- Email: [email protected]
EphA3 promotes malignant transformation of colorectal epithelial cells by upregulating oncogenic pathways
Accepted Manuscript EphA3 promotes malignant transformation of colorectal epithelial cells by upregulating oncogenic pathways Mingqi Li, Cheng Yang, Xin Liu, Liang Yuan, Fubin Zhang, Muhong Wang, Dazhuang Miao, Xinyue Gu, Shixiong Jiang, Binbin Cui, Jinxue Tong, MD, PhD, Zhiwei Yu, MD, PhD PII:
S0304-3835(16)30611-5
DOI:
10.1016/j.canlet.2016.10.004
Reference:
CAN 13061
To appear in:
Cancer Letters
Received Date: 10 July 2016 Revised Date:
26 September 2016
Accepted Date: 2 October 2016
Please cite this article as: M. Li, C. Yang, X. Liu, L. Yuan, F. Zhang, M. Wang, D. Miao, X. Gu, S. Jiang, B. Cui, J. Tong, Z. Yu, EphA3 promotes malignant transformation of colorectal epithelial cells by upregulating oncogenic pathways, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.10.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
EphA3 promotes malignant transformation of colorectal epithelial cells by upregulating oncogenic pathways
RI PT
Mingqi Li*, Cheng Yang*, Xin Liu*, Liang Yuan, Fubin Zhang, Muhong Wang,
Dazhuang Miao, Xinyue Gu, Shixiong Jiang, Binbin Cui, Jinxue Tong#, Zhiwei Yu#
SC
Affiliation: Department of Colorectal Surgery, The Affiliated Tumor Hospital of
M AN U
Harbin Medical University, Harbin, China
*These authors contributed equally to this work.
# Correspondence to:
TE D
ZhiWei Yu, MD, PhD, Department of Colorectal Surgery, The Affiliated Tumor Hospital of Harbin Medical University, Harbin, China Tel.: 86-0451-86298096, E-mail: [email protected]
EP
JinXue Tong, MD, PhD, Department of Colorectal Surgery, The Affiliated Tumor Hospital of Harbin Medical University, Harbin, China
AC C
Tel.: 86-0451-86298096, E-mail: [email protected]
1
ACCEPTED MANUSCRIPT
ABSTRACT Ephrin Type-A Receptor 3 (EphA3) belongs to the ephrin receptor subfamily of the
RI PT
protein tyrosine kinase family, and plays an important role in embryogenesis and neurogenesis. This study aimed to investigate the role of EphA3 in promoting malignant
SC
transformation of colorectal epithelial cells, and explore underlying molecular
M AN U
mechanisms. Colorectal cancer tissue specimens from 68 patients were analyzed for EphA3 expression. EphA3 expression levels were manipulated in rat colon epithelial cell lines. We found that EphA3 expression level in tumor tissues was associated with patient age (P = 0.015), tumor differentiation (P = 0.001), and lymph node metastasis (P =
TE D
0.039). Overexpression of EphA3 and its constitutively active mutants promoted colony formation, migration and invasion, and tumorigenicity of colon epithelial cells in nude
EP
mice. The cDNA and lncRNA microarray profiling data revealed that differentially
AC C
expressed genes and lncRNAs in EphA3 or mutant-transfected cells were associated with cell proliferation, invasion and angiogenesis. Our findings reveal the mechanisms underlying the oncogenic activities of EphA3 in colorectal cells, which could provide novel targets for the prevention, early diagnosis, and treatment of colorectal cancer. Keywords: EphA3; transformation; mutation; tumorigenicity; colorectal epithelial cells
2
ACCEPTED MANUSCRIPT
1. Introduction Colorectal cancer (CRC) is the third most common cancer in men and the second most
RI PT
common cancer in women worldwide, with over 1.2 million new cancer cases and more than 600,000 cancer-related deaths estimated globally in 2008[1–4]. The 5-year survival [5]
. Both genetic and environmental factors are
SC
rate of CRC with late-stage is only 12%
M AN U
involved in the development of CRC but the underlying mechanisms remain elusive [6–7]. Identification of novel diagnostic biomarkers and effective therapeutic targets for CRC is urgently needed.
Erythropoietin-producing hepatocellular (Eph) receptors, the largest known family of
TE D
tyrosine kinase receptors, mediate cell compartmentalization and directional cell migration during embryonic development[8]. Eph receptor subfamily consists of 16
EP
members that are divided into two groups based on sequence identity: EphA1–10 and
AC C
EphB1–6, which play important roles in cancer invasiveness [9-10]. The action of EphA3 in cancer seems contradictory: although tumor suppressor properties of EphA3 have been reported tumors
[8,11,12]
, EphA3 functions as an oncogene in different solid and hematopoietic
[13,14]
, and has been implicated in maintaining tumor-initiating cells in
glioblastoma and leukemia [13,15]. A previous study demonstrated that EphA3 is the sixth most frequently mutated gene in CRC [14]. However, the role of EphA3 in CRC has been
3
ACCEPTED MANUSCRIPT
poorly investigated. In this study, we employed microarray, proteomics and bioinformatics analyses to
RI PT
investigate the role of EphA3 in malignant transformation of colorectal epithelial cells. Our data demonstrate the role of EphA3 in upregulating angiogenesis-related
SC
pathways in CRC and provide novel insight into the mechanisms of CRC
2. Materials and Methods 2.1 Tissue specimens
M AN U
tumorigenesis.
TE D
Primary tumor specimens of colorectal cancer and adjacent normal mucosa were harvested from 68 patients who had undergone colectomy for pathologically
EP
diagnosed CRC at the Affiliated Tumor Hospital of Harbin Medical University (Harbin,
AC C
China) between March 2012 and March 2013. All tissue samples were immediately snap-frozen in liquid nitrogen a n d stored at -80°C. None of the patients received adjuvant chemoradiotherapy before surgery. The study protocol was approved by the Ethics Committee of Harbin Medical University, and all patients provided written informed consent.
4
ACCEPTED MANUSCRIPT
2.2 Immunohistochemistry Paraffin-embedded tissue blocks from 68 patients were retrieved from the Pathology
RI PT
Department, and 5 µm thick sections were prepared for standard immunohistochemistry with EnVisionTM immunohistochemistry methods. Briefly, the tissue sections were
SC
deparaffinized in xylene and re-hydrated through an ethanol gradient. Samples were
M AN U
blocked in 20% normal serum, and incubated overnight at 4°C with a primary antibody against EphA3 (sc-920, Santa Cruz Biotechnology; Santa Cruz, CA, USA) at a dilution of 1:100. The sections were washed with phosphate-buffered saline (PBS) and then incubated with biotin- or fluorescence-labeled secondary antibody at 4°C for 2 h. The
TE D
color reaction was processed with 3,3'-diaminobenzidine (DAB) solution, and then
AC C
2.3 Cell culture
EP
reviewed under a light microscope or inverted fluorescence microscope.
Rat colorectal YAMC, IMCE, YAMC-Ras, and IMCE-Ras cell lines were kindly provided by Dr. Zhenfeng Zhang of Ireland Cancer Center, Case Western Reserve University (Cleveland, OH, USA). These cell lines were generated from colonic epithelia of F1 immorto-APCmin/+ mouse hybrids [16]. The cells were maintained in RPMI 1640 medium supplemented with 5% heat-inactivated fetal bovine serum (FBS), 5 U/ml
5
ACCEPTED MANUSCRIPT
murine interferon gamma (IFN-γ), 100 U/ml penicillin and streptomycin, 5 µg/ml insulin, 5 µg/ml transferrin, and 5 ng/ml selenium acid in a humidified incubator with 5% CO2 at
RI PT
37°C.
SC
2.4 Reverse transcription PCR
M AN U
Total RNA was isolated from colorectal epithelial cell lines using Trizol reagent (Invitrogen, Carlsbad, CA, USA) and reverse transcribed into cDNA using Superscript First-Strand Synthesis System (Invitrogen) according to the manufacturer’s instructions. The
primers
were
as
follows:
and
5′5′-
TE D
ACTGGAATGGATTGTCAGCTCTCCATCCTCCTCCTTCTC-3′
EphA3
ACTCTCGAGTCAcagatcctcttctgagatgagtttttgttc(Myc)GAACACGGGAACTGGGCCA
EP
TTCTTTGATTGCG-3′; GAPDH 5′-GCCAAAAGGGTCATCATCTC -3′ and 5′-
AC C
GTAGAG GCAGGGATGATGTTC-3′. PCR conditions were initial denaturation at 95°C for 2 min, followed by 30 cycles of amplification at 95°C for 30 s, 55°C for 45 s, and 72°C for 1 min, and a final extension at 72°C for 15 min.
2.5 Western blot analysis
6
ACCEPTED MANUSCRIPT
Whole cell lysates were prepared and quantified using standard protocols, and then separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
RI PT
and electrophoretically transferred to nitrocellulose membranes. The membranes were blocked in blocking buffer and incubated with monoclonal anti-Myc antibody (Cell
SC
Signaling Technology, Danvers, MA, USA). The membranes were stripped and
M AN U
incubated with monoclonal anti-GAPDH antibody (Cell Signaling) to confirm equal loading.
2.6 Plasmid construction and transfection
TE D
A plasmid carrying mutated EphA3 cDNA was constructed using a GeneArt® SiteDirected Mutagenesis System (Invitrogen) according to the manufacturer’s instructions,
(impacts
the
ligand
domain)
was
constructed
AC C
cDNA
EP
and named pcDNA3.1-EphA3-Myc. The expression vector carrying EphA3-T37K
TCAATCTACTGGATTCAAAAAAAATTCAAGGGGA-3′
using and
primers
5'5'-
TTTTGAATCCAGTAGATTGACTTCATTGGA-3', the expression vector carrying EphA3-S792P (impacts the kinase domain) sequence was constructed using primers 5'AGATCCCAATCAGGTGGACACCACCAGAAGCTAT-3'
and
5'-
TGTCCACCTGATTGGGATCTTCCCTCCTCT-3'. All constructed plasmids were
7
ACCEPTED MANUSCRIPT
verified by DNA sequencing. IMCE cells were transfected with the plasmids using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. After 24 h,
RI PT
cells were passaged at 1:5 and selected with G418 (Sigma; St Louis, MO, USA) for 2 weeks at an increasing concentration from 200 to 800 µg/ml. The stably transfected
M AN U
SC
clones were harvested and expanded for subsequent experiments.
2.7 Cell viability assay
Cell viability was assayed using the CellTiter 96® AQueous One Solution Cell Proliferation Assay (CellTiter96; Promega, Madison, WI, USA) according to the manufacturer’s
TE D
instructions. Briefly, the cells were seeded into 96-well plates and cultured for up to 7 days. At the end of each period, 10 µL MTT solution was added and the cells were
EP
incubated for an additional 4 h, after which 150 µL dimethyl sulfoxide (DMSO) was
AC C
added to each well and mixed thoroughly. The optical density of each well was measured with a spectrophotometer (UV5100, Shanghai, China). The data are expressed as mean ± standard error of the mean (SE). The experiments were performed in triplicate and repeated at least three times.
2.8 Clonogenicity assay
8
ACCEPTED MANUSCRIPT
For colony formation assay, cells were plated in 6-well plates using a two-layer soft-agar system as described previously [17]. After 3 weeks of incubation at 37°C in 5% CO2, the
RI PT
colonies containing 50 cells or more were counted. All experiments were conducted at
SC
least three times in triplicate.
2.9 Wound healing assay
M AN U
Cell were plated in 12-well plates at 2 × 105 cells/well in 500 µL DMEM (containing 10% FCS). At 100% cell confluency, a scraped line (wound) was created using a pipette tip and scratch width was marked. Cells were washed with ice-cold PBS and further
TE D
cultured in 500 µL DMEM medium at 37°C for 24 h.Wound closure was imaged under
EP
an inverted microscope.
