Accepted Manuscript

TNF alpha signalling is associated with therapeutic responsiveness to vascular disrupting agents in endocrine tumors

Constanze Hantel, Alexandra Ozimek, Regia Lira, Bruno Ragazzon, Carsten Jäckel, Roman Frantsev, Martin Reincke, Jerôme Bertherat, Thomas Mussack, Felix Beuschlein

PII:	S0303-7207(15)30164-7
DOI:	10.1016/j.mce.2015.12.009
Reference:	MCE 9369

To appear in: Molecular and Cellular Endocrinology

Received Date: 8 September 2015

Revised Date: 1 December 2015

Accepted Date: 8 December 2015

R95H 0000-7209

Molecular and Cellular Endocrinology

Please cite this article as: Hantel, C., Ozimek, A., Lira, R., Ragazzon, B., Jäckel, C., Frantsev, R., Reincke, M., Bertherat, J., Mussack, T., Beuschlein, F., TNF alpha signalling is associated with therapeutic responsiveness to vascular disrupting agents in endocrine tumors, Molecular and Cellular Endocrinology (2016), doi: 10.1016/j.mce.2015.12.009.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

TNF alpha signalling is associated with therapeutic responsiveness to vascular disrupting agents in endocrine tumors

Constanze Hantel11, Alexandra Ozimek2, Regia Lira3, Bruno Ragazzon4, Carsten Jäckel5, Roman Frantsev1, Martin Reincke1, Jérôme Bertherat4, Thomas Mussack2, Felix Beuschlein1

1 Endocrine Research Unit, Medizinische Klinik und Poliklinik IV, Klinikum der Universität 8

München, Munich, Germany

2 Department of Surgery, Klinikum der Universität München, Munich, Germany

3 Pediatrics Department, FMRP-USP, Ribeirão Preto, Brazil

4 Institut Cochin, Inserm U1016, Cnrs UMR8104, Université Paris Descartes, Paris, France.

5 Clinical Biochemistry, Medizinische Klinik und Poliklinik IV, Munich, Germany

Correspondence to:

Dr. Constanze Hantel

Endocrine Research Unit, Medizinische Klinik und Poliklinik IV

Klinikum der Universität München

Ziemssenstr. 1

D-80336 Munich

Phone: +49 89 5160 2943 22 Fax: +49 89 5160 4467 23 E-mail: Constanze.Hantel@med.uni-muenchen.de

24 25 Short title: 26 TNFAIP3 and drug resistance in ACC

1 2 3 4 5 6 7

9 10 11 12 13 14 15 16 17 18 19 20 21

1 Key words:

2 Endocrine tumors, Tumor-Vascular-Disrupting-Agent, TNFa, TNFAIP3/A20

4 Abbreviations used:

5 ACC: Adrenocortical Carcinoma

6 GEP-NET: Neuroendocrine tumor of the gastroenteropancreatic system

7 IKK beta: Inhibitor of nuclear factor kappa-B kinase subunit beta

8 TNF-R1: TNF-receptor-1

9 TLR-4: Toll-like-receptor 4

10 Tumor-VDAs: Tumor-Vascular-Disrupting Agents

11 12 Declaration of interest:

13 The authors declare that there is no conflict of interest that could be perceived as prejudicing

14 15

the impartiality of the research reported.

Grants or fellowships supports:

16 17 This work was supported by the Wilhelm-Sander-Stiftung (2011.103.1) to C.H. and F.B. ASA404 was provided and financial contribution was made by Novartis. 18 19 20 Acknowledgements: 21 The authors are indebted to Igor Shapiro and Susanne Mentz for their excellent technical 22 support. Furthermore the authors thank Novartis Oncology for provision of ASA404 as a study drug and Dr. Silviu Sbiera and Sabine Herterich (University of Würzburg, Germany) for performing the STR-analyses on BON tumor cells. 23 24 25 26

1 Abstract

2 3

ASA404 (Vadimezan) belongs to a class of agents with disrupting properties against tumor 4 5 vasculature, which is partly mediated by TNFa-signalling. Preclinical and early clinical studies have indicated promising results for ASA404, while extended clinical trials performed 6 poorly. Our aim was to investigate the potential therapeutic applicability of ASA404 against 7 8 endocrine tumors. Moreover, as the reason for the unpredictable clinical anti-tumor activity of ASA 404 remained uncertain in previous studies, we compared two tumor models of 9 endocrine origin with different responses to ASA404 treatment. Specifically, we determined anti-tumoral effects in preclinical models of neuroendocrine tumors of the gastroenteropancreatic system (BON) and adrenocortical cancer (NCI-H295R) in vitro and in xenograft models in vivo. Upon treatment of tumor bearing mice significant anti-tumoral effects, an increase in TNFa as well as activation of TNFa-specific downstream signalling were evident in the BON tumor model while no comparable effects were detectable for NCI- H295R. We identified TNFAIP3/A20, a key molecule of an inhibitory feedback-loop downstream of TNF-receptor 1, CD40, Toll-like receptors, NOD-like receptors and the interleukin-1 receptor signalling cascades, as overexpressed in the adrenocortical carcinoma tumor model. Subsequent analyses of clinical patient samples confirmed a correlation between tumor TNFAIP3 expression levels and overall survival in patients with ACC. Taken together our findings provide evidence that modulation of TNFa-signalling could be of relevance both for the clinical course of ACC patients and as a marker of treatment response.

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

1 1. Introduction

2 3

Tumors arising from endocrine organs comprise a group of diseases with a wide spectrum of 4 functional and neoplastic properties. Among them, adrenocortical carcinomas (ACC) as well 5 as neuroendocrine tumors (NETs) of the gastroenteropancreatic system (GEP) represent both 6 rare tumors for which surgical resection is the primary therapeutic option that allows for 7 potential cure of the disease [1, 2]. However, a large proportion of patients either presents 8 with metastatic spread at initial work-up or with recurrence during follow-up examination. 9 While for NETs in recent years accumulating insights into molecular mechanisms led to the development of promising therapeutic strategies blocking VEGF and mTOR signalling [3], clinical studies with targeted therapies for patients with ACC were unfortunately mainly disappointing so far ([4, 5]). Thus, more insights into mechanisms of drug resistance are urgently needed. 10 11 12 13 14 15 16 17

Solid tumors cannot grow beyond a few mm in diameter nor metastasize without neo- vascularization. Furthermore, heterogeneous tumor perfusion, impaired vascular permeability as well as high interstitial pressure restrict the transfer of pharmacological agents from the circulation into tumors and metastases. For these reasons, in recent years tumor vasculature 18 itself has become an interesting target for cancer therapy [6]. Tumor-Vascular-Disrupting Agents (Tumor-VDAs) such as ASA 404 (vadimezan) differ from angiogenesis inhibitors by targeting the established blood vessel network within a tumor rather than preventing the growth of new blood vessels [7, 8]. Tumor VDAs have been shown to induce a rapid collapse in tumor blood flow resulting in extensive tumor-cell necrosis. Specifically, it has been demonstrated that combined treatments of ASA404 with taxanes such as paclitaxel result in enhanced therapeutic efficacy [6]. Although the exact mechanism of action caused by ASA 404 is unknown, the localized release of TNFa seems to play a major role in mediating

19 20 21 22 23 24 25 26 therapeutic efficacy [9]. TNFa itself is a cytokine which translates its effects mainly via the

1 TNF-receptor-1 (TNF-R1) and TNF-R1 stimulation is postulated to result in the formation of two signalling complexes which lead to induction of NF-KB and caspase 3 activity [10].

