation via mitochondrial dysfunction, together with the 561

impaired removal due to DJ-1 loss, resulted in the robust accumulation of ROS. 562

ROS played a central role in CPX-mediated CRC growth inhibition, as the scavenging 563

of ROS using NAC significantly abolished the anti-cancer effect of CPX (Figure 3). It 564

has been widely considered that therapeutic stimulation of ROS can be a potent 565

28

strategy to preferentially eliminate cancer cells [55, 56]. More intriguingly, in 566

addition to inducing apoptosis, CPX-induced ROS simultaneously provoke robust 567

autophagic flux, which generally acts as a stress-eliminating process. The role of 568

autophagy in response to therapeutic treatment is multifaceted [24]. The induction of 569

autophagy by therapeutic agents can be pro-death or pro-survival, contributing to the 570

anti-cancer efficacy or drug resistance [57-59]. In most cases, autophagy plays a 571

pro-survival role during anti-cancer treatments in various types of cancer [60]. This 572

cytoprotective autophagy impedes the effectiveness of cancer therapies. Growing 573

evidence has shown that autophagy inhibition could enhance the efficacy of 574

anti-cancer agents, suggesting autophagy as a promising target for cancer therapy 575

[61-64]. Indeed, in this study, the combination of CPX with ATG5 siRNA or CQ 576

treatment, which blocks autophagy initiation or autophagosome degradation, 577

respectively, resulted in remarkable decrease of cell proliferation and increase of 578

apoptosis compared with CPX treatment alone. In summary, we concluded that DJ-1 579

acts as the first protection for CRC, its loss results in ROS accumulation and 580

apoptosis. In this condition, cytoprotective autophagy was provoked by DJ-1 loss to 581

serve as the second protection. 582

In this study, we revealed that CPX induced DJ-1 downregulation, leading to ROS 583

accumulation, which mediated AMPK phosphorylation and autophagy induction. The 584

cytoprotective autophagy perhaps arose compensatively in response to DJ-1 loss, thus 585

implying a strategy of modulating autophagy in combination with therapeutic 586

29

treatment. Our study thus identifies DJ-1 as a target of CPX in CRC, and provides a 587

further insight into understanding the action and mechanism of cytoprotective 588

autophagy during cancer therapy. 589

Funding 590

This work was supported by grants from Chinese NSFC (81660499, 81872014, and 591

81401951), Chongqing Natural Science Foundation (cstc2016jcyjA0227), and 592

Scientific and Technological Research Program of Chongqing Municipal Education 593

Commission (Grant No. KJQN201800429). 594

Authors' Contributions 595

Conception and design: Y. Lei, C. Huang, H. Zhang; 596

Acquisition of data (provided animals, acquired and managed patients, provided 597

facilities, etc.): J. Zhou, L. Zhang, M. Wang, X. Feng, L. Yu, W. Gao; 598

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, 599

computational analysis): Y. Lei, J. Zhou, L. Zhang, M. Wang, J. Lan, C. Zhang, Y. Bu; 600

Writing, review, and/or revision of the manuscript: J. Zhou, L. Zhang, L. Zhou, Y. 601

Lei; 602

Study supervision: Y. Lei, H. Zhang. 603

Competing interests 604

The authors declare that they have no competing interests. 605

606

607

30

Reference 608

1. Siegel RL, Miller KD, Fedewa SA, Ahnen DJ, Meester RG, Barzi A, et al. Colorectal 609

cancer statistics, 2017. CA Cancer J Clin. 2017; 67:177-193. 610

2. Brenner H, Kloor M, Pox CP. Colorectal cancer. Lancet. 2014; 383: 1490-502. 611

3. Meyerhardt JA, Mayer RJ. Systemic therapy for colorectal cancer. N Engl J Med. 612

2005; 352: 476-87. 613

4. Adjuvant chemotherapy with oxaliplatin, in combination with fluorouracil plus 614

leucovorin prolongs disease-free survival, but causes more adverse events in people 615

with stage II or III colon cancer Abstracted from: Andre T, Boni C, Mounedji-Boudiaf L, 616

et al. Multicenter international study of oxaliplatin/5-fluorouracil/leucovorin in the 617

adjuvant treatment of colon cancer (MOSAIC) investigators. Oxaliplatin, fluorouracil, 618

and leucovorin as adjuvant treatment for colon cancer. N Engl J Med 619

2004;350:2343-51. Cancer Treat Rev. 2004; 30: 711-3. 620

5. Bardia A, Loprinzi C, Grothey A, Nelson G, Alberts S, Menon S, et al. Adjuvant 621

chemotherapy for resected stage II and III colon cancer: comparison of two widely 622

used prognostic calculators. Semin Oncol. 2010; 37: 39-46. 623

6. Dalerba P, Sahoo D, Paik S, Guo X, Yothers G, Song N, et al. CDX2 as a Prognostic 624

Biomarker in Stage II and Stage III Colon Cancer. N Engl J Med. 2016; 374: 211-22. 625

7. Niewerth M, Kunze D, Seibold M, Schaller M, Korting HC, Hube B. Ciclopirox 626

olamine treatment affects the expression pattern of Candida albicans genes encoding 627

virulence factors, iron metabolism proteins, and drug resistance factors. Antimicrob 628

Agents Chemother. 2003; 47: 1805-17. 629

8. Jue SG, Dawson GW, Brogden RN. Ciclopirox olamine 1% cream. A preliminary 630

review of its antimicrobial activity and therapeutic use. Drugs. 1985; 29: 330-41. 631

