Angiogenesis DOI 10.1007/s10456-012-9311-z

ORIGINAL PAPER

The experimental renal cell carcinoma model in the chick embryo Patricia Fergelot • Jean-Christophe Bernhard • Fabienne Soulet • Witold W. Kilarski • Ce´line Le´on • Nathalie Courtois • Colette Deminie`re John M. J. Herbert • Philipp Antczak • Francesco Falciani • Nathalie Rioux-Leclercq • Jean-Jacques Patard • Jean-Marie Ferrie`re • Alain Ravaud • Martin Hagedorn • Andreas Bikfalvi

•

Received: 4 June 2012 / Accepted: 29 September 2012 Ó Springer Science+Business Media Dordrecht 2012

Abstract The clear cell subtype of renal carcinoma (CCRCC) is highly vascularized and despite a slow progression rate, it is potentially a highly aggressive tumor. Although a doubling of median progression-free survival in CCRCC patients treated by targeted therapies has been observed, the fact that tumors escape after anti-VEGF treatment suggests alternative pathways. The chick chorioallantoic membrane (CAM) is a well-established model, which allows in vivo studies of tumor angiogenesis and the testing of anti-angiogenic molecules. However, only a few data exist on CCRCC grafted onto CAM. We aimed to validate herein the CAM as a suitable model for studying the development of CCRCC and the interactions with the surrounding stroma. Our study uses both CCRCC cell lines and fresh tumor samples after surgical resection. We Electronic supplementary material The online version of this article (doi:10.1007/s10456-012-9311-z) contains supplementary material, which is available to authorized users. P. Fergelot  J.-C. Bernhard  F. Soulet  W. W. Kilarski  C. Le´on  N. Courtois  M. Hagedorn  A. Bikfalvi INSERM U1029, Talence, France P. Fergelot  J.-C. Bernhard  F. Soulet  W. W. Kilarski  C. Le´on  N. Courtois  M. Hagedorn  A. Bikfalvi (&) Universite´ Bordeaux I, Avenue des Faculte´s, 33 405 Talence, France e-mail: [email protected]

demonstrate that in both cases CCRCC can be grafted onto the CAM, to survive and to induce an angiogenic process. We further provide insights into the transcriptional regulation of the model by performing a differential analysis of tumor-derived and stroma-derived transcripts. Keywords Renal cell carcinoma  Angiogenesis  Chick embryo  NAMPT  TGM4  MSLN  ANO1  CD3E

Introduction Renal cell carcinoma (RCC) accounts for approximately 3 % of human tumors. Clear-cell renal cell carcinoma (CCRCC) is the most prominent RCC subtype (75 %). The average age at diagnosis is 60, which indicates that CCRCC are slow progressing tumors. However, 30 % of patients have metastases

C. Deminie`re Service de Pathologie CHU de Bordeaux, Bordeaux, France J. M. J. Herbert  P. Antczak  F. Falciani Institute of Biomedical Research, University of Birmingham, Birmingham B15 2TT, UK N. Rioux-Leclercq  J.-J. Patard CNRS UMR 6061, Institut de Ge´ne´tique et De´veloppement, Universite´ Rennes 1, Rennes, France

P. Fergelot Laboratoire Maladies Rares: Ge´ne´tique et me´tabolisme, Universite´ Bordeaux Segalen, Bordeaux, France

A. Ravaud Service d’Oncologie Me´dicale, CHU de Bordeaux, Bordeaux, France

J.-C. Bernhard  J.-M. Ferrie`re Service d’Urologie, CHU de Bordeaux, Bordeaux, France

A. Ravaud  A. Bikfalvi CIC INSERM 0005, Universite´ Bordeaux, Bordeaux, France

123

Angiogenesis

at the time of diagnosis and an additional 30–40 % will develop metastasis after radical surgery. Thus CCRCC is potentially a highly aggressive tumor [1]. These tumors are highly vascularized and are characterized by the inactivation of the VHL gene, which occurs in about 80 % of CCRCC as a consequence of 3p or interstitial deletion and mutation or, less frequently, of promoter methylation [2–4]. The pVHL gene product is an E3 ubiquitin-ligase involved in the proteolytic processing of the Hypoxia inducible factors (HIF) a subunit. Accumulation of HIF 1 and 2 leads to a constitutive expression of critical genes involved in tumor angiogenesis, cell proliferation and migration (reviewed in [5]). VEGF is overexpressed in CCRCC. High VEGF expression is associated with high- grade tumors and poor survival [4, 6, 7]. Although the importance of tumor angiogenesis in CCRCC progression has been validated by the doubling of median progression-free survival in patients treated by anti VEGF or anti VEGFR and PDGFR targeted molecules [8, 9], the fact that tumors escape after anti-VEGF therapy suggests alternative pathways and a more complex regulation of endothelial cell survival. In fact, VHL loss of function is an early genetic event in tumor development. However, the proangiogenic phenotype of CCRCC is associated with a long period of latency. Moreover, an inverse correlation between VHL inactivation and VEGF expression and plasma levels has been detected in patients [4]; this suggests that VHL-independent mechanisms such as Stat3 [10] are involved in VEGF up regulation in advanced CCRCC. Hypoxia could also mediate progression through VHL-independent, HIF driven mechanisms [5]. Among them, the mTOR/akt pathway has been evaluated and a sequential approach using different classes of molecules are proposed [11], although neither VEGF nor mTOR targeted therapies are curative in CCRCC [12]. Apart from the pro-angiogenic mechanisms induced by tumor cells, the tumor microenvironment also contributes to angiogenesis such as immunocompetent cells that produce VEGF and other angiogenic cytokines [13]. Tumor-associated polynuclear neutrophils (TAN) are accumulated in CCRCC and may indicate tumor progression at an early stage [14]. However, despite increasing knowledge on the pathobiology of CCRCC, the relationship between angiogenesis and tumor progression, in the specific context of VHL inactivation is still poorly understood. It is, thus, important to develop models that better visualize the interaction of CCRCC with the surrounding stroma. The CAM is a well-established model, which allows in vivo studies of tumor angiogenesis and testing of antiangiogenic molecules [15–17]. Early studies have shown that embryonic rat kidney can be successfully implanted onto CAM [18]. However, to date, only a few data exist on RCC grafted onto CAM [19]. Furthermore, pharmacological treatment of renal cancer cell lines graft onto CAM [20] did not provide insights into CCRCC-induced angiogenesis.

