GATA4 inhibits cell differentiation and proliferation in pancreatic cancer.

Pancreatic ductal carcinoma (PDAC) is a common malignant tumor of the digestive system. GATA4 is one of the transcriptional regulatory factors, which regulates the development of endoderm-derived organs, including heart and gut. GATA4 may act as a putative tumor suppressor gene. However, the role of GATA4 in pancreatic carcinogenesis is not yet clarified. This study showed that GATA4 was highly expressed in pancreatic cancer tissues, and its expression level was positively related to the grade of pathological differentiation, suggesting that it may contribute to the progression of pancreatic neoplasia. Ectopic expression of GATA4 gene reduced cell viability and interference of GATA4 expression significantly increased the colony formation ability of pancreatic cancer cells. Furthermore, GATA4 inhibited tumor growth in xenograft mice. Agilent expression microarray profiling analysis indicated that the genes with significant levels of differential expression in GATA4 over-expressing cells were enriched in the cell differentiation process. Analysis of KEGG signaling pathway demonstrated that the regulated genes were partially enriched in MAPK and JAK-STAT signaling pathways. Re-expression of GATA4 up-regulated P53 gene expression. Our data indicate that GATA4 gene might play a role in cell proliferation and differentiation during the progression of pancreatic cancer.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is a common malignant tumor of the digestive system and currently ranks seventh in cancer related mortality. The predicted mortality in 2030 will be ranking as second in the USA[1]. In China, the incidence of pancreatic cancer is rising [2], and the prognosis of PDAC is poor. The 5-year survival rate after diagnosis is less than 8% [3].

A mutation in K-RAS oncogene is currently considered as the initiating factor in pancreatic cancer that exogenously stimulates and activates MAPKs, NF-κB, and PI3K signaling pathway based on gene mutation and epigenetics modification[4]. Together, these pathways promote cell proliferation, migration, and invasion and regulate the mucin release, local microenvironment, and angiogenesis, wherein the normal pancreatic cells gradually develop to invasive ductal adenocarcinoma [5].

The GATA factors are the members of the transcriptional regulatory factor families, which contain two conserved zinc finger DNA-binding domains that trigger the activation of WGATAR sequence [6]. GATA1, 2, and 3 play roles in the formation and differentiation of hematopoietic stem cells, whereas GATA4, 5, and 6 are involved in guiding the endoderm differentiation into digestive tract and respiratory tract epithelium, liver, pancreas, lung, bladder, urethra, and most epithelial cells[7,8]. Previous studies suggested that GATA4 was closely correlated with the tumors of the digestive system; for example, GATA4 acted in agreement with GATA6 and exerted synergistic effects on HNF-4, which was required for the development of gastric cancer (GC) [9]. The methylation of GATA4 gene promoter has been also detected in gastric, esophageal, and colon cancers [10-12]. The restored expression of GATA4 suppressed colorectal cancer cell proliferation [13]. Another study suggested that GATA4 regulated the expression of Bcl-2 in granulosa tumor cells and inhibited cell apoptosis [14]. Although recent studies had demonstrated that GATA4 was expressed in invasive pancreatic cancer, and the robust expression is related to gender (more in women than men) [15,16]. However, the exact function of GATA4 in pancreatic carcinogenesis is yet to be elucidated.

Therefore, the aim of the present study was to explore the expression of GATA4 in pancreatic cancer and study the effects of GATA4 on the physiological behaviors of pancreatic cancer cells both in vitro and in xenograft mouse model. In addition, the gene expression profile of GATA4 transfection in the pancreatic cancer cells and its related enrichment signaling pathway were analyzed to elucidate the possible molecular mechanism underlying the role of GATA4 in pancreatic carcinogenesis.

Material and methods

Cell culture

Human pancreatic cancer cell lines, CFPAC-1 and PANC-1 were obtained from the Chinese Academy of Sciences Cell Bank (Shanghai, China). The cells were cultured in RPMI 1640 or DMEM medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS), 50 units/mL penicillin and 50 mg/mL streptomycin in a humidified atmosphere at 5% CO2. The cells were passaged at 80% confluency, using trypsin (Sigma-Aldrich, St. Louis, MO, USA), into T-75 flasks (Sarstedt, Nümbrecht, Germany) (density of 1×106 cells/flask).