9. Eberhard Y, McDermott SP, Wang X, Gronda M, Venugopal A, Wood TE, et al. 632

Chelation of intracellular iron with the antifungal agent ciclopirox olamine induces 633

cell death in leukemia and myeloma cells. Blood. 2009; 114: 3064-73. 634

10. Zhou H, Shen T, Luo Y, Liu L, Chen W, Xu B, et al. The antitumor activity of the 635

fungicide ciclopirox. Int J Cancer. 2010; 127: 2467-77. 636

11. Yang J, Milasta S, Hu D, AlTahan A, Interiano R, Zhou J, et al. Targeting histone 637

demethylases in MYC-driven neuroblastomas with ciclopirox. Cancer Res. 2017; 638

77(17):4626-4638. 639

12. Kang JA, Kim S, Park M, Park HJ, Kim JH, Park S, et al. Ciclopirox inhibits Hepatitis 640

B Virus secretion by blocking capsid assembly. Nat Commun. 2019; 10: 2184. 641

13. Subissi A, Monti D, Togni G, Mailland F. Ciclopirox: recent nonclinical and clinical 642

data relevant to its use as a topical antimycotic agent. Drugs. 2010; 70: 2133-52. 643

14. Linden T, Katschinski DM, Eckhardt K, Scheid A, Pagel H, Wenger RH. The 644

antimycotic ciclopirox olamine induces HIF-1alpha stability, VEGF expression, and 645

angiogenesis. Faseb j. 2003; 17: 761-3. 646

15. Zhou H, Shen T, Shang C, Luo Y, Liu L, Yan J, et al. Ciclopirox induces autophagy 647

through reactive oxygen species-mediated activation of JNK signaling pathway. 648

31

Oncotarget. 2014; 5: 10140-50. 649

16. Minden MD, Hogge DE, Weir SJ, Kasper J, Webster DA, Patton L, et al. Oral 650

ciclopirox olamine displays biological activity in a phase I study in patients with 651

advanced hematologic malignancies. Am J Hematol. 2014; 89: 363-8. 652

17. Cao J, Lou S, Ying M, Yang B. DJ-1 as a human oncogene and potential 653

therapeutic target. Biochem Pharmacol. 2015; 93: 241-50. 654

18. Bonifati V, Rizzu P, van Baren MJ, Schaap O, Breedveld GJ, Krieger E, et al. 655

Mutations in the DJ-1 gene associated with autosomal recessive early-onset 656

parkinsonism. Science. 2003; 299: 256-9. 657

19. Hering R, Strauss KM, Tao X, Bauer A, Woitalla D, Mietz EM, et al. Novel 658

homozygous p.E64D mutation in DJ1 in early onset Parkinson disease (PARK7). Hum 659

Mutat. 2004; 24: 321-9. 660

20. Chan JY, Chan SH. Activation of endogenous antioxidants as a common 661

therapeutic strategy against cancer, neurodegeneration and cardiovascular diseases: 662

A lesson learnt from DJ-1. Pharmacol Ther. 2015; 156: 69-74. 663

21. Malhotra D, Thimmulappa R, Navas-Acien A, Sandford A, Elliott M, Singh A, et al. 664

Decline in NRF2-regulated antioxidants in chronic obstructive pulmonary disease 665

lungs due to loss of its positive regulator, DJ-1. Am J Respir Crit Care Med. 2008; 178: 666

592-604. 667

22. Lin X, Cook TJ, Zabetian CP, Leverenz JB, Peskind ER, Hu SC, et al. DJ-1 isoforms in 668

whole blood as potential biomarkers of Parkinson disease. Sci Rep. 2012; 2: 954. 669

23. Zhou J, Liu H, Zhang L, Liu X, Zhang C, Wang Y, et al. DJ-1 promotes colorectal 670

cancer progression through activating PLAGL2/Wnt/BMP4 axis. Cell death & disease. 671

2018; 9: 865. 672

24. Gewirtz DA. The four faces of autophagy: implications for cancer therapy. Cancer 673

Res. 2014; 74: 647-51. 674

25. Liu L, Yang M, Kang R, Wang Z, Zhao Y, Yu Y, et al. HMGB1-induced autophagy 675

promotes chemotherapy resistance in leukemia cells. Leukemia. 2011; 25: 23-31. 676

26. Auberger P, Puissant A. Autophagy, a key mechanism of oncogenesis and 677

resistance in leukemia. Blood. 2016; 129:547-552. 678

27. Wang K, Zhang T, Lei Y, Li X, Jiang J, Lan J, et al. Identification of ANXA2 (annexin 679

A2) as a specific bleomycin target to induce pulmonary fibrosis by impeding 680

TFEB-mediated autophagic flux. Autophagy. 2017: 1-56. 681

28. Xie N, Yuan K, Zhou L, Wang K, Chen HN, Lei Y, et al. PRKAA/AMPK restricts HBV 682

replication through promotion of autophagic degradation. Autophagy. 2016; 683

12:1507-20. 684

29. Shi SY, Lu SY, Sivasubramaniyam T, Revelo XS, Cai EP, Luk CT, et al. DJ-1 links 685

muscle ROS production with metabolic reprogramming and systemic energy 686

homeostasis in mice. Nat Commun. 2015; 6: 7415. 687

30. Klionsky DJ, Abdelmohsen K, Abe A, Abedin MJ, Abeliovich H, Acevedo Arozena A, 688

et al. Guidelines for the use and interpretation of assays for monitoring autophagy 689

(3rd edition). Autophagy. 2016; 12: 1-222. 690

32

31. Alpermann HG, Schütz E. [Studies on the pharmacology and toxicology of 691

ciclopiroxolamine (author's transl)]. Arzneimittelforschung. 1981; 31: 1328-32. 692