123

We aimed to validate the CAM as a suitable model for studying the development of CCRCC and the interactions with the surrounding stroma. Our study uses both CCRCC cell lines and fresh tumor samples after surgical resection. We demonstrate herein that in both cases CCRCC can be grafted onto the CAM, to survive and to induce the angiogenic process. We further provide some insights onto the transcriptional regulation that occurs in the tumor stroma at early stages of tumor development.

Materials and methods Human tumor samples and chicken embryos Renal tumor specimens were processed for pathology examination and banking immediately after nephrectomy at the Urologic Department of the university hospital of Bordeaux. Informed consent was obtained for all patients. Macroscopic examination of tumors was done by one Urologic pathologist (CD). Selected tumor fragments were immediately placed in culture medium (DMEM) without serum and stored on ice until deposition onto the CAM (maximum 2 h after surgery). Sample quality was further verified by histology using formalin-fixed paraffin sections stained with hematoxylin and eosin-safran, followed by light microscopy. Slides were reviewed by the pathologist (CD) and 6 conventional clear cell carcinomas were considered in this study. Histological parameters included tumor size (cm), tumor necrosis (present or absent) and nuclear Fuhrman grade [21]. Tumor stage was defined according to the international Union Against Cancer 2004 TNM classification [22]. Fertilized chicken eggs (Gallus gallus; E.A.R.L. Morizeau, Dangers, France) were incubated and handled as previously described [16]. Small tumor fragments mixed with fibrinogen (5 mg/ml; Sigma) were deposited on intact CAM. At E14 functional vessels were visualized by injection of India ink and treatment with benzyl benzoate benzyl acid (BBBA, vol/vol) according to [23]. Cell cultures The human RCC4, Caki-2 and 786-O cell lines were obtained from the European Collection of Cell Cultures (EATCC). They were cultured in Dulbecco’s modified Eagle’s medium (DMEM) 1 g/L glucose (Invitrogen) supplemented with 10 % fetal bovine serum, L-glutamine and antibiotics (penicillin/streptomycin). Cultures were incubated at 37 °C in 5 % CO2. On embryonic day 8 (E8), 3–5 million of 786-O, Caki-2 or RCC4 renal cancer cells in serum free medium (50 ll) were deposited on intact CAM. Tumor growth was

Angiogenesis

monitored on days 2, 3, 7 and 9. Digital photos were taken under a Nikon SMZ800 stereomicroscope.

Immunohistochemistry CCRCCs bearing CAMs were fixed in vivo with 4 % paraformaldehyde for 1 h at room temperature and cut into two fragments. One fragment was processed in the laboratory of the Department of Pathology, University Hospital of Bordeaux in parallel with non-implanted tumor fragments from the same patients and stained for CD31, CD34 and AE1/AE3 (pan cytokeratins) immunostaining (DAKO). The second half of CCRCCs bearing CAMs were immersed in 30 % sucrose overnight and stored at -80 °C.

786-O bearing CAMs were fixed in vivo with 4 % paraformaldehyde for 30 min at room temperature. Chick blood vessels were stained with biotinylated sambucus nigra (SNA) lectin (Vector laboratories) revealed with Vectastain ABC kit, (Vector laboratories). Confocal microscopy analysis Co-labelling experiments were performed on frozen sections (10 lm) of CCRCCs bearing CAMs using FITCconjugated SNA1 diluted 1:100 (Vector laboratories) and mouse anti-human CD31 antibody (DAKO) diluted 1:20 or a mouse anti-human desmin antibody (DAKO), diluted 1:500 cross-reacting with chick desmin. Alexa Fluor 546 goat anti-mouse antibody, diluted 1:1,000 (Invitrogen) was

Table 1 Characteristics of tumors and graft experiments Case

Primary tumor

Tumor graft

Loco-regional extension*

Nuclear grade

Histology

Egg development

Incubation time (days)

Angiogenesis

1

pT1a

N0

3

CCRCC**

E7

9

?

2

pT1b

N0

3

CCRCC

E8

7

?

3

pT1b

N0

2

CCRCC

E10

6

?

4

pT1b

N0

3

CCRCC

E10

6

?

5

pT3a

N0

3

CCRCC

E9

5

?

6

pT3b

N0

3

CCRCC

E9

9

?

* According to TNM stage [22] ** Clear cell renal cell carcinoma

Fig. 1 Implantation of human CCRCC on the chicken CAM. Numerous vessels are seen by biomicroscopy at the surface of tumor grafts. a surface view, b bottom view, c after Indian ink injection into

CAM vessels to visualize vessel perfusion. Histology of implanted tumors; d–e hematoxylin-eosin (HE) staining; f pancytokeratin staining of epithelial CRCC cells and lining CAM; g CD31 staining

123

Angiogenesis

used as a secondary antibody. Nuclei were visualized with DAPI (Invitrogen). Imaging was performed on a confocal microscope (SPS Leica) with a X63 objective. Transcriptome analysis

up regulated genes to the DAVID database using default analysis options [27, 28]. The most pertinent biologic enriched theme was selected for genes up regulated at day 3 and the two highest enriched themes are displayed for day 7. In silico gene expression studies

Total RNA from 786-O bearing CAM and paired control CAMs from the same eggs, was isolated at day 3 (D3) and day 7 (D7) post-implantation using the RNeasy kit Plus (Qiagen). RNA quantity and quality were assessed by optical density measurement and agarose gel electrophoresis. Three eggs from each condition were used for independent hybridisations. RNA quality assessment (bioanalyser, Agilent) and hybridization procedure were carried out at the Affymetrix transcriptomic platform of CHU St Eloi, Montpellier, France. Each of the labelled cRNAs were hybridized to both Affymetrix human (U133 Plus 2) and chicken GeneChips [24, 25]. Data were normalized using robust multiaverage normalization (RMA). SAM analysis was performed in parallel on human and chick datasets. A 13 % FDR was chosen for the chicken samples to match the human results. Hierarchical clustering was performed using Pearson correlation (TIGR MeV 4.0), [26].

Genes overexpressed in the course of the model were analyzed by a novel bioinformatics algorithm which assigns preferential endothelial expression of a given gene compared to other cell types [30, 31]. Only upregulated genes with human orthologs were included in the analysis.

