0.1

0.15

0.2

0.25

0.3

0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90

Absorbance@

590nm

Time(mins)

Glutamate

Malate

Isocitrate

Lactate

Co-culture PDAC tumor cells + CAFs respond to combination therapy of APX inhibition and metabolic inhibitor, CPI-613 (25 µM)

Arrows indicate treatment times.

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX3330 Combo

Days post plating

Fold

cha

nge

com

pare

d to

unt

reat

ed c

ontro

l

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX2009 Combo

APX3330=35 µM

Days post plating

APX2009=5 µM

-0.20

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX2014 Combo

Days post plating

APX2014=2.5 µM

Ctrl CPI-613 APX Combo

Co-culture PDAC tumor cells + CAFs respond to combination therapy of APX inhibition and metabolic inhibitor, CPI-613 (25 µM)

Arrows indicate treatment times.

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX3330 Combo

Days post plating

Fold

cha

nge

com

pare

d to

unt

reat

ed c

ontro

l

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX2009 Combo

APX3330=35 µM

Days post plating

APX2009=5 µM

-0.20

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX2014 Combo

Days post plating

APX2014=2.5 µM

Ctrl CPI-613 APX Combo

0

0.2

0.4

0.6

0.8

1

1.2

Day4 Day8 Day12

FoldCha

ngecompa

redtoSCR

G SCRG

SCRLowGSiGSiLowG

ScRNA-seq data of APE1 knockdown in normoxia and hypoxia

ScRNA-seq Differential Gene Expression Analysis

APEX1 Expression

Normoxia Hypoxia SCR SCR siAPE1 siAPE1

Fig 5: Validation of scRNA-seq in hypoxia. Expression of various Mitochondrial genes in normoxia and hypoxia, genes assessed via qRT-PCR in Pa03C cells. 3 independent experiments, showing average fold change +/- SD. *p<0.001 (Two-way ANOVA). All genes exhibited expression profiles consistent with scRNA-seq data.

DNA

CA9

JAK

STAT3Inhibitors(e.g.PG-S3-001)

STAT3

STAT3

IL#6%Stromal%Cells%

HIF1α

HIF-1

VEGFSignalingInhibitors(e.g.Bevacizumab)

IL-6targetedagents

JAK2inhibitors(e.g.Ruxolitinib)pH

Stabilization

UpstreamHIF1Inhibitors(variousmechanisms)

CA9inhibitors(e.g.SLC-0111)

VEGF

Figure2

CA9

APX3330

⬇O2

Angiogenesis

NFκB

NFκB

IκBαDegradationInhibitors(e.g.Bortezomib)

OtherTF’s(e.g.AP-1)

Figure 6: Functional effects of APE1 knockdown on tumor cell growth and metabolism

•  We previously used single cell RNA sequencing to identify 1950 differentially expressed genes (DEGs) between the scrambled control and siAPE1 PDAC cells in normoxia (Shah et al 2017).

•  Under hypoxic conditions, 2113 DEGs were identified between the scrambled and siAPE1 cells.

•  Ingenuity Pathway Analysis (IPA) identified over 100 pathways affected by APE1 knockdown under both conditions including metabolic pathways such as oxidative phosphorylation and mitochondrial function.

•  In our ex vivo 3D co-culture system, the effects of APE1 knockdown are amplified under low glucose conditions.

•  Using Biolog S-1 substrate plates, we identified multiple substrates from the Glycolysis, Glutaminase and Tricarboxylic Acid cycles that showed reduced usage in the presence of APE1 knockdown.

•  We combined APE1 redox specific inhibitor APX2014 (APX3330 analog) with mitochondrial metabolism inhibitors in our ex vivo 3D co-culture system using spheroids containing tumor cells and Cancer-associated Fibroblasts.

•  The combination of APX2014 with Glutaminase inhibitor CB-839 did not show a significant effect.

•  However, the combination of APX2014 with Dehydrogenase inhibitor CPI-613 demonstrated an additive enhancement effect in the tumor in our 3D co-culture system, without sensitizing CAFs.

Figure 7: APX2014 (APX3330 analog) in combination with mitochondrial metabolism

inhibitors

Targeting mitochondrial metabolism in Pancreatic Ductal Adenocarcinoma via blockade of the APE1/Ref-1 signaling cascade: using single-cell RNA sequencing to identify novel targets for combination drug therapies

Fenil Shah1, Nadia M. Atallah2, Michelle Grimard1, Chi Zhang4, Mark R. Kelley1,3,5, and Melissa L. Fishel1,3 1Indiana University School of Medicine, Department of Pediatrics, Wells Center for Pediatric Research; 2Purdue University, Center for Cancer Research; 3Indiana University School of Medicine, Department of

Pharmacology and Toxicology; 4Indiana University School of Medicine, Medical and Molecular Genetics; 5Indiana University School of Medicine, Department of Biochemistry and Molecular Biology; Indianapolis, IN

