Title: Novel candidate metastasis genes as putative drug targets for breast cancer

Cover Page

The handle http://hdl.handle.net/1887/20271 holds various files of this Leiden University dissertation.

Author: Roosmalen, Wies van

Title: Novel candidate metastasis genes as putative drug targets for breast cancer

Date: 2012-12-12

Chapter 4

An Imaging-based RNA-Interference Migration Screen Identifies the Splicing Related SRSF Protein Kinase 1 (SRPK1) as a Clinical Relevant Breast Cancer Metastasis Promoter

Wies van Roosmalen

, Sylvia E. Le Dévédec

, Ofra Golani

, Marcel Smid

, Mieke Timmermans

, Di Zi

, Marjo de Graauw

, Bart Jacobse

, Reshma Lalai

, Suha Naffar-Abu-Amara

, John W.M. Martens

, John A. Foekens

, Benjamin Geiger

, Bob van de Water

1

Division of Toxicology, LACDR, Leiden University, Leiden, Netherlands

2

Department of Molecular Cell Biology, Weizmann Institute of Science, Rehovot, Israel

3

Department of Medical Oncology, Erasmus University Medical Center, Daniel den Hoed Cancer Center, Rotterdam, Netherlands

Submitted

Abstract

Tumor cell migration is a key step underlying cancer cell dissemination and metastasis and is controlled by extracellular signaling-mediated dynamic cytoskeletal and cell matrix adhesion remodeling. Using a phagokinetic track (PKT) assay in combination with multi-parametric image analysis and highly motile H1299 adenocarcinoma cells, we screened for 1,429 upstream kinase signaling components and downstream adhesion and cytoskeletal regulators that determine tumor cell migratory behavior:

speed, directionality, and persistence. Thirty significant genes were validated by live cell imaging random tumor cell migration, which was associated with modulation of focal adhesion dynamics. For eight genes a significant association with metastasis free survival in breast cancer patients was observed, SHC1, SRPK1, NEK2, ITGB3BP and MAP3K8 being most significant. Also, high SRFS protein kinase 1 (SRPK1) protein expression on breast cancer tissue micro arrays was associated with poor disease outcome. SRPK1 expression was highest in basal-like breast cancer cell lines and depletion of SRPK1 inhibited breast cancer cell motility and focal adhesion dynamics. Finally, in an orthotopic mammary tumor metastasis model, stable knockdown of SRPK1 in 4T1 basal-like breast cancer cells did not affect primary tumor growth but almost completely prevented lung metastasis formation. This study provides a comprehensive information resource on the molecular determinants of tumor cell migration in close association with a clinical significant role in breast cancer progression.

Key words

RNA-interference screening, tumor cell migration, breast cancer, metastasis, SRPK1

4

Introduction

Cancer cell dissemination from the primary tumor causing distant organ metastasis formation remains the major reason for cancer deaths. Individual or cohesive tumor cell migration and invasion are critical steps in the dissemination process and involve the coordinated dynamic remodeling of cell matrix adhesion complexes (also called focal adhesions). Focal adhesions are dynamic protein complexes, consisting of over 150 different proteins, including signaling proteins as well as structural components, that form (in)direct functional interactions [1; 2]. Upstream signaling by proto- oncogenic growth factor signaling controls the dynamics of focal adhesion through the integrated phosphorylation and activation of focal adhesion-associated kinases and adapter proteins as well as the dynamic reorganization of the actin cytoskeletal network through regulation of the family of Rho-GTPases. These locally coordinated activities in cancer cells determine the directionality, speed and persistence of the cell motility and, thereby, cancer cell migration in vitro and metastatic phenotype in vivo.

Various genes involved in cell migration are associated with tumor cell motility and cancer metastasis. Increased signaling through EGF or HGF receptor family members promotes actin reorganization and cancer cell invasiveness and metastasis and is associated with poor disease outcome [3-5]. Alternative, activation of TGFbeta receptor induces an epithelial-to-mesenchymal transition of otherwise non-motile epithelial cancer cells [6]. Downstream effectors such as the Rho-GTPase RhoC stimulates invasiveness and cancer cell dissemination [7]. And also focal adhesion signaling molecules such as integrin beta1, focal adhesion kinase (FAK), Src and p130Cas have been associated with metastatic potential [8-11].

Although the role of some FA associated proteins in tumor cell migration

and invasion is well understood, a systematic approach in defining the functional

role of individual signaling and cell adhesion components in tumor cell migration

and metastatic disease is lacking. A limited RNA-interference screen and cDNA

gain-of-function screen were performed to identify kinases that regulate epithelial

collective cell migration or stimulate single cell migration [12; 13]. Since non-

motile and non-invasive cells were used, the identification of genes that drive the

highly motile behavior of aggressive cancer cells was impossible. Moreover, these

screens focused on kinases and several breast cancer associated genes, precluding

the identification of the individual role of cell adhesion and migration machinery

components.

Here we used an automated multiparametric and quantitative high content imaging-based RNA-interference screen to uncover the role of kinase signaling and cell adhesion components in different cell migratory behavior strategies, including migration speed, directionality and persistence. Using the well-characterized highly migratory H1299 cells as a model we screened 1,429 genes in total of which 136 genes were either stimulating or inhibiting tumor cell migration behavior. Of 30 genes validated by single siRNAs, eight showed a strong significant association with cancer metastasis free survival in breast cancer patients. We further focused on SRFS Protein Kinase 1 (SRPK1), which was highest expressed in aggressive cell lines of the basal-like breast cancer subtype. Depletion of SRPK1 in basal- like breast cancer cells inhibited their cell migration and affected the dynamics of focal adhesion. Importantly, SRPK1 was required for metastasis formation in a spontaneous orthotopic tumor metastasis mouse model.

Materials and Methods Cell culture

H1299 cells (ATCC-CRL-5803), MDA-MB-231 (ATCC-HTB-26), MDA-MB-417.5 (kindly provided by dr. Joan Massague) and 4T1 cells (ATCC-CRL-2539) were cultured in RPMI (GIBCO, Life Technologies, Carlsbad, CA, USA) supplemented with 10% FBS (PAA, Pasching, Austria) and 100 International Units/mL penicillin and 100 µg/

mL streptomycin (Invitrogen, Carlsbad, CA, USA). For live cell imaging random cell migration assays, phenol-red free culture medium was used. Cells were maintained in a 5% CO

humidified chamber at 37 ºC.

Antibodies

Mouse anti-SRPK1 was from BD Biosciences (Franklin Lakes, NJ, USA). Rabbit anti- ER and mouse anti-β-actin were from Santa Cruz (Santa Cruz, CA, USA). Mouse anti- tubulin was from Sigma Aldrich (St. Louis, MO, USA)

Transient siRNA mediated gene knockdown

Human Protein Kinases, Phosphatases and a custom designed Adhesome siRNA library were purchased in siGENOME format from Dharmacon (Lafayette, CO, USA).

Plates were diluted to 1 µM working concentration in complementary 1x siRNA buffer

in a 96-well plate format. A 50 nM reaction was performed according to manufacturer

guidelines. Complex time was 20 min and 2,400 cells were added. The plate was

overnight placed in the incubator and the next morning medium was refreshed. After

4

48-72 hr, cells were used for various assays.