Immunohistochemistry (IHC)

The expression of GATA4 in pancreatic tumors and normal pancreatic tissues was evaluated by IHC using an antibody against GATA4 (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) on commercial tissue arrays (Shanghai Outdo Biotech, Shanghai, China). The array contained 2 normal pancreas tissues, 4 chronic pancreatitis tissues, 19 PDAC specimens (tumor tissues and paired adjacent tissues), 4 metastatic pancreatic cancer tissues, 4 negative lymph nodes, and 8 positive lymph nodes tissues.

Briefly, the sections were dewaxed, hydrated, and washed. After neutralization of endogenous peroxidase and microwave-based antigen retrieval, the slides were preincubated with blocking serum, followed by overnight incubation with primary antibody. Subsequently, the sections were serially rinsed, incubated with the secondary antibody, and treated with HRP-conjugated streptavidin. The reactive products were visualized with 3, 3-diaminobenzidine tetrahydrochloride and counterstained with hematoxylin. The staining was assessed by two independent pathologists blinded to the origin of the samples. The staining intensity and extent of the stained area were graded according to the German semi-quantitative scoring system: staining intensity of the nucleus, cytoplasm, and/or membrane (no staining = 0; weak staining = 1; moderate staining = 2; strong staining = 3); the extent of stained cells (0% = 0, 1–24% = 1, 25–49% = 2, 50–74% = 3, 75–100% = 4). The final immunoreactive score (0–12) was determined by multiplying the intensity score with the extent of stained cells.

Plasmid construction

Full-length human GATA4 was amplified and subcloned into the pLV5-GFP lentivirus expression vector. The primers were 5′-GAGGATCCGCCACCATGGGGAAACTGCAGTCGAAG-3′ (F) and 5′-GATCTCGAGCTAGGACGGGTGGAAGTGGT-3′(R). The GATA4 shRNAs were designed using Oligo Designer 3.0 program and subcloned into pLV3-(shRNA)-GFP vector. The shRNA/siRNA sequence was as follows: 5 '-TTCTCCGAACGTGTCACGT-3'.

Retrovirus infection

After plasmid knockdown or over-expression of the GATA4 gene was successful, retroviruses were generated following the co-transfection of recombinant pLV5-GATA4 or pLV3-GATA4 RNAi plasmid with pRsv-REV, pMDlg-pRRE, pMD2G, and pLenOR-THM helper plasmid into 293T cells using the calcium phosphate method. The supernatants of the culture containing retroviruses were harvested 48 h post-transfection and passed through a 0.45-μm filter. Subsequently, the cells were seeded into 24-well plates before retroviral infection. Virus-containing supernatants were added to the cells for infection 24 h, and the medium was replaced every 24 h for 6 days consecutively.

Cell growth assay

Cells transfected with a control vector, p-GATA4 or p-GATA4 RNAi, were trypsinized and seeded at a density of 1×105 cells/well in 96-well plates, followed by proliferation for an additional 6 days. The proliferation was determined using the cell counting kit 8 (CCK-8) (Sigma) and absorbance measured at 450 nm and 630 nm with the EXL800 microimmunoanalyzer (BioTek, Burlington, VT, USA). The data were expressed as means ± SD. Experiments were performed in triplicate and repeated three times.

Flow cytometry analysis

Pancreatic cancer cells were transfected with a vector-control, p-GATA4 or p-GATA4 RNAi for 72 h. Subsequently, the attached and floating cells were harvested and fixed using 70% ethanol for a minimum of 48 h. Then, the cells were resuspended in 50 μg/mL propidium iodide (Sigma-Aldrich,) and 100 μg/mL RNase (Takara Biotechnology Inc, Shiga, Japan) for 30 min and analyzed using flow cytometry (FACSCalibur™ BD Biosciences, Franklin Lakes, NJ, USA).

Colony formation assay

Cells were trypsinized and the density adjusted to 1×106 cells/L. Then, the cells were resuspended in a medium containing 0.7% low-melting agarose and plated onto a solidified bottom layer medium containing 1.2% agarose in 6-well plates. The colonies formed at 2 weeks were stained with 1 mg/mL 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) for 6 h and images were captured.