Introduction

References Materials and Methods

Figure 5: Validation of Hypoxia Single Cell RNA Sequencing by qRT-PCR

Figure 2: APE1 siRNA knockdown results in a heterogeneous population

Figure 1: APE1 expression both in tumor cells and the stroma of patient tissue

•  Methods: We knocked down APE1 in patient derived PDAC cells and isolated single cells from the scrambled control and siAPE1 samples to extract RNA for sequencing. Cells were cultured in Normoxia and Hypoxia (1%O2). We tested single and combination studies of APX3330 in an ex vivo 3-Dimensional tumor-stroma model system using patient derived tumor cells along with patient derived cancer-associated fibroblasts (CAFs).

•  Cell lines used -  Low Passage Patient derived pancreatic cells: Pa03C (kindly provided by Anirban

Maitra) -  Cancer-Associated Fibroblasts: CAF19 (kindly provided by Anirban Maitra)

•  Immunohistochemistry (IHC) and Western Blotting -  APE1 antibody from Novus Biologicals

•  qRT-PCR -  Bio-Rad CFX96 Real Time Detection system

•  Spheroid Proliferation/Survival -  Thermo ArrayScan

•  Mitochondrial Function Assay -  Biolog S-1 Assay plates

Jiang, Y, Zhou S, Sandusky GE, Kelley MR and Fishel ML (2010). Reduced expression of DNA repair and redox signaling protein APE1/Ref-1 impairs human pancreatic cancer cell survival, proliferation, and cell cycle progression. Cancer Invest 28(9): 885-895. Fishel ML, Jiang Y, Rajeshkumar NV, Scandura G, Sinn AL, He Y, Shen C, Jones DR, Pollok KE, Ivan M, Maitra A, Kelley MR. (2011). Impact of APE1/Ref-1 Redox Inhibition on Pancreatic Tumor Growth. Molecular Cancer Therapeutics. Shah F, Logsdon D, Messmann RA, Fehrenbacher JC, Fishel ML, and Kelley MR (2017) Exploiting the Ref-1-APE1 node in cancer signaling and other diseases: from bench to clinic, NPJ Precision Oncology, 1(1), 19. Shah F, Goossens E, Atallah NM, Grimard M, Kelley MR and Fishel ML (2017) Characterizing gene expression changes and identifying novel pathways following APE1/Ref-1 knockdown in Pancreatic Ductal Adenocarcinoma using Single-cell RNA Sequencing, Molecular Oncology, 11(12):1711-1732.

APE1 Redox Activity Regulates Multiple Transcription Factor (TF) Signaling Involved In Pancreatic Cancer Survival and Drug Resistance; e.g. HIF-1, NFkB and STAT3. Its endonuclease activity is involved in DNA repair via the base excision repair pathway. The multifunctional nature of APE1 means it impacts multiple genes and pathways. APX3330 blocks the redox signaling function of APE1.

Figure 4B: Different Gene Expression patterns in Cancer-related Pathways between Normoxia and

Hypoxia

Fig 1: APE1 protein expression in (A) patient tissue, (IHC, red arrow= APE1 in tumor, black arrow= APE1 in stroma) and (B) Quantitation of APE1 staining in patient tumors (n=10). Both tumor and stroma express APE1 but tumor tissue is significantly higher in expression. (*p<0.05, Student’s t test)

Figure 3: Single Cell RNA Sequencing Analysis of PDAC cells following APE1 knockdown in

Normoxia and Hypoxia

Fig 3: Analysis of low passage patient derived PDAC cells (Pa03C) following APE1 knockdown and scRNA-seq in normoxia and hypoxia. The distribution of APE1 expression across the four sample groups is also shown. This allows us to identify differentially expressed genes in relation to APE1 expression per cell.

Fig 4B: Gene Expression Patterns in the Mitochondrial Dysfunction pathway in Hypoxia (1% O2). Genes with increased expression in APE1 knock-down cells are shown in pink whereas those with decreased expression are shown in green. Genes or complexes which were identified as differentially expressed are outlined in pink.

1 2 3Patient# Pancreatic ductal adenocarcinoma (PDAC) is the 4th leading cause of cancer related mortality in the United States, and is accompanied by a fibrotic phenotype that contributes to chemotherapeutic resistance. Combination therapies including Gemcitabine (GemzarTM) and sustained release, nab-paclitaxel (AbraxaneTM) and FOLFIRINOX (5-FU/leucovorin/irinotecan/oxaliplatin) offer modest improvement in survival, albeit at an increase in side effects. A novel PDAC target, Apurinic/apyrimidinic endonuclease/redox factor-1 (APE1/Ref-1 or APE1) is a multifunctional protein involved in repairing DNA damage via endonuclease activity and in redox regulation of transcription factors like HIF-1α, STAT3, NFκB and others. APE1 is essential for cell viability, which prevents generation of a stable APE1-knockout cell line. APE1 siRNA knockdowns are also transient, which means techniques like bulk RNA-seq are unable to generate a comprehensive list of genes regulated by APE1. Therefore, we utilized single-cell RNA sequencing to compare the transcriptomes of siAPE1 and scrambled control cells in low passage patient-derived PDAC cells in normoxic and hypoxic conditions to gain a more complete understanding of genes and pathways affected by APE1. Following Ingenuity Pathway Analysis, metabolic pathways such as oxidative phosphorylation and mitochondrial function were among the top 10 pathways identified. In this study we analyze the role of APE1 in mitochondrial metabolism. We also show how the combination of targeting APE1 redox function along with mitochondrial metabolism enhances tumor efficacy. Results & Summary