AC C
2.10 Transwell migration and invasion assay Transwell filter inserts (8-µm pore size) for 24-well plates were purchased from Costar (Cambridge, MA, USA). In brief, 5 × 104 cells/well in 500 µL DMEM (containing 1% FCS) were seeded in the upper chamber, and 500 µL DMEM (containing 10% FCS) was added to the lower chamber. Cells were incubated at 37°C for 24–36 h. Cells remaining on the surface of the upper chamber were removed by a cotton swab while
9
ACCEPTED MANUSCRIPT
cells migrated into the bottom of the filters were fixed with 4% formaldehyde and stained with 0.5% crystal violet. Cells were counted in five photographed fields.
RI PT
In the invasion assay, transwell filters were pre-coated with Matrigel, and the assay was performed following the migration assay protocol. Both cell migration and invasion
2.11 In vivo tumorigenic assay
M AN U
SC
assays were performed in triplicate and repeated three times.
Animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of the Affiliated Tumor Hospital of Harbin Medical University. Nude mice (5-6
TE D
weeks old) were maintained in a pathogen-free environment at experimental animal center of the Affiliated Tumor Hospital of Harbin Medical University, and divided into
EP
four groups: Group A, injected subcutaneously with 8 × 108 IMCE-neo cells; Group
AC C
B, injected subcutaneously with 8 × 108 IMCE-Wt-EphA3 cells; Group C, injected subcutaneously with 8 × 108 IMCE-EphA3-S792P cells; and Group D, injected subcutaneously with 8 × 108 IMCE-EphA3-T37K cells. Tumor volume (V, cm3) was evaluated after 4 weeks based on tumor length (l), width (w), and height (h): V= l × w × h × 0.5236.
10
ACCEPTED MANUSCRIPT
2.12 Microarray profiling and data processing IMCE (IMCE-neo cell), Wt-EphA3-transfected IMCE, and EphA3-T37K-transfected
RI PT
IMCE cells were subjected to microarray to establish differential gene expression profiles. mRNA microarray contained 23,420 probes corresponding to 17,298 genes.
SC
Probes matched to multiple genes were eliminated. Probe expression values represented
M AN U
gene expression levels, and values generated from multiple probe sets were averaged. lncRNA microarray included 34,735 probes. If the lncRNA matched to a single probe, its expression level was same as that of the probe; if the lncRNA matched to several probes, its expression level was the same as that of the longest probe.
TE D
Differentially expressed genes and lncRNAs between IMCE and IMCE-Wt-EphA3 or IMCE-EphA3-T37K ce lls were defined as those with fold changes ≥ 1.5. Genes close
EP
to lncRNAs were filtered using the human refSeq reference genome (hg19). Genes
AC C
located most closely to the region 500 kb upstream from the transcription start site (TSS) were considered to be potentially affected by lncRNAs.
2.13 Functional analysis of differentially expressed genes and lncRNAs Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was performed using DAVID Functional Annotation Bioinformatics Microarray Analysis
11
ACCEPTED MANUSCRIPT
(http://david.abcc.ncifcrf.gov/home.jsp) for differentially expressed genes, and the hosts or nearby genes of differentially expressed lncRNAs, respectively. Fisher’s exact test was
SC
adopted for GO_BP term enrichment analysis.
RI PT
applied to determine statistical significance. Enrichment Map plugin in Cytoscape was
M AN U
2.14 Construction of a protein-protein interaction sub-network for lncRNAs and mRNAs
An integrated protein-protein interaction (PPI) sub-network was established based on the following databases: Biomolecular Interaction Network Database (BIND), the Biological
TE D
General Repository for Interaction Data sets (BioGRID), the Database of Interacting Proteins (DIP), the Human Protein Reference Database (HPRD), IntAct, the Molecular
EP
INTeraction database (MINT), the mammalian PPI database of the Munich Information
AC C
Center on Protein Sequences (MIPS), PDZBase (a PPI database for PDZ-domains) and Reactome. The common mRNAs and the nearby genes for lncRNAs were set as seed genes. The sub-network was derived from the seed genes, nearby lncRNAs, and the genes connected to the seed genes in the PPI network. The network was constructed using Cytoscape (http://www.cytoscape.org/).
12
ACCEPTED MANUSCRIPT
2.15 Statistical analysis Statistical analysis was performed using SAS software (version 9.2). The differences
RI PT
between the groups were analyzed by paired t-test or χ2 test. P values less than 0.05 were
SC
considered statistically significant.
M AN U
3. Results
3.1 Upregulation of EphA3 expression in colorectal cancer tissues and colorectal epithelial cell lines
Immunohistochemistry (IHC) showed that EphA3 protein was expressed in 72% (49 of
TE D
68) colorectal cancer tissues with a broad range of colorectal cancer types, but was not expressed in paired normal colorectal mucosa (Fig. 1). EphA3 expression was associated
EP
with patient age (P = 0.015), tumor differentiation (P = 0.001), and lymph node
AC C
metastases (P = 0.039) (Table 1). RT-PCR analysis showed that endogenous EphA3 expression was very low in conditionally immortalized mouse colonic epithelial cell line IMCE cell line, but much higher in YAMC-Ras and IMCE-Ras cell lines with malignant phenotypes (Fig. 1F).
3.2 EphA3 overexpression promoted colony formation of colorectal epithelial cells
13
ACCEPTED MANUSCRIPT
We stably overexpressed EphA3 in IMCE cells by transfecting pcDNA3.1-EphA3-Myc, pcDNA3.1-EphA3-T37K-Myc and pcDNA3.1-EphA3-S792P-Myc into IMCE cells.
transfected IMCE cells, but not in the control (Fig. 1G).
RI PT
Western blot analysis showed that exogenous EphA3-Myc was highly expressed in
SC
MTT assay showed that there was no remarkable difference in cell viability among the
M AN U
transfected cells (Fig. 1H). Colony formation assay showed significant difference in colony formation between EphA3-tranfected cells and control cells, especially in those transfected with constitutively active EphA3-T37K mutant (Fig. 1I).
TE D
3.3 EphA3 overexpression promoted colorectal epithelial cell migration and invasion Wound-healing and transwell assays showed that EphA3 overexpression promoted IMCE
EP
cell migration and invasion. In particular, EphA3-T37K mutant induced even faster
AC C
wound closure and invasion (Fig. 2A,B). Furthermore, scanning electron microscope (SEM) micrographs revealed striking changes in IMCE cell morphology following WtEphA3, EphA3-T37K, and EphA3-S792P overexpression. We observed increased lamellipodia and ruffles in cells overexpressing Wt-EphA3, and increased filopodia-like protrusions on the surface of EphA3-T37K- and S792P-transfected cells (Fig. 2C).
14
ACCEPTED MANUSCRIPT
3.4 EphA3 promoted tumorigenicity of colorectal epithelial cells in nude mice Stable IMCE-neo-, IMCE-Wt-EphA3-, IMCE-EphA3-T37K-, and IMCE-EphA3-S792P-
RI PT
transfected cells were injected into athymic nude mice. All ten mice injected with IMCEEphA3-T37K cells formed tumors by day 13, whereas mice injected with IMCE-Wt-
SC
EphA3 or IMCE-EphA3-S792P cells formed tumors with a longer latency period. As the
M AN U
control, mice injected with IMCE-neo cells did not develop tumors (Fig. 3A,B). Tumor volume in mice injected with IMCE-EphA3-T37K cells was significantly increased compared to other cells (P < 0.001; Fig. 3C). Since aberrant cell proliferation is a major feature of the tumors, we then used proliferating cell nuclear antigen (PCNA) as cell
TE D
proliferation marker. Infiltrating lymphocytes in mice injected with IMCE-EphA3 but not IMCE-neo cells showed positive PCNA staining. In addition, positive PCNA staining
AC C
EP
was only significant in IMCE-EphA3-T37K cells compared to other cells (Fig. 3D, E).
3.5 Upregulated and downregulated mRNAs and lncRNAs between IMCE-neo and EphA3-T37K cells
We performed microarray analyses to investigate global mRNA and lncRNA patterns in IMCE-neo, IMCE-Wt-EphA3, and IMCE-EphA3-T37K cells (Fig. 4A,B). Although mRNA expression patterns among these three cell lines were very similar, a total of 607
15
ACCEPTED MANUSCRIPT
downregulated and 2,394 upregulated genes were identified between IMCE-neo and IMCE-EphA3-T37K cells by differential expression analysis with a fold change ≥ 1.5.
RI PT
GO and KEGG analyses showed that upregulated genes were mainly involved in transcription regulation and responses to hormone stimulus, and were enriched in eight Fig. S1A, B
. In contrast,
SC
pathways including MAPK and VEGF signaling pathways
M AN U
downregulated genes were involved in DNA packing, mitosis, gonad development, and other GO categories, and were enriched in three pathways including arginine and proline metabolism, apoptosis, and regulation of the actin cytoskeleton
Fig. S1C, D
.
A total of 668 downregulated and 2,769 upregulated lncRNAs were identified as
TE D
differentially expressed between IMCE-neo and IMCE-EphA3-T37K cells. Host or nearby genes for lncRNAs were identified and used for GO and KEGG enrichment
EP
analysis. Neighboring genes for upregulated lncRNAs were mainly involved in basic
AC C
cellular functions such as transmembrane transport, apoptosis, transcription, and muscle organ development, and were enriched in several pathways including cell adhesion junction, arrhythmogenic right ventricular cardiomyopathy and MAPK signaling pathway (Fig. S2A, B
. Downregulated lncRNAs were mainly important in
programmed cell death and negative transcription regulation, and were enriched in three
16
ACCEPTED MANUSCRIPT
pathways including aminoacyl-Trnat biosynthesis, steroid hormone biosynthesis, and endocytosis
Fig. S2C, D .
RI PT
Among the mRNAs and genes near lncRNAs, 242 were upregulated genes while 17 were downregulated between IMCE-neo and EphA3-T37K cells. Functional enrichment
SC
analysis indicated that these 259 genes were enriched in pathways such as response to
M AN U
hormone stimulus, carboxylic acid biosynthetic process, and some other GO categories (Table 2).
PPI network was integrated from eight PPI databases, consisting of 80,980 edges and 13,361 nodes. The 259 genes consistently regulated at the mRNA and lncRNA levels
TE D
were used as seed genes in the integrated network. In addition, the subnetwork included 2,474 nodes and 2,848 interactions, among which a small number of nodes were
EP
presented with high degrees (hub nodes) while the majority of the nodes were presented
AC C
with low degrees (Fig. 4C). The hub genes in the subnetwork included GRB2, BRCA1, NFKB1, SFN, FN1, WDR5, PDPK1, SH3GL3, DHX38, SREBF2, QKI, and PMS2 (Table 3).
3.6 Upregulated and downregulated mRNAs and lncRNAs between IMCE-neo and Wt-EphA3 cells
17
ACCEPTED MANUSCRIPT
Compared to IMCE-neo cells, IMCE-Wt-EphA3 cells had 820 downregulated and 1,768 upregulated genes. GO and KEGG analyses showed that upregulated genes were mainly
RI PT
involved in the regulation of DNA metabolic processes and RNA elongation, and were enriched in pathways such as lysine degradation and terpenoid backbone biosynthesis . Downregulated genes were mostly involved in ion transmembrane
SC
Fig. S3A, B
M AN U
transport, cell proliferation, chromatin assembly and some other basic GO categories, and were enriched in various pathways including focal adhesion and the regulation of the actin cytoskeleton
Fig. S3C, D
.
There were 784 downregulated and 2,428 upregulated lncRNAs between IMCE-neo and
TE D
IMCE-Wt-EphA3 cells. Neighboring genes for upregulated lncRNAs were mainly involved in basic cellular functions such as cell death, adhesion, and positive regulation
EP
of transcription, and were enriched in pathways including adhesion junction and p53 Fig. S4A, B
. Downregulated lncRNAs were mainly associated
AC C
signaling pathway
with the regulation of protein phosphorylation, cell proliferation, or gene expression, and were enriched in apoptosis pathway Fig. S4C, D
.