Stable shRNA mediated gene knockdown

First, 4T1 cells were transduced with a lentiviral GFP construct. After that, cells were transduced with Sigma MISSION shRNA constructs coding a non-targeting control sequence and 2 independent sequences for SRPK1 (shSRPK1-A: CCG GCC TGA TAG GAT CTG GCT ACA ACT CGA GTT GTA GCC AGA TCC TAT CAG GTT TTT, shSRPK1-B: CCG GGC GCC AGA GAT TAA TTG CAA TCT CGA GAT TGC AAT TAA TCT CTG GCG CTT TT). After puromycin selection, cells were FACS sorted for homogenous GFP expression.

Phagokinetic track assay

Glass bottom 96-well imaging plates (Whatman, GE Healtcare, Maidstone, UK) were overnight coated with 10 µg/µL fibronectin (Sigma Aldrich) at 4 ºC. Next day, plates were washed two times with PBS using a Hydroflex platewasher (Tecan, Männedorf, Switzerland). White 0.40 µm, CV = 6.8%, solid = 4.1% surfactant-free carboxylated beads (Molecular Probes, Invitrogen) were prepared by washing 1.84 mL/plate twice in PBS. After that, volume was adjusted to 4 mL with PBS and beads were stored at 4 ºC. The PBS on the fibronectin-coated plates was removed and 40 µL bead solution per well was added. The plate was then incubated for 1 hr at 37 ºC after which the plate was washed 7 times with PBS. Prior to cell seeding, PBS was replaced with culture medium. Knockdown cells were prepared 65 hr after transfection by washing them twice in PBS and addition of 50 µL trypsin/EDTA (Invitrogen). One part of the cell suspension was diluted in culture medium and plated on the bead-coated plates and placed in the incubator. After 7 hr migration time, cells were fixed using a 3%

paraformaldehyde solution and washed twice in PBS. Plates were then stored at 4 ºC for later imaging. For each transfection, duplicate bead plates were generated and the screening of the siRNA libraries was done in duplo.

PKT imaging and analysis

To image the bead-free tracks, an automatic screening microscope fitted with a 10x

/0.20 NA objective was used. Transmitted light signal was acquired for 25 images

per well and autofocus was made before each image. Montage images per well were

exported and analyzed using WIS-PhagoTracker software [12]. After that, obtained

data was firstly transformed by logarithm to base 2, in order to obtained normal

distribution. Next, normalization was performed using ratio of raw measurement to

the mean of the negative control (non-targeting siCtrl). After normalization, for each

individual of 4 replicates, its distance to the mean of the other 3 replicates was calculated and compared with standard deviation of the other 3 replicates. Outlier was identified when the distance was 3 times bigger than the standard deviation.

Outliers were removed before the hits identification. For hits identification, two tailed t-test to compare each condition (4 replicates) with the negative control was used and genes were ranked on P-value. Normalized values are average values of 4 replicates with removal of outliers.

Live cell imaging random cell migration assay

Glass bottom 96-well plates (PAA) were coated with 10 µg/µL fibronectin (Sigma Aldrich) for H1299 or 20 µg/µL collagen type 1 (isolated from rat tails) for MDA- MB-231 for 1 hr at 37 ºC. Cells were plated directly upon transfection procedure.

After 72 hr, medium was refreshed one more time and cells were placed on a NIKON Eclipse TE2000-E microscope fitted with a 37 ºC incubation chamber, 20x objective (0.75 NA, 1.00 WD) and perfect focus system. Automatically, 3 positions per well were defined and GFP signal was acquired every 390 seconds for a total imaging period of 12 hr. After the imaging period, plates were fixed using 4% formaldehyde and stored for later immunostaining. All data was converted and analyzed using custom made ImagePro Plus (MediaCybernetics, Bethesda, MA, USA) macros [14]. Cell migration was quantified by tracking of GFP signal in time using a custom ImageJ-based macro.

TIRF microscopy of focal adhesion dynamics

CELLview glass bottom dishes with four compartments were coated with 10 µg/µL fibronectin (Sigma Aldrich) for 1 hr at 37 ºC. H1299 cells stably expressing GFP-paxillin were used for this assay. Knockdown cells were seeded 24 hr after transfection and 48 hr later cells were placed on a Nikon Eclipse TE2000-E microscope fitted with a 37ºC incubation chamber, 60x oil objective (1.49 NA, 0.12 WD) and perfect focus system. Focal adhesions were visualized using TIRF with the 488 laser for one hr, with one min interval. The cytoplasmic GFP signal was acquired every 5 min using widefield fluorescence. Data was exported as .tif images and both channels were merged while making .avi files. FA dynamics was analyzed by making time color overlays using a custom made plug in for Image-Pro Plus (MediaCybernetics, Bethesda, MA, USA).

Gene expression profiling human breast cancer patients and cell lines

A total of 344 lymph node negative breast cancer patients who did not receive

4

any adjuvant systemic treatment (chemotherapy and/or endocrine therapy) were included. The dataset is available in National Center for Biotechnology Information/

Gene Expression Omnibus (GEO), entries GSE2034 and GSE5327. Clinical characteristics and treatment details were previously described [15-17]. The cohort consists of 221 estrogen receptor (ER) positive patients and 123 ER-negative patients (ER status was determined according to expression measured on microarray [15]. In total 118 patients developed a distant metastasis, which was counted as an event in the survival analysis. Patients who died without evidence of disease were censored at the time of last-follow-up. Gene expression data was obtained using the Affymetrix HG-U133A chips. Preprocessing was performed as described previously [16]. STATA, release 11 (StataCorp, Texas, USA), was used to calculate the Cox proportional hazards ratio (HR). Log2-transformed gene expression data were evaluated as continuous variables with metastasis free survival times as endpoint.

For the Kaplan–Meier survival curves, log-rank tests were used to test the equality of survivor functions across two groups. For each gene, an optimal cutpoint was established in the expression data, such that the resulting two groups showed the biggest difference in survival curves (low log-rank p-value). This was determined using the srd algorithm in STATA. A two-sided P value of 0.05 was considered statistically significant.

For the breast cancer cell lines, microarray data of Affymetrix HG-U133A chips were used. The expression data of 39 BC cell lines are available in GEO (series GSE16795). Additional details are described by Hollestelle et al [18].

Tissue microarray analysis SRPK1

In this study a tissue microarray (TMA) was used, which included all formalin-fixed

paraffin-embedded (FFPE) primary tumors of 1,350 consecutive breast cancer

patients who entered the Erasmus Medical Center Rotterdam during the years 1985

to 2000 for local treatment of their primary disease, and from whom complete

clinical follow-up information was available. For the design of the TMA and a detailed

description of the cohort we refer to Timmermans et al (in preparation). To allow

the analysis of pure prognosis without confounding effects of adjuvant therapy

only the 562 lymph-node negative (LNN) patients who did not receive adjuvant

systemic therapy were included in our statistical analyses. From all specimen, tumor

grading according to Scarff, Bloom & Richardson and histology was available as

well as the expression of clinical relevant marker such as ER, PgR, HER2, EGFR,

and MIB1 protein expression determined by IHC (Timmermans et al, in preparation).