Functional annotation clustering (Gene ontology)

Real-time qPCR

To gain insight into biological themes associated with the growth of tumors in the course of our model, we submitted

Reverse transcription was performed from 1 lg total RNA using the Quantitect Reverse Transcription (Qiagen).

a

b

40µ

300µ

c

d

The Oncomine database was queried for probesets detecting chick stromal genes overexpressed on day 7 after implantation [29]. The following Oncomine search parameters were applied: Cancer versus normal and genes overexpressed in renal cell carcinoma, without modifying Oncomine default search restrictions. Endothelial enrichment of experimental RCC stromal genes

50µ

*

80µ

300µ

Fig. 2 Immunostaining of non-implanted tumors (a, c) and implanted tumors (b, d) of two different types (a, b, pT3a; c, d, pT1b). Higher magnification demonstrates nucleated erythrocytes in CD34-positive

123

100µ

vessels (b, arrow). At the edge of the tumor CD34-low vessels are seen (d, arrow). Chicken CAM vessels are CD34 negative (*)

Angiogenesis

Specific primer pairs were designed with PrimerBlast (NCBI: http://www.ncbi.nlm.nih.gov/tools/primer-blast/) for human amplification hCXCL8 (hIL8, NM_000584) forward primer: 50 TGTGAAGGTGCAGTTTTGCCAAGG and reverse primer: 50 GTTGGCGCAGTGTGGTCCACTC, hNAMPT (NM_005746.2) forward primer: 50 AGGGTT ACAAGTTGCTGCCACCTT and reverse primer: 50 AC

a

CD31

CTCCACCAGAACCGAAGGCAA. Primer specificity was assessed using species-specific in silico PCR (UCSC: http:// genome.ucsc.edu/cgi-bin/hgPcr) and Megablast (NCBI: http://www.ncbi.nlm.nih.gov/blast/). Human HPRT and bactin where chosen as reference genes using geNormTM (http://medgen.ugent.be/genorm/); bactin forward primer: 50 CGTACCACTGGCATC GTGAT and reverse primer:

b

DESMIN

SNA-1

SNA-1

**

DAPI

DAPI

* MERGED

MERGED

c

DESMIN

SNA-1

DAPI

MERGED

Fig. 3 Confocal analysis of microvessels in implanted CCRCC (a–c). Triple labeling with CD31 (a) or desmin (b and c) revealed with Alexa Fluor456, FITC conjugated SNA1 (a–c) and DAPI (a–c). CD31-positive cells along with a SNA1-positive cell in a mosaic vessel (a, arrow). Desmin and SNA1 co-labelling of chicken vessels

(b, arrow with *). SNA1 negative structure covered with desmin low cells or desmin positive cells (b, arrow with **) corresponding to human vessels. c desmin-positive vessels and neutrophil-like inflammatory cells, revealed by a strong SNA1 labelling, at branching points

123

Angiogenesis

50 GTGTTGGCGTACAGGTCTTT, HPRT forward primer: 50 CCAGACAAGTTTGTTGTAGG and reverse primer: 50 TCCAAACTCAACTTGAACTC. Real-time PCR was performed on a Step One plus (Applied biosystems) with the Mesa Blue qPCR Master Mix Plus (Eurogentec). Expression values were determined using DCt and the geometric mean of reference genes values as normalization factor [32]. The Mann–Whitney test (XLSTAT 2011, Addinsoft) was used to compare at day3 and day7 post-implantation human CXCL8 or PBEF1/NAMPT expression.

Results CCRCC implantation on CAM Clear cell renal cell carcinoma (CCRCC) arises from proximal tubular cells and is characterized by the overproduction of angiogenic factors such as VEGF. In order to evaluate the relevance of the chick CAM as a model for

CCRCC angiogenesis and invasion, we implanted 6 tumor samples from CCRCC patients onto the CAM. Chick embryos were cultured for 8–10 days and tumor fragments (2 mm 9 2 mm 9 3 mm) from macroscopically homogeneous parts of the tumor specimen, without visible necrosis, were implanted on the CAM. Specimen description is summarized in Table 1. Tumor take and development was followed by biomicroscopy. Stroma and tumor cells were maintained as long as 9 days after grafting on the CAM. In all cases, chick vasculature penetrated tumor graft and new vessels were visible at the surface of tumor grafts 4 days after implantation (Fig. 1a, b). Indian ink injection into a CAM vessel confirmed that the chick vasculature penetrated into the tumor fragments (Fig. 1c). Standard histology of implanted tumors clearly showed cell nests in contact with a fibrous axis, containing a moderate to abundant vasculature and some inflammatory cells (Fig. 1d, e). Tumor grafts were covered by a cylindric epithelium that corresponds to the CAM ectodermic membrane (Fig. 1f). A necrotic zone was observed in 3

Fig. 4 Implantation and growth of the CCRCC cell line 786-0 on CAM. a day 3 after implantation; b day 7 after implantation; c day 9 after implantation (upper view); e day 9 after implantation (bottom view)

123

Angiogenesis

cases and haemorrhage in 1 case. Pan cytokeratins staining of tumor cells confirmed their epithelial origin (Fig. 1f). Immunostaining for CD31 revealed nucleated erythrocytes from the chick in CD31? human vessels (Fig. 1g). This indicates that CD31? vessels are still functional and are connected to the CAM vasculature. Immunostaining for CD34 of non-implanted or CAM-implanted fragments from five CCRCC (cases 2–6) was performed. The tumor stroma was CD34? (Fig. 2a–d). An abundant vascular stroma was present inside tumor nests similar to nonimplanted tumors. In fragments from advanced-stage tumors (pT3), vessels were composed of relatively large and abundant capillaries (Fig. 2a, b). In small, non-invasive tumors (pT1), a dense network of thin capillaries was seen (Fig. 2c, d). CAM vessels were CD34- (Fig. 2d). We observed CD34 immunostaining at the edge of the graft, close to CAM vessels, that was much weaker than inside the tumor nests, (Fig. 2d). This may correspond to mosaic vessels.

a

Phenotype of tumor vessels We therefore undertook a more complete characterization of the vessel phenotype by confocal microscopy. Intratumor vessels were stained with anti-desmin antibodies or with anti-CD31 to detect human vessels and co-labeled with SNA1 isolectin to detect chick vessels (Fig. 3a–c). CD31 antibodies only recognized vessels of human origin and SNA isolectin vessels of chicken origin. Anti-desmin antibodies labeled vessels of both species. A subset of vessels were SNA1-positive and thus of avian origin (Fig. 3a, b). These vessels were mostly hybrid (human and chicken) (Fig. 3a) as shown by the presence of SNA1 positive endothelial cells in the close vicinity of CD31 positive cells. In tumor fragments, desmin-positive cells were covering SNA1-positive as well as SNA1-negative vessels. This indicates that tumor and chicken capillaries were almost uniformly desmin-positive (Fig. 3b) and raises the possibility that coverage of human vessels by

b

c

100µ

100µ

d

50µ

Fig. 5 Histoloy of tumors derived from the implantation of 786-O cells. a biomicroscopy; b–c HE staining: in b the ingrowth of tumor in the CAM stroma is clearly seen; c another view of the tumor with numerous vessels (higher magnification is shown below). d SNA1

