Supplementary Data
The WIP1 Oncogene Promotes Progression and Invasion of Aggressive Medulloblastoma Variants
Meghan C. Buss1, Marc Remke4, Juhyun Lee1, Khanjan Gandhi3, Matthew J. Schniederjan2,3, Marcel
Kool5, Paul A. Northcott5, Stefan M. Pfister5,6, Michael D. Taylor4, and Robert C. Castellino1,2,3
Department of Pediatrics, 1Aflac Cancer and Blood Disorders Center, 2Children’s Healthcare of Atlanta,
1
3
Winship Cancer Institute, 3Emory University School of Medicine, Atlanta, GA, USA; 4Division of
Neurosurgery, Arthur and Sonia Labatt Brain Tumour Research Center, and Program in Developmental
and Stem Cell Biology, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada;
5
German Cancer Research Center, 6Department of Pediatric Hematology, Oncology and Immunology,
Heidelberg University Hospital, Heidelberg, Germany.
Corresponding Author: Robert C. Castellino, HSRB, 1760 Haygood Drive, N.E., Room E394, Atlanta,
GA 30322. Telephone: (404) 727-2078. Fax: (404) 727-4455. E-mail: rccaste@emory.edu.
Figure Legends
Supplementary Figure 1. D556-WIP1 stable clones express markers of Group 3 and 4
medulloblastomas. Western blotting of whole cell lysates of D556 medulloblastoma cells with stable
expression of an empty vector (pcDNA), wild-type WIP1, or phosphatase-dead WIP1 (WIP1 D314A).
Blots were probed with antibodies against the Group 3 and 4 medulloblastoma markers, NPR3 and
KCNA1, respectively. -actin was used as a loading control. Protein expression was quantified from nearinfrared fluorescence and normalized to expression of -actin, the loading control. The relative amount of
each protein is expressed as a ratio of normalized protein expression in a particular lane, to normalized
expression of protein in a D556-pcDNA stable clone. Western blotting was repeated at least three times
with similar results.
1|Page
Supplementary Figure 2. High WIP1 expression is associated with up-regulation of genes involved
in cancer invasion and metastasis in D556 medulloblastoma cells. Up-regulated expression of cancer
invasion and metastasis genes in D556 cells with stable expression of WIP1 (D556-WIP1), versus stable
expression of an empty vector (D556-pcDNA), on Affymetrix U133 +2.0 gene expression microarrays.
Up-regulated genes, including CXCR4 and MMP9, are shown in red. Down-regulated genes, including
ADCY1, are shown in green.
Supplementary Figure 3. Stimulation with SDF1 promotes growth of WIP1 high-expressing
medulloblastoma cells. Viability of D556 clones by (A) CellTiter-Glo and (B) WST-1 assays, 48 hours
following serum starvation and CXCL12 (SDF1, 1 g/mL) stimulation. *, p < 0.005. (C) Viability of
D425 and Med8A cells by CellTiter-Glo assays, 48 hours following serum starvation and SDF1
stimulation. *, p < 0.005. Error bars, standard deviation. Experiments were repeated at least three times.
Supplementary Figure 4. WIP1 knock-down promotes Ser339 phosphorylation of CXCR4 only in
the cytoplasmic fraction of D556 WIP1-stable medulloblastoma cells. Western blotting of (A) nuclear
(N) and cytoplasmic (C) cell lysates from D556-pcDNA, D556-WIP1, and D556-WIP1 D314A stable
clones, and (B) whole cell lysates from D425 cells, 48 hours following infection with control (-; shNC) or
EGFP-shWIP1 (+) lentivirus at multiplicity of infection (MOI) of 1.5. Western blots in (B) show protein
expression in D425 cells before and 30 minutes after stimulation with 100 ng/mL CXCL12. Protein bands
that correspond to serine 339-phosphorylated and total CXCR4 on western blots were measured by nearinfrared fluorescence, normalized to -actin expression, and are shown relative to expression in EGFPshNC-infected cells. Lamin B1 and eEF2 expression were used as nuclear and cytoplasmic controls,
respectively. Experiments were repeated twice.