According to guidelines HER2++ cases were additionally assessed by fluorescence in situ hybridization (FISH). The Medical Ethical Committee of the Erasmus Medical Center Rotterdam, the Netherlands, approved our study design (MEC 02.953). This retrospective study was conducted in accordance with the Code of Conduct of the Federation of Medical Scientific Societies in the Netherlands (http:/www.federa.

org/). Median age of the patients in the analysis at the time of surgery (258 pre- and 304 postmenopausal patients) was 54 years (range, 26 – 92 years). Two hundred ninety-one patients (52%) underwent mastectomy while the remaining had breast- conserving therapy. Metastasis-free survival (MFS) was defined as the time from diagnosis to the first distant metastasis or for patients still surviving, as the time from diagnosis to the last contact. Of the 562 patients included in the analysis of MFS, 143 were counted as failures for the entire cohort, 115 for the 460 ER positive and 28 failures for the 102 ER negative subgroups analyzed. Patients diagnosed with secondary contralateral breast cancer during follow-up were censored for MFS at the date of diagnosis of the secondary breast cancer. Patients who died without evidence of disease and were censored at last follow-up in the analysis of MFS. Slides from the TMA were cut (4 mm) and stained for SRPK1 (BD Biosciences) with the EnVision™ method (Dako, Glostrup, Denmark). After staining, slides were converted to digital images with a Virtual Slide Scanner (NanoZoomer 2.0-HT, Hamamatsu, Japan) and uploaded into our TMA database (Distiller, Slidepath, Ireland). All images were scored manually for staining intensity (0=negative, 1=weak, 2=moderate, 3=strong). Only staining on invasive breast tumor cells was recorded. Mean of the positive stained breast tumor cells was calculated from the total cell count of the three separate images of the cores.

Western Blotting

Cells were scraped in ice-cold TSE (10 nM Tris-HCL, 250 mM sucrose, 1mM EGTA pH 7.4) supplemented with inhibitors. After sonication of cell lysates, protein concentration was determined by Bio-Rad protein assay (Hercules, CA, USA) using IgG as internal standard. 30 mg of total cellular protein was separated on 7.5 or 10% SDS-PAGE and transferred to PVDF membranes (Millipore, Billerica, USA). Blots were blocked in 5%

w/v bovine serum albumin in TBST (0.5 M NaCl, 20 mM Tris-HCl, 0.05% v/v Tween20

pH 7.4) and probed with primary antibody (overnight, 4ºC) followed by incubation

with secondary antibody either Dylight-649 or Horseradish peroxidase-coupled

(Jackson ImmunoResearch Laboratories Inc, West Grove, PA, USA). Dylight-649 was

directly visualized by scanning on Typhoon imager 9400 (Amersham Biosciences,

4

Uppsala, Sweden). HRP probes were first activated by incubation with Enhanched Chemiluminescence Plus reagent (Amersham Biosciences).

Immunofluorescence staining

Cells were fixed after imaging for RCM in 4% paraformaldehyde in PBS for 10 min at room temperature. Coverslips were blocked in TBP (0.1 % v/v Triton-X and 0.5% w/v bovine serum albumin in PBS) for 1 hr at room temperature and overnight incubated with primary antibody, followed by 1 hr incubation of fluorescently labeled secondary antibody. Next, cells were incubated with 2 µg/mL Hoechst 33258 (Sigma Aldrich) to visualize nuclei. Cells were imaged using a Nikon Eclipse TE 2000-E confocal microscope fitted with a 20x objective (0.75 NA, 1.00 WD) and 4 times zoom.

Orthotopic mouse breast cancer model

Experiment with 4T1 was done as previously described [19]. In short, female Rag2

γc

mice were orthotopically injected with 100,000 4T1-GFP cells in the fourth mammary gland (10-11 animals per group). Tumor growth was monitored over a 3-week period and then mice were sacrificed and tumor and lungs were isolated.

Lungs were injected with Indian ink to count surface metastasis.

Statistical analysis

Student’s t test was used to determine significant differences between two means (P<0.05 or P<0.01). Values are represented as mean ± sem. Significant differences are marked in the graphs.

Results

RNA-interference screening identifies novel regulators of tumor cell migration To discover novel regulators of tumor cell migration we used the highly migratory and

well-characterized human non-small lung cancer cell line H1299 in a high-content imaging-based so-called phagokinetic track (PKT) assay [12] in which cells migrate on top of a monolayer of beads and bead-free tracks indicate the migratory path.

Multi-parametric image analysis on individual cell tracks allowed the quantitative

assessment of the migratory strategy of individual cells by determining eight

different quantitative markers for cell migration (e.g. ‘Total Area’, ‘Net Area’, ‘Axial

Ratio’, ‘Major Axis’, ‘Minor Axis’, ‘Roughness’, ‘Perimeter’ and ‘Solidity’) [12]. Cell

migration is mediated by cell adhesion components and involves upstream regulation

mostly through protein kinase-mediated signaling. We used siGENOME smartpool

libraries (Dharmacon) targeting all kinases (779) and phosphatases (198) and a custom established adhesome library composed all cell adhesion components and cytoskeletal regulators (576), in total 1,429 genes that were all screened in two independent experiments in duplicate plates (see schematic overview in Fig 1A).

Candidate genes affecting H1299 cell migration were identified based on statistical comparison with non-targeting siControl. Four main parameters (‘Net Area’, ‘Axial Ratio’, ‘Minor Axis’ and ‘Roughness’) that mainly determine cell behavior were selected for hit calling. P-values for each parameter and each gene were calculated using a two tailed t-test and the cut-off value for hit calling was set at P-value

< 0.001 in at least one of the four mentioned parameters and visually represented in scatter plots (see Fig 1B and full table Supplemental Table 1). This method allowed the selection of strong candidate genes that upon knockdown: enhance (ACVRL1, NGEF) or inhibit (SRPK1, FHOD1) migration; diminish directionality (IKBKE, CAPN7) or migration strategy (ROR1, GIT1) (Fig 1C and D). In total we identified 136 genes that affected H1299 cell migration. Unsupervised hierarchical clustering of all 136 genes and all eight migration parameters defined four large clusters of genes with common role in a particular migration strategies. For example, genes that strongly increased the roughness (cluster 1) of tracks also had decreased overall migratory capacity. Genes that do increase the overall migratory track (cluster 4) typically have a increase axial ratio, but not an increased roughness, indicating that these cells have increased cell migration speed and directionality (Fig 1E and SFig 2)

Figure 1: RNA-interference screen identifies novel tumor cell migration regulators. A) Schematic

representation screen set up and follow up. Starting upper left, clockwise: Screening of 1,429

individual smartpools (kinases, phosphatases and adhesion genes), making use of the

phagokinetic track assay (PKT) in H1299 to examine the effect of knockdown of a gene on cell migration. In a

deconvolution screen, 30 out 64 tested genes were validated with at least 3/4 single siRNAs. Expression

of validated genes was correlated with metastasis free survival in a breast cancer patient cohort, which

resulted in eight clinical relevant genes. B) For each tested smartpool in the primary screen the P-value is

calculated for each parameter. The extremes of the graph indicate the significant smartpools (P<0.001),

visualized as yellow and blue data points. A negative score indicates a decrease compared to siGFP

control. Examples of identified genes in the kinase and phosphatase library are shown in C) and

results for the adhesion related genes in D). Normalized values (± st dev) are written underneath each

image (full table in SFig 1). E) Unsupervised hierarchical clustering on the 136 identified genes revealed

signatures for specific migratory behavior such as increased Net Area combined with increased Axial

Ratio, or decreased Net Area combined with increased Roughness (full table SFig 2).