50µ

peroxydase immunostaining of tumor, vessels are clearly positive. In addition, single cells are also found to be positive localized mainly at the periphery

123

Angiogenesis

invasion inside the CAM mesoderm was a constant feature. However, the allantoic cavity was not invaded. The tumor stroma was composed of neovessels and neutrophile-like inflammatory cells comprising chicken leucocyte-like heterophiles (Fig. 5d). We next set out to analyze modifications of the transcriptome by comparing day 3 (E11) and 7 (E15) postimplantation using human and chicken microarrays. We previously demonstrated that no significant cross-hybridization (5 %) occurs with human RNA on chicken microarrays [24]. In order to identify differentially expressed genes that could be involved in the interactions between tumor cells and the stroma, we focused on genes that vary during the time-course of angiogenesis and 786-O growth (day 3 vs day 7). We performed chicken microarray and human microarray using the same samples. In chicken microarray, we found 512 differentially expressed genes. Among them, 398 genes were up-regulated at day 3 (Online Resource 1) and 114 were up-regulated at day 7 post implantation (Online Resource 2). Among the most differentially expressed chicken genes, we identified a cluster of up-regulated genes at day 7 including TGM4, CTSG, MSLN, EPB41L4B (EHM2) TMEM16A (ANO1) and ADIPOQ (Fig. 6a).

pericyte-like cells are derived from the chick. We also found a number of SNA1-positive macrophage/neutrophile-like cells at branching points. Furthermore, some of these were localized at sites where intussusception might occur. Our data indicate that besides host vessel remodeling, tumor vessels are maintained, connected to host vessels and perfused. We next examined the crosstalk between tumor cells and the CAM stroma. To this aim, we used the VHL-inactivated CCRCC cell lines, RCC4 and 786-O and also VHL wild-type CAKI2 cells. RCC4 cells never induced an angiogenic response and only a few capillaries were observed with CAKI 2 cells. No significant tumor growth was observed 10 days after implantation onto the CAM. Only 786-O cells grew on the CAM and induced the development of a dense capillary network (Fig. 4a–d). After 3 days, 30 % of eggs implanted with 786-O cells were developing with a concomitant vessel ingrowth (Fig. 4a). At day 7, a small angiogenic tumor was visible in 80 % of the samples analyzed (Fig. 4b). After 9 days, tumors reached 3 mm in diameter and were highly vascularized (Fig. 4c, d). Standard hematoxylin staining showed invading nests of tumor cells exhibiting a CCRCC phenotype inside the mesoderm (Fig. 5a–c). Tumor cell

a

b

D3 D7 DNA metabolic process M phase of mitotic cell cycle mitosis positive regulation of leukocyte …

BP

positive regulation of lymphocyte … antigen processing and presentation chromosome, centromeric region intracellular non-membrane-…

CC

MHC protein complex extracellular region 0

Fig. 6 Human 786-0 cells and chicken CAM differentially expressed genes: a hierarchical clustering of the most differentially expressed chicken genes (fold change [ 5) between D3 and D7 post-implantation. b functional annotation of chicken genes. Retrieved words for

123

5

10

15

20

upregulated genes at D3 and D7 respectively, according to Gene Ontology pathway (BP and cellular compartment (CC)). For each word, bar represents the percentage of annotated genes

Angiogenesis Table 2 Human 786-O genes up-regulated at day 3 versus day 7 post-implantation on CAM Probe set

Gene symbol

Fold change

Gene name

209774_x_at

CXCL2

33,4

Chemokine (C-X-C motif) ligand 2

1554997_a_at

PTGS2

32

Prostaglandin-endoperoxide synthase 2 (prostaglandin G/H synthase and cyclooxygenase)

204748_at

PTGS2

26,1

207850_at

CXCL3

17,9

Prostaglandin-endoperoxide synthase 2 (prostaglandin G/H synthase and cyclooxygenase) Chemokine (C-X-C motif) ligand 3

205207_at

IL6

17,2

Interleukin 6 (interferon, beta 2)

204470_at

CXCL1

16,2

Chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha)

210229_s_at

CSF2

14,7

Colony stimulating factor 2 (granulocyte–macrophage)

202859_x_at

IL8

11,2

Interleukin 8

211506_s_at

IL8

10,6

Interleukin 8

205680_at

MMP10

9,2

Matrix metallopeptidase 10 (stromelysin 2)

39402_at

IL1B

7,7

Interleukin 1, beta

202672_s_at

ATF3

6,9

Activating transcription factor 3

217739_s_at

PBEF

4,2

Pre-B cell colony-enhancing factor

215223_s_at

SOD2

3,7

Superoxide dismutase 2, mitochondrial

213524_s_at

G0S2

3,7

G0/G1switch 2

218810_at

ZC3H12A

3,2

Zinc finger CCCH-type containing 12A

206239_s_at

SPINK1

2,8

Serine peptidase inhibitor, Kazal type 1

Functional annotation clustering using DAVID associated 157 out of 197 genes upregulated more than two-fold with known biological functions. The most prominent theme at the early phase of tumor implantation at day three was cell cycle and associated genes (7.33-fold enrichment, Online Resource 3a). This concerned 17 genes (10.19 %), including two kinesin family member genes, KIF11 and KIF4A, which have recently been found to be induced by VEGF in vivo on the CAM (Exertier et al. MS in preparation). At day 7, in 57 out of 84 genes up-regulated more than twofold, the main themes were immune response (2.12-fold enrichment, 9 genes, 15.8 % of annotated genes; Online Resource 3b) and extracellular matrix-associated genes (2.02-fold enrichment, 9 genes, 15.8 % of annotated genes; Online Resource 3c). Among the 398 probe sets detected as up regulated genes at day 3, 336 had human orthologs. For these 336 genes, a preferential endothelial expression was predicted for 117 candidates (35 %; Online Resource 4a). At day 7, 72 of 114 genes had human orthologs and for 27 genes endothelial enrichment was predicted (37.5 %, Online Resource 4b). Biological functions were associated exclusively with early (day 3) or late condition (day 7) (Fig. 6b), suggesting that different cell types or cells at different differentiation steps are predominant in these two conditions. Relative overexpression of cell cycle genes at day 3 reflects the abundance of proliferating vessels in the early tumor stroma whereas infiltration of chick MHC-expressing cells is predominant at day 7 post-implantation.