32. Korting HC, Grundmann-Kollmann M. The hydroxypyridones: a class of 693

antimycotics of its own. Mycoses. 1997; 40: 243-7. 694

33. Mihailidou C, Papakotoulas P, Papavassiliou AG, Karamouzis MV. Superior 695

efficacy of the antifungal agent ciclopirox olamine over gemcitabine in pancreatic 696

cancer models. Oncotarget. 2018; 9: 10360-74. 697

34. Bernier KM, Morrison LA. Antifungal drug ciclopirox olamine reduces HSV-1 698

replication and disease in mice. Antiviral Res. 2018; 156:102-106. 699

35. Clement PM, Hanauske-Abel HM, Wolff EC, Kleinman HK, Park MH. The 700

antifungal drug ciclopirox inhibits deoxyhypusine and proline hydroxylation, 701

endothelial cell growth and angiogenesis in vitro. Int J Cancer. 2002; 100: 491-8. 702

36. Shen T, Shang C, Zhou H, Luo Y, Barzegar M, Odaka Y, et al. Ciclopirox inhibits 703

cancer cell proliferation by suppression of Cdc25A. Genes Cancer. 2017; 8: 505-16. 704

37. Wu J, Liu H, Zhang G, Gu L, Zhang Y, Gao J, et al. Antileukemia Effect of Ciclopirox 705

Olamine Is Mediated by Downregulation of Intracellular Ferritin and Inhibition 706

β-Catenin-c-Myc Signaling Pathway in Glucocorticoid Resistant T-ALL Cell Lines. PLoS 707

ONE. 2016; 11: e0161509. 708

38. Alpermann HG, Schutz E. Studies on the pharmacology and toxicology of 709

ciclopiroxolamine (author's transl). Arzneimittel-Forschung. 1981; 31: 1328-32. 710

39. Kellner HM, Arnold C, Christ OE, Eckert HG, Herok J, Hornke I, et al. 711

Pharmacokinetics and biotransformation of the antimycotic drug ciclopiroxolamine in 712

animals and man after topical and systemic administration. Arzneimittel-Forschung. 713

1981; 31: 1337-53. 714

40. Shen T, Huang SL. Repositioning the Old Fungicide Ciclopirox for New Medical 715

Uses. Curr Pharm Design. 2016; 22: 4443-50. 716

41. Weir SJ, Wood R, Schorno K, Brinker AE, Ramamoorthy P, Heppert K, et al. 717

Preclinical Pharmacokinetics of Fosciclopirox, a Novel Treatment for Urothelial 718

Cancers in Rats and Dogs. The Journal of pharmacology and experimental 719

therapeutics. 2019. 720

42. Mihailidou C, Papakotoulas P, Schizas D, Papalampros A, Vailas M, Felekouras E, 721

et al. Effect of ciclopirox olamine in immunotherapy effect by stimulating 722

immunogenic cell death in pancreatic cancer. Ann Oncol. 2017; 28: 1. 723

43. Kim RH, Peters M, Jang Y, Shi W, Pintilie M, Fletcher GC, et al. DJ-1, a novel 724

regulator of the tumor suppressor PTEN. Cancer cell. 2005; 7: 263-73. 725

44. Davidson B, Hadar R, Schlossberg A, Sternlicht T, Slipicevic A, Skrede M, et al. 726

Expression and clinical role of DJ-1, a negative regulator of PTEN, in ovarian 727

carcinoma. Hum Pathol. 2008; 39: 87-95. 728

45. Mémin E, Hoque M, Jain MR, Heller DS, Li H, Cracchiolo B, et al. Blocking eIF5A 729

modification in cervical cancer cells alters the expression of cancer-related genes and 730

suppresses cell proliferation. Cancer Res. 2014; 74: 552-62. 731

46. Clements CM, McNally RS, Conti BJ, Mak TW, Ting JP. DJ-1, a cancer- and 732

33

Parkinson's disease-associated protein, stabilizes the antioxidant transcriptional 733

master regulator Nrf2. Proc Natl Acad Sci USA. 2006; 103: 15091-6. 734

47. Zhou W, Freed CR. DJ-1 up-regulates glutathione synthesis during oxidative stress 735

and inhibits A53T alpha-synuclein toxicity. J Biol Chem. 2005; 280: 43150-8. 736

48. Hod Y. Differential control of apoptosis by DJ-1 in prostate benign and cancer 737

cells. J Cell Biochem. 2004; 92: 1221-33. 738

49. Junn E, Taniguchi H, Jeong BS, Zhao X, Ichijo H, Mouradian MM. Interaction of 739

DJ-1 with Daxx inhibits apoptosis signal-regulating kinase 1 activity and cell death. 740

Proc Natl Acad Sci USA. 2005; 102: 9691-6. 741

50. Im JY, Lee KW, Junn E, Mouradian MM. DJ-1 protects against oxidative damage by 742

regulating the thioredoxin/ASK1 complex. Neurosci Res. 2010; 67: 203-8. 743

51. Saitoh M, Nishitoh H, Fujii M, Takeda K, Tobiume K, Sawada Y, et al. Mammalian 744

thioredoxin is a direct inhibitor of apoptosis signal-regulating kinase (ASK) 1. Embo j. 745

1998; 17: 2596-606. 746

52. Zhu H, Liao SD, Shi JJ, Chang LL, Tong YG, Cao J, et al. DJ-1 mediates the 747

resistance of cancer cells to dihydroartemisinin through reactive oxygen species 748

removal. Free Radic Biol Med. 2014; 71: 121-32. 749

53. Irrcher I, Aleyasin H, Seifert EL, Hewitt SJ, Chhabra S, Phillips M, et al. Loss of the 750

Parkinson's disease-linked gene DJ-1 perturbs mitochondrial dynamics. Hum Mol 751

Genet. 2010; 19: 3734-46. 752

54. Thomas KJ, McCoy MK, Blackinton J, Beilina A, van der Brug M, Sandebring A, et 753

al. DJ-1 acts in parallel to the PINK1/parkin pathway to control mitochondrial 754

function and autophagy. Hum Mol Genet. 2011; 20: 40-50. 755

55. Trachootham D, Alexandre J, Huang P. Targeting cancer cells by ROS-mediated 756

mechanisms: a radical therapeutic approach? Nat Rev Drug Discov. 2009; 8: 579-91. 757