4

Inhibition of cell migration diminishes focal adhesion dynamics

To address off-target effects in the primary screen, we first performed a deconvolution PKT screen, testing four individual sequences next to the standard smartpool. Taking the likelihood of candidate metastasis genes in consideration, we selected 64 genes out of 136 candidate genes based on their strongest affect per parameter as well as their inhibitory effect on cell migration upon knockdown. In total, 30 high confidence genes were defined based on the observation that the effect in the primary screen was confirmed in at least 3/4 single sequences and smartpool in the secondary screen (Fig 2A). The strongest validated hits included overall inhibitors of cell migration (Net Area): LCK, MALT1, MYO9B, RAC1, ROS1, STYK1 and SRPK1;

genes involved in cell migration mode (Axial Ratio, Minor Axis and Roughness):

SH3KBP1, IKBKE, PVRL3, ITGB3BP and MYO15A. To confirm the effect on cell migration in the PKT assay, we selected all 30 high confidence genes and subjected them to an automated multiparametric high-content live cell imaging random cell migration assay that provided additional information on actual cell behavior during cell migration (Fig 2B). Thus, while depletion of SRPK1 and ITGB3BP both inhibited cell migration, ITGB3BP knockdown cell lines still had dynamic membrane ruffling, which was in agreement with an increase in Roughness value in the PKT assay, which was associated with enhanced membrane activity and formation of multiple protrusions. Dynamic focal adhesions (FA) are essential for cell migration and therefore we systematically monitored FA organization for these 30 high confidence genes (data not shown). In general, for genes that increased cell migration, FA size was decreased and for genes that upon knockdown inhibited migration, FA size was increased. The role in the control of FA dynamics was further evaluated for several genes in H1299 cells stably expressing GFP-tagged paxillin. In control conditions, in a timeframe of 60 min, a continuous polarized remodeling of FA occurred at the front of the cell. In contrast depletion of SRPK1 or MAP2K2, which both inhibited

Figure 2: Validated high confidence hits affect random cell migration in association with modulated focal adhesion dynamics. A) 64 selected genes were tested in a deconvolution PKT screen with 4 single siRNAs/genes. Validated high confidence genes (confirmed effect primary screen in at least 3/4 single siRNA and smartpool) are highlighted in yellow (increased compared to control) or blue (decreased).

B) Hits for each parameter were selected and tested in a random cell migration assay. C) Focal adhesion

(FA) dynamics were studied using TIRF imaging of GFP-paxillin (PXN) expressing H1299 cells. Decreased

cell migration after knockdown of SRPK1 and MAP2K2 was related to increased stabilization of FA as

indicated by the presence of ‘white’ FAs in the time color overlay. Knockdown of IKBKE, affecting Minor

Axis, does not affect cell migration per se, but resulted in increased cell area, reflecting the increase in

Minor Axis in the PKT screen.

4

cell migration, stabilized FA structures at the entire cell periphery and hardly any reorganization occurred. Yet knockdown of IKBKE, which caused a depolarization of lamilipodia formation (see Fig 2B) was associated with enhanced focal adhesion remodeling along the entire cell border (Fig 2C). Together these data indicated that the change in the cell migratory behavior by the different candidate migration genes is largely dominated by affecting cell matrix adhesion restructuring.

Genes affecting cell migration are associated with breast cancer metastasis free survival

Cell migration, together with enhanced proliferation, survival and invasion, is one of the hallmarks of cancer and potentially contributes to metastasis formation [20].

Figure 3: Clinical relevance of candidate tumor cell migration hits in breast cancer metastasis

free survival. A) Gene expression data of a lymph node negative breast cancer patient cohort (n=344)

without prior treatment was used. Expression of identified tumor cell migration genes was correlated

with metastasis free survival (MFS) and Hazard Ratio (HR) as well as Cox P-values were calculated

for estrogen receptor (ER) negative (n=123) and positive (n=221) BC patients, as well as all patients

combined. Eight out of 30 validated genes showed clinical relevance (P<0.05, highlighted in bold). B)

Kaplan-Meier curves are shown for genes with highest HR for MFS drawn after determination of the

optimal cut off point. Low expression is represented by blue line (0), and high expression by red line (1).

4

Some of our candidate genes have previously been described to be associated with cancer progression, including SHC1 [21], MSTR1 [22; 23] and ROR1 [24]. To make the clinical translation for our 30 high confidence genes we took advantage of a gene expression dataset from a 344 breast cancer (BC) patient cohort that were lymph node negative at time of presentation in the clinic [15]. We determined the correlation between expression levels of the identified genes and metastasis free survival and Hazard Ratio (HR) and Cox P-value were calculated. Out of our 30 high confidence genes, ITGB3BP, LCK, MAP2K2, MAP3K8, NEK2, ROS1, SHC1 and SRPK1 showed significant clinical association with disease progression. The expression level of those genes (low versus high, split by optimal cut-off point) significantly influenced the metastasis free survival (MFS) of either estrogen receptor (ER) negative (n=123), ER positive (n=221) or all profiled BC patients (Cox P-value < 0.05, table Fig 3A). Kaplan-Meier (KM) survival curves further stressed

Figure 4: SRPK1 protein levels correlate with poor prognosis in breast cancer patients. A) Tissue

micro array of lymph node negative breast cancer patients (n=562) without prior treatment was stained

for SRPK1 and scored. Examples of SRPK1 intensity weak (1), moderate (2) and strong (3) in the TMA

are shown for ER positive and ER negative samples. Scale bar 200 µm. B) Increased expression of

SRPK1 correlated with decreased metastasis free survival in ER positive patients (n=460, P<0.05 (i)),

but not ER negative patients (n= 102, (ii)). For all patients combined, SRPK1 expression was associated

with worse prognosis (P<0.05 (iii)).

the relation between the expression of hits found in the migration screen and poor clinical outcome for BC patients (Fig 3B). Increased expression of NEK2 and SRPK1 in ER positive BC patients is related to decreased MFS and expression of ITGB3BP and MAP3K8 in ER negative BC patients. High expression level of SHC1 is correlated with poor prognosis in all BC patients. Next to their expression in BC patient data, we also examined the expression of the tumor cell migration genes in a panel of 22 luminal-like and 13 basal-like human BC cell lines and calculated the expression ratio of basal- over luminal-like cell lines (SFig 3). Interestingly, most genes showed enhanced expression in basal-like cells, which are ER negative and more aggressive compared to luminal-like, mostly ER positive, cells.