Gene expression profiles analysis

Total cellular RNA was isolated from the stably transfected cells and the control using mirVanaTM RNA Isolation Kit (Applied Biosystems p/n AM1556, Foster city, CA) according to the manufacturer's instructions. RNA quality and quantity were assessed by agarose gel electrophoresis (1%) and spectrophotometric analysis (260/280 ratio). First strand complementary DNA (cDNA) was synthesized by reverse transcription with the T7 promoter primers. Subsequently, the cDNA was subjected to biotinylated cRNA nucleotide reaction in vitro transcription using T7 RNA Polymerase and Cy3-CTP. The labeling of cRNA, purification, and fragmentation were preformed according to the manufacturer's instructions (QIAGEN RNeasy Mini Purification cRNA Kit); the purified cRNA was hybridized on Agilent expression microarray chip, followed by washing, scanning, and analysis.

Western blot

Protein from pancreatic cancer cells was collected 48 hours after transfection and the soluble proteins were resolved on SDS-polyacrylamide gel and electrophoretically transferred to a polyvinylidene fluoride membrane (Millipore, Bedford, MA). Western blot was performed as described previously [17]. Antibodies were diluted according to manufacturer’s instructions. The primary antibodies were GATA4 (Santa Cruz, CA, USA), P53 (ZSGB-BIO, Beijing, China), ki67 (Santa Cruz, CA, USA), β-actin, (Beyotime, Nanjing, China) and GAPDH (ZSGB-BIO, Beijing, China). The data were normalized to β-actin or GAPDH.

In vivo tumorigenicity

GATA4 stably overexpressed or vectors-control cells (2×106cells/0.2 mL phosphate-buffered saline (PBS)) were subcutaneously injected into the axilla flank of 6-week-old male BALB/c nude mice. The tumor size was measured every 4 days for 2 weeks beginning 5 days after implantation. The tumor volume was calculated according to the following formula 1: V = (a2×b)/2, where V, volume (mm3); a, small diameter (mm); b, large diameter (mm). The relative tumor volume (RTV) was calculated according to the following formula 2: RTV = Vt/V0, where Vo indicated random grouping (D0) and Vt was the tumor volume for each measurement (DN) of tumor volume. All of the experimental procedures were performed in accordance with protocols approved by the Institutional Animal Care and Research Advisory Committee of the Chinese PLA General Hospital.

Apoptosis detection

TUNEL assay was used to detect apoptosis of the resected tumors and normal organs. The apoptosis of positive cells in five random viewpoints (×200) was estimated, and the mean and standard deviation were calculated for statistical analysis.

Statistical analysis

Data were expressed as the mean ± standard error as appropriate. All statistical analyses were performed using SPSS 17.0 for Windows (SPSS, Inc., Chicago, IL, USA). P-values for dichotomous variables were two-tailed and based on the Pearson’s test or the Pearson’s test with continuity correction. Continuous variables conformed to normal distributions were analyzed by Student's t-test. P<0.05 was considered as a statistically significant difference.

Results

GATA4 expression is positively associated with the developmental process of pancreatic cancer

The expression of GATA4 was firstly assessed using a tissue microarray with pancreatic samples by IHC. In contrast to the adjacent normal tissues, most of the pancreatic cancer samples were stained positive for GATA4 in the cytoplasm and partly in the nuclei of pancreatic cancer cells. The chronic pancreatitis samples were stained moderately positive for GATA4, whereas the normal pancreatic samples did not show any positive staining (Fig 1A–1C).

Fig 1

GATA4 is highly expressed in pancreatic cancer tissues.

(A, B, C) Representative images of immunohistochemistry (IHC) staining showing GATA4 expression in different pancreatic tissues: (A) Normal pancreatic tissue. (B) Chronic pancreatitis tissues. (C) Pancreatic cancer tissues. (D) High staining of GATA4 in pancreatic cancer tissues. (E)Low staining of GATA4 in pancreatic cancer tissues. Magnification: ×200. (F) The staining index of GATA4 in pancreatic cancer tissues and its relationship with cell differentiation. *P<0.05.

All the adjacent normal tissue samples (19/31) were stained negative (0) for GATA4, however, 70.97% of the pancreatic cancer samples (22/31) exhibited a high GATA4 staining index (4–12) (Table 1). Subsequently, the association between staining index of the GATA4 and clinicopathologic features were analyzed. A total of 31 cases in this tissue array, 1 with no age data, 1 with no gender data, 4 with no node metastasis data, 1 with no distant metastasis data, 12 with no differentiation data. The result were shown in Table 2, staining index of the GATA4 was associated with poor differentiation (p = 0.037) (Fig 1D–1F), indicating a positive relationship between the expression of GATA4 and the developmental process of pancreatic cancer. However, no significant association was found between the staining index of GATA4 and other clinicopathological features, such as age, gender, tumor size, lymph node metastasis, distant metastasis, and clinical stage (Table 2), implying that the expression level of GATA4 was an independent indicator of pancreatic cancer.