56. Gorrini C, Harris IS, Mak TW. Modulation of oxidative stress as an anticancer 758

strategy. Nat Rev Drug Discov. 2013; 12: 931-47. 759

57. Kondo Y, Kanzawa T, Sawaya R, Kondo S. The role of autophagy in cancer 760

development and response to therapy. Nat Rev Cancer. 2005; 5: 726-34. 761

58. Sui X, Chen R, Wang Z, Huang Z, Kong N, Zhang M, et al. Autophagy and 762

chemotherapy resistance: a promising therapeutic target for cancer treatment. Cell 763

death & disease. 2013; 4: e838. 764

59. Kumar A, Singh UK, Chaudhary A. Targeting autophagy to overcome drug 765

resistance in cancer therapy. Future Med Chem. 2015; 7: 1535-42. 766

60. Maes H, Rubio N, Garg AD, Agostinis P. Autophagy: shaping the tumor 767

microenvironment and therapeutic response. Trends Mol Med. 2013; 19: 428-46. 768

61. Thorburn A, Thamm DH, Gustafson DL. Autophagy and cancer therapy. Mol 769

Pharmacol. 2014; 85: 830-8. 770

62. Janku F, McConkey DJ, Hong DS, Kurzrock R. Autophagy as a target for anticancer 771

therapy. Nat Rev Clin Oncol. 2011; 8: 528-39. 772

63. Yang ZJ, Chee CE, Huang S, Sinicrope FA. The role of autophagy in cancer: 773

therapeutic implications. Mol Cancer Ther. 2011; 10: 1533-41. 774

34

64. Amaravadi RK, Lippincott-Schwartz J, Yin XM, Weiss WA, Takebe N, Timmer W, et 775

al. Principles and current strategies for targeting autophagy for cancer treatment. 776

Clin Cancer Res. 2011; 17: 654-66. 777

778

779

780

781

782

783

784

785

786

787

788

789

790

791

792

793

794

795

796

797

35

Figure legends 798

Figure 1. CPX inhibits proliferation and induces apoptosis in CRC cells 799

A. HCT116, DLD-1, RKO, HT29，SW480，HIEC and NCM460 cells were treated 800

with the indicated concentrations of CPX for 24 hours and cell viability was 801

determined by CCK8 kit. B. Cell proliferation rate was analyzed by clone formation 802

assay. HCT116 and SW480 cells were treated with the indicated concentration of 803

CPX for 24 hours, after treatment, cells were seeded into 6-well plates for two weeks 804

and colony numbers were quantified. C. EdU assay of CRC cells treated with the 805

indicated concentration of CPX for 24 hours. The EdU incorporation was quantitated. 806

Scale bar, 100 μm. D. Analysis of dehydrogenase (LDH) release in HCT116 and 807

SW480 cells subjected to indicated concentration of CPX for 24 hours. E. 808

Morphology of HCT116 and SW480 cells following treatment with or without 20 μM 809

CPX for 24 hours. Scale bar, 100 μm. F. HCT116 and SW480 cells treated with 20 810

μM CPX for 24 hours were fixed, stained with Annexin V/PI, and then analyzed by 811

flow cytometry. The apoptosis rate was quantitated. G. Immunoblot analysis of PARP 812

in HCT116 and SW480 cells. Data are means ± s.d. and are representative of 3 813

independent experiments. *, P < 0.05, **, P < 0.01, ***, P < 0.001. Statistical 814

significance compared with respective control groups. 815

Figure 2. CPX-induced downregulation of DJ-1 is involved in anti-CRC effects of 816

CPX 817

36

A. Volcano plot analysis of DEGs in HCT116 and SW480 cells treated with or 818

without 20 μM CPX for 24 hours. Red areas indicate significant overexpression and 819

green areas indicate significant downregulation. B. qRT-PCR was conducted in 820

HCT116 and SW480 cells treated with or without 20 μM CPX for 24 hours to 821

determine DJ-1 mRNA expression. Statistical significance compared with respective 822

control groups. C. Immunoblot analysis of DJ-1 expression in CRC cells treated with 823

or without the indicated concentration of CPX for 24 hours. D-E. Colony formation 824

assay of CRC cells stably overexpressing DJ-1 and Vector treated with indicated 825

concentrations of CPX. F. Cell viability assay was performed by CCK8 kit. G-J. 826

HCT116 cells stably overexpressing DJ-1 and Vector were subcutaneously injected 827

into nude mice(n=6). Tumor volumes were monitored at indicated time points (G), 828

and the weight of tumors were measured at time of sacrificed (H). (I) The image of 829

isolated tumors derived from mice treated with vehicle or CPX (25 mg/kg/day). (J) 830

Representative images of immunohistochemical staining of PCNA and quantification 831

of relative intensity of PCNA staining in xenografts. Scale bar, 100 μm. *, P < 0.05, 832

**, P < 0.01, ***, P < 0.001. 833

Figure 3. ROS induced by DJ-1 downregulation are responsible for anti-CRC 834

effects of CPX 835

A. ROS level was analyzed by DCFH-DA staining via a fluorescence microplate 836

Reader (Thermo scientific) in CRC cells treated with the indicated concentrations of 837

CPX. Statistical significance compared with respective control groups. B. MitoSOX 838

37

Red was used to evaluate ROS level in mitochondria upon CPX treatment. Scale bar, 839

25μm. C. ROS level was determined by flow cytometry in CRC cells treated with or 840

without 20 μM CPX for 24 hours overexpressing Myc-DJ-1 and Vector. D. Colony 841

formation assay of CRC cells treated with the indicated concentrations of CPX in the 842

presence or absence of 5mM NAC. E. EdU assay of CRC cells treated with or without 843