Increased SRPK1 protein levels are correlated with poor prognosis in breast cancer patients

Recently RNA splicing has become of increasing interest in cancer progression [25]. Therefore we further focused on SRPK1, a splicing factor kinase that showed increased expression in the more aggressive basal-like cell lines. A role for SRPK1 in breast cancer cell migration and metastasis has not been demonstrated. Gene expression profiling revealed a correlation between SRPK1 mRNA expression level and BC disease outcome. To confirm this correlation on protein level, we performed a staining for SRPK1 of a tissue micro array containing samples of 562 lymph node negative BC patients. Intensity levels were scored (weak, moderate, strong (Fig 4A)) and correlated to disease progression. While only 19 % of the ER positive patients (n=460) scored positive for SRPK1 expression, the increased SRPK1 protein levels in these BC patients correlated significantly with shorter MFS. This was in full agreement with the gene expression analysis above (see Fig 3). In the smaller set of ER negative BC patients (n=102), such correlation was not observed, but here the majority of patients (61 %) showed already increased SRPK1 protein levels. When all patients were analyzed regardless of their ER status, increased SRPK1 protein expression was significantly associated with poor disease outcome (Fig 4B).

SRPK1 knockdown in ER negative breast cancer cells reduces their migratory capacity and metastatic potential

Since SRPK1 expression significantly correlated with poor MFS outcome in BC

patients we wanted to define the role of SRPK1 in breast cancer metastasis. First we

determined SRPK1 protein levels in a panel of human BC cell lines. Although SRPK1

protein was detected in all cell lines, its expression was significantly higher in the

4

Figure 5: Knockdown of SRPK1 in human ER negative BC cells blocks cell migration and results in a

more epithelial-like morphology. A) Western blot analysis showed the increased expression of SRPK1

in ER negative human breast cancer cells compared to ER positive cells (P<0.05). B) Transient knock-

down of SRPK1 in MDA-MB-231 and MDA-MB-417.5 blocked cell migration in a PKT assay. C) Transient

knockdown inhibited random cell migration of MDA-MB-231 and -417.5 cells (P<0.01). D) Knockdown

MDA-MB-417.5 cells were stained for actin with rhodamin phalloidin (red) and pPXN to visualize focal

adhesions (green). After depletion of SRPK1, cells had a more epithelial-like morphology.

more aggressive ER negative basal-like BC cells (Fig 5A). Next we determined the role of SRPK1 in cell migration in the ER negative and highly migratory MDA-MB-231 cell line and its lung metastatic variant MDA-MB-417.5. Depletion of SRPK1 significantly decreased the migration velocity in both cell lines in the PKT assay (Fig 5B) as well as the live cell imaging random cell migration assay (Fig 5C). This decrease in cell migration was accompanied by a switch from a mesenchymal phenotype with strong protrusion formation containing multiple focal adhesions under control conditions,

Figure 6: SRPK1 is required for breast cancer metastasis in vivo. A) 4T1 cells were transduced with

lentiviral shRNA specific for two independent sequences of SRPK1 (shSRPK1-A and SRPK-B) and a non-

targeting shRNA (shCtrl). SRPK1 knockdown was confirmed by Western Blot. B) 4T1 cells were injected

in 6-8 week old Rag2

gc

mice (n=10-11) and tumor growth was followed over time. Tumor growth

(Bi) and weight (Bii) was similar in all groups. C) Three weeks post injection, mice were sacrificed and

lungs were isolated. GFP positive tumor cells in the lung were visualized (Ci)). Lungs were injected with

Indian ink (Cii) and surface metastases were quantified. Knockdown of SRPK1 significantly reduced

the number of metastases in lung (P<0.01(Ciii)). The small lobe was processed for immunohistological

analysis and stained with H&E (iv). D) Within the BC patient gene expression data set, only patients who

metastasize to a single site were selected and this was related to SRPK1 expression (i). KM-curves between

patients without relapse and patients with lung-relapse (optimal cutpoint) were compared. High SRPK1

expression is associated with significantly more lung metastasis (ii).

4

to a resting cell phenotype with no membrane protrusion and strong focal adhesion at the cell periphery under SRPK1 knockdown condition, similar as observed for H1299 cells (compare Fig 5D and Fig 2C).

Finally we determined the role for tumor cell associated SRPK1 expression in BC metastasis in vivo. We used the well-characterized mouse ER negative basal- like mammary carcinoma cell line 4T1 that metastases to lung in a orthotopic tumor metastasis mouse model. We generated two individual lentiviral shRNA stable knockdown 4T1 cell lines that showed over 85 % depletion of SRPK1 protein levels (Fig 6A). SRPK1 knock down did not affect tumor primary tumor growth at the mammary fat pad (Fig 6B). In contrast, while control 4T1 cells formed numerous metastatic foci in lungs, an almost complete eradication of the metastatic potential of 4T1 cells to colonize the lungs was observed for both shRNA constructs (Fig 6C). Given this strong effect of SRPK1 on lung colonization, within the available BC patient gene expression data, we selected patients that showed metastasis to a single site and wondered whether homing of the cells to lung was related to SRPK1 expression (Fig 6Di). Primary breast tumors with increased SRPK1 levels showed preference for homing to lung and brain. Comparison of KM-curves between patients without relapse and patients with lung-relapse (optimal cut-off point) demonstrated that high SRPK1 expression is significantly associated with lung metastasis (P<0.05, Fig 6Dii). All data combined indicates that one of our newly identified tumor cell migration regulators SRPK1 is required for BC progression and metastasis with strong clinical prognosis. Our RNA-interference based screening combined with our translational efforts is a strong and successful strategy to identify novel candidate anti-cancer drug targets.

Discussion

Here we used a multiparametric quantitative high content imaging-based RNA- interference screen to identify novel regulators of tumor cell migration. Since tumor cell migration is likely controlled by enhanced oncogenic (receptor-mediated) signaling that drives the dynamic rearrangement of the actin cytoskeletal network and cell matrix adhesions we focused on kinases, phosphatases and all adhesion- related genes. Our work led to the identification of eight clinically relevant candidate metastasis genes that affect cell migration and focal adhesion turnover, of which SRPK1 plays a major role in breast cancer metastasis to lung.

The phagokinetic track assay allows us to quantitatively assess different

modes of cell migration. We focused in particular on hits that affected the

parameters Net Area, Minor Axis, Axial Ratio and Roughness. These turned out to be the most independent parameters associated with different migratory phenotypes in relation to either enhanced or decreased migration, which is essential for a translation to cancer metastasis. Most of the high confidence hits decreased the overall Net Area, indicative of inhibition of cell migration as validated by the random cell migration assay. Interestingly, several receptor tyrosine kinases (LTK, MST1R/

RON, ROS1 and STYK1) and likely downstream effectors including adapters (SHC1), non-receptor tyrosine kinase or MAPK family members (LCK, MAP2K2, MAP3K8 and MAK3K14) and Rho GTPases and upstream regulators (RAC1, NGEF and MYO9B) were identified. While none of these genes affected cell survival, they apparently seem important regulators in cancer cell motility in H1299 cells. While only two out of the five genes that we re-evaluated for Roughness were validated, other genes in the primary screen including NGEF, RBJ, ARHGEF12, PTRH2, GJA5, GPC1, GSTM1, HRAS, ITGAD, ITGB3BP, LGALS3, MARCKS, MYO15A, PELO, PPP1R13B and TAF1L all resulted in increased Roughness (Fig 1E and SFig 1). Since enhanced Roughness will involve enhanced lamellar activity, it is remarkable that several of the identified Roughness hits have previously been linked to Rac1, an important driver of lamellipodia formation. These genes include MARCKS [26], HRAS [27; 28], ARHGEF12 [29] and NGEF [30]. Importantly, ITG3BP (also known as TAP20) had the strongest effect on Roughness and blocked cell migration upon knockdown (Fig 2) and its expression levels in breast cancer were associated with poor MFS outcome in ER negative breast cancer (see Fig 3).