Table 1

The staining index of GATA4 in different tumor stage of pancreatic cancer.

Tumor stage	n	GATA4, 0-+, n(%)	GATA4,++-+++, n (%)
1A	3	2 (66.67)	1 (33.33)
1B	3	2 (66.67)	1 (33.33)
2A	4	0	4 (100)
2B	12	2 (16.67)	10(83.33)
4	9	3 (33.33)	6 (66.67)
total	31	9(29.03)	22(70.97)

no staining, 0; weak staining, +; moderate staining, ++; strong staining, +++

Table 2

Relationship between GATA4 scores of pancreatic cancer tissues and clinicopathologic features.

Variable	GATA4 scores	p-value
Age(y)    <50(n = 4)    ≥50(n = 26)	9.25±1.897.38±2.77	0.208
Gender    male(n = 24)    female(n = 6)	8.08±2.655.83±2.40	0.069
Tumor size(cm)    ≤2(n = 4)    >2(n = 27)	7.00±2.007.59±2.87	0.695
T    T1(n = 4)    T2(n = 16)    T3(n = 7)	7.00±2.007.37±3.098.85±2.27	0.445
N    N0(n = 10)    N1(n = 17)	7.70±2.917.70±2.78	0.996
M    M0(n = 21)    M1(n = 9)	7.71±2.877.22±2.73	0.666
Stage    1.2A(n = 10)    2B.4(n = 21)	7.70±2.917.43±2.75	0.803
Tumor differentiation    well/moderate(n = 11)    poor (n = 8)	7.00±2.579.50±2.07	0.037*

Data presented as mean±SD Statistically significant (*P-value <0.05)

Ectopic expression of GATA4 gene reduces cell proliferation and colony formation in pancreatic cancer cells

Next we study the role of GATA4 in pancreatic carcinogenesis in vitro. The expression vectors harboring the GATA4 gene, as well as, the empty vector control were transfected into CFPAC-1 cells, respectively. The stably transfected cells were obtained after 6 days. (Fig 2A). MTT assay showed that restored expression of GATA4 gene reduced cell proliferation as compared to that of the control cells (P<0.05) and the expression of ki67 protein was decreased compared to that of the control cells. (Fig 2B). Flow cytometric analysis indicated that both early apoptotic status and dead cells have no difference compared with the control cells. (Fig 2C). The colony formation ability was decreased compared to that of the control cells (30.33±2.52Vs.45.00±3.61, P<0.05) (Fig 2D).

Fig 2

GATA4 over-expression reduces cell proliferation and colony formation in pancreatic cancer cells.

(A) Fluorescent microscopic images showing stable expression of GATA4 in CFPAC-1 cells. Magnification: ×100 (B) CCK-8 activity assay showing the growth curves of cells stably expressing GATA4 or vector control and Western blotting identified the expression of ki67 protein was decreased compared to that of the control cells. (C) 7-ADD and Annexin-V positive staining results for detecting apoptosis and necrosis of cells stably expressing GATA4 or vector control. (D) Colony formation assay of cells stably expressing GATA4 or vector control. Magnification: ×100 pGATA4: stable expression plasmid of GATA4, pcon: vector control, *P<0.05.

Interference of GATA4 expression significantly increases the colony formation ability of pancreatic cancer cells

Then, the pGATA4-RNAi plasmid and the blank control was respectively transfected into the pancreatic cancer CFPAC-1 cells or PANC-1 cells (Fig 3A). The result of MTT assay exhibited that the cell proliferation of both cell lines transfected with pGATA4-RNAi plasmid had no significant difference as compared to that of the control cells (Fig 3B). Flow cytometry demonstrated that early apoptotic and dead cells had no change by GATA4i compared with the control of CFPAC-1 cells or PANC-1 cells (Fig 3C). In additional, reduction of GATA4 significantly increased the colony formation ability of both of pancreatic cancer cell lines (P<0.05) (Fig 3D).