Six genes were significantly overexpressed in RCC compared to normal kidney tissue in our experimental RCC model (Fig. 7). TGM4, MSLN, CD3E are novel RCC markers, TMEM16A (ANO1) has been shown to be associated with head and neck squamous cell carcinoma [33] whereas CD74 and ANGPTL2 have been already suspected to be associated with RCC progression [34, 35]. Human microarray mostly revealed upregulated genes at day 3. Functional annotation clustering using GO terms highlighted cytokine activity, inflammatory response, response to wounding and neutrophil chemotaxis (enrichment score: 4.57). Interestingly, the term angiogenesis had an enrichment score of 1.72. Chemokines and cytokines including CXCL2 (33 fold), CXCL3 (18 fold), CXCL6 (17 fold), and stress-induced gene PTGS2 (26 and 32 fold) were found among the most up-regulated genes (Table 2). CXCL1 and CXCL8, which are involved like CXCL3 in neutrophil chemotaxis and PBEF1 (Pre-B cell colony Enhancing factor)/NAMPT (nicotinamide phosphoribosyltransferase)/Visfatin, which is involved in the positive regulation of cell proliferation, were also upregulated respectively 16, 11 and 4 fold at day 3. The differential expression of human CXCL8 (IL8) and PBEF1/NAMPT was confirmed by real-time PCR analysis (IL8: 6.9 ± 4 at D3 vs 0.13 ± 0.04 at D7, n = 5, p \ 0.05 and PBEF1/ NAMPT: 4.4 ± 1.8 at D3 vs 0.85 ± 0.35 at D7, n = 5, p \ 0.05). Simultaneous up-regulation of transcription factors CSF2 (15 fold) and ATF3 (7 fold) was consistent with their role in activation of cell proliferation.

123

Angiogenesis

TGM4

MSLN

P=2.2-30

Clear Cell Renal Cell Carcinoma (n=23)

Kidney (n=23)

P=1.8-16

Kidney (n=23)

TMEM16A (ANO1)

Clear Cell Renal Cell Carcinoma (n=23)

CD3E

P=5.48-10

Renal Cortex (n=10)

Renal Tissue (n=1)

Hederitary Clear Cell Renal Cell Carcinoma (n=32)

P=1.27-5

Fetal Kidney (n=2)

CD74

Kidney (n=3)

Clear Cell Renal Cell Carcinoma (n=26)

ANGPTL2

P=1.38-16 P=5.51-7

Kidney (n=10)

Clear Cell Renal Cell Carcinoma (n=10)

Fig. 7 Graphical representation of genes over expressed in the course of the model which are also over expressed in RCC patients compared to normal kidney tissue. TGM4, MSLN, TMEM16A (ANO1) and

123

Renal Cortex (n=10)

Renal Tissue Non-hederitary Clear (n=1) Cell Renal Cell Carcinoma (n=27)

CD3E are novel RCC genes, whereas CD74 and ANGPTL2 have already be suspected to be associated with cancer progression

Angiogenesis

Discussion In this article, we described a model for human clear cell renal carcinoma in the chick embryo and studied morphological and molecular characteristics of the tumorstroma interaction. To this end, we used both fresh tumor samples from patients and the VHL-/- CCRCC 786-O cell line. Tumor fragments were engrafted onto the CAM and developed over 9 days. Engraftment leads to vascular remodeling, irrespective of the loco-regional development and Fuhrmann grade of non-grafted tumors. We observed chick vessels from the CAM penetrating into the tumor stroma and hybrid vessels were formed by anastomosis between chick and human capillaries. These were perfused from the chick vasculature since nucleated chick erythrocytes were found inside human CD31-positive vessels. Human vessels were detected using antibodies for human CD31 and CD34; chicken vessels were detected with SNA lectin. Hybrid vessels were visible at the edge of the tumor grafts in the vicinity of CAM vessels. We found that human tumor vessels, in the CAM implanted with fresh human tumor samples, were covered with desminpositive pericyte-like cells as seen for chick vessels in another experimental CAM tumor model [24]. It is generally accepted that coverage of vessels by desmin-positive cells allows vessel stabilisation and survival of the vasculature. In CCRCC, tumor vessels are generally desminnegative despite the fact that pericytes are present [36]. Thus, coverage by desmin-positive pericytes of tumor vessels in our model is most likely derived from the chick. Such hybrid vessels have already been described in chickquail hybrids. In this case, endothelial cells were derived from the host and pericytes from the grafted tissue [37]. Anastomosis to the chick vasculature and survival of preexisting vessels has been already described for engraftment of rat embryonic kidney [18]. This suggests that similar mechanisms may also be involved in the case of tumor implantation. We also implanted the CCRCC cell line 786-O in the CAM, in order to analyze more precisely gene regulations that may occur between tumor cells and the stroma. This analysis could not be performed with fresh human tumor specimens because they also contain, besides tumor cells, a human stroma that would contaminate our analysis. Implantation of a human cell line onto a chick stroma has the advantage that differential gene expression analysis can be performed, because of only small cross-hybridization between human and chick RNA in the transcriptome analysis (5 % cross-hybridization). The 786-O cell line was successfully implanted on the CAM and a strong angiogenic response was observed. This angiogenic response to 786-O cells implantation has been described earlier with use of Matrigel [20]. Here we show that direct contact is