5mM NAC in the presence or absence of 20 μM CPX for 24 hours. The EdU 844

incorporation was quantitated. Scale bar, 100 μm. F. The CCK8 assay determined cell 845

viability of CRC cells treated with the indicated concentrations of CPX in the 846

presence or absence of 5mM NAC for 24 hours. G. Immunoblot analysis of PARP 847

expression in HCT116 and SW480 cells treated with CPX (20 μM) or vehicle and/or 848

with NAC (5mM). H. HCT116 and SW480 cells treated with CPX (20 μM) or vehicle 849

and/or with NAC (5mM) were fixed, stained with Annexin V/PI, and then analyzed 850

by flow cytometry. The development of apoptosis was quantitated. Data are means ± 851

s.d. and are representative of 3 independent experiments. *, P < 0.05, **, P < 0.01, 852

***, P < 0.001. 853

Figure 4. CPX induces mitochondrial dysfunction to promote ROS accumulation 854

in CRC cells 855

A. Transmission electron microscopy images from HCT116 cells treated with or 856

without 20 μM CPX. Scale bar, 1 μm. B. Mitotracker Green was used to analyze 857

mitochondrial mass independent of mitochondrial membrane potential (MMP) 858

alteration of CRC cells treated with or without 20 μM CPX for 24 hours via a 859

38

fluorescence microplate Reader (Thermo scientific). Statistical significance compared 860

with respective control groups. C. The number of mitochondrial DNA copies was 861

examined by qRT-PCR in HCT116 and SW480 cells treated with or without 20 μM 862

CPX for 24 hours. Statistical significance compared with respective control groups. D. 863

CRC cells were transfected with empty Vector or Myc-DJ-1 plasmid for 24 hours, 864

followed by treatment with or without 20 μM CPX for 24 hours, then seeded in plates 865

for 12 hours before OCR analysis. Basal OCR was analyzed under XF Base Medium 866

containing 10 mM glucose, 1 mM sodium pyruvate and 2 mM glutamine (pH was 867

adjusted to 7.40 ± 0.05 at 37°C) (n = 3 per group). E. The glycolysis stress test was 868

performed under XF Base Medium containing 2 mM glutamine (pH was adjusted to 869

7.40 ± 0.05 at 37°C) after injection of 10 mM glucose, 1μM oligomycin (an inhibitor 870

of mitochondrial membrane adenosine triphosphate synthase) and 50 mM 871

2-deoxyglucose (2-DG, a non-metabolizable glucose analog that inhibits glycolysis) 872

in HCT116 cells treated as in (D) (n = 3 per group). F and G. Heat map shows 873

normalized intensity values for genes involved in oxidative phosphorylation (F) and 874

glycolysis assessment (G) after 20 μM CPX treatment with a P value < 0.05. H and I. 875

qRT-PCR analysis the expression of genes involved in oxidative phosphorylation (H) 876

and glycolysis (I) after 20 μM CPX treatment in HCT116 cells. This experiment was 877

carried out in triplicate. Data are means ± s.d. *, P < 0.05, **, P < 0.01, ***, P < 878

0.001. 879

Figure 5. CPX induces autophagy in CRC cells. 880

39

A. Immunoblot analysis of LC3, Atg5, Beclin1, and p62 expression in CRC cells 881

treated with indicated concentrations of CPX for 24 hours. B. The formation of 882

endogenous LC3 puncta was shown and quantitated by immunofluorescence analysis 883

in cells treated with or without 20 μM CPX for 24 hours. Scale bar, 20 μm. C. 884

Autophagic vesicles detected by transmission electron microscope in HCT116 cells 885

treated as in (B). Scale bar, 2 μm. D. Left, autophagy was measured by acridine 886

orange staining of cells treated as in (B). Right, total number of acidic vesicular 887

organelles (AVO) per cell. Scale bar, 10 μm. E. Representative images of IHC 888

analysis for LC3 expression in HCT116 xenografts collected from vehicle or 889

CPX-treated mice. Scale bar, 50 μm. F. The protein expression level of LC3 in 890

vehicle- or CPX-treated mice bearing HCT116 xenografts was examined by 891

immunoblot analysis. G. Co-immunoprecipitation analysis of the interaction between 892

Beclin1 and Bcl-2 in CRC cells treated with or without 20 μM CPX for 24 hours. 893

Data are means ± s.d. *, P < 0.05, **, P < 0.01, ***, P < 0.001. 894

Figure 6. CPX promotes autophagy flux in CRC cells. 895

A. CRC cells were treated with vehicle, CPX (20 μM), CQ (10 μM), or in 896

combination for 24 hours. Immunoblot analysis was used to detect the expression of 897

LC3 and p62. B. CRC cells were treated with CPX (20 μM) with or without Baf A1 898

(100 nM) for 24 hours. The expression of LC3 and p62 were examined by 899

immunoblotting. C. CRC cells were treated with E64D (10μg/ml) and PepA (10μg/ml) 900

in the presence or absence of CPX (20 μM) for 24 hours. The expression of LC3 and 901

40

p62 were examined by immunoblotting. D. The accumulation of LC3 puncta was 902

examined by immunofluorescent analysis of cells treated with CQ (10 μM) in the 903

presence or absence of CPX (20 μM) for 24 hours. Scale bar, 20 μm. The number of 904

LC3 puncta was quantitated. E. Immunofluorescence analysis of cells transiently 905

transfected with tandem mRFP-GFP-tagged LC3 and treated with vehicle, CPX (20 906