Fig 3

GATA4 interference increases colony forming ability in pancreatic cancer cells.

(A) Fluorescent microscopic images showing the Interference of GATA4 expression in CFPAC-1 and PANC-1 cells. (B) CCK-8 activity assay showing the growth curves of cells transfected with GATA4 interference or control vector. (C) 7-ADD and Annexin-V positive staining results for detecting apoptosis of cells transfected with GATA4 interference or control vector. (D) Colony formation assay of cells transfected with GATA4 interference or control vector. Magnification: ×100 pGATA4RNAi:GATA4 interference vector, pcon: control vector *P<0.05.

Analysis of DEGs and its enrichment signaling pathway after GATA4 transfection into pancreatic cancer cells

Gene profiling is a gene detection method, especially suitable for the screening of differentially expressed genes (DEGs) [18]. The bioinformatics approach combined with the expression profiling techniques will obtain DEGs and the related signaling pathway. To explore the mechanism of GATA4 effects on the biology behavior of pancreatic cancer cells, agilent expression microarray profiling was used and screened out 6533 genes of significant levels of differential expression in GATA4-overexpressed cell lines as compared to pancreatic cancer cells CFPAC1 (>2-fold change; P<0.05); 2623 genes were significantly up-regulated and 3910 were significantly down-regulated in GATA4-overexpressing cell lines. The DEGs genes ontology (GO) analysis was performed using the GO software. The DEGs were classified into three functional groups: molecular function group, biological process group, and cellular component group (S1 Fig). As shown in Table 3, all the three groups demonstrated that the genes were primarily enriched in cell differentiation, cell junction, receptor activity, and signal transduction process. These results showed that most of the DEGs affected by GATA4 were significantly enriched in the cell differentiation process.

Table 3

Gene analysis and significant enriched GO terms of differential expressed genes (DEGs) after GATA4-overexpressing in pancreatic cancer cells.

Group	Term	Description	Gene Count	P-value
Biological process group	GO:0007165	signal transduction	174	4.05e-02
	GO:0007275	multicellular organismal development	150	4.26e-03
	GO:0050896	response to stimulus	114	2.7e-04
	GO:0055085	transmembrane transport	114	4.01e-03
	GO:0006811	ion transport	101	5.52e-04
	GO:0030154	cell differentiation	83	1.89e-02
	GO:0007608	sensory perception of smell	80	1.78e-03
	GO:0045944	positive regulation of transcription from RNA polymerase II promoter	76	4.73e-02
	GO:0007596	blood coagulation	76	9.38e-04
	GO:0007399	nervous system development	74	9.38e-04
	GO:0007186	G-protein coupled receptor protein signaling pathway	65	2.52e-04
	GO:0007268	synaptic transmission	59	3.78e-03
Cellular component group	GO:0016021	integral to membrane	641	4.66e-03
	GO:0005886	plasma membrane	575	2.48e-08
	GO:0005576	extracellular region	301	1.86e-02
	GO:0005887	integral to plasma membrane	191	3.83e-06
	GO:0030054	cell junction	96	6.99e-06
	GO:0045202	synapse	59	3.14e-04
	GO:0031410	cytoplasmic vesicle	51	1e-02
	GO:0045211	postsynaptic membrane	38	3.46e-04
	GO:0005929	cilium	24	4.98e-02
	GO:0045121	membrane raft	23	2.64e-02
	GO:0045095	cilium	22	4.05e-03
	GO:0005882	intermediate filament	21	1.81e-02
Molecular function group	GO:0004872	receptor activity	248	1.36e-07
	GO:0005509	calcium ion binding	110	1.3e-02
	GO:0043565	sequence-specific DNA binding	95	1.12e-02
	GO:0004930	G-protein coupled receptor activity	83	7.97e-07
	GO:0004984	olfactory receptor activity	75	1.09e-02
	GO:0005215	transporter activity	47	1.48e-02
	GO:0005216	ion channel activity	34	1.99e-03
	GO:0005244	voltage-gated ion channel activity	30	2.3e-02
	GO:0020037	heme binding	29	7.16e-03
	GO:0008237	metallopeptidase activity	20	2.27e-02
	GO:0005179	hormone activity	18	4.58e-02
	GO:0004497	monooxygenase activity	16	2.02e-03