APE1

Ku 70

SCR 10nM 50nM

10nM SCR 10nM siAPE1

Fig 2: APE1 protein expression in PaCa-2 PDAC cells following siRNA knockdown using (A) Western Blot and (B) IHC. Panel B shows differential knockdown within the cells. Black arrow= Minimal reduction in APE1, Red arrow=No reduction in APE1.

A

A B

B

A

AB

Figure 4A: Different Gene Expression patterns in Cancer-related Pathways between Normoxia

and Hypoxia

Fig 4A: Gene Expression Patterns in the Mitochondrial Dysfunction pathway in Normoxia. Genes with increased expression in APE1 knock-down cells are shown in pink whereas those with decreased expression are shown in green. Genes or complexes which were identified as differentially expressed are outlined in pink.

Fig 7: (A & B) Low passage patient derived PDAC cells (Pa03C, red) in co-culture with cancer-associated fibroblasts (CAF19, green) treated with 2.5 µM APX2009 in combination 25 µM CPI-613. (C) 3D spheroid co-culture assays, where cells were treated with 35 µM APX3330 in combination with 10 µM CB-839. Tumor intensity measured and images captured using Thermo Arrayscan

A

Fig 6: (A) 3D spheroid assays using patient derived PDAC cells (Pa03C) grown using media with different glucose concentrations. G = 4.5 g/L Glucose, Low G = 1 g/L. (n=3; *p<0.001, **p<0.0001). Tumor intensity measured using Thermo ArrayScan. (B) Mitochondrial function assessed using different metabolic substrates. Kinetic assay performed using Biolog S-1 plates, OD @ 590nm measured every 90 seconds for 90 minutes.

B

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

NDUFA1 NDUFS4 UQCRC1 SURF1 COX15 NNT PRDX5

FoldCha

nge

SCRNorm SiNorm SCRHyp SiHyp* * * * * * *

Complex 1 Complex 3 Complex 4 Membrane Mito Redox

B

0.1

0.15

0.2

0.25

0.3

0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90

Absorbance@

590nm

Time(mins)

SCR siAPE1

*

**

Co-culture PDAC tum

or cells + CAFs respond to combination therapy

of APX inhibition and metabolic inhibitor, CPI-613 (25 µ

M)

Arrow

s indicate treatment tim

es.

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8

12 14

Tumor Intensity P

a03C+C

AF (n=3)

Ctrl

CP

I-613 A

PX

3330 C

ombo D

ays post plating Fold change compared to untreated control

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8

12 14

Tumor Intensity P

a03C+C

AF (n=3)

Ctrl

CP

I-613 A

PX

2009 C

ombo

AP

X3330=35 µM

Days post plating

AP

X2009=5 µM

-0.20

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8

12 14

Tumor Intensity P

a03C+C

AF (n=3)

Ctrl

CP

I-613 A

PX

2014 C

ombo D

ays post plating

AP

X2014=2.5 µM

Ctrl C

PI-613 A

PX

Com

bo

C

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)CtrlCPI-613APX3330Combo

Days post plating

Fold

cha

nge

com

pare

d to

unt

reat

ed c

ontro

l

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

CtrlCPI-613APX2009Combo

APX3330=35 µM

Days post plating

APX2009=5 µM

-0.20

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

CtrlCPI-613APX2014Combo

Days post plating

APX2014=2.5 µM

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

4 8 12 14

TumorIntensityPa03C+CAF(n=1)

Ctrl

CB-839

APX2014

Combo

Co-culture PDAC tumor cells + CAFs respond to combination therapy of APX inhibition and metabolic inhibitor, CPI-613 (25 µM)

Arrows indicate treatment times.

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX3330 Combo

Days post plating

Fold

cha

nge

com

pare

d to

unt

reat

ed c

ontro

l

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX2009 Combo

APX3330=35 µM

Days post plating

APX2009=5 µM

-0.20

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

4 8 12 14

Tumor Intensity Pa03C+CAF (n=3)

Ctrl CPI-613 APX2014 Combo

Days post plating

APX2014=2.5 µM

Ctrl CPI-613 APX Combo