3 TNFAIP3 is the key component of an inducible feedback loop that inhibits TNFa-stimulated NF-KB-activity and restricts TNFa-induced apoptosis [11, 12]. In addition, TNFAIP3 represents an interesting link in the complex interplay of different receptors involved in innate

4 5

6 immune response and cell survival as it delimits also NF-KB signalling downstream of CD40, 7 Toll-like receptors (TLRs), NOD-like receptors (NLRs) and the interleukin-1 receptor (IL-1R; [12]). Multiple studies have linked TNFAIP3 with inflammatory, autoimmune and malignant 8 9 10

diseases [12]. Moreover, TNFAIP3 has been found to be involved in drug resistance against tamoxifen and alkylating agents in breast cancer and glioblastoma cells [13, 14]. The investigation of tumor samples of these entities furthermore revealed a poor prognostic status for those with high TNFAIP3 levels. In accordance with these findings, TNFAIP3 has been proposed as predictive biomarker and therapeutic modulator for these malignant neoplasms. Our data provide evidence that TNFAIP3 could also represent an interesting biomarker and novel candidate for therapeutic intervention of adrenocortical carcinoma.

11 12 13

14 15 16

2. Materials and Methods

3 2.1 Cell experiments

Culturing, media and reagents

NCI-H295R and BON cells were cultured as previously described [15, 16]. TNFa (T 0157)

4 5 6 from Sigma-Aldrich (Steinheim, Germany) and ASA 404 was kindly provided by Novartis

7 (Basel, Suisse). If not stated otherwise 0.1ug/ml TNFa incubated for 24 hours has been used

8 for in vitro experiments.

9 The utilized cell line NCIh295R has been obtained from ATCC and furthermore recently authentified. BON cells have not been yet deposited in a comparable way in a cell bank,

10 11 12

however the cell line and xenografts have been originally described and characterized in numerous publications as exemplified here [17, 18]. Moreover, recent STR (short tandem

13 repeat) analyses revealed comparable results of two independent laboratories.

D5S818	D13S317	D7S820	D16S539	vWA	TH01	TPOX	CSF1PO	AMEL
9,12	11,12	9,9	10,11	18,19	8,8	9,9	10,11	X, Y

14 15

2.2 Caspase 3/7 assay

16 Cells were seeded on a 96 well white polystyrene plate (MaxiSorp M, Nunc, Langenselbold, 17 18 Germany) and incubated with TNFa (0.1 and 1µg/ml) or ASA404 (100 and 250u g/ml) for 24 hours. Detection bas been carried out as previously described [15].

19 2.3 IKK beta ELISA

21 22 23 24

20 For quantification of IKK beta (Inhibitor of nuclear factor kappa-B kinase subunit beta) activity the PathScan® Phospho-IKK® (Ser177/181) Sandwich ELISA Kit from Cell Signalling Technology® (Beverly, CA) based on the measurement of Ser177 and Ser181 phosphorylation in the activation loop of IKK, was used. 2 x 10° BON and NCI-H295R cells, respectively, were seeded in triplicates on 10cm dishes and cultivated overnight. After

medium removal cells were rinsed briefly with PBS and then incubated for 3 hours with 1 2 medium alone or medium containing 0.1µ g/ml TNFa. ELISA was performed according to the manufacturer’s protocol. The detection was carried out in a SPECTRA microplate reader from Tecan (Crailsheim, Germany).

3 4 5

2. 4 Animal Experiments

For in vivo experiments female athymic NMRI nu/nu mice (6 - 8 weeks) were housed under pathogen-free conditions in a temperature controlled isolation unit with 12-hours light and dark cycles and food and water ad libitum as also described elsewhere [15, 16]. All experiments were carried out following protocols approved by the Regierung von Oberbayern and in accordance with the German guidelines for animal studies. For therapeutic experiments, 15 x 10° BON or NCI-H295R cells in a volume of 200 ul PBS were inoculated for tumor development subcutaneously into the neck of each mouse. At days 11 and 15 after BON and NCI-H295R tumor induction therapeutic treatments were given intraperitoneally in 15 dosages of 20mg/kg body weight Paclitaxel and/or 22.5 mg/kg ASA404. 24 hours after treatment mice (n=4-5) were sacrificed, the tumors excised and paraffin embedded. For Esta investigation of TNFa secretion BON and NCI-H295R tumor-bearing mice (n=4-5) were treated with either NaCl or ASA404 (22.5 mg/kg bodyweight). Two hours after administration mice were sacrificed, trunc blood collected for serum analysis and a piece of each tumor was snap frozen for real time PCR analyses. The rest of the tumors was subsequently minced and incubated for further three hours on 48 well plates containing 1ml of serum free DMEM supplemented with Polymyxin B [50mg/l] at 37 ℃. Preparations were transferred into 1.5 ml tubes followed by a centrifugation step for 10 minutes at 14 000 rpm. Supernatants were collected in fresh tubes and stored at 80℃ until analysis. TNFa ELISA was performed following manufacturers protocol (Mouse TNFa ELISA Kit, Becton

16 17 18 19 20 21 22 23 24 25

6 7 8 9 10 11 12 13 14

ACCEPTED MANUSCRIPT

1 Dickinson, Heidelberg, Germany). The detection was carried out in a SPECTRA microplate

2 reader from Tecan (Crailsheim, Germany).

