Captions for supplementary figures
Supplementary figure 1: Figure 1: Effect of BVZ on the production of VEGF and
CXCL8 by different RCC cells
RCC cells were treated or not for 24 h or one week with 500 ng/ml of BVZ. The amounts of
VEGF and CXCL8 present in cell supernatants were detected with an ELISA. Data are
presented as the mean  SD. Statistical differences are presented **p <0.01; ***p <0.001.
Supplementary figure 2: CXCL8 exerts an autocrine loop on 786-O cells
(A) 786-O cells were serum deprived for 24 h then stimulated with CXCL8 (50 or 100 ng/ml)
for 24 h. The percentage of cells has been evaluated with MTT assays and was plotted +
SEM. The condition without any treatment is considered as the reference value (100 %).
These data represent the mean + SEM of four independent experiments. (***) represent p
value < 0.001.
(B) 786-O cells were transduced with lentivirus expressing control or two independent shRNA
targeting
CXCL8
(A:
Clone
ID:
NM_000584.2-285s1c1;
B:
Clone
ID;
NM_000584.2281s1c1, Sigma, France). Cell populations were obtained after selection with
puromycin (5 g/ml). The proliferative capacity of these cells was tested using the MTT
assay. Data are presented as the mean fold increase  SD. Statistical differences in the fold
increase of tumor cells isolated from control sh-RNA-transduced cells were taken as reference
values. *p<0.05; **p <0.01, ***p< 0.001 compared to cells of control PBS-treated mice.
(C) 106 of the indicated cells were cultured for 24 h in fresh medium. CXCL8 were quantified
in cell supernatants using an ELISA. Data are presented as the mean  SD (***p< 0.001).
(D) RCC cells were serum deprived and incubated in presence or absence of anti-CXCL8
antibodies (500 ng/ml). Their proliferative capacity was tested using the MTT assay after 24h
of incubation. Statistical differences are indicated (* p<0.05; **p< 0.01).
Supplementary figure 3: BVZ accelerates RCC-10 and Caki-2 xenograft tumor growth
(A) 3106 RCC-10LUC+ cells were subcutaneously inoculated in nude mice (n = 10 per group).
Seven days after inoculation, all mice developed tumors and were treated weekly with PBS
for the control or 150 g BVZ. Bioluminescence was measured weekly as described in the
experimental procedure. Data are presented as mean  SD. Statistical differences between the
size of control and treated tumors are presented: *p<0.05.
(B) LYVE-1 immuno-staining (green) revealed the lymphatic network. Nuclei are stained
with DAPI (blue). Lymphatic vessels with lumens (L) were observed in the centre and near
the periphery of BVZ-treated tumors.
(C) 3106 Caki-2LUC+ cells were subcutaneously inoculated in nude mice (n = 10 per group).
Seven days after inoculation, all mice developed tumors and were treated weekly with PBS
for the control or 150 g BVZ. Bioluminescence was measured weekly as described in the
experimental procedure. Data are presented as mean  SD. Statistical differences between the
size of control and treated tumors are presented: *p<0.05; **p <0.01.
Supplementary figure 4: BVZ and paclitaxel effects on the vasculature of tumors.
The tumor vasculature in each experimental group (control, BVZ weekly or bi-weekly, antiCXCL 8, BVZ plus anti-CXCL8, paclitaxel, paclitaxel plus BVZ, paclitaxel plus anti-CXCL
8 and paclitaxel plus BVZ plus anti-CXCL 8) was evaluated by CD31 immuno-staining and
coverage of the vessels using anti-SMA antibody as presented on figure 2 and
supplementary figure 4. Vascular density (vessels/mm2) and the number of vessels covered by
SMA labelled cells were determined by using the Image J program. Both parameters,
determined in placebo treated tumors were considered as the reference value (100%).
Quantification (means ± SD) resulted from analysis of four independent tumors and
considered at least ten fields for each tumour. Statistically significant differences are
indicated: *p<0.05; **p <0.01; ***p <0.001.
(A) Quantification of the experimental groups presented on figure 2
(B) Quantification of the experimental groups presented on supplementary figure 4.
Supplementary figure 5: BVZ treatment results in an improved tumor perfusion
(A) Differences in tumor vasculature of mice treated with PBS or BVZ.
(B) Sections of tumors from PBS or BVZ-treated mice were labelled with hematoxylin and
eosin. Ten fields by tumors (PBS or BVZ-treated) have been analyzed. Of these, no
enlarged vessels containing blood cells have been observed in the control tumors.
Representative images are shown in the figure.
Supplementary figure 6: Paclitaxel increases CXCL8 gene transcription and mRNA
stability
786-O cells were transfected with CXCL8 minimal promoter (Sparmann, and Bar-Sagi, 2004)
or 3’UTR luciferase reporter genes. This last vector was obtained by inserting the region
corresponding to the CXCL8 3’ UTR obtained by PCR on 786-O genomic DNA instead of
the corresponding VEGF 3’UTR region in the already described TK luc vector (EssafiBenkhadir, K. et al, 2007). Cells were treated with 1 M of paclitaxel for 24 hours. Luciferase
activity was measured and reported as the percentage of activity compared to the untreated
condition. n = 3; data are presented as mean  SD. Statistically significant differences are
shown: **p <0.01; ***p<0.001.
Supplementary figure 7: Paclitaxel treatment modifies tumor vascularization but do not
affect lymphangiogenesis
(A) Vessels and -smooth muscle actin (SMA) were detected using anti-CD31 antibody
immuno-staining (green) and anti-SMA antibody (red) respectively. Tumor sections were
counterstained with DAPI (blue).
(B) The lymphatic network was stained using anti LYVE-1 antibody and nuclei were stained
with DAPI (blue). Lymphatic vessels with lumens (L) were observed in the centre and near
the periphery of paclitaxel plus BVZ-treated tumors.
Supplementary figure 8: Decreased in hypoxia of tumors treated with paclitaxel plus
BVZ. Increased growth capacity of 786-OLUC+ cells derived from these tumors
(A) Tumor extracts from four independent PBS, paclitaxel or paclitaxel + BVZ treated
tumors were tested for the presence of carbonic anhydrase IX by western blot. Total ERK is
used as a loading control for normalization. Statistical differences are indicated (**p< 0.01).
(B) Cells from four independent PBS, BVZ + Paclitaxel or BVZ + Paclitaxel + Anti-CXCL8treated tumors were serum deprived and incubated in presence or absence of anti-CXCL8
antibodies (500 ng/ml). Their proliferative capacity was tested using the MTT assay after 24 h
of incubation. Statistical differences are indicated. (NS: non-significant).