Supplementary Figure 5. WIP1 knock-down inhibits localization of CXCR4 to the plasma
membrane of WIP1 high-expressing medulloblastoma cells. D556-WIP1 stable clones were plated at
3.5 x 104 cells per chamber. 48 hours after cells were infected with negative control (EGFP-shNC) or
WIP1 shRNA lentiviral particles, cells were fixed with 4% paraformaldehyde. Cells were permeabilized
2|Page
with 0.3% Triton-X, incubated in 0.3% H2O2 in MeOH, incubated with 10% normal goat serum in PBS
containing 0.3% Triton-X, then blocked with 5% BSA for 30 minutes, and incubated overnight at 4°C,
with primary antibody, diluted 1:100 in 5% BSA. Immunofluorescence (IF) was performed for EGFP,
WIP1, and CXCR4. Cells were washed with buffer TNT (1 mM Tris/HCl, 5 mM NaCl, 1%Tween-20)
and incubated with goat anti-rabbit Alexa Fluor 555 secondary antibody, at a dilution 1:250. Slides were
mounted using Hard Set Mounting Media containing DAPI, which defines the cell nucleus, and were
stored at 4°C in the dark. Images were acquired using an EVOS-fl fluorescent microscope. The left –most
“Merged” column refers to the merge of IF for GFP, WIP1, and DAPI. The Scale bars, 100 m.
Experiments were repeated at least twice.
Supplementary Figure 6. CXCR4-mediated growth of D556-WIP1 stable medulloblastoma cells
depends on the presence of functional p53. 1x105 D556-pcDNA and D556-WIP1 cells were plated in 6well plates and, 24 hours later, were transfected with 10nM scrambled, negative control siRNA (siNC) or
p53 siRNA. After a 24 hour incubation, cells were serum starved. (A) To determine the number of viable
cells, cells were serum starved for 8 hours and then stimulated with vehicle (-) or 1μg/mL SDF1α for 48
hours in serum-free media. After 48 hours, cells were harvested and counted by trypan blue exclusion,
using standard methods. (B) For short-term stimulation, cells were serum starved for 24 hours and then
stimulated with vehicle (-) or 100ng/mL CXCL12 (SDF1α) for 0, 2, or 10 minutes. Whole cell lysates
were blotted for expression of serine 473-phosphorylated (pAKT), total AKT, p53, and -actin as a
loading control. (C, E) To determine the number of viable cells, cells were serum starved for 8 hours and
treated with either 4µM Nutlin-3a or 25nM RITA and stimulated with 1μg/mL SDF1α for 48 hours in
serum-free media. After 48 hours, cells were harvested and counted by trypan blue exclusion, using
standard methods. (D) For short-term stimulation, cells were serum starved and treated with 4µM Nutlin3a for 24 hours. Cells were stimulated with 100ng/mL SDF1α for 0, 2, or 10 minutes. Whole cell lysates
were blotted for expression of pAKT, p53, and -actin as a loading control.*, p < 0.005. Error bars,
standard deviation. Experiments were repeated at least three times.
3|Page
Supplementary Figure 7. p53 mutant Daoy cells with stable high WIP1 expression do not exhibit
increased invasion in response to the chemoattractant SDF1. Medium containing 1g/mL CXCL12
(SDF1α; +) or vehicle (-) was added to the bottom of each well in a Matrigel Invasion Chamber, followed
by placement of inserts. 2.5x104 Daoy-pcDNA or Daoy-WIP1 stable clones were added to each insert and
incubated for 24 hours. Inserts were fixed in 100% methanol, stained with crystal violet, and counted at
40X magnification. Bar graphs represent the average number of migrated cells per high-power field
(HPF) in triplicate measurements of 10 representative HPFs per cell line. Error bars, standard deviation.
Experiments were repeated twice.
Supplementary Figure 8. D556-pcDNA xenografts exhibit weak, focal expression of CXCR4.
Representative immunohistochemistry for CXCR4 in the cerebellum of two different mice xenografted
with D556-pcDNA stable medulloblastoma clones. Magnification 40x; Scale bars, 20 m.