Among the mRNA and the near lnRNA, 142 were upregulated and 32 were downregulated between IMCE-neo and IMCE-Wt-EphA3 cells. Functional enrichment analysis indicated that these 174 genes were enriched in the pathway of cell cycle
18
ACCEPTED MANUSCRIPT
(Table 4). We mapped these 174 genes into the integrated network as seed genes. Then the sub-network was constructed by interactions of these seed genes and nearby lncRNAs
SC
in this subnetwork included WDR5, NFX1, MYH9 (Table 5).
RI PT
(Fig. 4D). We found 1174 nodes and 1169 interactions in the sub-network. The hub genes
M AN U
3.7 Identification of genes and pathways for tumorigenicity
To identify genes and pathways that are associated with tumorigenicity, we defined differential expression of mRNAs or lncRNAs between IMCE-neo and Wt-EphA3 as Set 1, and that between IMCE-neo and EphA3- T37K as Set 2. Since Wt-EphA3 cells had
TE D
lower tumorigenicity than EphA3-T37K cells, we considered the overlap of Set 1 and Set 2 as the candidates for low tumorigenicity and the genes in Set 2 but not in Set 1 as the
EP
candidates for high tumorigenicity. Consequently, we identified 812 upregulated and 300
AC C
downregulated candidate genes associated with weak tumorigenicity (Fig. 5A). These genes were mainly involved in cell cycle regulation, muscle adaptation, terpenoid backbone biosynthesis, nitrogen metabolism, RIG-I-like receptor signaling pathway, and melanoma development (Fig. 5B,C). On the other hand, we identified 1,582 upregulated and 307 downregulated candidate genes linked to high tumorigenicity (Fig. 5A). These
19
ACCEPTED MANUSCRIPT
genes were involved in cell adhesion, GTPase activity, and metabolic processes such as lipid metabolism and VEGF signaling (Fig. 5D,5E).
RI PT
Similarly, we found 1,380 upregulated and 376 downregulated lncRNAs associated with weak tumorigenicity (Fig. 6A). The genes nearby these lncRNAs were mainly enriched
SC
for common cell functions such as transcription, apoptosis, and neuronal differentiation.
M AN U
Two pathways, namely axon guidance and cell adhesion molecules, were associated with weak oncogenesis (Fig. 6B,6C). We also found “nearby or hosted genes” of the 1,386 upregulated and 292 downregulated lncRNAs in the high tumorigenicity group. These genes can mediate neuronal differentiation, negative regulation of transcription,
TE D
programmed cell death, and other functions for gene pathways related to cancer progression such as MAPK signaling pathway, endometrial cancer, prostate cancer, and
AC C
EP
other pathways linked to cancer (Fig. 6D, 6E).
4. Discussion
EphA3 plays important role in embryo development and regulates cell adhesion, mobility, and growth.[8] EphA3 was overexpressed or mutated in a variety of human cancers including colorectal cancer, melanoma, and glioma[8,11,12,14,18–20] A previous study demonstrated that EphA3 mutations in lung tissues promoted lung cancer
20
ACCEPTED MANUSCRIPT
development, and an EphA3 mutation-associated gene signature was associated with poor patient survival.[12] In this study, we first confirmed EphA3 protein overexpression
RI PT
in colorectal cancer tissue specimens, and then manipulated its expression in rat colorectal epithelial cell lines. We found that EphA3 overexpression was significantly
SC
associated with patient age, tumor differentiation, and lymph node metastasis. In
M AN U
addition, the overexpression of EphA3 and its constitutively active mutants significantly promoted colony formation, migration and invasion of colorectal epithelial cells. Furthermore, EphA3 overexpression, in particular the active mutants, significantly promoted the tumorigenicity of colorectal epithelial cells in nude mouse xenograft model.
TE D
At the molecular level, EphA3 differentially regulated the expression of genes and lncRNAs that were associated with cancer- and angiogenesis-related gene pathways.
EP
Thus, our current study provides additional evidence for oncogenic activity of EphA3 in
AC C
colorectal cancer. Additional studies are needed to investigate molecular mechanisms underlying EphA3 overexpression and oncogenic activities in colorectal cancer. Several studies have assessed EphA3 mutations in different human cancer tissues and demonstrated that EphA3 is a tumor suppressor gene
[8, 11,12,14]
. Other studies have
indicated that EphA3 expression may inhibit tumor malignant behaviors in vitro[21–23]. However, a recent ex vivo study showed that EphA3 expression was upregulated in
21
ACCEPTED MANUSCRIPT
gastric carcinoma tissues and was associated with tumor TNM stage, angiogenesis, and poor prognosis[24]. In colorectal cancer, EphA3 is reported to be overexpressed in tissues
RI PT
and associated with poor prognosis[13]. These data indicate that EphA3 has different roles in different human malignancies, which may depend on EphA3 protein localization[8],
SC
downstream signaling pathway [25, 26], or active or inactive mutation in EphA3 gene [12]. In
M AN U
this study, we mutated EphA3 at T37K and S792P sites as active mutants and found that both Wt-EphA3 and S792P mutants were able to induce colorectal cell tumorigenesis, whereas T37K mutant had a strong oncogenic activity. Indeed, a previous study suggested that EphA3 does not possess the characteristics of a classical tumor suppressor
TE D
gene, as the inactivation of both alleles is required to induce carcinogenesis[27]. If EphA3 has tumor suppressor activities that depend on ephrin and kinase activity, its tumor-
EP
promoting activities that depend on crosstalk with other signaling molecules do not
AC C
require ephrin binding or kinase activity[25, 26]. In this case, mutations that impair ephrin binding and kinase activity could shift the balance towards tumor-promoting effects of EphA3. Moreover, mutated EphA3 may have dominant negative effects and inhibit the function of wild-type receptor. In this study we further explored the underlying molecular events after the overexpression of wild-type and active mutated EphA3 in colorectal epithelial cells. We
22
ACCEPTED MANUSCRIPT
found that upregulated mRNAs between IMCE-neo and EphA3-T37K cells mainly enriched transcription regulation and the response to hormonal stimulus in eight
RI PT
pathways, including cytosolic DNA-sensing pathway, MAPK and VEGF signaling pathways. Upregulated angiogenesis-related genes such as VEGF are known to promote [30]
. MAPK signaling
SC
tumor progression, tumor cell growth, invasion, or metastasis
M AN U
pathway regulates cell proliferation and differentiation[31]. In contrast, the downregulated genes between IMCE-neo and EphA3-T37K cells were significantly concentrated in DNA packing, mitosis, gonad development and other GO terms and enriched in three pathways including those related to arginine and proline metabolism, apoptosis, and the
TE D
regulation of the actin cytoskeleton. The neighboring genes in upregulated lncRNAs were mainly involved in transmembrane transport, apoptosis, transcription and muscle
EP
development, and were enriched in several pathways, including adherent junction,
AC C
ARVC, pathways in cancer, MAPK signaling pathway. However, downregulated lncRNAs were mainly concentrated in programmed cell death, negative regulation of transcription, and enriched in three pathways aminoacyl-Trna biosynthesis, steroid hormone biosynthesis and endocytosis. However, it remains to be determined how EphA3 regulates these gene pathways to promote colorectal epithelial cell malignant transformation in vitro and in vivo. Further mechanistic studies on the association of
23
ACCEPTED MANUSCRIPT
EphA3 with these pathways will help provide insight into oncogenic potential of EphA3 in CRC.
RI PT
In summary, our study provides both in vitro and in vivo evidence supporting oncogenic activities of EphA3 in colorectal epithelial cells. Our data demonstrate that EphA3 is
SC
overexpressed in colorectal cancer tissues and induces colorectal epithelial cell malignant
M AN U
transformation. EphA3 overexpression may serve as a biomarker for early diagnosis of colorectal cancer and the prediction of tumor progression.
Acknowledgments
TE D
This work was supported by the National Natural Science Foundation of China (No. 81272703), the Natural Science Foundation of HeiLongJiang Province (No. QC08C56)
EP
and the Scientific Research Foundation for Returned Scholars, Department of Education
AC C
of HeiLongJiang Province (No. 1154h17).
Conflict of interest
The authors declare that they have no conflicts of interest.
24
ACCEPTED MANUSCRIPT
References 1. Karsa LV, Lignini TA, Patnick J, Lambert R, Sauvaget C. The dimensions of the
RI PT
CRC problem. Best Pract Res Clin Gastroenterol. 2010; 24: 381-396.
2. Ho AS, Chen CH, Cheng CC, Wang CC, Lin HC, Luo TY, `et al. Neutrophil elastase
SC
as a diagnostic marker and therapeutic target in colorectal cancers. Oncotarget.
M AN U
2014; 5(2): 473-480.
3. Saika K, Sobue T. Cancer statistics in the world. Gan To Kagaku Ryoho. 2013; 40: 2475–2480.
4. Center MM, Jemal A, Ward E. International trends in colorectal cancer incidence
TE D
rates. Cancer Epidemiol Biomarkers Prev. 2009; 18:1688-1694. 5. Jemal A, Siegel R, Ward E, Hao Y, Xu J, et al. Cancer statistics, 2008. CA Cancer J
Markowitz SD, Bertagnolli MM. Molecular Origins of Cancer: Molecular Basis of
AC C
6.
EP
Clin. 2008; 58 : 71-96.
Colorectal Cancer. N Engl J Med. 2009; 361 (25): 2449–2460.
7.
Muzny DM, Bainbridge MN, Chang K, Dinh HH, Drummond JA, Fowler G, et al. Comprehensive molecular characterization of human colon and rectal cancer. Nature. 2012; 487: 330-337.
8.
Lisabeth EM, Fernandez C, Pasquale EB. Cancer somatic mutations disrupt functions
25
ACCEPTED MANUSCRIPT
of the EphA3 receptor tyrosine kinase through multiple mechanisms. Biochemistry. 2012; 51: 1464-1475.
RI PT
9. Batlle E, Bacani J, Begthel H, Jonkheer S, Gregorieff A, van de Born M, Malats N, Sancho E, Pawson T, Gallinger S, Pals S, Clevers H. EphB receptor activity
SC
suppresses colorectal cancer progression. Nature. 2005;435:1126-1130.
M AN U
10. Li S, Ma Y, Xie C, Wu Z, Kang Z Fang Z, Su B, Guan M. EphA6 promotes angiogenesis and prostate cancer metastasis and is associated with human prostate cancer
progression.
10.18632/oncotarget.4088.
Oncotarget.
2015;
6:22587-22597.
doi:
TE D
11. Wood LD, Calhoun ES, Silliman N, Ptak J, Szabo S, et al. Somatic mutations of GUCY2E, EPHA3, and NTRK3 in human cancers. Hum Mutat. 2006; 27: 1060-1061.
EP
12. Zhuang G, Song W, Amato K, Hwang Y, Lee K, Boothby M. Effects of cancer-
AC C
associated EphA3 mutations on lung cancer. J Natl Cancer Inst. 2012; 104:1182– 1197.
13. XiHQ Zhao P. Clinicopathological significance and prognostic value of EphA3 and CD133 expression in colorectal carcinoma. J Clin Pathol. 2011; 64:498-503.
14. Sjöblom T, Jones S, Wood LD. The consensus coding sequences of human breast and colorectal cancers. Science. 2006; 314(5797): 268-274.
26
ACCEPTED MANUSCRIPT
15. Keane N, Freeman C, Swords R， Giles FJ. EPHA3 as a novel therapeutic target in the hematological malignancies. Expert Rev Hematol. 2012; 5:325-340.
RI PT
16. Whitehead RH, Joseph JL. Derivation of conditionally immortalized cell lines containing the Min mutation from the normal colonic mucosa and other tissues of an
SC
“Immortomouse”/Min hybrid. Epithelial Cell Bio. 1994 Jul; 3(3): 119–125.
M AN U
17. Munker R, Norman A, Koeffler HP. Vitamin D compounds. Effect on clonal proliferation and differentiation of human myeloid cells. J Clin Invest. 1986; 78: 424–430.