Dynamic remodeling of cell matrix adhesion complexes/focal adhesions is essential for cell migration [31]. Indeed we found that almost all genes that affected cell migration also affected the organization of FA in the H1299 cells (data not shown). Previously 44 kinases/phosphatases/adhesion-related genes were already identified in a RNA-interference screen in HeLa cells that affected the structural organization of FAs [32]. Hardly any overlap was observed between the different genes that affect cell migration in H1299 cells and FA organization in HeLa cells.

This is most likely related to large differences in migratory behavior between our

highly motile H1299 cells and the more static HeLa cells as well as a role for different

oncogenic programs in these cell types. Regardless of cell differences, IKBKE

(inhibitor of NFkB kinase subunit epsilon) was identified in both screens as a strong

hit. Interestingly, it was also identified in a separate screen in MCF10-A and impaired

the collective cell migration [13]. Detailed analysis of FA dynamics in H1299 cells

demonstrated that knockdown of IKBKE strongly affected the Axial Ratio and caused

4

a circular cell track, indicating a loss of persistent cell migration, disturbed the polarized distribution of FA, but not the dynamics of the FA. This in contrast to SRPK1 and MAPK2K2 (also known as MEK2) that showed almost complete inhibition of FA dynamics in association with loss of any cell motility. While our screens provides a detailed comprehensive information source on the role of individual genes in cell migration and the relation to FA organization, it remains to be established whether the stable FA structures are cause or consequence of the inhibition of cell migration.

An important strength of our current work is that we systematically determined the relationship between genes that affect tumor cell migration and metastatic disease. We have used expression profiling of a breast cancer (BC) patient cohort to determine the clinical relevance of the hits found in the migration screen. We established that ITGB3BP, LCK, MAP2K2, MAP3K8, NEK2, ROS1, SHC1 and SRPK1 are strongly associated with metastasis free survival. Some of these genes are established modulators of BC progression. In particular, activation of the signaling adapter Src homology 2 domain containing transforming protein 1 (SHC1), which acts directly downstream of extracellular signals, is associated with poor BC patient prognosis [33]. Lymphocyte-specific protein tyrosine kinase (LCK) was highly expressed in estrogen receptor (ER) negative tumors compared to ER positive tumors.

High membrane LCK expression was significantly associated with improved survival [34]. This is in agreement with our data in which we also showed that patients with high LCK levels have increased MFS compared to low levels of LCK (Fig 3A). Given the fact that LCK inhibited migration in H1299 cells, this contradictory association with MFS is possibly related to a different cancer cell type or differential regulation of LCK activity in vitro versus the in vivo condition.

SRSF protein kinase 1 (SRPK1) strongly associated with MFS in BC patients, which was confirmed in orthotopic breast cancer models. SRPK1 overexpression has been associated with increasing tumor grade in several other cancers [35].

SRPK1 belongs to the serine-arginine protein kinase family, a relative new subfamily

within the serine-threonine kinases. SRPK1 has a role in both constitutive as

well as alternative splicing by regulating intracellular location of splicing factors

[36]. Downregulation of SRPK1 resulted in altered splicing of MAP2K2 leading to

imbalanced mitogen-activated protein kinase pathway signaling in various tumor cell

lines [35]. Interestingly, we also identified MAP2K2 as a migratory regulator in our

screen. Knockdown of both SRPK1 as well as MAP2K2 resulted in inhibition of cell

migration. Intriguingly, in contrast to SRPK1, high levels of MAP2K2 were associated

with better BC patient survival, although this could be related to other splice variants.

More recently, SRPK1 was shown to regulate VEGF splicing in podocytes and thereby induced angiogenesis in association with cancer progression [37]. However, in our hands no difference in the CD31-positive blood vessel density was observed in SRPK1 depleted tumors, excluding the possibility of perturbed angiogenesis (SFig 4).

SRPK1 is reported to be part of so called spliceosomes. Within these organelles, genes are regulated through the splicing of pre-mRNA into mRNA. While the spliceosome consists of 141 core components, it co-purified with over 200 proteins, which included the catalytic snRNPs (small nuclear ribonucleoproteins) and members of the SRSF protein family [38]. SRPK1 is one of these noncore proteins, which are the presumed link between the spliceosome and other cellular machineries such as transcription factors [39]. SRPK1 regulates SRSF proteins by phosphorylation upon which a nuclear translocation of SRSF proteins is induced resulting in spliceosome assembly [40]. The best described interaction is that of SRPK1 and SRSF1. SRSF1 is reported as a oncogenic splicing factor, which is a direct transcriptional target of MYC and is frequently upregulated in cancer [41]. In addition, SRSF1 regulates apoptosis by splicing of BIM and BIN1 into non pro-apoptotic isoforms and proliferation to promote mammary epithelial cell transformation [42].

This indicates that splicing events are underlying cancer initiation and progression.

This was also highlighted in a recent study in which an epithelial-to-mesenchymal transition (EMT)-driven alternative splicing program was shown to be involved in human BC and modulation of cellular phenotype [43]. We therefore anticipate that SRPK1-dependent metastasis formation is linked to alternatively splicing of genes that (in)directly affect cancer cell migratory and invasive progams. Further detailed next generation RNA-sequencing is required to identify the gene networks that are modulated by SRPK1 through splicing.

In summary, using a high-content imaging-based RNA-interference migration screen 30 high confidence tumor cell migration regulators were identified.

Additionally, we identified the splicing related factor SRPK1 as an important clinical

relevant metastasis modifying factor. We showed that high SRPK1 levels in BC

patients was associated with poor disease prognosis. Importantly, despite clear

differences between cell culture conditions and the 3D microenvironment in the

intact tumor tissue, a systematic analysis of in vitro screening data with clear clinical

cancer progression end-point is a powerful method to identify novel likely drug target

candidates.

4

Acknowledgements

We thank Zvi Kam for technical help during image acquisition and helpful discussions.

We thank Vasiliki-Maria Rogkoti for help with the WB samples of the human BC cell

lines. This work was financially supported by grants from the Dutch Cancer Society

(UL2007-3860), the EU FP7 Health Programs MetaFight project (Grant agreement

no.201862) and Systems Microscopy NoE project (Grant agreement no.258068)

and EMBO short-term fellowship (ASTF 109-2009).

Supplemental data

4

Supplemental Figure 1:

Identification of 136

novel cell migration

regulators. Details 136 genes

identified in the screening of

1,429 individual smartpools

(kinases, phosphatases and

adhesion genes), making

use of the phagokinetic track

assay (PKT) in H1299 to

examine the effect of knock-

down of a gene on cell

migration. Normalized values

are calculated for each gene

and each parameter, with

siControl set to 1. Red

indicates decrease compared

to control, green indicated

increase. P-values are calcu-

lated using a two tailed t-test.