Supplementary Figure 9. D556-WIP1 xenografts exhibit unequivocal infiltration along perivascular
spaces. Representative immunohistochemistry for phospho-AKT (Ser473) in medulloblastomas resected
from two different mice xenografted with D556-WIP1 stable medulloblastoma clones. Note the tumor cell
infiltration (arrows) along perivascular spaces. Magnification 20x; Scale bars, 50 m.
Supplementary Figure 10. GRK5 knock-down rescues the effects of transfection of wild-type GRK5
in D556-WIP1 stable medulloblastoma cells. 5x105 D556-WIP1 cells were plated in 6-well plates and,
24 hours later, were transfected with 1μg of an empty vector (EV; pcDNA3) or with pcDNA3 containing
wild-type GRK5, using Lipofectamine 2000. (A) After 24 hours of transfection, cells were transduced
with negative control shRNA (shNC) or GRK5 shRNA lentivirus particles, at a multiplicity of infection
(MOI) of 1.5, along with 8μg/μL polybrene. Cells were harvested 48 hours after transduction and counted
by trypan blue exclusion, using standard methods. (B) Whole cell lysates were run on western blots and
probed for expression of GRK5 (using an antibody that detects GRK4, 5, and 6), serine 339-
4|Page
phosphorylated, and total CXCR4, and -actin as a loading control. *, p < 0.05. Error bars, standard
deviation. Experiments were repeated at least three times.
Supplementary Figure 11. Transduction of the phosphorylation mutant CXCR4-S339A into D425
medulloblastoma cells suppresses growth and responsiveness to SDF1 stimulation. 1x106 D425
cells were plated in 6-wells plates. After 24 hours, cells were serum starved and transduced with negative
control shRNA (shNC; -), WIP1 shRNA (shWIP1), HA-tagged empty vector, and/or HA-tagged CXCR4S339A lentivirus particles at a multiplicity of infection (MOI) of 1.5, as well as 8 μg/μL polybrene. Cells
were stimulated with 1μg/mL SDF1α for 48 hours in serum-free media. After 48 hours, cells were
harvested and counted by trypan blue exclusion, using standard methods. *, p < 0.05. Error bars, standard
deviation. Experiments were repeated twice.
Supplementary Figure 12. Validation of GRK5 knock-down in D556 stable medulloblastoma clones.
GRK5 mRNA expression in D556-pcDNA or D556-WIP1 stable medulloblastoma clones by quantitative
real-time, RT-PCR, 72 hours following infection with control pLKO empty vector (shNC) or with one of
two non-overlapping GRK5 shRNA lentiviruses. Expression was normalized to expression in D556pcDNA cells infected with pLKO empty vector control lentivirus. Error bars, standard error of the mean.
Experiments were performed in triplicate and repeated at least three times.
Supplementary Figure 13. GRK5 knock-down inhibits Ser339 phosphorylation of CXCR4 and
increases cell surface expression of CXCR4 in D556 medulloblastoma cells. (A) Immunofluorescence
for CXCR4 and DAPI in D556-pcDNA cells 48 hours following infection with scrambled, negative
control shRNA (shNC) or shGRK5 (#2) lentivirus at a multiplicity of infection (MOI) of 1.5. Scale bars,
100 m. (B) Western blotting of nuclear (N) and cytoplasmic (C) cell lysates from D556-pcDNA and
D556-WIP1 stable clones for serine 339-phosphorylated and total CXCR4, 48 hours following infection
with scrambled, negative control shRNA (-) or shGRK5 (#2) lentivirus at MOI 1.5. Western bands were
quantified using near-infrared fluorescence, relative to expression in negative control shRNA-infected
5|Page
cells, and normalized to expression of -actin. Expression is shown relative to the cytoplasmic fraction of
negative control, shRNA-infected D556-pcDNA or D556-WIP1 cells. Lamin B1 and eEF2 expression are
shown as nuclear and cytoplasmic controls, respectively. Experiments were repeated with reproducible
results twice.
6|Page