18. Easty DJ, Bennett DC. Protein tyrosine kinases in malignant melanoma. Melanoma
TE D
Res. 2000; 10:401-411.
19. Wykosky J, Debinski W. The EphA2 receptor and ephrinA1 ligand in solid tumors:
EP
function and therapeutic targeting. Mol Cancer Res. 2008; 6(12):1795-1806.
AC C
20. Greenman C, Stephens P, Smith R, Dalgliesh GL, Hunter C, Bignell G. Patterns of somatic mutation in human cancer genomes. Nature. 2007; 446: 153-158.
21. Lee DJ, Schonleben F, Banuchi VE, Qiu W, Close LG, et al. Multiple tumorsuppressor genes on chromosome 3p contribute to head squamous cell carcinoma tumorigenesis. Cancer Biol Ther. 2010; 10:689-693. 22. Dottori M, Down M, Huttmann A, Fitzpatrick DR, Boyd AW. Cloning and
27
ACCEPTED MANUSCRIPT
characterization of EphA3 (Hek) gene promoter: DNA methylation regulates expression in hematopoietic tumor cells. Blood. 1999; 94:2477-2486.
RI PT
23. Guan M, Liu L, Zhao X, Wu Q, Yu B, et al. Copy number variations of EphA3 are associated with multiple types of hematologic malignancies. Clin Lymphoma
SC
Myeloma Leuk. 2011; 11:50-53.
M AN U
24. Xi NQ, Wu XS, Wei B, Chen L. Aberrant expression of EphA3 in gastric carcinoma: correlation with tumor angiogenesis and survival. J Gastroenterol. 2012; 47: 785-794.
25. Pasquale EB. Eph receptors and ephrins in cancer: bidirectional signaling and
TE D
beyond. Nat Rev Cancer. 2010; 10:165-180.
26. Miao H, LiDQ, Mukerjee A, Guo H, Petty A, et al. EphA2 mediates ligand-
EP
dependent inhibition and ligandindependent promotion of cell migration and invasion
AC C
via a reciprocal regulatory loop with Akt. Cancer cell. 2009; 16:9-20. 27. Lahtela J, Pradhan B, Narhi K, Hemmes A, Sarkioja M, et al. The putative tumor suppressor gene EphA3 fails to demonstrate a crucial role in murine lun tumorigenesis or morphogenesis. Dis Model Mech. 2015; 8(4):393-401.
28. Hoek K, Rimm DL, Williams KR, Zhao H, Ariyan S, et al. Expression profiling reveals novel pathways in the transformation of melanocytes to melanomas. Cancer
28
ACCEPTED MANUSCRIPT
Res. 2004; 64:5270-5282. 29. Valsesia A, Rimoldi D, Martiner D, Ibberson M, Benaglio P, et al. Network-guided
RI PT
analysis of genes with altered somatic copy number and gene expression reveals pathways commonly perturbed in metastatic melanoma. Plos One. 2011; 6: e18369.
SC
30. Lee SH, Jeong D, Han YS, Baek MJ. Pivotal role of vascular endothelial
M AN U
growthfactor pathway in tumor angiogenesis. Ann Surg Treat Res. 2015; 89(1):1-8. 31. Rovida E, Stecca B. Mitogen-activated progein kinases and Hedgehog-GLI signaling in cancer: A crosstalk providing therapeutic opportunities? Semin Cancer Biol. 2015;
AC C
EP
TE D
35:154-167.
29
ACCEPTED MANUSCRIPT
Figure legends
RI PT
Figure 1. EphA3 was overexpressed in CRC tissues and promoted colony formation of colorectal epithelial cell cells. Immunohistochemistry staining of EphA3 in normal
SC
mucosae (A), differentiated adenocarcinoma (B,C), mucinous adenocarcinoma (D), and
M AN U
signet-ring cell carcinoma (E). RT-PCR analysis of EphA3 mRNA level in different cells (F). Western blot analysis of EphA3 protein in IMCE cells stably transfected with EphA3 wild-type and mutant constructs (G). MTT assay of IMCE cells stably transfected with EphA3 wild-type and mutant constructs (H). Colony formation assay of IMCE cells
TE D
stably transfected with EphA3 wild-type and mutant constructs (I).
EP
Figure 2. EphA3 promoted the migration and invasion of colorectal epithelial cells.
AC C
A. Stably transfected cells were subjected to wound healing assay. IMCE-EphA3-T37K had significantly higher rates of colorectal epithelial cell migration. B. Stably transfected cells were subjected to Transwell assay. IMCE-EphA3-T37K induced higher rates of cell invasion. C. SEM analysis of the morphology of stably transfected cells. Cells were fixed in 2.5% glutaraldehyde for 24 h, dehydrated with a graded ethanol series, and then dried in CO2. The cell samples were then mounted in aluminum and were sputter-coated with
30
ACCEPTED MANUSCRIPT
gold for SEM using a FEI Quanta 200 FEG SEM (FEI, Hillsboro, Oregon, USA).
RI PT
Figure 3. EphA3 promoted the growth of colorectal epithelial cells in nude mouse model. A. Nude mouse xenograft assay. Stable IMCE-neo, IMCE-Wt-EphA3, IMCE-
SC
EphA3-S792P, and IMCE-EphA3-T37K-transfected cells were subcutaneously injected
M AN U
into athymic nude mice and the mice were sacrificed after 80 days of cell injection. B. Tumor-free mice in three groups. C. Tumor volume in three groups. D. Typical images of PCNA staining of xenografts in three groups. E. Quantitation of PCNA staining of
TE D
xenografts in three groups.
Figure 4. Microarray profiling of global mRNA and lncRNA expression and the
EP
interaction network. A. Circos v0.62 was used to profile the changes of global mRNA
AC C
and lncRNA expression in stable IMCE-neo, Wt-EphA3, and EphA3-T37K-transfected cells. B. Quantitation of global expression levels of mRNA and lncRNA in the cells. C. The differentially expressed lncRNAs could form a sub-network in EphA3-T37K cells. (D. The differentially expressed lncRNAs could form a subnet work in Wt-EphA3 cells.
Figure 5. Functional analysis of mRNAs between IMCE-Wt-EphA3 and IMCE-
31
ACCEPTED MANUSCRIPT
EphA3-T37K. There were 812 upregulated and 300 downregulated genes associated with weak tumorigenicity (A), and these genes were mainly enriched in four gene
RI PT
pathways (B and C). There were 1,582 upregulated and 307 downregulated genes associated with high tumorigenicity (A), and these genes were mainly enriched in ten
M AN U
SC
gene pathways (D and E).
Figure 6. Functional analysis of LncRNAs between IMCE-Wt-EphA3 and IMCEEphA3-T37K. There were 1,380 upregulated and 376 downregulated lncRNAs associated with weak tumorigenicity (A), and they were mainly enriched in two gene
TE D
pathways (B and C). There were 1,386 upregulated and 292 downregulated lncRNAs associated with high tumorigenicity (A), and they were mainly enriched in eight gene
AC C
EP
pathways (D and E).
32
ACCEPTED MANUSCRIPT Table 1. Association of EphA3 expression with clinicopathological features of colorectal cancer patients EphA3 Character
N
Positive
Negative
< 60
21
11
≥ 60
47
39
Male
52
37
Female
16
12
Gross Type 1
30
2
22
3
16
15
10
15
7
10
4
0
0
0
26
13
13
42
36
6
14
7
7
54
42
12
Positive
8
5
3
Negative
60
44
16
TE D
Well Moderate
EP
Poor
0.76
4
20
Differentiation
0.015
9
M AN U
Sex
10
SC
Age (yrs.)
(n=19)
RI PT
(n=49)
p value
0.28
0.001
LN metastasis
AC C
Positive Negative
0.039
Distant metastasis
0.52
ACCEPTED MANUSCRIPT Table 2. Gene enriched function by EphA3-T37K Description
Count
P value
GO:0009725
Response to hormone stimulus
13
0.0042
GO:0048545
Response to steroid hormone stimulus
9
0.0046
GO:0046394
Carboxylic acid biosynthetic process
8
0.0051
GO:0016053
Organic acid biosynthetic process
8
0.0051
GO:0009719
Response to endogenous stimulus
13
0.0090
AC C
EP
TE D
M AN U
SC
RI PT
Term
ACCEPTED MANUSCRIPT Table 3. List of hub genes for EphA3-T37K lncRNA
Regulation
Degree
GRB2
RP11-16C1.3
Up
522
BRCA1
BRCA1
Up
207
NFKB1
AF213884.2
Up
166
SFN
RP1-50O24.6
Up
156
FN1
AC012462.1
Up
WDR5
XLOC_007596
Up
PDPK1
RP11-20I23.8
Up
SH3GL3
SH3GL3
Up
DHX38
AK055364
Up
SREBF2
RP5-821D11.7
QKI
XLOC_005531
PMS2
PMS2
TANK
AK027541
ITGB3BP
ITGB3BP
UBTF
RI PT
Gene
86
71
M AN U
SC
50
47
47
47
Up
43
Up
42
Up
41
Up
41
RP5-882C2.2
Up
40
WASL
AL110181
Up
39
ID2
AC011747.7
Up
37
GSN
RP11-477J21.6
Up
34
XLOC_011384
Up
29
XLOC_007069
Up
27
SHANK2
ANO1-AS2
Up
24
AURKA
XLOC_013807
Up
24
ANAPC10
ANAPC10
Up
23
NR2C2
BC092452
Up
23
UBE2G2
AL773604.8
Up
23
AC C
PLAT
EP
MEF2A
TE D
Up
ACCEPTED MANUSCRIPT Table 4. Gene enriched function by Wt-EphA3 Description
Count
P value
GO:0022403
Cell cycle phase
11
0.0047
GO:0022402
Cell cycle process
13
0.0056
GO:0000279
M phase
9
0.010
GO:0051726
Regulation of cell cycle Regulation of transcription from RNA
GO:0006357
9
0.011
14
0.015
AC C
EP
TE D
M AN U
SC
polymerase II promoter
RI PT
Term
ACCEPTED MANUSCRIPT Table 5. List of hub genes for Wt-EphA3
Degree
XLOC_007596
Up
71
NFX1
RP11-255A11.13
Up
54
MYH9
RP4-633O19__A.1
Up
44
SMARCA5
GUSBP5
Up
PRKAR2A
PRKAR2A-AS1
Down
BRE
XLOC_001407
Up
MAD2L1
RP11-96A1.5
FSCN1
ZNF815P
AURKA
XLOC_013807
ANAPC10
ANAPC10
NR2C2
BC092452
MAP2K6
AC003051.1
SC
WDR5
EP AC C
RI PT
Regulation
40
40
35
Up
32
Down
28
M AN U
lncRNA
Up
24
Up
23
Up
23
Up
20
TE D
Gene
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Ephrin Type-A Receptor 3 (EphA3) belongs to the ephrin receptor subfamily of the protein tyrosine kinase family, and altered EphA3 expression or activity is associated with carcinogenesis, including colorectal cancer. This study assessed the expression of EphA3 in colon cancer, using both colorectal cancer tissue specimens and rat colon epithelial cell lines. The results showed that EphA3 overexpression in tumor tissues was associated with patient age, tumor differentiation, and lymph node metastasis. The expression of EphA3 and its constitutively active mutants promoted colony formation, migration, and invasion, and induced tumorigenicity of colon epithelial cells in nude mice. In addition, cDNA and lncRNA microarray profiling data revealed that differentially expressed genes and lncRNAs in EphA3-expressing cells were associated with cell proliferation, invasion, and angiogenesis. The findings of our study reveal additional mechanisms underlying the oncogenic effects of EphA3 on colorectal cells, which could provide novel targets for the prevention, early diagnosis, and treatment of colorectal cancer.
ACCEPTED MANUSCRIPT Conflict of interest statement This work was supported by the National Natural Science Foundation of China (Grant No 81272703), the Natural Science Foundation of HeiLongJiang Province (Grant No
RI PT
QC08C56) and the Scientific Research Foundation for Returned Scholars, Department of Education of HeiLongJiang Province (Grant No. 1154h17).