3 4 5 6

2.5 Histology and immunohistochemistry

Paraffin-embedded sections were rehydrated and incubated with blocking buffer containing 3 % BSA (Roche Diagnostics, Mannheim, Germany), 5% goat serum (Jackson 7 8 9 10 11 12 13 14 ImmunoResearch Laboratories, West Grove, PA), and 0.5 % Tween 20 for 15 min. For the specific stainings either monoclonal mouse anti-human Ki67 (DakoCytomation, Denmark; 1:200 in BB), purified rat anti-mouse CD31 (Pharmingen, NJ, USA; 1:30 in BB), polyclonal rabbit anti-human TLR-4 (Acris antibodies, San Diego, CA, USA) antibodies or DeadEnd colorimetric TUNEL system (Promega, Madison, WI; following the manufacturers protocol) were used and incubated overnight at 4 ℃. After rinsing for 15 min in PBS, secondary antibody (goat anti-mouse biotinylated IgG; Santa Cruz, CA, biotin-SP-conjugated AffiniPure goat anti-rat, Jackson Immuno Research Lab, CA or biotinylated anti-rabbit, Vector 15 Laboratories, Burlingame, CA, USA) was applied for 30 min at room temperature. Bound 16 17 primary antibody was visualized using the Vectatstain ABC Kit (Vector Laboratories, Burlingame, CA, USA) according to the manufacturer’s protocol with incubation for 30 min 18 followed by 3,3’-diaminobenzidine (Sigma-Aldrich, Steinheim, Germany) staining. For quantification 6 high power fields (HPF, 0.307 mm2, 400x magnification)/ tumor were investigated and quantified for either proliferation index (% Ki67 positive cells), relative amount of apoptotic cells (%TUNEL positive cells) or microvessel density (number of CD31 positive cells). Histological examination of tissue slides was analyzed by hematoxylin-eosin staining (Sigma-Aldrich, Steinheim, Germany).

19 20 21 22 23 24 25 26

1 2.6 cDNA microarray analysis

The human affymetrix gene 1.1 ST array has been performed by the Kompetenzzentrum Fluoreszente Bioanalytik, University of Regensburg, Germany on samples of ± 0.1ug/ml

2 3 4

TNFa incubated for 24 hours has been used for in vitro experiments. TNFa treated BON and 5 NCI-H295R cells (NCI-H295R as singlicate, NCI-H295R + TNFa and BON ± TNFa as

6 duplicates). Complete data for this study has been uploaded to the GEO repository under GSE64250.

Quality control and clustering analyses of the array data were carried out using the MADMAX (Management and Analysis Database for Multi-platform microArray experiments) platform (https://madmax.bioinformatics.nl, University of Wageningen). The same platform was used to normalize the data using the RMA algorithm and annotate individual genes based on Gene Ontology terms [19].

2.7 Real-Time PCR analysis and Western Blot

For detailed Real-Time PCR analyses TNFa treated and not treated human BON and NCI- H295R cells (± TNFa in triplicates) and tumors (NaCl and ASA404 treated, n=4-5) were used for RNA extraction (SV Total RNA Isolation system, Promega) and reverse transcription (RevertAid™ H Minus First Strand cDNA Synthesis Kit, Fermentas). For Real-Time PCR analyses we utilized the EvaGreen® reaction mix (Bio-Rad, Munich, Germany) in the Stratagene Mx3000PTM Cycler (Agilent Technologies, Waldbronn, Germany). Human primer sequences or catalogue numbers were as follows: TNF-receptor 1 forward: 5’-ACC AAG TGC CAC AAA GGA AC-3’ and TNF-receptor 1 reverse: 5’-CTG CAA TTG AAG CAC TGG AA-3’ (263 bp amplicon size), TNFAIP3 (PPH00063A, Qiagen, Hilden, Germany), CXCL8 (VBC Genomics, Phadia: Uppsala, Sweden), Irak2 (origene/amsbio, Abingdon, UK, #HP205425 TNFa, and NFKBIA (VHPS-9415 and VHPS-6180, Biomol, Hamburg,

24 25 26 Germany). Quantification was adjusted using the housekeeping gene GAPDH for human

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

1 samples forward: 5’-AGC CTC CCG CTT CGC TCT CT-3’ and reverse: 5’-CCA GGC GCC

2 CAA TAC GAC CA-3’ (141 bp amplicon size).

Protein extraction and Western Blot were in general performed as described [20]. 20µg

4 protein preparations were run on Mini PROTEAN TGX Stain-Free Precast gels (Bio-Rad, 5 6 Hercules, CA, USA). TNFAIP3 (1:5000, Novus Biologicals, Cambridge, UK) and ß-actin antibody (1:5000; Sigma-Aldrich Corp., Saint Louis, U.S.A.) were used as primary 7 antibodies. Quantification was achieved by the measurement of the pixels and adjacent analyses using the imaging program ImageJ (NIH).

2.8 mRNA expression profiling in patient samples:

Two cohorts of adrenocortical tumors were used: Cochin cohort included 47 ACC and 4 normal adrenals (N.Ad) (Gene Expression Omnibus (GEO) dataset GSE49280 and ArrayExpress dataset E-TABM-311) [21] and TCGA cohort (79 ACC, http://gdac.broadinstitute.org/). For TCGA cohort, mRNA sequencing data were extracted from Broad Institute GDAC Firehose (TCGA data version 2015_03_26) and all calculations were performed on Log2 values of RSEM normalised read counts. Differential expression was measured with moderated t-test (limma R package). Survival curves were obtained by the Kaplan-Meier method. Differences in survival were assessed with the log-rank test. All P values were adjusted using the Benjamini-Hochberg correction method.

2.9 Statistical analysis

Statistical significance was determined using the t-test for a two group comparison or One- Way-ANOVA for more groups (Bonferroni for overall or Dunnett adjusted for direct comparison with a control group, Prizm software, Houston, TX). Statistical significance is

denoted as stars (*, p<0.05; ** , p<0.01; *** , p<0.001) in the figures if not stated otherwise.

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

1 3. Results

3.1 Investigation of ASA404 induced anti-tumor effects in preclinical models for 3

endocrine tumors in vivo

In a first step we investigated the therapeutic potential of ASA404 against endocrine tumors 6 7 in preclinical xenograft models for neuroendocrine tumors of the gastroenteropancreatic system (BON) and adrenocortical carcinoma (NCI-H295R). It is known that ASA404s 8 mechanism of action is working in two parts highly supported by the immune system: an early period mediated by macrophages, resulting in case of therapeutic efficacy in 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

hemorrhagic necrosis, and a second line involving T-cells mediating long-term effects. In accordance with these findings, it has been reported, that short-term therapeutic efficacy is detectable independently from the utilized mouse host (nude/athymic versus syngeneic [22]) while long-term therapeutic effects are not translated in immunodeficient mouse models. In accordance with these previous reports, we chose a short-term therapeutic setting and subsequent quantification for the utilized xenograft-models in athymic nude mice. Twenty four hours after treatment with ASA404 (A), paclitaxel (P), the combined administration (A+P) or NaCl as control, tumors were excised and subesquently analyzed for non-vital areas (H&E), proliferation (Ki67), apoptosis (TUNEL) and vascularization (CD31). While A and A+P treated BON tumors exhibited extensive necrotic areas (NaCl: 8.3±1.7%, P: 13.0±2.9%, A: 24.9±7.5%, A+P: 79.1+24.5%; p<0.01 versus NaCl; Figure 1A, left), showed a highly significant reduction in tumor cell proliferation (Ki67-Indices/ HPF/ tumor: NaCl: 60.2±0.8%, P: 54.8±0.8%, A: 52.0±0.7%, A+P: 48.0±0.7%; p<0.001 versus NaCl; Figure 1B, left), induction of apoptosis (TUNEL positive cells: NaCl: 7.2±1.2, P: 12.0±1.8, A: 30.9±3.8, A+P: 31.3±4.4; p<0.001 versus NaCl; Figure 1C, left) and disruption of tumor blood vessels (CD31 positive cells/ HPF/ tumor: NaCl: 11.9±0.7, P: 12.5±1.2, A: 6.9+0.9, A+P: 5.5±0.4; p<0.001