sufficient to induce angiogenesis although tumor growth was relatively slow with macroscopic appearance of new vessels at 3 days after implantation. The variable ability of CCRCC cell lines to grow on the CAM may be explained by different intracellular mechanisms that are responsible for growth promotion and angiogenesis. For example, HIFa isoforms are differentially expressed in CCRCC cell lines. A Recent analysis of HIF expression in CCRCC revealed two groups of VHLinactivated tumors. These included only HIF2a expressing tumors, and tumor that express both, HIF1a and HIF2a [38]. Each group displayed different gene expression profiles when compared to wild-type VHL CCRCC. In tumors expressing only HIF2a, genes that control cell division are mainly active, leading to larger tumors. Moreover, opposite effects of HIF isoforms are observed [39]. This may explain the failure of RCC4- cells, which are expressing both HIF1a and HIF2a, to grow on the CAM. On the other hand, 786-O cells are known to only express HIF2 [40, 41] and are able to grow in the CAM environment. Analysis of the transcriptome of 786-O cells implanted on the CAM revealed a high percentage of cell cycle, inflammation and immune system related genes. In the developing CAM, the sprouting phase of angiogenesis normally culminates at E6 [42]. The fact that cell cycle genes detected by chicken microarray constituted the vast majority of up-regulated genes at day 3 (E11) are indicative of stroma cell proliferation. It is of interest to note that by using selective profiling of stroma and tumor genes, a detailed understanding of the stroma and especially endothelial reaction to tumor cell growth could be established. Up to 37.5 % of stromal genes are suspected to be expressed in endothelial cells, which reflects the high vascularization status of RCC and further underlines the abundance of potential novel therapeutic targets in the tumor vasculature. In particular, a strong proliferative response of the stroma during the early phase of tumor growth could be evidenced, followed by an activation of immune response genes and extracellular matrix-associated genes. This shift in genetic programs in the course of our model is likely to reflect a more intricate interaction of tumor cells with the host after a first growth phase. Several novel genes, which are activated at the later stage of the model, are also over expressed in patient samples. These include TGM4, MSLN, CD3E, TMEM16A (ANO1), CD74 and ANGPTL2. TMEM16A (ANO1) has been shown to be associated with head and neck squamous cell carcinoma [33], whereas CD74 and ANGPTL2 have been already suspected to be associated with RCC progression [34, 35]. TGM4, MSLN, CD3E are novel RCC markers. Genes encoding human cytokines were expressed during tumor expansion and invasion in the CAM. CXCL6 (IL6) and CXCL8 (IL8) genes expression paralleled expression

123

Angiogenesis

of cell-cycle genes and preceded the up-regulation of chicken MHC genes. These two chemokines are known mediators of leucocyte recruitment and induction of angiogenesis [43, 44]. In our model, we found a dense population of leucocyte-like cells surrounding 786-O cell nests. At day 7, this massive infiltrate of inflammatory cells was associated with up-regulation of cathepsin G, a marker of neutrophil activation [45]. Thus by promoting inflammation, 786-O cells favor extravasation of leucocyte-like cells that could in turn amplify the angiogenic process induced by tumor cells [46]. Interestingly, the 786-O gene set overlapped a previously published endothelial signature [47]. The tumour gene expression profile appeared very similar to a cluster of endothelial genes up-regulated in response to VEGF and IL1, containing ATF3, CXCL2, NFKBIZ, PTGS2 and IL8. IL8 production is induced in human microvascular endothelial cells overexpressing HIF2a [48]. In human CCRCC, IL8 is overexpressed by tumor cells and could mediate drug resistance [49]. Besides CXCL8 gene expression at day 3 post-implantation in tumour cells, we also found a high level of PBEF transcripts. PBEF1 is also known as Nicotinamide phosphoribosyltransferase (NAMPT) and the adipocyte-derived cytokine (adipokine) Visfatin. NAMPT catalyzes the ratelimiting step of nicotinamide adenine dinucleotide (NAD?) synthesis and controls cell survival and inflammation [50, 51]. Exogenous PBEF1/NAMPT/Visfatin has proangiogenic activity, regulates endothelial IL6 production through Stat3 in vitro [52] and can induce the mTOR pathway in HUVEC cells [53]. PBEF1 is upregulated at the mRNA level in various tumor types such as prostate cancers [54] and correlated with poor survival in breast cancer [55]. Recently, an adipogenic signature was identified in CCRCC by comparison of low stage tumours and normal kidney, as a model of renal tumorigenesis. The PBEF1 gene was up-regulated in low stage (I and II), localized tumors [56]. How the adipogenic phenotype of CCRCC impacts the development of the tumor vasculature remains to be investigated. It is of note that normal fat tissue induces angiogenesis when grafted on CAM [57]. In our model we show that PBEF1/NAMPT/Visfatin gene is early expressed by tumor cells during CAM invasion, together with IL6 and IL8. In 786-O cells PBEF-1 mRNA, as shown in several cells types [58], could be induced by the c-MYC oncoprotein, a HIF-2a responsive gene in CCRCC [37]. In summary, we described herein an experimental renal cell carcinoma model in the CAM using fresh tumor specimen and a CCRCC cell line and provided a morphological characterization of the model. Furthermore, we performed a transcriptome analysis in this model that differentiates between signals derived from tumor and the stroma. These signals could amplify HIF effects to stimulate vascular remodeling. Our data showed that the CAM

123

model may be useful to decipher tumor-stroma cell interactions in CCRCC, a crucial step for the understanding of resistance to targeted therapies. Acknowledgments This work was supported by grants from the Agence National de la Recherche (ANR), and Ligue National contre le Cancer. PF was a recipient of fellowship for a sabattical from the university Rennes and INSERM (‘‘De´le´gation’’). WK was a recipient of a fellowship from the Lefoulon-Delalande Foundation, JCB was a recipient of a Master Fellowship from the University Bordeaux II. Conflict of interest

The authors declare no conflict of interest.

References 1. Rabinovitch RA, Zelefsky MJ, Gaynor JJ, Fuks Z (1994) Patterns of failure following surgical resection of renal cell carcinoma: implications for adjuvant local and systemic therapy. J Clin Oncol 12:206–212 2. Gnarra JR, Tory K, Weng Y, Schmidt L, Wei MH, Li H, Latif F, Liu S, Chen F, Duh FM et al (1994) Mutations of the VHL tumour suppressor gene in renal carcinoma. Nat Genet 7:85–90 3. Banks RE, Tirukonda P, Taylor C, Hornigold N, Astuti D, Cohen D, Maher ER, Stanley AJ, Harnden P, Joyce A, Knowles M, Selby PJ (2006) Genetic and epigenetic analysis of von HippelLindau (VHL) gene alterations and relationship with clinical variables in sporadic renal cancer. Cancer Res 66:2000–2011 4. Patard JJ, Rioux-Leclercq N, Masson D, Zerrouki S, Jouan F, Collet N, Dubourg C, Lobel B, Denis M, Fergelot P (2009) Absence of VHL gene alteration and high VEGF expression are associated with tumour aggressiveness and poor survival of renalcell carcinoma. Br J Cancer 101:1417–1424 5. Baldewijns MM, van Vlodrop IJ, Vermeulen PB, Soetekouw PM, van Engeland M, de Bruı¨ne AP (2010) VHL and HIF signalling in renal cell carcinogenesis. J Pathol 221:125–138 6. Na X, Wu G, Ryan CK, Schoen SR, di’Santagnese PA, Messing EM (2003) Overproduction of vascular endothelial growth factor related to von Hippel-Lindau tumor suppressor gene mutations and hypoxia-inducible factor-1 alpha expression in renal cell carcinomas. J Urol 170:588–592 7. Jacobsen J, Grankvist K, Rasmuson T, Bergh A, Landberg G, Ljungberg B (2004) Expression of vascular endothelial growth factor protein in human renal cell carcinoma. BJU Int 93:297–302 8. Motzer RJ, Hutson TE, Tomczak P, Michaelson MD, Bukowski RM, Rixe O, Oudard S, Negrier S, Szczylik C, Kim ST, Chen I, Bycott PW, Baum CM, Figlin RA (2007) Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med 356:115–124 9. Figlin RA, Hutson TE, Tomczak P, Michaelson MD, Bukowski RM, Negrier S, Huang X, Kim ST, Chen I, Motzer RJ (2008) Overall survival with sunitinib versus interferon (IFN)-alfa as first-line treatment of metastatic renal cell carcinoma (mRCC). J Clin Oncol 26:5024 10. Horiguchi A, Oya M, Shimada T, Uchida A, Marumo K, Murai M (2002) Activation of signal transducer and activator of transcription 3 in renal cell carcinoma: a study of incidence and its association with pathological features and clinical outcome. J Urol 168:762–765 11. Patard JJ, Pignot G, Escudier B, Eisen T, Bex A, Sternberg C, Rini B, Roigas J, Choueiri T, Bukowski R, Motzer R, Kirkali Z, Mulders P, Bellmunt J (2011) ICUD-EAU international