AC C
EP
TE D
M AN U
SC
The authors declare that they have no conflicts of interest.
S0304-3835(16)30611-5
DOI:
10.1016/j.canlet.2016.10.004
Reference:
CAN 13061
To appear in:
Cancer Letters
Received Date: 10 July 2016 Revised Date:
26 September 2016
Accepted Date: 2 October 2016
Please cite this article as: M. Li, C. Yang, X. Liu, L. Yuan, F. Zhang, M. Wang, D. Miao, X. Gu, S. Jiang, B. Cui, J. Tong, Z. Yu, EphA3 promotes malignant transformation of colorectal epithelial cells by upregulating oncogenic pathways, Cancer Letters (2016), doi: 10.1016/j.canlet.2016.10.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
EphA3 promotes malignant transformation of colorectal epithelial cells by upregulating oncogenic pathways
RI PT
Mingqi Li*, Cheng Yang*, Xin Liu*, Liang Yuan, Fubin Zhang, Muhong Wang,
Dazhuang Miao, Xinyue Gu, Shixiong Jiang, Binbin Cui, Jinxue Tong#, Zhiwei Yu#
SC
Affiliation: Department of Colorectal Surgery, The Affiliated Tumor Hospital of
M AN U
Harbin Medical University, Harbin, China
*These authors contributed equally to this work.
# Correspondence to:
TE D
ZhiWei Yu, MD, PhD, Department of Colorectal Surgery, The Affiliated Tumor Hospital of Harbin Medical University, Harbin, China Tel.: 86-0451-86298096, E-mail: [email protected]
EP
JinXue Tong, MD, PhD, Department of Colorectal Surgery, The Affiliated Tumor Hospital of Harbin Medical University, Harbin, China
AC C
Tel.: 86-0451-86298096, E-mail: [email protected]
1
ACCEPTED MANUSCRIPT
ABSTRACT Ephrin Type-A Receptor 3 (EphA3) belongs to the ephrin receptor subfamily of the
RI PT
protein tyrosine kinase family, and plays an important role in embryogenesis and neurogenesis. This study aimed to investigate the role of EphA3 in promoting malignant
SC
transformation of colorectal epithelial cells, and explore underlying molecular
M AN U
mechanisms. Colorectal cancer tissue specimens from 68 patients were analyzed for EphA3 expression. EphA3 expression levels were manipulated in rat colon epithelial cell lines. We found that EphA3 expression level in tumor tissues was associated with patient age (P = 0.015), tumor differentiation (P = 0.001), and lymph node metastasis (P =
TE D
0.039). Overexpression of EphA3 and its constitutively active mutants promoted colony formation, migration and invasion, and tumorigenicity of colon epithelial cells in nude
EP
mice. The cDNA and lncRNA microarray profiling data revealed that differentially
AC C
expressed genes and lncRNAs in EphA3 or mutant-transfected cells were associated with cell proliferation, invasion and angiogenesis. Our findings reveal the mechanisms underlying the oncogenic activities of EphA3 in colorectal cells, which could provide novel targets for the prevention, early diagnosis, and treatment of colorectal cancer. Keywords: EphA3; transformation; mutation; tumorigenicity; colorectal epithelial cells
2
ACCEPTED MANUSCRIPT
1. Introduction Colorectal cancer (CRC) is the third most common cancer in men and the second most
RI PT
common cancer in women worldwide, with over 1.2 million new cancer cases and more than 600,000 cancer-related deaths estimated globally in 2008[1–4]. The 5-year survival [5]
. Both genetic and environmental factors are
SC
rate of CRC with late-stage is only 12%
M AN U
involved in the development of CRC but the underlying mechanisms remain elusive [6–7]. Identification of novel diagnostic biomarkers and effective therapeutic targets for CRC is urgently needed.
Erythropoietin-producing hepatocellular (Eph) receptors, the largest known family of
TE D
tyrosine kinase receptors, mediate cell compartmentalization and directional cell migration during embryonic development[8]. Eph receptor subfamily consists of 16
EP
members that are divided into two groups based on sequence identity: EphA1–10 and
AC C
EphB1–6, which play important roles in cancer invasiveness [9-10]. The action of EphA3 in cancer seems contradictory: although tumor suppressor properties of EphA3 have been reported tumors
[8,11,12]
, EphA3 functions as an oncogene in different solid and hematopoietic
[13,14]
, and has been implicated in maintaining tumor-initiating cells in
glioblastoma and leukemia [13,15]. A previous study demonstrated that EphA3 is the sixth most frequently mutated gene in CRC [14]. However, the role of EphA3 in CRC has been
3
ACCEPTED MANUSCRIPT
poorly investigated. In this study, we employed microarray, proteomics and bioinformatics analyses to
RI PT
investigate the role of EphA3 in malignant transformation of colorectal epithelial cells. Our data demonstrate the role of EphA3 in upregulating angiogenesis-related
SC
pathways in CRC and provide novel insight into the mechanisms of CRC
2. Materials and Methods 2.1 Tissue specimens
M AN U
tumorigenesis.
TE D
Primary tumor specimens of colorectal cancer and adjacent normal mucosa were harvested from 68 patients who had undergone colectomy for pathologically
EP
diagnosed CRC at the Affiliated Tumor Hospital of Harbin Medical University (Harbin,
AC C
China) between March 2012 and March 2013. All tissue samples were immediately snap-frozen in liquid nitrogen a n d stored at -80°C. None of the patients received adjuvant chemoradiotherapy before surgery. The study protocol was approved by the Ethics Committee of Harbin Medical University, and all patients provided written informed consent.
4
ACCEPTED MANUSCRIPT
2.2 Immunohistochemistry Paraffin-embedded tissue blocks from 68 patients were retrieved from the Pathology
RI PT
Department, and 5 µm thick sections were prepared for standard immunohistochemistry with EnVisionTM immunohistochemistry methods. Briefly, the tissue sections were
SC
deparaffinized in xylene and re-hydrated through an ethanol gradient. Samples were
M AN U
blocked in 20% normal serum, and incubated overnight at 4°C with a primary antibody against EphA3 (sc-920, Santa Cruz Biotechnology; Santa Cruz, CA, USA) at a dilution of 1:100. The sections were washed with phosphate-buffered saline (PBS) and then incubated with biotin- or fluorescence-labeled secondary antibody at 4°C for 2 h. The
TE D
color reaction was processed with 3,3'-diaminobenzidine (DAB) solution, and then
AC C
2.3 Cell culture
EP
reviewed under a light microscope or inverted fluorescence microscope.
Rat colorectal YAMC, IMCE, YAMC-Ras, and IMCE-Ras cell lines were kindly provided by Dr. Zhenfeng Zhang of Ireland Cancer Center, Case Western Reserve University (Cleveland, OH, USA). These cell lines were generated from colonic epithelia of F1 immorto-APCmin/+ mouse hybrids [16]. The cells were maintained in RPMI 1640 medium supplemented with 5% heat-inactivated fetal bovine serum (FBS), 5 U/ml
5
ACCEPTED MANUSCRIPT
murine interferon gamma (IFN-γ), 100 U/ml penicillin and streptomycin, 5 µg/ml insulin, 5 µg/ml transferrin, and 5 ng/ml selenium acid in a humidified incubator with 5% CO2 at
RI PT
37°C.
SC
2.4 Reverse transcription PCR
M AN U
Total RNA was isolated from colorectal epithelial cell lines using Trizol reagent (Invitrogen, Carlsbad, CA, USA) and reverse transcribed into cDNA using Superscript First-Strand Synthesis System (Invitrogen) according to the manufacturer’s instructions. The
primers
were
as
follows:
and
5′5′-
TE D
ACTGGAATGGATTGTCAGCTCTCCATCCTCCTCCTTCTC-3′
EphA3
ACTCTCGAGTCAcagatcctcttctgagatgagtttttgttc(Myc)GAACACGGGAACTGGGCCA
EP
TTCTTTGATTGCG-3′; GAPDH 5′-GCCAAAAGGGTCATCATCTC -3′ and 5′-
AC C
GTAGAG GCAGGGATGATGTTC-3′. PCR conditions were initial denaturation at 95°C for 2 min, followed by 30 cycles of amplification at 95°C for 30 s, 55°C for 45 s, and 72°C for 1 min, and a final extension at 72°C for 15 min.
2.5 Western blot analysis
6
ACCEPTED MANUSCRIPT
Whole cell lysates were prepared and quantified using standard protocols, and then separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
RI PT
and electrophoretically transferred to nitrocellulose membranes. The membranes were blocked in blocking buffer and incubated with monoclonal anti-Myc antibody (Cell
SC
Signaling Technology, Danvers, MA, USA). The membranes were stripped and
M AN U
incubated with monoclonal anti-GAPDH antibody (Cell Signaling) to confirm equal loading.
2.6 Plasmid construction and transfection
TE D
A plasmid carrying mutated EphA3 cDNA was constructed using a GeneArt® SiteDirected Mutagenesis System (Invitrogen) according to the manufacturer’s instructions,
(impacts
the
ligand
domain)
was
constructed
AC C
cDNA
EP
and named pcDNA3.1-EphA3-Myc. The expression vector carrying EphA3-T37K
TCAATCTACTGGATTCAAAAAAAATTCAAGGGGA-3′
using and
primers
5'5'-
TTTTGAATCCAGTAGATTGACTTCATTGGA-3', the expression vector carrying EphA3-S792P (impacts the kinase domain) sequence was constructed using primers 5'AGATCCCAATCAGGTGGACACCACCAGAAGCTAT-3'
and
5'-
TGTCCACCTGATTGGGATCTTCCCTCCTCT-3'. All constructed plasmids were
7
ACCEPTED MANUSCRIPT
verified by DNA sequencing. IMCE cells were transfected with the plasmids using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. After 24 h,
RI PT
cells were passaged at 1:5 and selected with G418 (Sigma; St Louis, MO, USA) for 2 weeks at an increasing concentration from 200 to 800 µg/ml. The stably transfected
M AN U
SC
clones were harvested and expanded for subsequent experiments.
2.7 Cell viability assay
Cell viability was assayed using the CellTiter 96® AQueous One Solution Cell Proliferation Assay (CellTiter96; Promega, Madison, WI, USA) according to the manufacturer’s
TE D
instructions. Briefly, the cells were seeded into 96-well plates and cultured for up to 7 days. At the end of each period, 10 µL MTT solution was added and the cells were
EP
incubated for an additional 4 h, after which 150 µL dimethyl sulfoxide (DMSO) was
AC C
added to each well and mixed thoroughly. The optical density of each well was measured with a spectrophotometer (UV5100, Shanghai, China). The data are expressed as mean ± standard error of the mean (SE). The experiments were performed in triplicate and repeated at least three times.
2.8 Clonogenicity assay
8
ACCEPTED MANUSCRIPT
For colony formation assay, cells were plated in 6-well plates using a two-layer soft-agar system as described previously [17]. After 3 weeks of incubation at 37°C in 5% CO2, the
RI PT
colonies containing 50 cells or more were counted. All experiments were conducted at
SC
least three times in triplicate.
2.9 Wound healing assay
M AN U
Cell were plated in 12-well plates at 2 × 105 cells/well in 500 µL DMEM (containing 10% FCS). At 100% cell confluency, a scraped line (wound) was created using a pipette tip and scratch width was marked. Cells were washed with ice-cold PBS and further
TE D
cultured in 500 µL DMEM medium at 37°C for 24 h.Wound closure was imaged under
EP
an inverted microscope.
AC C
2.10 Transwell migration and invasion assay Transwell filter inserts (8-µm pore size) for 24-well plates were purchased from Costar (Cambridge, MA, USA). In brief, 5 × 104 cells/well in 500 µL DMEM (containing 1% FCS) were seeded in the upper chamber, and 500 µL DMEM (containing 10% FCS) was added to the lower chamber. Cells were incubated at 37°C for 24–36 h. Cells remaining on the surface of the upper chamber were removed by a cotton swab while
9
ACCEPTED MANUSCRIPT
cells migrated into the bottom of the filters were fixed with 4% formaldehyde and stained with 0.5% crystal violet. Cells were counted in five photographed fields.