μM), CQ (10 μM), or in combination for 24 hours. Scale bar, 10 μm. The ratio of red 907

puncta indicating autolysosome (GFP-/RFP+) versus yellow puncta indicating 908

autophagosome (GFP+/RFP+) was quantitated. F. Immunofluorescence analysis of the 909

co-localization of endogenous LC3 and LAMP1 in CRC cells treated with vehicle, 910

CPX (20 μM), CQ (10 μM), or in combination for 24 hours. Scale bar, 20 μm. Data 911

are means ± s.d. and are representative of 3 independent experiments. *, P < 0.05, **, 912

P < 0.01, ***, P < 0.001. 913

Figure 7. Inhibition of autophagy augments the anti-CRC effects of CPX. 914

A and B. The cell viability of CRC cells treated with indicated concentrations of CPX 915

in the presence or absence of 10 μM CQ for 24 hours was detected by CCK8 assays. 916

C. The cell viability of CRC cells transfected with siNC or siATG5 for 24 hours, 917

followed by treatment with or without 20 μM CPX for another 24 hours was detected 918

by CCK8 assays. D. EdU assay of CRC cells treated with or without 10 μM CQ in the 919

presence or absence of 20 μM CPX for 24 hours. The EdU incorporation was 920

quantitated. Scale bar, 100 μm. E and F. HCT116 cells and SW480 cells were treated 921

with indicated concentrations of CPX in the presence or absence of 10 μM CQ. Cell 922

41

proliferation was detected by colony formation assay. G. CRC cells were transfected 923

with siNC or siATG5 for 24 hours, followed by treatment with or without 20 μM CPX 924

for another 24 hours. Cells were fixed, stained with Annexin V/PI, and then analyzed 925

by flow cytometry. The development of apoptosis was quantitated. Data are means ± 926

s.d. and are representative of 3 independent experiments. *, P < 0.05, **, P < 0.01, 927

***, P < 0.001. 928

Figure 8. Transcriptional inhibition of DJ-1 by RUNX1 929

A. Venn diagram analysis of transcription factors was performed between HCT116 930

Ctrl VS CPX and SW480 Ctrl VS CPX. B. Heat map representation of normalized 931

intensity values for transcription factors binding to DJ-1 promoter in HCT116 and 932

SW480 cells treated with or without 20 μM CPX. C. qRT-PCR was conducted in 933

HCT116 and SW480 cells to determine the mRNA levels of genes from (B). D. 934

HCT116 cells were transfected with siRUNX1 for 24 hours, followed by treatment 935

with or without 20 μM CPX for 24 hours. The protein levels of DJ-1 were examined 936

by Immunoblotting. E. HCT116 cells were transfected with siNC or siRUNX1 for 24 937

hours, mRNA expression of DJ-1 and RUNX1 was examined by qRT-PCR. F. 938

Immunoblot analysis of DJ-1 and RUNX1 expression in HCT116 cells transiently 939

transfected with Vector or Flag-RUNX1 plasmid for 24 hours, followed by treatment 940

with or without 20 μM CPX for another 24 hours. G. Luciferase-reporter studies on 941

DJ-1 promoter in HCT116 cells transiently transfected with Vector or Flag-RUNX1 942

plasmid in the presence or absence of 20 μM CPX. H. Chromatin from HCT116 cells 943

42

transfected with RUNX1 plasmid for 48 hours was used for ChIP analysis with 944

RUNX1 antibody, followed by qRT-PCR to amplify different RUNX1-binding site 945

regions of DJ-1 promoter. Enrichments were calculated relative to IgG. The qRT-PCR 946

primer sequences were shown in Supplementary Table 4. Data are means ± s.d. and 947

are representative of 3 independent experiments. *, P < 0.05, **, P < 0.01, ***, P < 948

0.001. 949

950

951

952

953

0 5 10 200

20

40

60

80

100

120

Clo

ne N

um

bers

HCT116

SW480

* *

*** ***

*** ***

(µM)

0.01.

02.

03.

04.

05.

06.

07.

08.

09.

0

0

20

40

60

80

100

120

CPX (µM)

Cell

via

bili

ty (

% o

f contr

ol)

HCT116

HT29

DLD-1

RKO

0.5 1 5 10 20 30

SW480

40 50

HIEC

NCM460

0

HCT116

SW480

Figure 1. CPX inhibits proliferation and induces apoptosis in CRC cells

A B D

C

HC

T1

16

SW

48

0

CPX Ctrl

Annexin V-FITC

PI

F

E

HCT116 SW480

02

01

05

DAPI MergeEdU EdU DAPI Merge

CPX Ctrl

HC

T1

16

SW

48

0

G

0 5 10 20 250

10

20

30

40

50

60

LD

H r

ele

ase (

% o

f to

tal)

HCT116SW480

CPX (µM)

***

***

******

****

******

HCT116 SW4800

10

20

30

40

50

% D

evelo

pm

ent

of

apopto

sis Ctrl

CPX

******

Cleaved-PARP

Cleaved-PARP

CPX (μM) 0 5 10 20

β-actin

PARP

PARP

β-actin

HC

T1

16

SW

48

0

116kD

116kD

89kD

89kD

43kD

43kD

cleaved-PARP/β-actin 1 1 2.9 3.3

cleaved-PARP/β-actin 1 1.3 1.6 2.5

0 5 10 200

20

40

60

80

100

120

EdU

incorp

ora

tion (

%)

HCT116SW480

** **

(µM)

****

****

02.5 5 10 20 30 40

0

20

40

60

80

100

120

CPX (µM)

Cell

via

bili

ty (

% o

f contr

ol)

VectorDJ-1

DJ-

1

Vec

tor

DJ-

1 + C

PX

CPX

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Tum

or

weig

ht

(g) ***

**

Figure 2. CPX-induced downregulation of DJ-1 is involved in anti-CRC

effects of CPX

D

C

G H

F HCT116

Ctrl CPX

Ve

cto

rD

J-1

J

PC

NA

I

Ctr

lC

PX

Ve

cto

rD

J-1

DJ-1

Ve

cto

r

B

Ve

cto

rD

J-1

HCT116

Ctrl CPX 20μM CPX 10μM

02.5 5 10 20 30 40

0

20

40

60

80

100

120

CPX (µM)