Angiogenesis

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22. 23.

24.

25.

26.

27.

28.

29.

consultation on kidney cancer 2010: treatment of metastatic disease. Eur Urol 60:684–690 Coppin C, Kollmannsberger C, Le L, Porzsolt F, Wilt TJ (2011) Targeted therapy for advanced renal cell cancer (RCC): a Cochrane systematic review of published randomised trials. BJU Int 108:1556–1563 Houghton AM (2010) The paradox of tumor-associated neutrophils: fueling tumor growth with cytotoxic substances. Cell Cycle 9:1732–1737 Jensen HK, Donskov F, Marcussen N, Nordsmark M, Lundbeck F, von der Maase H (2009) Presence of intratumoral neutrophils is an independent prognostic factor in localized renal cell carcinoma. J Clin Oncol 27:4709–4717 Kunzi-Rapp K, Genze F, Ku¨fer R, Reich E, Hautmann RE, Gschwend JE (2001) Chorioallantoic membrane assay: vascularized 3-dimensional cell culture system for human prostate cancer cells as an animal substitute model. J Urol 166:1502–1507 Hagedorn M, Javerzat S, Gilges D, Meyre A, de Lafarge B, Eichmann A, Bikfalvi A (2005) Accessing key steps of human tumor progression in vivo by using an avian embryo model. Proc Natl Acad Sci USA 102:1643–1648 Ribatti D (2008) Chick embryo chorioallantoic membrane as a useful tool to study angiogenesis. Int Rev Cell Mol Biol 270:181–224 Ausprunk DH, Knighton DR, Folkman J (1975) Vascularization of normal and neoplastic tissues grafted to the chick chorioallantois. Role of host and preexisting graft blood vessels. Am J Pathol 79:597–618 Okamura K, Tsuji Y, Shimoji T, Miyake K (1995) Growth of human tumor xenografts on chorioallantoic membrane of chick embryo. Hinyokika Kiyo 41:163–170 Yalcin M, Bharali DJ, Lansing L, Dyskin E, Mousa SS, Hercbergs A, Davis FB, Davis PJ, Mousa SA (2009) Tetraidothyroacetic acid (tetrac) and tetrac nanoparticles inhibit growth of human renal cell carcinoma xenografts. Anticancer Res 29: 3825–3831 Fuhrman SA, Lasky LC, Limas C (1982) Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 6:655–663 AJCC (2002) Cancer staging manual, 6th edn. Springer, Berlin Kilarski WW, Samolov B, Petersson L, Kvanta A, Gerwins P (2009) Biomechanical regulation of blood vessel growth during tissue vascularization. Nat Med 15:657–664 Dumartin L, Quemener C, Laklai H, Herbert J, Bicknell R, Bousquet C, Pyronnet S, Castronovo V, Schilling MK, Bikfalvi A, Hagedorn M (2010) Netrin-1 mediates early events in pancreatic adenocarcinoma progression, acting on tumor and endothelial cells. Gastroenterology 138:1595–1606 Park ES, Kim SJ, Kim SW, Yoon SL, Leem SH, Kim SB, Kim SM, Park YY, Cheong JH, Woo HG, Mills GB, Fidler IJ, Lee JS (2011) Cross-species hybridization of microarrays for studying tumor transcriptome of brain metastasis. Proc Natl Acad Sci USA 108:17456–17461 Saeed AI, Bhagabati NK, Braisted JC, Liang W, Sharov V, Howe EA et al (2006) TM4 microarray software suite. Methods Enzymol 411:134–193 Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA (2003) DAVID: database for annotation, visualization, and integrated discovery. Genome Biol 4:P3 Huang DW, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4:44–57 Rhodes DR, Yu J, Shanker K, Deshpande N, Varambally R, Ghosh D, Barrette T, Pandey A, Chinnaiyan AM (2004) ONCOMINE: a cancer microarray database and integrated datamining platform. Neoplasia 6:1–6