RI PT
In the invasion assay, transwell filters were pre-coated with Matrigel, and the assay was performed following the migration assay protocol. Both cell migration and invasion
2.11 In vivo tumorigenic assay
M AN U
SC
assays were performed in triplicate and repeated three times.
Animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of the Affiliated Tumor Hospital of Harbin Medical University. Nude mice (5-6
TE D
weeks old) were maintained in a pathogen-free environment at experimental animal center of the Affiliated Tumor Hospital of Harbin Medical University, and divided into
EP
four groups: Group A, injected subcutaneously with 8 × 108 IMCE-neo cells; Group
AC C
B, injected subcutaneously with 8 × 108 IMCE-Wt-EphA3 cells; Group C, injected subcutaneously with 8 × 108 IMCE-EphA3-S792P cells; and Group D, injected subcutaneously with 8 × 108 IMCE-EphA3-T37K cells. Tumor volume (V, cm3) was evaluated after 4 weeks based on tumor length (l), width (w), and height (h): V= l × w × h × 0.5236.
10
ACCEPTED MANUSCRIPT
2.12 Microarray profiling and data processing IMCE (IMCE-neo cell), Wt-EphA3-transfected IMCE, and EphA3-T37K-transfected
RI PT
IMCE cells were subjected to microarray to establish differential gene expression profiles. mRNA microarray contained 23,420 probes corresponding to 17,298 genes.
SC
Probes matched to multiple genes were eliminated. Probe expression values represented
M AN U
gene expression levels, and values generated from multiple probe sets were averaged. lncRNA microarray included 34,735 probes. If the lncRNA matched to a single probe, its expression level was same as that of the probe; if the lncRNA matched to several probes, its expression level was the same as that of the longest probe.
TE D
Differentially expressed genes and lncRNAs between IMCE and IMCE-Wt-EphA3 or IMCE-EphA3-T37K ce lls were defined as those with fold changes ≥ 1.5. Genes close
EP
to lncRNAs were filtered using the human refSeq reference genome (hg19). Genes
AC C
located most closely to the region 500 kb upstream from the transcription start site (TSS) were considered to be potentially affected by lncRNAs.
2.13 Functional analysis of differentially expressed genes and lncRNAs Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was performed using DAVID Functional Annotation Bioinformatics Microarray Analysis
11
ACCEPTED MANUSCRIPT
(http://david.abcc.ncifcrf.gov/home.jsp) for differentially expressed genes, and the hosts or nearby genes of differentially expressed lncRNAs, respectively. Fisher’s exact test was
SC
adopted for GO_BP term enrichment analysis.
RI PT
applied to determine statistical significance. Enrichment Map plugin in Cytoscape was
M AN U
2.14 Construction of a protein-protein interaction sub-network for lncRNAs and mRNAs
An integrated protein-protein interaction (PPI) sub-network was established based on the following databases: Biomolecular Interaction Network Database (BIND), the Biological
TE D
General Repository for Interaction Data sets (BioGRID), the Database of Interacting Proteins (DIP), the Human Protein Reference Database (HPRD), IntAct, the Molecular
EP
INTeraction database (MINT), the mammalian PPI database of the Munich Information
AC C
Center on Protein Sequences (MIPS), PDZBase (a PPI database for PDZ-domains) and Reactome. The common mRNAs and the nearby genes for lncRNAs were set as seed genes. The sub-network was derived from the seed genes, nearby lncRNAs, and the genes connected to the seed genes in the PPI network. The network was constructed using Cytoscape (http://www.cytoscape.org/).
12
ACCEPTED MANUSCRIPT
2.15 Statistical analysis Statistical analysis was performed using SAS software (version 9.2). The differences
RI PT
between the groups were analyzed by paired t-test or χ2 test. P values less than 0.05 were
SC
considered statistically significant.
M AN U
3. Results
3.1 Upregulation of EphA3 expression in colorectal cancer tissues and colorectal epithelial cell lines
Immunohistochemistry (IHC) showed that EphA3 protein was expressed in 72% (49 of
TE D
68) colorectal cancer tissues with a broad range of colorectal cancer types, but was not expressed in paired normal colorectal mucosa (Fig. 1). EphA3 expression was associated
EP
with patient age (P = 0.015), tumor differentiation (P = 0.001), and lymph node
AC C
metastases (P = 0.039) (Table 1). RT-PCR analysis showed that endogenous EphA3 expression was very low in conditionally immortalized mouse colonic epithelial cell line IMCE cell line, but much higher in YAMC-Ras and IMCE-Ras cell lines with malignant phenotypes (Fig. 1F).
3.2 EphA3 overexpression promoted colony formation of colorectal epithelial cells
13
ACCEPTED MANUSCRIPT
We stably overexpressed EphA3 in IMCE cells by transfecting pcDNA3.1-EphA3-Myc, pcDNA3.1-EphA3-T37K-Myc and pcDNA3.1-EphA3-S792P-Myc into IMCE cells.
transfected IMCE cells, but not in the control (Fig. 1G).
RI PT
Western blot analysis showed that exogenous EphA3-Myc was highly expressed in
SC
MTT assay showed that there was no remarkable difference in cell viability among the
M AN U
transfected cells (Fig. 1H). Colony formation assay showed significant difference in colony formation between EphA3-tranfected cells and control cells, especially in those transfected with constitutively active EphA3-T37K mutant (Fig. 1I).
TE D
3.3 EphA3 overexpression promoted colorectal epithelial cell migration and invasion Wound-healing and transwell assays showed that EphA3 overexpression promoted IMCE
EP
cell migration and invasion. In particular, EphA3-T37K mutant induced even faster
AC C
wound closure and invasion (Fig. 2A,B). Furthermore, scanning electron microscope (SEM) micrographs revealed striking changes in IMCE cell morphology following WtEphA3, EphA3-T37K, and EphA3-S792P overexpression. We observed increased lamellipodia and ruffles in cells overexpressing Wt-EphA3, and increased filopodia-like protrusions on the surface of EphA3-T37K- and S792P-transfected cells (Fig. 2C).
14
ACCEPTED MANUSCRIPT
3.4 EphA3 promoted tumorigenicity of colorectal epithelial cells in nude mice Stable IMCE-neo-, IMCE-Wt-EphA3-, IMCE-EphA3-T37K-, and IMCE-EphA3-S792P-
RI PT
transfected cells were injected into athymic nude mice. All ten mice injected with IMCEEphA3-T37K cells formed tumors by day 13, whereas mice injected with IMCE-Wt-
SC
EphA3 or IMCE-EphA3-S792P cells formed tumors with a longer latency period. As the
M AN U
control, mice injected with IMCE-neo cells did not develop tumors (Fig. 3A,B). Tumor volume in mice injected with IMCE-EphA3-T37K cells was significantly increased compared to other cells (P < 0.001; Fig. 3C). Since aberrant cell proliferation is a major feature of the tumors, we then used proliferating cell nuclear antigen (PCNA) as cell
TE D
proliferation marker. Infiltrating lymphocytes in mice injected with IMCE-EphA3 but not IMCE-neo cells showed positive PCNA staining. In addition, positive PCNA staining
AC C
EP
was only significant in IMCE-EphA3-T37K cells compared to other cells (Fig. 3D, E).
3.5 Upregulated and downregulated mRNAs and lncRNAs between IMCE-neo and EphA3-T37K cells
We performed microarray analyses to investigate global mRNA and lncRNA patterns in IMCE-neo, IMCE-Wt-EphA3, and IMCE-EphA3-T37K cells (Fig. 4A,B). Although mRNA expression patterns among these three cell lines were very similar, a total of 607
15
ACCEPTED MANUSCRIPT
downregulated and 2,394 upregulated genes were identified between IMCE-neo and IMCE-EphA3-T37K cells by differential expression analysis with a fold change ≥ 1.5.
RI PT
GO and KEGG analyses showed that upregulated genes were mainly involved in transcription regulation and responses to hormone stimulus, and were enriched in eight Fig. S1A, B
. In contrast,
SC
pathways including MAPK and VEGF signaling pathways
M AN U
downregulated genes were involved in DNA packing, mitosis, gonad development, and other GO categories, and were enriched in three pathways including arginine and proline metabolism, apoptosis, and regulation of the actin cytoskeleton
Fig. S1C, D
.
A total of 668 downregulated and 2,769 upregulated lncRNAs were identified as
TE D
differentially expressed between IMCE-neo and IMCE-EphA3-T37K cells. Host or nearby genes for lncRNAs were identified and used for GO and KEGG enrichment
EP
analysis. Neighboring genes for upregulated lncRNAs were mainly involved in basic
AC C
cellular functions such as transmembrane transport, apoptosis, transcription, and muscle organ development, and were enriched in several pathways including cell adhesion junction, arrhythmogenic right ventricular cardiomyopathy and MAPK signaling pathway (Fig. S2A, B
. Downregulated lncRNAs were mainly important in
programmed cell death and negative transcription regulation, and were enriched in three
16
ACCEPTED MANUSCRIPT
pathways including aminoacyl-Trnat biosynthesis, steroid hormone biosynthesis, and endocytosis
Fig. S2C, D .
RI PT
Among the mRNAs and genes near lncRNAs, 242 were upregulated genes while 17 were downregulated between IMCE-neo and EphA3-T37K cells. Functional enrichment
SC
analysis indicated that these 259 genes were enriched in pathways such as response to
M AN U
hormone stimulus, carboxylic acid biosynthetic process, and some other GO categories (Table 2).
PPI network was integrated from eight PPI databases, consisting of 80,980 edges and 13,361 nodes. The 259 genes consistently regulated at the mRNA and lncRNA levels
TE D
were used as seed genes in the integrated network. In addition, the subnetwork included 2,474 nodes and 2,848 interactions, among which a small number of nodes were
EP
presented with high degrees (hub nodes) while the majority of the nodes were presented
AC C
with low degrees (Fig. 4C). The hub genes in the subnetwork included GRB2, BRCA1, NFKB1, SFN, FN1, WDR5, PDPK1, SH3GL3, DHX38, SREBF2, QKI, and PMS2 (Table 3).
3.6 Upregulated and downregulated mRNAs and lncRNAs between IMCE-neo and Wt-EphA3 cells
17
ACCEPTED MANUSCRIPT
Compared to IMCE-neo cells, IMCE-Wt-EphA3 cells had 820 downregulated and 1,768 upregulated genes. GO and KEGG analyses showed that upregulated genes were mainly
RI PT
involved in the regulation of DNA metabolic processes and RNA elongation, and were enriched in pathways such as lysine degradation and terpenoid backbone biosynthesis . Downregulated genes were mostly involved in ion transmembrane
SC
Fig. S3A, B
M AN U
transport, cell proliferation, chromatin assembly and some other basic GO categories, and were enriched in various pathways including focal adhesion and the regulation of the actin cytoskeleton
Fig. S3C, D
.
There were 784 downregulated and 2,428 upregulated lncRNAs between IMCE-neo and
TE D
IMCE-Wt-EphA3 cells. Neighboring genes for upregulated lncRNAs were mainly involved in basic cellular functions such as cell death, adhesion, and positive regulation
EP
of transcription, and were enriched in pathways including adhesion junction and p53 Fig. S4A, B
. Downregulated lncRNAs were mainly associated
AC C
signaling pathway
with the regulation of protein phosphorylation, cell proliferation, or gene expression, and were enriched in apoptosis pathway Fig. S4C, D
.