The analysis of the DEGs signaling pathway enrichment was conducted using online platforms of KEGG Pathway. The regulated genes were mainly enriched in the neuroactive ligand-receptor interaction, olfactory transduction, MAPK signaling pathway, calcium signaling pathway, glutamatergic synapse, JAK-STAT signaling pathway, osteoclast differentiation, dilated cardiomyopathy, bile secretion, taste transduction, primary immunodeficiency, and histidine metabolism (Fig 4, Table 4). Furthermore, the pathway enrichment analysis showed that MAPK signaling pathway consisted of 50 DEGs including P53 gene (Fig 4, S2 Fig). These DEGs were mainly associated with proliferation and differentiation, whereas the JAK-STAT signaling pathway consisted of 28 DEGs (S3 Fig), which were mainly associated with apoptosis and cell cycle. Furthermore, western blotting suggested that p53 expression was up-regulated after GATA4 transfection.

Fig 4

The significantly enriched pathways after GATA4-overexpressed in pancreatic cancer cells.

(A) DEGs functional and signaling pathway enrichment were conducted using the online platform of KEGG pathway. Significantly enriched pathway ratio: the enriched genes in the pathway/ all enriched genes. (B) The expression of p53 was up-regulated by GATA4- overexpressed using western blotting identification.

Table 4

The significantly enriched pathway terms after GATA4-overexpressed in pancreatic cancer cells.

Description	Related Genes	P-value
Neuroactive ligand-receptor interaction	72	1.19e-09
Olfactory transduction	70	2.69e-03
MAPK signaling pathway	50	5.44e-03
Calcium signaling pathway	41	1.46e-04
Glutamatergic synapse	35	7.88e-06
Jak-STAT signaling pathway	28	4.66e-02
Osteoclast differentiation	24	4.29e-02
Dilated cardiomyopathy	18	4.24e-02
Bile secretion	17	8.92e-03
Taste transduction	16	6.92e-04
Primary immunodeficiency	9	3.29e-02
Histidine metabolism	8	3.53e-02

GATA4 inhibits tumor growth in xenograft mice

To further explore the effect of GATA4 on the growth of pancreatic cancer in vivo, we established the xenograft mouse model in which, GATA4 was over-expressed or only control cells. The tumor relative volume was smaller in xenograft mice injected with GATA4 overexpressed CFPAC-1 cells than those of control cells (6.19±1.99 Vs. 10.95±5.87, P = 0.038), the expression of ki67 protein was decreased as compared with the control cells(Fig 5A–5C). TUNEL assay indicated that the immune scores of cell apoptosis did not show significant difference between the two groups (234.52 ± 48.94 Vs. 208.02 ± 65.24, P>0.05) (Fig 5D and 5E). These results implied that the inhibitory effect of GATA4 on the tumor growth was mainly dependent on the reduced cell proliferation.

Fig 5

GATA4 inhibits tumor formation in xenograft mice.

(A) Representative results of xenograft tumors in nude mice for GATA4 over-expressed and control cells, The volume of tumors in nude mice measured at the 14 days after injection of GATA4 over-expressed or control pancreatic cancer cells. (B) Subcutaneous tumor growth curves of xenograft mice injected with GATA4 over-expressed and control cells at different time points and Western blotting identified the expression of ki67 protein was decreased compared to that of the control cells. (C) Representative photographs of IHC analysis of cell apoptosis in xenograft mice in the tumors caused by injection of GATA4 over-expressed or control cells. Magnification: ×200. (D) The quantitative results of TUNEL analysis detecting cell apoptosis in xenograft mice injected with GATA4 over-expressed or control vector. pGATA4: stable over-expression vector of GATA4, pcon: control vector, *P<0.05.

Discussion

In the current study, GATA4 was highly expressed in pancreatic cancer tissues. Its higher expression was associated with poor differentiation (p = 0.037), suggesting its expression was positively related to the degree of tumor development. It was well known that the initial step of PDAC development is the formation of pancreatic intraepithelial neoplasia (Pan-IN lesions), combined with inflammation and desmoplasia[19]. Bincy et al reported a high fat diet induced pancreatic inflammation, and then fibrotic stroma and numbers of PanINs in LSL-Kras/Ela-CreERT mice (with oncogenic Kras) [20] .while chronic pancreatitis may arise more earlier than Pan-IN lesions. Our study indicated GATA4 was not only increased in early pancreatic lesions such as chronic pancreatitis, but also in late PDAC, supporting its potential usefulness in the development of PDAC. No significant association was found between the staining index of GATA4 and other clinicopathological features excluded cell differentiation implying that the expression level of GATA4 was an independent indicator of pancreatic cancer. However, GATA4 immunoreactivity was detected in the cytoplasm and only partly in the nuclei, which was indiscrepancy with the previously report that GATA4 staining in the nuclei of pancreatic cancer, ovarian cancer and breast cancer [15,21,22]. We conferred that it was closely correlated with tumor heterogeneity [23]. Whether GATA4 transfers from nuclei to cytoplasm needs further study.