Cell

via

bili

ty (

% o

f contr

ol)

VectorDJ-1

SW480

0 3 6 9 12 15 18 210

200

400

600

800

1000

1200

Tum

or

siz

e (

mm

3)

Vector

CPX

days

DJ-1

DJ-1 + CPX

***

***

***

***

***

50 10 20

β-actin

CPX (μM）

DJ-1

β-actin

DJ-1

HC

T1

16

SW

48

0

43kD

43kD

23kD

23kD

E

DJ-1

Ve

cto

r

Ctrl

SW480

CPX 20μM CPX 10μM

0 5 10 200.0

0.3

0.6

0.9

1.2

1.5HCT116SW480

Rela

tive E

xpre

ssio

n o

f D

J-1

******

******

****

A

Log2 (FC)

-lo

g1

0(F

DR

)

significant

UP Down Normal

Log2 (FC)

-lo

g1

0(F

DR

)

HCT116 SW480

DJ-1/β-actin 1 0.9 0.5 0.3

DJ-1/β-actin 1 0.8 0.4 0.3

0 5 10 200

10

20

30

40

50

60

RO

S (

% o

f contr

ol)

HCT116

SW480

CPX (µM)

******

******

****

0NAC

CPX

CPX +

NAC

0

20

40

60

80

100

120E

dU

incorp

ora

tion (

%)

HCT116SW480

***

E

HCT116 SW480

Ctr

lC

PX

+N

AC

NA

CC

PX

DAPI MergeEdU EdU DAPI Merge

CPX 20μM Ctrl

Ctr

lN

AC

SW480

CPX 10μM D

B

Mito

So

x

Ctrl CPX

HC

T1

16

SW

48

0

C

F

A

Figure 3. ROS induced by DJ-1 downregulation are responsible for anti-CRC

effects of CPX

HCT116

Ctr

lN

AC

Ctrl CPX 20μM CPX 10μM

0 5 10 200

20

40

60

80

100

120

Cell

via

bili

ty (

% o

f contr

ol) Ctrl

NAC*

CPX (µM)

***

0 5 10 200

20

40

60

80

100

120

Cell

via

bili

ty (

% o

f contr

ol) Ctrl

NAC

*

CPX (µM)

***

0NAC

CPX

CPX +

NAC

0

10

20

30

40

% D

evelo

pm

ent

of

apopto

sis

HCT116SW480

******

HCT116 SW4800

20

40

60

80

100

CtrlCPX

Myc-DJ-1Myc-DJ-1 + CPX

******

RO

S (

% o

f contr

ol)

** **

SW480

G

β-actin

PARP

PARP

β-actin

CPX

NAC

+-

- -- ++ +

Cleaved-PARP

Cleaved-PARP

HC

T1

16

SW

48

0

116kD

116kD

89kD

89kD

43kD

43kD

cleaved-PARP/β-actin 1 2.9 1.2 1.1

cleaved-PARP/β-actin 1 3.2 1.1 1

HCT116

H

0 5 10 200

20

40

60

80

100

120

Mitochondrial m

ass

(%

of

contr

ol)

HCT116SW480

***

***

(µM)

**

**

***

***

B

FCPX +- - +

HCT116 SW480

RBP1

BOP1

CISD3

CYB5R3

DHODH

G6PC3

GCDH

HMMR

KISS1R

MAT2A

MRPL35

MRPL40

MRPS14

MRPS28

MSTO1

NDUFAF3

UCK2

UCP2

TSFM

0.0

0.3

0.6

0.9

1.2

1.5

Rela

tive E

xpre

ssio

n

CtrlCPX

G

HCT116 SW480

CPX +- - +

I

Figure 4. CPX induces mitochondrial dysfunction to promote ROS

accumulation in CRC cells

A

Ctr

lC

PX

0 5 10 200

20

40

60

80

100

120

mtD

NA

copy n

um

ber

(%

of

contr

ol)

HCT116SW480

******

(µM)

****

******

ENO2

ENO3

ALDOA

HK2

PGK1

0

1

2

3

4

5

Rela

tive E

xpre

ssio

n CtrlCPX

H

C

ED

Ctrl

CPX

DJ-

1

DJ-

1 + C

PX

0

50

100

150

200

OC

R (

mpH

/min

)

*** **

***

HCT116

Ctrl

CPX

DJ-

1

DJ-

1 + C

PX

0

50

100

150

200

OC

R (

mpH

/min

)

*** **

***

SW480

Figure 5. CPX induces autophagy in CRC cells

Animal No.

LC3 I

LC3 II

β-actin

92 3 4 5 6 7 81 10 11 12

Ctrl CPX

DE

Ctrl CPX

LC3

F

SW

48

0H

CT

11

6

Ctrl CPX

HCT116 SW4800

10

20

30

40

50

% D

evelo

pm

ent

of

AV

O CtrlCPX

******

C

CPX Ctrl

LC3 DAPI Merge

CP

XC

trl

LC3 DAPI Merge

HCT116 SW480

B

HCT116 SW4800

10

20

30

40

50

No.

of

LC

3 p

uncta

/cell

CtrlCPX

******

CPX (μM)

A

0 10 20

LC3 I

LC3 II

β-actin

HCT116

5

Atg5

Beclin 1

P62

0 5 10 20

SW480

16kD

14kD

62kD

43kD

60kD

55kD

16kD

14kD

43kDBeclin 1

Beclin 1

Inp

ut

β-actin

Bcl2

IP: B

eclin

1

CPX - -+ +

HCT116 SW480

G

43kD

60kD

60kD

28kD

LC3 II/β-actin 1 1.7 3.2 4.4

Atg5/β-actin 1 1.3 1.9 2.5

Beclin 1/β-actin 1 5.4 5.8 7.5

P62/β-actin 1 0.6 0.3 0.1

1 1.9 2.0 3.4

1 1.1 1.0 1.9

1 1.5 1.3 2.1

1 0.6 0.5 0.3

LC3 II/β-actin 1 1.3 1.4 1.2 1.5 1.6 2.3 3.4 2.1 2.6 2.5 1.2

Ctrl

CPX

CQ

CPX +

CQ

Ctrl

CPX

CQ

CPX +

CQ

0

10

20

30

40

50

60

70N

o.