25 26 versus NaCl; Figure 1D, left), no such treatment related effects were detectable in NCI-

4 5

1 H295R xenografts (necrotic areas: NaCl: 0.9±0.6%, P: 1.0±1.0%, A: 0.7±0.7%, A+P:

2 6.0±3.8%; Ki67-Indices/ HPF/ tumor: NaCl: 46.6±1.0%, P: 43.7±1.5%, A: 46.6±1.6%, A+P:

3 45.6±0.9%; p>0.05; TUNEL positive cells: NaCl: 9.3±0.9, P: 21.6±2.2, A: 9.3±1.2, A+P:

4 6.5±1.2; CD31 positive cells/ HPF/ tumor: NaCl: 19.8±1.5, P: 16.6±1.7, A: 19.7+2.4, A+P: 5 26.3±1.7;Figure 1, right rows).

Necrotic area [%]

100

Ki 67 index [%]

NaCl

A + P

NaCl

A+P

NaCI

NaCl

A + P

NaCl

A + P

TUNEL positive cells/HPF/tumor

100

50-

CD31 positive cells/HPF/tumor

NaCl

A + P

NaCl

A + P

NaCl

A + P

NaCl

A + P

7 8

9 Figure 1: Quantification of necrotic areas in NaCl, Paclitaxel (P), ASA404 (A) and combined 10 (A+P) treated BON (left) and NCI-H295R tumors (right) as well as representative pictures of 11 H&E stained sections (A). Investigation of proliferation (B), apoptosis (C) and microvessels 12 (D) for BON (left) and NCI-H295R (right) tumors. Stars denote significance over NaCl 13 treated controls (*, p<0.05; ** , p<0.01; *** , p<0.001).

14 15

1 3.2 Induction of TNFa secretion and signalling

2 In a next step we investigated TNFa levels upon ASA404 treatment. Thereby, we observed a

3 significant TNFa increase in serum and tumors in the BON model (Figure 2 A and B), while

4 5

for NCI-H295R only a trend towards increased TNFa levels was detectable. Notably, also stimulation with exogenous TNFa induced highly significant upregulation of TNF mRNA in BON cells while this effect was blunted in NCI-H295R cells (Figure 2C).

1500

3000

1.00-

TNFa [pg/ml serum]

[pg TNFa./mg tumor]

TNF Expression [Relative Quantity]

0.75

1000

TNFa

2000

Caspase 3/7 activity [% of basal]

NCI ø NCIASA E 150 ** ** NCIh295 100 - - 250 - - a

0.50

500

1000

0.25

0.00

BON ø

BON ASA

NCI ø

NCI ASA

BON ø

BON ASA

BON ø

BON TNFa.

NCI ø

NCI TNF&

1000-

Active IKK beta [% of basal]

750

100

500

250

BON

BON controls

BON TNFO.

NCI controls

NCI TNF&

ASA [µg/ml]

100

250

TNFa [µg/ml]

0.1

9 10 11

Figure 2: Investigation of treatment related effects on serum (A) and intra-tumor (B) TNFa levels in BON and NCI-H295R (NCI) tumor bearing mice. Endogenous TNF expression (C) following external stimulation with TNFa in vitro. TNFa dependent caspase 3/7 (D) and IKK beta (E) activity in both tumor models. Stars denote significance over untreated controls (*, p<0.05; ** , p<0.01; *** , p<0.001).

12 13 14 15

To further highlight differences in cellular sensitivity, we investigated TNFa dependent

16 effects on apoptosis and NFKB-pathway activation. In keeping with the in vivo findings and

17 independent of blood vessel disruption or differences in ASA404-induced TNFa secretion

1 2

NCI-H295R cells were less responsive to the TNFa treatment in comparison to BON cells (Figure 2D). Moreover, while IKK beta (Inhibitor of nuclear factor kappa-B kinase subunit beta) activity was significantly increased in BON cells upon TNFa treatment, this was abolished in NCI-H295R cells (Figure 2E).

3.3 Identification of dysregulated signalling in adrenocortical carcinoma cells upon TNFa stimulation

To identify molecular alterations in downstream signalling upon TNFa stimulation, we 9 performed a gene-array expression analysis on RNA of BON and NCI-H295R cells in the presence and absence of TNFa. A complete list of involved pathways resulting from a KEGG pathway analysis and a list of all altered genes (>1.5 fold) is provided in Supplemental Table 1. Table 1 summarizes the most highly up-regulated pathways and candidates related to cytokines/chemokines, cell adhesion or TNF, TLR-signalling, NOD and RIG-I-like-receptor signalling (TNF, CXCL8, CXCL1, CCL2, TNFRSF9, IRAK2, ICAM-1, BIRC3, NFKBIA, TNFAIP3/A20) which were confirmed in additional Real time PCR studies (Supplemental Figure 1).

10 11 12 13 14 15 16 17

PATHWAYS		BON			NCI
ENTREZ GENE ID		COUNTS	P-VALUE	FOLD INCREASE	COUNTS	P-VALUE	FOLD INCREASE
Cytokine-cytokine receptor interaction		30	0.000000036		-	not significant
3604 tumor necrosis factor receptor superfamily, member 9	TNFRSF9			14,9			2,2
6347 chemokine (C-C motif) ligand 2	CCL2			3,7			10,2
7124 tumor necrosis factor (TNF superfamily, member 2)	TNF			7,8			-
chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity,
2919 alpha)	CXCL1			7,1			-
3576 chemokine (C-X-C motif) ligand 8 (interleukin 8)	CXCL8			6,7			3
Pathways in cancer		24	0.0014		9	0.0047
nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor,
4792 alpha	NFKBIA			9,9			2,6
330 baculoviral IAP repeat-containing 3	BIRC3			9			2,9
3576 chemokine (C-X-C motif) ligand 8 (interleukin 8)	CXCL8			6,7			3
MAPK signalling pathway		19	0.0071		6	not significant
7124 tumor necrosis factor (TNF superfamily, member 2)	TNF			7,8			-
Chemokine signalling pathway		17	0.001		6	0.0013
6347 chemokine (C-C motif) ligand 2	CCL2			3,7			10,2
nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor,
4792 alpha	NFKBIA			9,9			2,6
chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity,
2919 alpha)	CXCL1			7,1			-
3576 chemokine (C-X-C motif) ligand 8 (interleukin 8)	CXCL8			6,7			3
Focal adhesion		16	0.0055		-	not significant
330 baculoviral IAP repeat-containing 3	BIRC3			9			2,9
Cell adhesion molecules (CAMs)		15	0.00025		-	not significant
3383 intercellular adhesion molecule 1	ICAM1			12,5			-
ECM-receptor interaction		13	0.000038
Antigen processing and presentation		13	0,000038		-	not significant
Toll-like receptor signalling pathway		12	0.00016		6	0.0013
nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor,
4792 alpha	NFKBIA			9,9			2,6
7124 tumor necrosis factor (TNF superfamily, member 2)	TNF			7,8			-