Supplemental Figure 2: Unsupervised hierarchical clustering 136 genes identified in the

primary screen. Unsupervised hierarchical clustering on the 136 identified genes revealed 4 clusters with

signatures for specific migratory behavior such as increased Net Area combined with increased Axial

Ratio (cluster 4), or decreased Net Area combined with increased Roughness (cluster 1).

4

Supplemental Figure 3: Breast cancer cell lines data mRNA expression migration hits.

The expression of the clinical relevant tumor cell migration genes in a panel of 22 luminal-like and 13 basal-like human BC cell lines was determined and the expression ratio of basal- over luminal-like cell lines was calculated.

Supplemental Figure 4: SRPK1 knockdown does not affect tumor vascularization.

Immuno-histological analysis of CD31 expression in 4T1 control and SRPK1 knockdown

tumors revealed no differences in tumor vascularization.

References

1. Zaidel-Bar, R., & Geiger, B. (2010). The switchable integrin adhesome. J Cell Sci, 123(Pt 9), 1385- 1388.

2. Zaidel-Bar, R., Itzkovitz, S., Ma’ayan, A., Iyengar, R., & Geiger, B. (2007). Functional atlas of the integrin adhesome. Nat Cell Biol, 9(8), 858-867.

3. Hynes, N. E., & Lane, H. A. (2005). ERBB receptors and cancer: the complexity of targeted inhibitors.

Nat Rev Cancer, 5(5), 341-354.

4. Rijken, P. J., Hage, W. J., van Bergen en Henegouwen, P. M., Verkleij, A. J., & Boonstra, J. (1991).

Epidermal growth factor induces rapid reorganization of the actin microfilament system in human A431 cells. J Cell Sci, 100 ( Pt 3), 491-499.

5. Xue, C., Wyckoff, J., Liang, F., Sidani, M., Violini, S., Tsai, K. L., Zhang, Z. Y., Sahai, E., Condeelis, J., &

Segall, J. E. (2006). Epidermal growth factor receptor overexpression results in increased tumor cell motility in vivo coordinately with enhanced intravasation and metastasis. Cancer Res, 66(1), 192-197.

6. Zhang, H. J., Wang, H. Y., Zhang, H. T., Su, J. M., Zhu, J., Wang, H. B., Zhou, W. Y., Zhang, H., Zhao, M.

C., Zhang, L., & Chen, X. F. (2011). Transforming growth factor-beta1 promotes lung adenocarcinoma invasion and metastasis by epithelial-to-mesenchymal transition. Mol Cell Biochem, 355(1-2), 309-314.

7. Rosenthal, D. T., Iyer, H., Escudero, S., Bao, L., Wu, Z., Ventura, A. C., Kleer, C. G., Arruda, E. M., Garikipati, K., & Merajver, S. D. (2011). p38gamma promotes breast cancer cell motility and metastasis through regulation of RhoC GTPase, cytoskeletal architecture, and a novel leading edge behavior. Cancer Res, 71(20), 6338-6349.

8. Danen, E. H. (2005). Integrins: regulators of tissue function and cancer progression. Curr Pharm Des, 11(7), 881-891.

9. Di Stefano, P., Leal, M. P., Tornillo, G., Bisaro, B., Repetto, D., Pincini, A., Santopietro, E., Sharma, N., Turco, E., Cabodi, S., & Defilippi, P. (2011). The adaptor proteins p140CAP and p130CAS as molecular hubs in cell migration and invasion of cancer cells. Am J Cancer Res, 1(5), 663-673.

10. van Nimwegen, M. J., Verkoeijen, S., van Buren, L., Burg, D., & van de Water, B. (2005). Requirement for focal adhesion kinase in the early phase of mammary adenocarcinoma lung metastasis formation.

Cancer Res, 65(11), 4698-4706.

11. Zhang, X. H., Wang, Q., Gerald, W., Hudis, C. A., Norton, L., Smid, M., Foekens, J. A., & Massague, J. (2009). Latent bone metastasis in breast cancer tied to Src-dependent survival signals. Cancer Cell, 16(1), 67-78.

12. Naffar-Abu-Amara, S., Shay, T., Galun, M., Cohen, N., Isakoff, S. J., Kam, Z., & Geiger, B. (2008).

Identification of novel pro-migratory, cancer-associated genes using quantitative, microscopy-based screening. PLoS One, 3(1), e1457.

13. Simpson, K. J., Selfors, L. M., Bui, J., Reynolds, A., Leake, D., Khvorova, A., & Brugge, J. S. (2008).

Identification of genes that regulate epithelial cell migration using an siRNA screening approach. Nat Cell Biol, 10(9), 1027-1038.

14. van Roosmalen, W., Le Devedec, S. E., Zovko, S., de Bont, H., & van de Water, B. (2011). Functional screening with a live cell imaging-based random cell migration assay. Methods Mol Biol, 769, 435-448.

15. Foekens, J. A., Atkins, D., Zhang, Y., Sweep, F. C., Harbeck, N., Paradiso, A., Cufer, T., Sieuwerts, A.

M., Talantov, D., Span, P. N., Tjan-Heijnen, V. C., Zito, A. F., Specht, K., Hoefler, H., Golouh, R., Schittulli, F., Schmitt, M., Beex, L. V., Klijn, J. G., & Wang, Y. (2006). Multicenter validation of a gene expression-based prognostic signature in lymph node-negative primary breast cancer. J Clin Oncol, 24(11), 1665-1671.

16. Wang, Y., Klijn, J. G., Zhang, Y., Sieuwerts, A. M., Look, M. P., Yang, F., Talantov, D., Timmermans, M., Meijer-van Gelder, M. E., Yu, J., Jatkoe, T., Berns, E. M., Atkins, D., & Foekens, J. A. (2005). Gene- expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet, 365(9460), 671-679.

17. Yu, J. X., Sieuwerts, A. M., Zhang, Y., Martens, J. W., Smid, M., Klijn, J. G., Wang, Y., & Foekens, J.

4

A. (2007). Pathway analysis of gene signatures predicting metastasis of node-negative primary breast cancer. BMC Cancer, 7, 182.

18. Hollestelle, A., Nagel, J. H., Smid, M., Lam, S., Elstrodt, F., Wasielewski, M., Ng, S. S., French, P. J., Peeters, J. K., Rozendaal, M. J., Riaz, M., Koopman, D. G., Ten Hagen, T. L., de Leeuw, B. H., Zwarthoff, E.

C., Teunisse, A., van der Spek, P. J., Klijn, J. G., Dinjens, W. N., Ethier, S. P., Clevers, H., Jochemsen, A. G., den Bakker, M. A., Foekens, J. A., Martens, J. W., & Schutte, M. (2010). Distinct gene mutation profiles among luminal-type and basal-type breast cancer cell lines. Breast Cancer Res Treat, 121(1), 53-64.