30. Herbert JM, Stekel D, Sanderson S, Heath VL, Bicknell R (2008) A novel method of differential gene expression analysis using multiple cDNA libraries applied to the identification of tumour endothelial genes. BMC Genomics 9:153 31. Javerzat S, Franco M, Herbert J, Platonova N, Peille AL, Pantesco V, De Vos J, Assou S, Bicknell R, Bikfalvi A, Hagedorn M (2009) Correlating global gene regulation to angiogenesis in the developing chick extra-embryonic vascular system. PLoS ONE 4:e7856 32. Vandesompele J, Preter KD, Pattyn F, Poppe B, Roy NV, Paepe AD, Speleman F (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3:research0034.1–0034.11 33. Duvvuri U, Shiwarski DJ, Xiao D, Bertrand C, Huang X, Edinger RS, Rock JR, Harfe BD, Henson BJ, Kunzelmann K et al (2012) TMEM16A induces MAPK and contributes directly to tumorigenesis and cancer progression. Cancer Res 72:3270–3281 34. Currie MJ, Gunningham SP, Turner K, Han C, Scott PA, Robinson BA, Chong W, Harris AL, Fox SB (2002) Expression of the angiopoietins and their receptor Tie2 in human renal clear cell carcinomas; regulation by the von Hippel-Lindau gene and hypoxia. J Pathol 198:502–510 35. Liu YH, Lin CY, Lin WC, Tang SW, Lai MK, Lin JY (2008) Upregulation of vascular endothelial growth factor-D expression in clear cell renal cell carcinoma by CD74: a critical role in cancer cell tumorigenesis. J Immunol 181:6584–6594 36. Yao X, Qian CN, Zhang ZF, Tan MH, Kort EJ, Yang XJ, Resau JH, Teh BT (2007) Two distinct types of blood vessels in clear cell renal cell carcinoma have contrasting prognostic implications. Clin Cancer Res 13:161–169 37. Stein J, Drenckhahn D, Nehls V (1996) Development of pericytelike cells during angiogenesis in quail chick chimeras as detected by combined Feulgen reaction and immunohistochemistry. Ann Anat 178:153–158 38. Gordan JD, Lal P, Dondeti VR, Letrero R, Parekh KN, Oquendo CE, Greenberg RA, Flaherty KT, Rathmell WK, Keith B, Simon MC, Nathanson KL (2008) HIF-alpha effects on c-Myc distinguish two subtypes of sporadic VHL-deficient clear cell renal carcinoma. Cancer Cell 14:435–446 39. Keith B, Johnson RS, Simon MC (2011) HIF1a and HIF2a: sibling rivalry in hypoxic tumour growth and progression. Nat Rev Cancer 12:9–22 40. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, Wykoff CC, Pugh CW, Maher ER, Ratcliffe PJ (1999) The tumour suppressor protein VHL targets hypoxiainducible factors for oxygen-dependent proteolysis. Nature 399:271–275 41. Raval RR, Lau KW, Tran MG, Sowter HM, Mandriota SJ, Li JL, Pugh CW, Maxwell PH, Harris AL, Ratcliffe PJ (2005) Contrasting properties of hypoxia-inducible factor 1 (HIF-1) and HIF2 in von Hippel-Lindau-associated renal cell carcinoma. Mol Cell Biol 25:5675–5686 42. Djonov V, Schmid M, Tschanz SA, Burri PH (2000) Intussusceptive angiogenesis: its role in embryonic vascular network formation. Circ Res 86:286–292 43. Wei LH, Kuo ML, Chen CA, Chou CH, Lai KB, Lee CN, Hsieh CY (2003) Interleukin-6 promotes cervical tumor growth by VEGF-dependent angiogenesis via a STAT3 pathway. Oncogene 22:1517–1527 44. Li A, Dubey S, Varney ML, Dave BJ, Singh RK (2003) IL-8 directly enhanced endothelial cell survival, proliferation, and matrix metalloproteinases production and regulated angiogenesis. J Immunol 170:3369–3376 45. Shamamian P, Schwartz JD, Pocock BJ, Monea S, Whiting D, Marcus SG, Mignatti P (2001) Activation of progelatinase A (MMP-2) by neutrophil elastase, cathepsin G, and proteinase-3:

123

Angiogenesis

46.

47.

48.

49.

50.

51. 52.

a role for inflammatory cells in tumor invasion and angiogenesis. J Cell Physiol 189:197–206 Zijlstra A, Seandel M, Kupriyanova TA, Partridge JJ, Madsen MA, Hahn-Dantona EA, Quigley JP, Deryugina EI (2006) Proangiogenic role of neutrophil-like inflammatory heterophils during neovascularization induced by growth factors and human tumor cells. Blood 107:317–327 Schweighofer B, Testori J, Sturtzel C, Sattler S, Mayer H, Wagner O, Bilban M, Hofer E (2009) The VEGF-induced transcriptional response comprises gene clusters at the crossroad of angiogenesis and inflammation. Thromb Haemost 102:544–554 Florczyk U, Czauderna S, Stachurska A, Tertil M, Nowak W, Kosakowska M, Poellinger L, Joskowicz A, Loboda A, Dulak J (2011) Opposite effects of HIF-1a and HIF-2a on the regulation of IL-8 expression in endothelial cells. Free Radic Biol Med 51:1882–1892 Huang D, Ding Y, Zhou M, Rini BI, Petillo D, Qian CN, Kahnoski R, Futreal PA, Furge KA, Teh BT (2010) Interleukin-8 mediates resistance to antiangiogenic agent sunitinib in renal cell carcinoma. Cancer Res 70:1063–1071 Galli M, Van Gool F, Rongvaux A, Andris F, Leo O (2010) The nicotinamide phosphoribosyltransferase: a molecular link between metabolism, inflammation and cancer. Cancer Res 70:8–11 Burgos ES (2011) NAMPT in regulated NAD biosynthesis and its pivotal role in human metabolism. Curr Med Chem 18:1947–1961 Kim JY, Bae YH, Bae MK, Kim SR, Park HJ, Wee HJ, Bae SK (2009) Visfatin through STAT3 activation enhances IL-6

123

53.

54.

55.

56.

57.

58.

expression that promotes endothelial angiogenesis. Biochim Biophys Acta 1793:1759–1767 Park JW, Kim WH, Shin SH, Kim JY, Yun MR, Park KJ, Park HY (2011) Visfatin exerts angiogenic effects on human umbilical vein endothelial cells through the mTOR signaling pathway. Biochim Biophys Acta 1813:763–771 Wang B, Hasan MK, Alvarado E, Yuan H, Wu H, Chen WY (2011) NAMPT overexpression in prostate cancer and its contribution to tumor cell survival and stress response. Oncogene 30:907–921 Lee YC, Yang YH, Su JH, Chang HL, Hou MF, Yuan SS (2011) High visfatin expression in breast cancer tissue is associated with poor survival. Cancer Epidemiol Biomarkers Prev 20:1892–1901 Tun HW, Marlow LA, von Roemeling CA, Cooper SJ, Kreinest P, Wu K, Luxon BA, Sinha M, Anastasiadis PZ, Copland JA (2010) Pathway signature and cellular differentiation in clear cell renal cell carcinoma. PLoS ONE 5:e10696 Ledoux S, Queguiner I, Msika S, Calderari S, Rufat P, Gasc JM, Corvol P, Larger E (2008) Angiogenesis associated with visceral and subcutaneous adipose tissue in severe human obesity. Diabetes 57:3247–3257 Menssen A, Hydbring P, Kapelle K, Vervoorts J, Diebold J, Lu¨scher B, Larsson LG, Hermeking H (2012) The c-MYC oncoprotein, the NAMPT enzyme, the SIRT1-inhibitor DBC1, and the SIRT1 deacetylase form a positive feedback loop. Proc Natl Acad Sci USA 109:E187–E196