Among the mRNA and the near lnRNA, 142 were upregulated and 32 were downregulated between IMCE-neo and IMCE-Wt-EphA3 cells. Functional enrichment analysis indicated that these 174 genes were enriched in the pathway of cell cycle
18
ACCEPTED MANUSCRIPT
(Table 4). We mapped these 174 genes into the integrated network as seed genes. Then the sub-network was constructed by interactions of these seed genes and nearby lncRNAs
SC
in this subnetwork included WDR5, NFX1, MYH9 (Table 5).
RI PT
(Fig. 4D). We found 1174 nodes and 1169 interactions in the sub-network. The hub genes
M AN U
3.7 Identification of genes and pathways for tumorigenicity
To identify genes and pathways that are associated with tumorigenicity, we defined differential expression of mRNAs or lncRNAs between IMCE-neo and Wt-EphA3 as Set 1, and that between IMCE-neo and EphA3- T37K as Set 2. Since Wt-EphA3 cells had
TE D
lower tumorigenicity than EphA3-T37K cells, we considered the overlap of Set 1 and Set 2 as the candidates for low tumorigenicity and the genes in Set 2 but not in Set 1 as the
EP
candidates for high tumorigenicity. Consequently, we identified 812 upregulated and 300
AC C
downregulated candidate genes associated with weak tumorigenicity (Fig. 5A). These genes were mainly involved in cell cycle regulation, muscle adaptation, terpenoid backbone biosynthesis, nitrogen metabolism, RIG-I-like receptor signaling pathway, and melanoma development (Fig. 5B,C). On the other hand, we identified 1,582 upregulated and 307 downregulated candidate genes linked to high tumorigenicity (Fig. 5A). These
19
ACCEPTED MANUSCRIPT
genes were involved in cell adhesion, GTPase activity, and metabolic processes such as lipid metabolism and VEGF signaling (Fig. 5D,5E).
RI PT
Similarly, we found 1,380 upregulated and 376 downregulated lncRNAs associated with weak tumorigenicity (Fig. 6A). The genes nearby these lncRNAs were mainly enriched
SC
for common cell functions such as transcription, apoptosis, and neuronal differentiation.
M AN U
Two pathways, namely axon guidance and cell adhesion molecules, were associated with weak oncogenesis (Fig. 6B,6C). We also found “nearby or hosted genes” of the 1,386 upregulated and 292 downregulated lncRNAs in the high tumorigenicity group. These genes can mediate neuronal differentiation, negative regulation of transcription,
TE D
programmed cell death, and other functions for gene pathways related to cancer progression such as MAPK signaling pathway, endometrial cancer, prostate cancer, and
AC C
EP
other pathways linked to cancer (Fig. 6D, 6E).
4. Discussion
EphA3 plays important role in embryo development and regulates cell adhesion, mobility, and growth.[8] EphA3 was overexpressed or mutated in a variety of human cancers including colorectal cancer, melanoma, and glioma[8,11,12,14,18–20] A previous study demonstrated that EphA3 mutations in lung tissues promoted lung cancer
20
ACCEPTED MANUSCRIPT
development, and an EphA3 mutation-associated gene signature was associated with poor patient survival.[12] In this study, we first confirmed EphA3 protein overexpression
RI PT
in colorectal cancer tissue specimens, and then manipulated its expression in rat colorectal epithelial cell lines. We found that EphA3 overexpression was significantly
SC
associated with patient age, tumor differentiation, and lymph node metastasis. In
M AN U
addition, the overexpression of EphA3 and its constitutively active mutants significantly promoted colony formation, migration and invasion of colorectal epithelial cells. Furthermore, EphA3 overexpression, in particular the active mutants, significantly promoted the tumorigenicity of colorectal epithelial cells in nude mouse xenograft model.
TE D
At the molecular level, EphA3 differentially regulated the expression of genes and lncRNAs that were associated with cancer- and angiogenesis-related gene pathways.
EP
Thus, our current study provides additional evidence for oncogenic activity of EphA3 in
AC C
colorectal cancer. Additional studies are needed to investigate molecular mechanisms underlying EphA3 overexpression and oncogenic activities in colorectal cancer. Several studies have assessed EphA3 mutations in different human cancer tissues and demonstrated that EphA3 is a tumor suppressor gene
[8, 11,12,14]
. Other studies have
indicated that EphA3 expression may inhibit tumor malignant behaviors in vitro[21–23]. However, a recent ex vivo study showed that EphA3 expression was upregulated in
21
ACCEPTED MANUSCRIPT
gastric carcinoma tissues and was associated with tumor TNM stage, angiogenesis, and poor prognosis[24]. In colorectal cancer, EphA3 is reported to be overexpressed in tissues
RI PT
and associated with poor prognosis[13]. These data indicate that EphA3 has different roles in different human malignancies, which may depend on EphA3 protein localization[8],
SC
downstream signaling pathway [25, 26], or active or inactive mutation in EphA3 gene [12]. In
M AN U
this study, we mutated EphA3 at T37K and S792P sites as active mutants and found that both Wt-EphA3 and S792P mutants were able to induce colorectal cell tumorigenesis, whereas T37K mutant had a strong oncogenic activity. Indeed, a previous study suggested that EphA3 does not possess the characteristics of a classical tumor suppressor
TE D
gene, as the inactivation of both alleles is required to induce carcinogenesis[27]. If EphA3 has tumor suppressor activities that depend on ephrin and kinase activity, its tumor-
EP
promoting activities that depend on crosstalk with other signaling molecules do not
AC C
require ephrin binding or kinase activity[25, 26]. In this case, mutations that impair ephrin binding and kinase activity could shift the balance towards tumor-promoting effects of EphA3. Moreover, mutated EphA3 may have dominant negative effects and inhibit the function of wild-type receptor. In this study we further explored the underlying molecular events after the overexpression of wild-type and active mutated EphA3 in colorectal epithelial cells. We
22
ACCEPTED MANUSCRIPT
found that upregulated mRNAs between IMCE-neo and EphA3-T37K cells mainly enriched transcription regulation and the response to hormonal stimulus in eight
RI PT
pathways, including cytosolic DNA-sensing pathway, MAPK and VEGF signaling pathways. Upregulated angiogenesis-related genes such as VEGF are known to promote [30]
. MAPK signaling
SC
tumor progression, tumor cell growth, invasion, or metastasis
M AN U
pathway regulates cell proliferation and differentiation[31]. In contrast, the downregulated genes between IMCE-neo and EphA3-T37K cells were significantly concentrated in DNA packing, mitosis, gonad development and other GO terms and enriched in three pathways including those related to arginine and proline metabolism, apoptosis, and the
TE D
regulation of the actin cytoskeleton. The neighboring genes in upregulated lncRNAs were mainly involved in transmembrane transport, apoptosis, transcription and muscle
EP
development, and were enriched in several pathways, including adherent junction,
AC C
ARVC, pathways in cancer, MAPK signaling pathway. However, downregulated lncRNAs were mainly concentrated in programmed cell death, negative regulation of transcription, and enriched in three pathways aminoacyl-Trna biosynthesis, steroid hormone biosynthesis and endocytosis. However, it remains to be determined how EphA3 regulates these gene pathways to promote colorectal epithelial cell malignant transformation in vitro and in vivo. Further mechanistic studies on the association of
23
ACCEPTED MANUSCRIPT
EphA3 with these pathways will help provide insight into oncogenic potential of EphA3 in CRC.
RI PT
In summary, our study provides both in vitro and in vivo evidence supporting oncogenic activities of EphA3 in colorectal epithelial cells. Our data demonstrate that EphA3 is
SC
overexpressed in colorectal cancer tissues and induces colorectal epithelial cell malignant
M AN U
transformation. EphA3 overexpression may serve as a biomarker for early diagnosis of colorectal cancer and the prediction of tumor progression.
Acknowledgments
TE D
This work was supported by the National Natural Science Foundation of China (No. 81272703), the Natural Science Foundation of HeiLongJiang Province (No. QC08C56)
EP
and the Scientific Research Foundation for Returned Scholars, Department of Education
AC C
of HeiLongJiang Province (No. 1154h17).
Conflict of interest
The authors declare that they have no conflicts of interest.
24
ACCEPTED MANUSCRIPT
References 1. Karsa LV, Lignini TA, Patnick J, Lambert R, Sauvaget C. The dimensions of the
RI PT
CRC problem. Best Pract Res Clin Gastroenterol. 2010; 24: 381-396.
2. Ho AS, Chen CH, Cheng CC, Wang CC, Lin HC, Luo TY, `et al. Neutrophil elastase
SC
as a diagnostic marker and therapeutic target in colorectal cancers. Oncotarget.
M AN U
2014; 5(2): 473-480.
3. Saika K, Sobue T. Cancer statistics in the world. Gan To Kagaku Ryoho. 2013; 40: 2475–2480.
4. Center MM, Jemal A, Ward E. International trends in colorectal cancer incidence
TE D
rates. Cancer Epidemiol Biomarkers Prev. 2009; 18:1688-1694. 5. Jemal A, Siegel R, Ward E, Hao Y, Xu J, et al. Cancer statistics, 2008. CA Cancer J
Markowitz SD, Bertagnolli MM. Molecular Origins of Cancer: Molecular Basis of
AC C
6.
EP
Clin. 2008; 58 : 71-96.
Colorectal Cancer. N Engl J Med. 2009; 361 (25): 2449–2460.
7.
Muzny DM, Bainbridge MN, Chang K, Dinh HH, Drummond JA, Fowler G, et al. Comprehensive molecular characterization of human colon and rectal cancer. Nature. 2012; 487: 330-337.
8.
Lisabeth EM, Fernandez C, Pasquale EB. Cancer somatic mutations disrupt functions
25
ACCEPTED MANUSCRIPT
of the EphA3 receptor tyrosine kinase through multiple mechanisms. Biochemistry. 2012; 51: 1464-1475.
RI PT
9. Batlle E, Bacani J, Begthel H, Jonkheer S, Gregorieff A, van de Born M, Malats N, Sancho E, Pawson T, Gallinger S, Pals S, Clevers H. EphB receptor activity
SC
suppresses colorectal cancer progression. Nature. 2005;435:1126-1130.
M AN U
10. Li S, Ma Y, Xie C, Wu Z, Kang Z Fang Z, Su B, Guan M. EphA6 promotes angiogenesis and prostate cancer metastasis and is associated with human prostate cancer
progression.
10.18632/oncotarget.4088.
Oncotarget.
2015;
6:22587-22597.
doi:
TE D
11. Wood LD, Calhoun ES, Silliman N, Ptak J, Szabo S, et al. Somatic mutations of GUCY2E, EPHA3, and NTRK3 in human cancers. Hum Mutat. 2006; 27: 1060-1061.
EP
12. Zhuang G, Song W, Amato K, Hwang Y, Lee K, Boothby M. Effects of cancer-
AC C
associated EphA3 mutations on lung cancer. J Natl Cancer Inst. 2012; 104:1182– 1197.
13. XiHQ Zhao P. Clinicopathological significance and prognostic value of EphA3 and CD133 expression in colorectal carcinoma. J Clin Pathol. 2011; 64:498-503.
14. Sjöblom T, Jones S, Wood LD. The consensus coding sequences of human breast and colorectal cancers. Science. 2006; 314(5797): 268-274.
26
ACCEPTED MANUSCRIPT
15. Keane N, Freeman C, Swords R， Giles FJ. EPHA3 as a novel therapeutic target in the hematological malignancies. Expert Rev Hematol. 2012; 5:325-340.
RI PT
16. Whitehead RH, Joseph JL. Derivation of conditionally immortalized cell lines containing the Min mutation from the normal colonic mucosa and other tissues of an
SC
“Immortomouse”/Min hybrid. Epithelial Cell Bio. 1994 Jul; 3(3): 119–125.
M AN U
17. Munker R, Norman A, Koeffler HP. Vitamin D compounds. Effect on clonal proliferation and differentiation of human myeloid cells. J Clin Invest. 1986; 78: 424–430.
18. Easty DJ, Bennett DC. Protein tyrosine kinases in malignant melanoma. Melanoma
TE D
Res. 2000; 10:401-411.