19. Le Devedec, S. E., van Roosmalen, W., Maria, N., Grimbergen, M., Pont, C., Lalai, R., & van de Water, B. (2009). An improved model to study tumor cell autonomous metastasis programs using MTLn3 cells and the Rag2(-/-) gammac (-/-) mouse. Clin Exp Metastasis, 26(7), 673-684.

20. Hanahan, D., & Weinberg, R. A. (2011). Hallmarks of cancer: the next generation. Cell, 144(5), 646- 674.

21. Rajendran, M., Thomes, P., Zhang, L., Veeramani, S., & Lin, M. F. (2010). p66Shc--a longevity redox protein in human prostate cancer progression and metastasis : p66Shc in cancer progression and metastasis. Cancer Metastasis Rev, 29(1), 207-222.

22. Liu, H. S., Hsu, P. Y., Lai, M. D., Chang, H. Y., Ho, C. L., Cheng, H. L., Chen, H. T., Lin, Y. J., Wu, T. J., Tzai, T. S., & Chow, N. H. (2010). An unusual function of RON receptor tyrosine kinase as a transcriptional regulator in cooperation with EGFR in human cancer cells. Carcinogenesis, 31(8), 1456-1464.

23. Wagh, P. K., Zinser, G. M., Gray, J. K., Shrestha, A., & Waltz, S. E. (2012). Conditional deletion of beta-catenin in mammary epithelial cells of Ron receptor, Mst1r, overexpressing mice alters mammary tumorigenesis. Endocrinology, 153(6), 2735-2746.

24. Rebagay, G., Yan, S., Liu, C., & Cheung, N. K. (2012). ROR1 and ROR2 in Human Malignancies:

Potentials for Targeted Therapy. Front Oncol, 2, 34.

25. Padgett, R. A. (2012). New connections between splicing and human disease. Trends Genet, 28(4), 147-154.

26. Kalwa, H., Sartoretto, J. L., Sartoretto, S. M., & Michel, T. (2012). Angiotensin-II and MARCKS: a hydrogen peroxide- and Rac1-dependent signaling pathway in vascular endothelium. J Biol Chem.

27. Appledorn, D. M., Dao, K. H., O’Reilly, S., Maher, V. M., & McCormick, J. J. (2010). Rac1 and Cdc42 are regulators of HRasV12-transformation and angiogenic factors in human fibroblasts. BMC Cancer, 10, 13.

28. Lim, I. K., Hong, K. W., Kwak, I. H., Yoon, G., & Park, S. C. (2001). Translocational inefficiency of intracellular proteins in senescence of human diploid fibroblasts. Ann N Y Acad Sci, 928, 176-181.

29. Jaiswal, M., Gremer, L., Dvorsky, R., Haeusler, L. C., Cirstea, I. C., Uhlenbrock, K., & Ahmadian, M.

R. (2011). Mechanistic insights into specificity, activity, and regulatory elements of the regulator of G-protein signaling (RGS)-containing Rho-specific guanine nucleotide exchange factors (GEFs) p115, PDZ-RhoGEF (PRG), and leukemia-associated RhoGEF (LARG). J Biol Chem, 286(20), 18202-18212.

30. Shamah, S. M., Lin, M. Z., Goldberg, J. L., Estrach, S., Sahin, M., Hu, L., Bazalakova, M., Neve, R.

L., Corfas, G., Debant, A., & Greenberg, M. E. (2001). EphA receptors regulate growth cone dynamics through the novel guanine nucleotide exchange factor ephexin. Cell, 105(2), 233-244.

31. Huttenlocher, A., & Horwitz, A. R. (2011). Integrins in cell migration. Cold Spring Harb Perspect Biol, 3(9), a005074.

32. Winograd-Katz, S. E., Itzkovitz, S., Kam, Z., & Geiger, B. (2009). Multiparametric analysis of focal adhesion formation by RNAi-mediated gene knockdown. J Cell Biol, 186(3), 423-436.

33. Ursini-Siegel, J., & Muller, W. J. (2008). The ShcA adaptor protein is a critical regulator of breast cancer progression. Cell Cycle, 7(13), 1936-1943.

34. Elsberger, B., Fullerton, R., Zino, S., Jordan, F., Mitchell, T. J., Brunton, V. G., Mallon, E. A., Shiels, P. G.,

& Edwards, J. (2010). Breast cancer patients’ clinical outcome measures are associated with Src kinase family member expression. Br J Cancer, 103(6), 899-909.

35. Hayes, G. M., Carrigan, P. E., & Miller, L. J. (2007). Serine-arginine protein kinase 1 overexpression is

associated with tumorigenic imbalance in mitogen-activated protein kinase pathways in breast, colonic, and pancreatic carcinomas. Cancer Res, 67(5), 2072-2080.

36. Ghosh, G., & Adams, J. A. (2011). Phosphorylation mechanism and structure of serine-arginine protein kinases. FEBS J, 278(4), 587-597.

37. Amin, E. M., Oltean, S., Hua, J., Gammons, M. V., Hamdollah-Zadeh, M., Welsh, G. I., Cheung, M. K., Ni, L., Kase, S., Rennel, E. S., Symonds, K. E., Nowak, D. G., Royer-Pokora, B., Saleem, M. A., Hagiwara, M., Schumacher, V. A., Harper, S. J., Hinton, D. R., Bates, D. O., & Ladomery, M. R. (2011). WT1 mutants reveal SRPK1 to be a downstream angiogenesis target by altering VEGF splicing. Cancer Cell, 20(6), 768-780.

38. Hegele, A., Kamburov, A., Grossmann, A., Sourlis, C., Wowro, S., Weimann, M., Will, C. L., Pena, V., Luhrmann, R., & Stelzl, U. (2012). Dynamic protein-protein interaction wiring of the human spliceosome.

Mol Cell, 45(4), 567-580.

39. Agafonov, D. E., Deckert, J., Wolf, E., Odenwalder, P., Bessonov, S., Will, C. L., Urlaub, H., & Luhrmann, R. (2011). Semiquantitative proteomic analysis of the human spliceosome via a novel two-dimensional gel electrophoresis method. Mol Cell Biol, 31(13), 2667-2682.

40. Giannakouros, T., Nikolakaki, E., Mylonis, I., & Georgatsou, E. (2011). Serine-arginine protein kinases:

a small protein kinase family with a large cellular presence. FEBS J, 278(4), 570-586.

41. Das, S., Anczukow, O., Akerman, M., & Krainer, A. R. (2012). ONCOGENIC SPLICING FACTOR SRSF1 IS A CRITICAL TRANSCRIPTIONAL TARGET OF MYC. Cell Rep, 1(2), 110-117.

42. Anczukow, O., Rosenberg, A. Z., Akerman, M., Das, S., Zhan, L., Karni, R., Muthuswamy, S. K., &

Krainer, A. R. (2012). The splicing factor SRSF1 regulates apoptosis and proliferation to promote mammary epithelial cell transformation. Nat Struct Mol Biol, 19(2), 220-228.

43. Shapiro, I. M., Cheng, A. W., Flytzanis, N. C., Balsamo, M., Condeelis, J. S., Oktay, M. H., Burge, C. B.,

& Gertler, F. B. (2011). An EMT-driven alternative splicing program occurs in human breast cancer and

modulates cellular phenotype. PLoS Genet, 7(8), e1002218.