3 4 5 6 7 8

3576 chemokine (C-X-C motif) ligand 8 (interleukin 8)	CXCL8			6,7			3
RIG-I-like receptor signalling pathway		12	0.00091		5	0.0026
nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor,
4792 alpha	NFKBIA			9,9			2,6
7124 tumor necrosis factor (TNF superfamily, member 2)	TNF			7,8			-
3576 chemokine (C-X-C motif) ligand 8 (interleukin 8)	CXCL8			6,7			3
ptor NOD-like receptor signalling pathway		10	0.00032		6	0.00014
7128 tumor necrosis factor, alpha-induced protein 3	TNFAIP3			13,9			1,9
nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor,
4792 alpha	NFKBIA			9,9			2,6
330 baculoviral IAP repeat-containing 3	BIRC3			9			2,9
7124 tumor necrosis factor (TNF superfamily, member 2)	TNF			7,8			-
Apoptosis		9	0.012		-	not significant
nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor,
4792 alpha	NFKBIA			9,9			2,6
330 baculoviral IAP repeat-containing 3	BIRC3			9			2,9
7124 tumor necrosis factor (TNF superfamily, member 2)	TNF			7,8			-
3656 interleukin-1 receptor-associated kinase 2	IRAK2			5,8			-
Natural killer cell mediated cytotoxicity		10	0.049		-	not significant
3383 intercellular adhesion molecule 1	ICAM1			12,5			-
7124 tumor necrosis factor (TNF superfamily, member 2)	TNF			7,8			-
TGF-beta signalling pathway		8	0.036
7124 tumor necrosis factor (TNF superfamily, member 2)	TNF			7,8		PER	-
p53 signalling pathway		7	0.035		-	not significant
Cytosolic DNA-sensing pathway		6	0.047		-	not significant
	NFKBIA			9,9			2,6

3.4 Specific IRAK2 and TNFAIP3 induction by TNFa

4 From the previous list we identified two key molecules in TNFa synthesis and signalling, 5 differently regulated upon TNFa stimulation: IRAK2 and TNFAIP3. IRAK2 kinase has 6 recently implicated to mediate long-term expu RAK2 induction of TNFa downstream of TLR-signalling 7 In fact, IRAK2 expression was highly inducible by TNFa treatment and showed sustained expression over a time course of several hours specifically in BON cells (Figure 3A). 8

9 Moreover, while TNFAIP3/A20 expression was as expected significantly inducible upon TNFa stimulation in BON cells, basal overexpression and diminished inducibility was found in treatment resistant NCI-H295R cells (Figure 3B). The same tendencies were detectable for

12 TNFAIP3 protein levels in both tumor models (Figure 3C).

IRAK2 [Relative Quantity]

王

24h

BON controls

BON TNFa

NCI controls

NCI TNFa

TNFAIP3 Expression [Relative Quantity]

BON Ø BON TNFa

NCI ø

NCI TNFa

1.00

TNFAIP3 Protein [Relative Quantity]

0.75

0.50

0.25

0.00

BON ø

BON TNFa

NCI ø

NCI TNFa

TNFAIP3

Actin - beta

CRIPT

Figure 3: Endogenous TNF expression (G) following external stimulation with TNFa in vitro 3

4 as well as IRAK2 expression after 1, 3, 9 and 24 hours of TNFa stimulation (H) in BON and

1 NCI-H295R cells in vitro. Real Time PCR analysis of TNFAIP3/A20 expression for both

2 tumor entities after TNFa treatment in vitro (A) and Western Blot analysis of TNFAIP3/A20

3 protein under the same conditions (B). Stars represent significance vs. untreated controls (*, p<0.05; ** , p<0.01; *** , p<0.001).

3.5 Correlation of TNFAIP3 expression in ACC samples with clinical outcome

Of note, subsequent analyses of all identified candidates listed in Table 1 indicated exclusively for TNFAIP3 an overexpression in ACC compared to normal adrenals(N.Ad.) and show a correlation with patient survival. Indeed, high TNFAIP3 expression was significantly associated with poor overall survival in ACC, whether analysed as a continuous variable (TCGA cohort, p=0.0017, hazard ratio (HR)=1.63, 95% confidence interval (CI)=1.2-2.2), stratified on tumor stage (I-II vs III-IV ; p=0.025, HR=1.46, CI=1.05-2.03) or after dichotomisation between ACC with low and high TNFAIP3 expression (Cochin and TCGA cohorts, Figure 4).

14 15

high : 22 cases low : 25 cases

Thigh : 40 cases

low : 39 cases

0.8

Expression (relative to N.Ad)

frequency

0.6

Expression

frequency

0.6

0.4

-1

0.2

-2

logrank test

p-value = 0.033

logrank test p-value = 1.1e-04

N.Ad

low

high

100

150

low

high

100

150

ACC

times (months)

ACC

times (months)

17 18 19 20

Figure 4: TNFAIP3 expression and overall suvival after dichotomisation between ACC with

low and high expression in Cochin (A and B) and TCGA cohorts (C and D). N.Ad = normal adrenals

4 5 6 7 8 9 10 11 12 13

4. Discussion

1 2 3

8 9 10

Recently, studies involving the vascular disrupting agent ASA404 have indicated anti-tumor 4 5 efficacy in preclinical models, provided promising results in initial clinical trials, but performed poorly in further clinical studies [23-25]. The reason for differential tumor 6 7 responsiveness had remained uncertain. Herein we demonstrate that ASA404 has high therapeutic activity in a tumor model for GEP-NETs. In agreement with previous reports on other solid tumors, combination with taxane-based agents augmented efficacy of ASA404 treatment [6, 9]. In clear contrast, no comparable effect was detectable in adrenocortical NCI- H295R tumors. As the release of TNFa followed by induction of local inflammation seems to 11 play a major role in mediating therapeutic efficacy of VDAs [9], we examined ASA404 dependent TNFa release. Of note, also chemotherapy and radiotherapy induced anti-tumor effects have been recently recognized to be, at least in part, dependent on local immune responses [26, 27]. Moreover, very recently it has been reported that the administration of doxorubicin in tumor-bearing mice stimulates the secretion of TNFa in tumors responding to anthracyclines [28].