19. Wykosky J, Debinski W. The EphA2 receptor and ephrinA1 ligand in solid tumors:
EP
function and therapeutic targeting. Mol Cancer Res. 2008; 6(12):1795-1806.
AC C
20. Greenman C, Stephens P, Smith R, Dalgliesh GL, Hunter C, Bignell G. Patterns of somatic mutation in human cancer genomes. Nature. 2007; 446: 153-158.
21. Lee DJ, Schonleben F, Banuchi VE, Qiu W, Close LG, et al. Multiple tumorsuppressor genes on chromosome 3p contribute to head squamous cell carcinoma tumorigenesis. Cancer Biol Ther. 2010; 10:689-693. 22. Dottori M, Down M, Huttmann A, Fitzpatrick DR, Boyd AW. Cloning and
27
ACCEPTED MANUSCRIPT
characterization of EphA3 (Hek) gene promoter: DNA methylation regulates expression in hematopoietic tumor cells. Blood. 1999; 94:2477-2486.
RI PT
23. Guan M, Liu L, Zhao X, Wu Q, Yu B, et al. Copy number variations of EphA3 are associated with multiple types of hematologic malignancies. Clin Lymphoma
SC
Myeloma Leuk. 2011; 11:50-53.
M AN U
24. Xi NQ, Wu XS, Wei B, Chen L. Aberrant expression of EphA3 in gastric carcinoma: correlation with tumor angiogenesis and survival. J Gastroenterol. 2012; 47: 785-794.
25. Pasquale EB. Eph receptors and ephrins in cancer: bidirectional signaling and
TE D
beyond. Nat Rev Cancer. 2010; 10:165-180.
26. Miao H, LiDQ, Mukerjee A, Guo H, Petty A, et al. EphA2 mediates ligand-
EP
dependent inhibition and ligandindependent promotion of cell migration and invasion
AC C
via a reciprocal regulatory loop with Akt. Cancer cell. 2009; 16:9-20. 27. Lahtela J, Pradhan B, Narhi K, Hemmes A, Sarkioja M, et al. The putative tumor suppressor gene EphA3 fails to demonstrate a crucial role in murine lun tumorigenesis or morphogenesis. Dis Model Mech. 2015; 8(4):393-401.
28. Hoek K, Rimm DL, Williams KR, Zhao H, Ariyan S, et al. Expression profiling reveals novel pathways in the transformation of melanocytes to melanomas. Cancer
28
ACCEPTED MANUSCRIPT
Res. 2004; 64:5270-5282. 29. Valsesia A, Rimoldi D, Martiner D, Ibberson M, Benaglio P, et al. Network-guided
RI PT
analysis of genes with altered somatic copy number and gene expression reveals pathways commonly perturbed in metastatic melanoma. Plos One. 2011; 6: e18369.
SC
30. Lee SH, Jeong D, Han YS, Baek MJ. Pivotal role of vascular endothelial
M AN U
growthfactor pathway in tumor angiogenesis. Ann Surg Treat Res. 2015; 89(1):1-8. 31. Rovida E, Stecca B. Mitogen-activated progein kinases and Hedgehog-GLI signaling in cancer: A crosstalk providing therapeutic opportunities? Semin Cancer Biol. 2015;
AC C
EP
TE D
35:154-167.
29
ACCEPTED MANUSCRIPT
Figure legends
RI PT
Figure 1. EphA3 was overexpressed in CRC tissues and promoted colony formation of colorectal epithelial cell cells. Immunohistochemistry staining of EphA3 in normal
SC
mucosae (A), differentiated adenocarcinoma (B,C), mucinous adenocarcinoma (D), and
M AN U
signet-ring cell carcinoma (E). RT-PCR analysis of EphA3 mRNA level in different cells (F). Western blot analysis of EphA3 protein in IMCE cells stably transfected with EphA3 wild-type and mutant constructs (G). MTT assay of IMCE cells stably transfected with EphA3 wild-type and mutant constructs (H). Colony formation assay of IMCE cells
TE D
stably transfected with EphA3 wild-type and mutant constructs (I).
EP
Figure 2. EphA3 promoted the migration and invasion of colorectal epithelial cells.
AC C
A. Stably transfected cells were subjected to wound healing assay. IMCE-EphA3-T37K had significantly higher rates of colorectal epithelial cell migration. B. Stably transfected cells were subjected to Transwell assay. IMCE-EphA3-T37K induced higher rates of cell invasion. C. SEM analysis of the morphology of stably transfected cells. Cells were fixed in 2.5% glutaraldehyde for 24 h, dehydrated with a graded ethanol series, and then dried in CO2. The cell samples were then mounted in aluminum and were sputter-coated with
30
ACCEPTED MANUSCRIPT
gold for SEM using a FEI Quanta 200 FEG SEM (FEI, Hillsboro, Oregon, USA).
RI PT
Figure 3. EphA3 promoted the growth of colorectal epithelial cells in nude mouse model. A. Nude mouse xenograft assay. Stable IMCE-neo, IMCE-Wt-EphA3, IMCE-
SC
EphA3-S792P, and IMCE-EphA3-T37K-transfected cells were subcutaneously injected
M AN U
into athymic nude mice and the mice were sacrificed after 80 days of cell injection. B. Tumor-free mice in three groups. C. Tumor volume in three groups. D. Typical images of PCNA staining of xenografts in three groups. E. Quantitation of PCNA staining of
TE D
xenografts in three groups.
Figure 4. Microarray profiling of global mRNA and lncRNA expression and the
EP
interaction network. A. Circos v0.62 was used to profile the changes of global mRNA
AC C
and lncRNA expression in stable IMCE-neo, Wt-EphA3, and EphA3-T37K-transfected cells. B. Quantitation of global expression levels of mRNA and lncRNA in the cells. C. The differentially expressed lncRNAs could form a sub-network in EphA3-T37K cells. (D. The differentially expressed lncRNAs could form a subnet work in Wt-EphA3 cells.
Figure 5. Functional analysis of mRNAs between IMCE-Wt-EphA3 and IMCE-
31
ACCEPTED MANUSCRIPT
EphA3-T37K. There were 812 upregulated and 300 downregulated genes associated with weak tumorigenicity (A), and these genes were mainly enriched in four gene
RI PT
pathways (B and C). There were 1,582 upregulated and 307 downregulated genes associated with high tumorigenicity (A), and these genes were mainly enriched in ten
M AN U
SC
gene pathways (D and E).
Figure 6. Functional analysis of LncRNAs between IMCE-Wt-EphA3 and IMCEEphA3-T37K. There were 1,380 upregulated and 376 downregulated lncRNAs associated with weak tumorigenicity (A), and they were mainly enriched in two gene
TE D
pathways (B and C). There were 1,386 upregulated and 292 downregulated lncRNAs associated with high tumorigenicity (A), and they were mainly enriched in eight gene
AC C
EP
pathways (D and E).
32
ACCEPTED MANUSCRIPT Table 1. Association of EphA3 expression with clinicopathological features of colorectal cancer patients EphA3 Character
N
Positive
Negative
< 60
21
11
≥ 60
47
39
Male
52
37
Female
16
12
Gross Type 1
30
2
22
3
16
15
10
15
7
10
4
0
0
0
26
13
13
42
36
6
14
7
7
54
42
12
Positive
8
5
3
Negative
60
44
16
TE D
Well Moderate
EP
Poor
0.76
4
20
Differentiation
0.015
9
M AN U
Sex
10
SC
Age (yrs.)
(n=19)
RI PT
(n=49)
p value
0.28
0.001
LN metastasis
AC C
Positive Negative
0.039
Distant metastasis
0.52
ACCEPTED MANUSCRIPT Table 2. Gene enriched function by EphA3-T37K Description
Count
P value
GO:0009725
Response to hormone stimulus
13
0.0042
GO:0048545
Response to steroid hormone stimulus
9
0.0046
GO:0046394
Carboxylic acid biosynthetic process
8
0.0051
GO:0016053
Organic acid biosynthetic process
8
0.0051
GO:0009719
Response to endogenous stimulus
13
0.0090
AC C
EP
TE D
M AN U
SC
RI PT
Term
ACCEPTED MANUSCRIPT Table 3. List of hub genes for EphA3-T37K lncRNA
Regulation
Degree
GRB2
RP11-16C1.3
Up
522
BRCA1
BRCA1
Up
207
NFKB1
AF213884.2
Up
166
SFN
RP1-50O24.6
Up
156
FN1
AC012462.1
Up
WDR5
XLOC_007596
Up
PDPK1
RP11-20I23.8
Up
SH3GL3
SH3GL3
Up
DHX38
AK055364
Up
SREBF2
RP5-821D11.7
QKI
XLOC_005531
PMS2
PMS2
TANK
AK027541
ITGB3BP
ITGB3BP
UBTF
RI PT
Gene
86
71
M AN U
SC
50
47
47
47
Up
43
Up
42
Up
41
Up
41
RP5-882C2.2
Up
40
WASL
AL110181
Up
39
ID2
AC011747.7
Up
37
GSN
RP11-477J21.6
Up
34
XLOC_011384
Up
29
XLOC_007069
Up
27
SHANK2
ANO1-AS2
Up
24
AURKA
XLOC_013807
Up
24
ANAPC10
ANAPC10
Up
23
NR2C2
BC092452
Up
23
UBE2G2
AL773604.8
Up
23
AC C
PLAT
EP
MEF2A
TE D
Up
ACCEPTED MANUSCRIPT Table 4. Gene enriched function by Wt-EphA3 Description
Count
P value
GO:0022403
Cell cycle phase
11
0.0047
GO:0022402
Cell cycle process
13
0.0056
GO:0000279
M phase
9
0.010
GO:0051726
Regulation of cell cycle Regulation of transcription from RNA
GO:0006357
9
0.011
14
0.015
AC C
EP
TE D
M AN U
SC
polymerase II promoter
RI PT
Term
ACCEPTED MANUSCRIPT Table 5. List of hub genes for Wt-EphA3
Degree
XLOC_007596
Up
71
NFX1
RP11-255A11.13
Up
54
MYH9
RP4-633O19__A.1
Up
44
SMARCA5
GUSBP5
Up
PRKAR2A
PRKAR2A-AS1
Down
BRE
XLOC_001407
Up
MAD2L1
RP11-96A1.5
FSCN1
ZNF815P
AURKA
XLOC_013807
ANAPC10
ANAPC10
NR2C2
BC092452
MAP2K6
AC003051.1
SC
WDR5
EP AC C
RI PT
Regulation
40
40
35
Up
32
Down
28
M AN U
lncRNA
Up
24
Up
23
Up
23
Up
20
TE D
Gene
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Ephrin Type-A Receptor 3 (EphA3) belongs to the ephrin receptor subfamily of the protein tyrosine kinase family, and altered EphA3 expression or activity is associated with carcinogenesis, including colorectal cancer. This study assessed the expression of EphA3 in colon cancer, using both colorectal cancer tissue specimens and rat colon epithelial cell lines. The results showed that EphA3 overexpression in tumor tissues was associated with patient age, tumor differentiation, and lymph node metastasis. The expression of EphA3 and its constitutively active mutants promoted colony formation, migration, and invasion, and induced tumorigenicity of colon epithelial cells in nude mice. In addition, cDNA and lncRNA microarray profiling data revealed that differentially expressed genes and lncRNAs in EphA3-expressing cells were associated with cell proliferation, invasion, and angiogenesis. The findings of our study reveal additional mechanisms underlying the oncogenic effects of EphA3 on colorectal cells, which could provide novel targets for the prevention, early diagnosis, and treatment of colorectal cancer.
ACCEPTED MANUSCRIPT Conflict of interest statement This work was supported by the National Natural Science Foundation of China (Grant No 81272703), the Natural Science Foundation of HeiLongJiang Province (Grant No
RI PT
QC08C56) and the Scientific Research Foundation for Returned Scholars, Department of Education of HeiLongJiang Province (Grant No. 1154h17).
AC C
EP
TE D
M AN U
SC
The authors declare that they have no conflicts of interest.