Since GATA4 was high expressed in PANC-1 cells and low expressed in CFPAC-1 cells [16]. The restored expression of GATA4 in CFPAC-1 cells reduced cell proliferation and colony formation in vitro and xenograft mice model in vivo were in according with the earlier report that re-expression of GATA4 suppressed colorectal cancer cell proliferation [13]. Flow cytometric analysis indicated whether GATA4 over-expression or inference had no effect on both early apoptotic status and dead cells in these two pancreatic cancer cell lines. This result had some difference compared with the previously report such that GATA4 regulated the expression of Bcl-2 and inhibited cell apoptosis in granulosa tumor cells [14].

GATA4 was highly expressed in pancreatic cancer tissues and its higher expression was associated with poor differentiation (p = 0.037). However, these in vitro results all suggest that GATA4 acted as a negative regulator in the developmental process of pancreatic cancer. Therefore, the positive relationship between the expression of GATA4 and the developmental process of pancreatic cancer might be a negative feedback of organism, which attempted to inhibit cancerous cell differentiation and proliferation by elevating the expression of GATA4. However, the exact role of overexpressed GATA4 in the regulation of cell differentiation and proliferation in human pancreatic cancer still needs be further determined.

The previous study showed GATA4 and tyrosine kinase receptor ERBB2 formed a negative feedback regulatory loop in breast cancer cells [24], in our study, gene expression analyses of pancreatic cancer cell transfection with GATA4 gene had identified several sets of coregulated genes harbored in the progression of pancreatic cancer accounting for DNA replication, cell cycle and apoptosis[25]. Further integrated bioinformatics analysis indicated a majority of DEGs were significantly enriched in cell differentiation processes, which were in accordance with the above data that GATA4 overexpression inhibited cell differentiation and proliferation. Some regulated genes were enriched in MAPK and JAK-STAT signaling pathways, further demonstrating that GATA4 took part in pancreatic carcinogenesis. In this study, there are 50 genes enriched in MAPK signaling pathway were identified [26]. Among them, p53 acted as a downstream effector of MAPK signaling pathway. As a well-known inhibitor of cell proliferation, p53 also promoted cell senescence to prevent the initiation of cancer [27]. Recent reports also showed that the positive expression of p53 was related to the late stage of pancreatic carcinoma [28-31].

Taken together, our study showed that GATA4 was highly expressed in pancreatic cancer tissues, and its expression was positively related to the degree of pathological process, indicating GATA4 is an independent indicator of pancreatic cancer. The re-expression of GATA4 inhibited tumor growth in vitro and in vivo. The regulated genes of GATA4 over-expressing were partially enriched in the MAPK and JAK-STAT signaling pathways, including p53, these results illustrated that GATA4 might play a role in pancreatic cancer biology behaviors. However, downstream effect of GATA4-MAPK signaling pathway remained poorly characterized. In order to determine potential of our findings for clinical applications in pancreatic cancer, further study with molecular mechanisms of GATA4 anti-tumor is needed.

Gene analysis and significantly enriched GO terms of differential expressed genes (DEGs) after GATA4 over-expressing in pancreatic cancer cells.

GO analysis classified the DEGs into 3 groups: (A) biological process group, (B) cellular component group, (C) molecular function group. Significantly enriched GO terms of DEGs based on their functions, ratio: the enriched genes in the function/the all enriched genes.

(TIF)

Click here for additional data file.

Significantly enriched genes in MAPK signaling pathway after GATA4 over-expressing in pancreatic cancer cells.

Blue: related genes in the pathway.

(TIF)

Click here for additional data file.

Enrichment genes in the JAK-STAT signaling pathway after GATA4 over-expressing in pancreatic cancer cells.

Blue: related genes in the pathway.

(TIF)

Click here for additional data file.