The observed increase in intra-tumor and serum TNFa demonstrated increased levels for TNFa synthesis in the BON tumor model upon ASA404 treatment. Furthermore, we detected a self-pertaining autocrine loop with TNFa induced expression of tumor derived human TNFa. NFKB-activation is mostly self-limiting while synthesis of pro-inflammatory cytokines such as TNFa can continue for prolonged period of times as sustained cytokine production is 22 23 important in the maintenance of an inflammatory state [29]. Recently it has been shown that TLR-mediated inflammation falls into an early and a late phase and that IRAK2 is a critical 24 mediator for sustained TNFa production [30]. In accordance with this model, we found a significant IRAK2 induction in TNFa stimulated BON cells over 24 hours with a peak in the 25 26 early phase after 3 hours [30, 31], while this effect was not detectable in TNFa treated NCI-

12 13 14 15 16 17 18 19 20 21

1 H295R cells. Thus, differences in this autocrine TNFa related augmentation system might contribute to the ASA404 treatment resistance in the NCI-H295R model. Both, TNF-R1 and TLR-4 dependent NFKB- activation, is known to induce expression of pro-inflammatory cytokines and chemokines. Accordingly, in addition to expression of TNFa itself, we detected high and TNFa dependent up-regulation of CXCL8 and CXCL1, while this effect was absent or comparably lower for TNFa-treated NCI-H295R cells. Expression of ICAM-1 is also known to be regulated by TNFa [32].

4 5

8 9 10

12 13

In contrast, higher up-regulation of the chemokine CCL2 was present in NCI-H295R cells (Table 1 and Figure 5). Interestingly, the CCL2/CCR2 axis has been recently shown to coordinate survival and motility in breast cancer cells [33], tumor progression in breast cancer xenografts [34] and to protect prostate cancer cell survival from autophagic cell death [35]. Thereby, CCL2 responses to TNFa could modify tumor behavior. Similarly, inhibitor of apoptosis (IAP) proteins such as BIRC3 (c-IAP2) have been recognized to modulate TNFa- mediated cell death as well as TLR-signalling, innate immunity and NFKB-pathways [36]. As c-IAP 1 and 2 are responsible for RIP1 ubiquitination, less TNFa induced BIRC3 expression in NCI-H295R could well contribute to downstream differences in NFKB and caspase activation in both tumor models.

14 15 16 17 18 TNFAIP3/A20 is of particular interest for the interpretation of the current experiments because of its dual role in NFKB-activation and cell death [37]: Initially, TNFAIP3 had been identified in endothelial cells as a primary response upon TNFa treatment mediating resistance to TNFa cytotoxicity [38]. Furthermore, it has recently been recognized as an anti- inflammatory protein with a role not only in TNF-R1 but also IL-1, CD40, TLR and B cell- receptor induced NFKB-signalling [11, 39]. TNFAIP3 is an integral part of a negative feedback loop that mediates transient and cyclic pathway activation in the context of continuous TNFa stimulation [40]. Accordingly, we found TNFa-dependent TNFAIP3/A20 induction in the BON tumor model while this was absent in the ACC tumor model.

19 20 21 22

23 24 25 26

1 Interestingly, basal TNFAIP3 RNA and protein levels were highly elevated in the NCI- 2 H295R tumor model in line with a potential cause for the observed drug resistance (as 3 schematically illustrated in Figure 5). We identified similarly altered expression profiles, also 4 5 for TNFRSF9 [41], NFKBIA [42] and the direct interaction partner of TNFAIP3 TNIP1 [43] (and Supplemental table 1). Signalling via TNFSF9 (4-IBBL) and its receptor TNFRSF9 (4- 6 IBB) upregulates survival genes, enhances cell division, induces cytokine production, and 7 prevents activation-induced cell death in T cells [41] and an involvement of the latter has been 8 described in the negative feedback loop downstream of TNF-R1 and TLR-4.

9 As our experiments indicated a dysregulation for this loop in the tumor model for ACC while this appeared to be active in the neuroendocrine tumor model, human ACC tumor samples

10 11 12 13

where investigated to verify a putative correlation of these candidates regarding overall survival. However, in subsequent expression analyses of ACC tumor specimen from all identified candidates only for TNFAIP3 significant correlation with poor outcome could be found (Figure 4 C and D).

14 15 16

Taken together, we provide evidence that ASA404 holds promise in the treatment of GEP- NETs. Furthermore, we identified a negative feedback loop, including TNFAIP3/A20, TNIP1 17 and NFKBIA, to be dysregulated in the ACC tumor model and provide a correlation between tumor TNFAIP3 expression levels and overall survival in patients with ACC. Thus, in addition to insights into mechanisms of drug action our findings provide an interesting marker for treatment response that could translate into individualized therapeutic strategies.

18 19 20 21

TNF a / TL-R / CD40 / NOD / IL-R Stimulation

TNFAIP3

NFKB + Caspase Activities

B TNF a / TL-R / CD40 / NOD / IL-R Stimulation

TNFAIP3

NFKB + Caspase Activities

1 2

3 Figure 5: Schematic illustration of the putative TNFAIP3 loop leading to the differences in therapeutic response observed for both tumor models: A) The stimulation of the functional signaling axis downstream of TNFa-stimulation as well as of other upstream targets of TNFAIP3 including Toll-like (TL), CD40, NOD and interleukin (IL) receptors (R, 1) is leading to the activation of NFKB and Caspase-activation (2). Simultaneously, the induction of TNFAIP3 and resulting inhibition of NFKB and Caspase-activation (4) enable the intact cell to counteract to and balance these stress stimuli up to a certain point. B) Impaired signaling axis (1) due to the basal overexpression of TNFAIP3 (2) leads to a basal lack in inducibility and transmission of death signals resulting in therapeutic resistance (3).

4 5 6 7 8 9 10 11 12

13 14 15 16 17

18 19 20 21

1 5. References:

1. Fassnacht, M., M. Kroiss, and B. Allolio, Update in adrenocortical carcinoma. J Clin Endocrinol Metab, 2013. 98(12): p. 4551-64.

2. Modlin, I.M., et al., Gastroenteropancreatic neuroendocrine tumours. Lancet Oncol, 2008. 9(1): p. 61-72.

3. Cella, C.A., et al., Dual inhibition of mTOR pathway and VEGF signalling in neuroendocrine neoplasms: From bench to bedside. Cancer Treat Rev, 2015. 41(9): p. 754-60.