LC

3 p

uncta

/cell

GFP+/RFP+GFP-/RFP+

***

**

***

******

*

*

*

*****

******

**

**

**

Figure 6. CPX promotes autophagy flux in CRC cells

EHCT116 SW480

D LC3

CP

X +

CQ

CQ

DAPI Merge LC3 DAPI Merge

Ctr

lC

PX

CP

X+

CQ

CQ

RFP-LC3 GFP-LC3 Merge RFP-LC3 GFP-LC3 Merge

HCT116 SW480

Ctr

lC

PX

LAMP1 LC3DAPI Merge

SW

48

0

Ctr

lC

PX

LAMP1 LC3DAPI MergeF

HC

T1

16

HCT116 SW480

CQ

CP

X+

CQ

CP

X+

CQ

CQ

B

LC3 I

Bafilomycin A1 -

+ -

CPX - + -

LC3 II

β-actin

p62

LC3 I

LC3 II

β-actin

p62

HC

T1

16

+ +

SW

48

0

16kD

14kD

62kD

43kD

16kD

14kD

62kD

43kD

LC3 I

CQ -

+

-CPX - + -

LC3 II

β-actin

p62

LC3 I

LC3 II

p62

β-actin

HC

T1

16

A

+ +

16kD

14kD

62kD

43kD

SW

48

0

16kD

14kD

62kD

43kD

C CPX - + - + E64D + PepA - - + +

HC

T1

16

LC3 I

LC3 I

LC3 II

LC3 II

p62

β-actin

p62

β-actin

16kD

14kD

62kD

43kD

SW

48

0

16kD

14kD

62kD

43kD

HCT116 SW4800

20

40

60

80

100

120

siNC siNC + CPXsiATG5 siATG5 + CPX

*** ***

Cell

via

bili

ty (

% o

f contr

ol)

Figure 7. Inhibition of autophagy augments the anti-CRC effect of CPX

CB

D

HCT116 SW480

Ctr

lC

PX

+C

QC

QC

PX

DAPI MergeEdU EdU DAPI Merge

A HCT116 SW480

HCT116 SW480

PI

Annexin V-FITC

siN

CsiA

TG

5

Ctrl CPX CPXCtrl

Ctrl

Ctr

lC

Q

SW480

CPX 20μM CPX 10μM E F

G

Ctr

lC

Q

HCT116

Ctrl CPX 20μM CPX 10μM

0 5 10 200

20

40

60

80

100

120

Cell

via

bili

ty (

% o

f contr

ol) Ctrl

CQ

**

CPX (µM)

*

0 CPX CQ CPX + CQ0

20

40

60

80

100

120

EdU

incorp

ora

tion (

%)

HCT116SW480

**

HCT116 SW4800

10

20

30

40

50

siNC siNC + CPXsiATG5 siATG5 + CPX

*** ***

% D

evelo

pm

ent

of

apopto

sis

0 5 10 200

20

40

60

80

100

120

Cell

via

bili

ty (

% o

f contr

ol) Ctrl

CQ *

CPX (µM)

**

0 20 40 60 80 100 120

pGL3-basic

DJ-1 promoter

DJ-1 promoter + Flag-RUNX1

DJ-1 promoter + Flag-RUNX1 + CPX

Relative luciferase activity

***

***

SREBF1

ETV4

RUNX1

FOXA

1

BHLH

E40EH

F

NR4A

2

ATF3JU

N

EGR3

0

1

2

3

4

5

Rela

tive E

xpre

ssio

n

CtrlCPX

IgG

regi

on 1

regi

on 2

regi

on 3

regi

on 4

regi

on 5

regi

on 6

0

1

2

3

4

5

6

7

rela

tive o

ccupancy

********

EGR3

RUNX1

NR4A

2

ATF3

SREBF1

NFIL

3

FOXA1

BHLH

E40JU

N

ETV4

0

1

2

3

4

5

6

7

8

9

10

Rela

tive E

xpre

ssio

n

CtrlCPX

A

C

Figure 8. Transcriptional inhibition of DJ-1 by RUNX1

HCT116 CPX

HCT116 Ctrl

SW480 Ctrl

SW480 CPX

B

Ge

ne

Exp

ressio

n

-1.5

-0.75

0

0.75

1.5

β-actin

RUNX1

CPX - -+ +

Vector RUNX1

DJ-1

E F

SW480

55kD

23kD

43kD

siNC siRUNX10

20

40

60

80

100

120

Rela

tive E

xpre

ssio

n o

f R

UN

X1 HCT116

SW480

HCT116 SW4800

30

60

90

120

150

180

Rela

tive E

xpre

ssio

n o

f D

J-1

siNCsiRUNX1

**

***

G H

RUNX1/β-actin 1 1.8 1.8 2.3

DJ-1/β-actin 1 0.5 0.5 0.5

HCT116

D

CPX - -+ +

siNC siRUNX1

β-actin

RUNX1

DJ-1

D

RUNX1/β-actin 1 1.9 0.3 0.4

DJ-1/β-actin 1 0.4 1.3 0.8