4. Fassnacht, M., et al., Linsitinib (OSI-906) versus placebo for patients with locally advanced or metastatic adrenocortical carcinoma: a double-blind, randomised, phase 3 study. Lancet Oncol, 2015. 16(4): p. 426-35.

5. Else, T., et al., Adrenocortical carcinoma. Endocr Rev, 2014. 35(2): p. 282-326.

6. McKeage, M.J., The potential of DMXAA (ASA404) in combination with docetaxel in advanced prostate cancer. Expert Opin Investig Drugs, 2008. 17(1): p. 23-9.

7. Tozer, G.M., C. Kanthou, and B.C. Baguley, Disrupting tumour blood vessels. Nat Rev Cancer, 2005. 5(6): p. 423-35.

8. Rehman, F. and G. Rustin, ASA404: update on drug development. Expert Opin Investig Drugs, 2008. 17(10): p. 1547-51.

9. Baguley, B.C., Antivascular therapy of cancer: DMXAA. Lancet Oncol, 2003. 4(3): p. 141-8.

10. Lavrik, I., A. Golks, and P.H. Krammer, Death receptor signaling. J Cell Sci, 2005. 118(Pt 2): p. 265-7.

11. Boone, D.L., et al., The ubiquitin-modifying enzyme A20 is required for termination of Toll-like receptor responses. Nat Immunol, 2004. 5(10): p. 1052-60.

12. Ma, A. and B.A. Malynn, A20: linking a complex regulator of ubiquitylation to immunity and human disease. Nat Rev Immunol, 2012. 12(11): p. 774-85.

13. Vendrell, J.A., et al., A20/TNFAIP3, a new estrogen-regulated gene that confers tamoxifen resistance in breast cancer cells. Oncogene, 2007. 26(32): p. 4656-67.

14. Bredel, M., et al., Tumor necrosis factor-alpha-induced protein 3 as a putative regulator of nuclear factor-kappaB-mediated resistance to 06-alkylating agents in human glioblastomas. J Clin Oncol, 2006. 24(2): p. 274-87.

15. Hantel, C., et al., Anti insulin-like growth factor I receptor immunoliposomes: a single formulation combining two anticancer treatments with enhanced therapeutic efficiency. J Clin Endocrinol Metab, 2010. 95(2): p. 943-52.

16. Hantel, C., et al., Liposomal doxorubicin-based treatment in a preclinical model of adrenocortical carcinoma. J Endocrinol, 2012. 213(2): p. 155-61.

17. Evers, B.M., et al., The human carcinoid cell line, BON. A model system for the study of carcinoid tumors. Ann N Y Acad Sci, 1994. 733: p. 393-406.

18. Evers, B.M., et al., Establishment and characterization of a human carcinoid in nude mice and effect of various agents on tumor growth. Gastroenterology, 1991. 101(2): p. 303-11.

19. Irizarry, R.A., et al., Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics, 2003. 4(2): p. 249-64.

20. Lichtenauer, U.D., et al., Pre-B-cell transcription factor 1 and steroidogenic factor 1 synergistically regulate adrenocortical growth and steroidogenesis. Endocrinology, 2007. 148(2): p. 693-704.

21. Assie, G., et al., Integrated genomic characterization of adrenocortical carcinoma. Nat Genet, 2014. 46(6): p. 607-12.

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

22. Jassar, A.S., et al., Activation of tumor-associated macrophages by the vascular disrupting agent 5,6-dimethylxanthenone-4-acetic acid induces an effective CD8+ T- cell-mediated antitumor immune response in murine models of lung cancer and mesothelioma. Cancer Res, 2005. 65(24): p. 11752-61.

23. McKeage, M.J., et al., Phase II study of ASA404 (vadimezan, 5,6-dimethylxanthenone- 4-acetic acid/DMXAA) 1800mg/m(2) combined with carboplatin and paclitaxel in previously untreated advanced non-small cell lung cancer. Lung Cancer, 2009. 65(2): p. 192-7.

24. Fruh, M., et al., Carboplatin and paclitaxel plus ASA404 as first-line chemotherapy for extensive-stage small-cell lung cancer: a multicenter single arm phase II trial (SAKK 15/08). Clin Lung Cancer, 2013. 14(1): p. 34-9.

25. Conlon, J., et al., Mouse, but not human STING, binds and signals in response to the vascular disrupting agent 5,6-dimethylxanthenone-4-acetic acid. J Immunol, 2013. 190(10): p. 5216-25.

26. Dudek, A.M., et al., Inducers of immunogenic cancer cell death. Cytokine Growth Factor Rev, 2013. 24(4): p. 319-33.

27. Garay, R.P., et al., Cancer relapse under chemotherapy: why TLR2/4 receptor agonists can help. Eur J Pharmacol, 2007. 563(1-3): p. 1-17.

28. Ma, Y., et al., Tumor necrosis factor is dispensable for the success of immunogenic anticancer chemotherapy. Oncoimmunology, 2013. 2(6): p. e24786.

29. Kang, Y.J., et al., Cell surface 4-1BBL mediates sequential signaling pathways ‘downstream’ of TLR and is required for sustained TNF production in macrophages. Nat Immunol, 2007. 8(6): p. 601-9.

30. Kawagoe, T., et al., Sequential control of Toll-like receptor-dependent responses by IRAK1 and IRAK2. Nat Immunol, 2008. 9(6): p. 684-91.

31. Meylan, E. and J. Tschopp, IRAK2 takes its place in TLR signaling. Nat Immunol, 2008. 9(6): p. 581-2.

32. Couffinhal, T., et al., Tumor necrosis factor-alpha stimulates ICAM-1 expression in human vascular smooth muscle cells. Arterioscler Thromb, 1993. 13(3): p. 407-14.

33.

Fang, W.B., et al., CCL2/CCR2 chemokine signaling coordinates survival and motility of breast cancer cells through Smad3 protein- and p42/44 mitogen-activated protein kinase (MAPK)-dependent mechanisms. J Biol Chem, 2012. 287(43): p. 36593-608.

34. Li, M., et al., A role for CCL2 in both tumor progression and immunosurveillance. Oncoimmunology, 2013. 2(7): p. e25474.

35. Roca, H., Z.S. Varsos, and K.J. Pienta, CCL2 is a negative regulator of AMP- activated protein kinase to sustain mTOR complex-1 activation, survivin expression, and cell survival in human prostate cancer PC3 cells. Neoplasia, 2009. 11(12): p. 1309-17.

36. de Almagro, M.C. and D. Vucic, The inhibitor of apoptosis (IAP) proteins are critical regulators of signaling pathways and targets for anti-cancer therapy. Exp Oncol, 2012. 34(3): p. 200-11.

37. Beyaert, R., K. Heyninck, and S. Van Huffel, A20 and A20-binding proteins as cellular inhibitors of nuclear factor-kappa B-dependent gene expression and apoptosis. Biochem Pharmacol, 2000. 60(8): p. 1143-51.

38. Opipari, A.W., Jr., et al., The A20 zinc finger protein protects cells from tumor necrosis factor cytotoxicity. J Biol Chem, 1992. 267(18): p. 12424-7.

39. Catrysse, L., et al., A20 in inflammation and autoimmunity. Trends Immunol, 2013.

40. Coornaert, B., I. Carpentier, and R. Beyaert, A20: central gatekeeper in inflammation and immunity. J Biol Chem, 2009. 284(13): p. 8217-21.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

5 6

7 8

41. Schaer, D.A., D. Hirschhorn-Cymerman, and J.D. Wolchok, Targeting tumor-necrosis factor receptor pathways for tumor immunotherapy. J Immunother Cancer, 2014. 2: p. 7.

42. Verstrepen, L., et al., ABINs: A20 binding inhibitors of NF-kappa B and apoptosis signaling. Biochem Pharmacol, 2009. 78(2): p. 105-14.

43. Jacobs, M.D. and S.C. Harrison, Structure of an IkappaBalpha/NF-kappaB complex. Cell, 1998. 95(6): p. 749-58.

ACCEPTED MANUS CRIPT

ACCEPTED MANUSCRIPT

FC = 11.1

150

CXCL8 Expression [Relative Quantity]

CXCL1 Expression [Relative Quantity]

FC = 8.5

100

FC = 4.4

n. a.

BON ø

BONTNFa.

NCI ø

NCI TNF&

BON ø

BON TNF&

NCI ø

NCI TNF&

0.75

0.0150

FC = 38.1

ICAM1 Expression [Relative Quantity]

FC = 4.3

CCL2 Expression [Relative Quantity]

0.0125

0.50

0.0100

0.0075

0.001

0.25

FC = 1.9

FC = 11.7

0.00

0.000

BON ø

BONTNFa.

NCI ø

NCI TNFa.

BON ø

BONTNFo.

NCI ø

NCI TNFa.

FC = 12.7

TNFRSF9 Expression [Relative Quantity]

0.075

FC = 2

BIRC3 Expression [Relative Quantity]

0.050

FC =24.3

FC = 5.8

0.025

0.000

BON ø

BON TNF&.

NCI ø

NCI TNF&

BON ø

BONTNF&

NCI ø

NCI TNF&

FC = 15.6

NFkBIA-expression [Relative Quantity]

FC =2.2

TNIP1-expression [Relative Quantity]

FC =3.3

FC = 1.1

BON ø

BON TNFa.

NCI ø

NCI TNF&

BON @ BON TNFo. NCI ø NCI TNF&

Supplemental Figure 1: 2

3 Real Time PCR analysis of chemokine (C-X-C motif) ligand 8 (CXCL8, A), chemokine (C-X-

4 C motif) ligand 1 (CXCL1, B), intercellular adhesion molecule 1 (ICAM-1, C), chemokine (C-

factor receptor superfamily member 9 (TNFRSF9, F), nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor alpha (NFKBIA, G) and TNFAIP3 interacting protein 1(TNIP1, H) expression in untreated and TNF a treated BON and NCI-H295R (NCI) cells. Stars denote significance between untreated vs. treated cells. Pound signs represent significant differences in basal expression levels between both tumor models (BON untreated vs. NCI untreated).

C motif) ligand 2 (CCL2, D), baculoviral IAP repeat containing 3 (BIRC3, E), tumor necrosis 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

ACCEPTED MANUS CRI(BON UL. TOP

18 19 20 21 22 23 24 25 26

ENTREZ GENE ID			BON	NCI
Pathways in cancer
4792	nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha baculoviral IAP repeat-containing	NFKBIA	9,9	2,6
330	3	BIRC3	9	2,9
330	interleukin
3576	8	IL8	6,7	3
4791	nuclear factor of kappa light polypeptide gene enhancer in B-cells 2 (p49/p100)	NFKB2	5,1	4,
3918	laminin, gamma 2	LAMC2	4,3
3725	jun oncogene	JUN	3,5
7477	wingless-type MMTV integration site family, member 7B signal transducer and activator of	WNT7B	2,9
7477		WNT7B
6776	transcription 5A	STAT5A	2,8
11191	phosphatase and tensin homolog; phosphatase and tensin homolog pseudogene 1	PTENP1		2,4
3685	integrin, alpha V (vitronectin receptor, alpha polypeptide, antigen CD51)	ITGAV	2,3
7187	TNF receptor-associated factor 3	TRAF3	2	2,3
7480	wingless-type MMTV integration site family, member 10B	WNT10B	2,2
652	bone morphogenetic protein 4	BMP4	2,2
MAPK signaling	pathway
	tumor necrosis factor (TNF superfamily,
7124	member 2)	TNF	7,8
	v-rel reticuloendotheliosis viral oncogene
5971	homolog B	RELB	5,4	5
4791	nuclear factor of kappa light polypeptide gene enhancer in B-cells 2 (p49/p100)	NFKB2	5,1	4,1
3725	jun oncogene	JUN	3,5
3304	heat shock 70kDa protein 1B	HSPA1B	2,7
6237	related RAS viral (r-ras) oncogene homolog	RRAS	2,6
11221	dual specificity phosphatase 10	DUSP10	2,3
3303	heat shock 70kDa protein 1A	HSPA1A	2,3
	mitogen-activated protein kinase kinase
1326	kinase 8	MAP3K8	2	2,2
3306	heat shock 70kDa protein 2	HSPA2	2
Chemokine signaling pathway
6347	chemokine (C-C motif) ligand 2	CCL2	3,7	10,2
4792	nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha	NFKBIA	9,9	2,6
2919	chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha)	CXCL1	7,1
	interleukin
3576	8	IL8	6,7	3
	chemokine (C-X-C motif) ligand
3627	10	CXCL10	2,5	3,8
2920	chemokine (C-X-C motif) ligand 2 chemokine (C-X3-C motif) ligand	CXCL2	3,4
6376	1	CX3CL1	2,6
	chemokine (C-X-C motif) ligand
6373	11	CXCL11	2,5
	chemokine (C-X-C motif) ligand
9547	14	CXCL14	2,2
Toll-like receptor signaling pathway
4792	nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha	NFKBIA	9,9	2,6
7124	tumor necrosis factor (TNF superfamily, member 2) interleukin	TNF	7,8
3576	8	IL8	6,7	3
9641	inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase epsilon chemokine (C-X-C motif) ligand	IKBKE	5,4
3627	10	IGHE	2,5	3,8
3725	jun oncogene	JUN	3,5
3725	chemokine (C-X-C motif) ligand
6373	11	CXCL11	2,5
7187	TNF receptor-associated factor 3	TRAF3	2	2,3