Article Text

Download PDFPDF

Original article
Gastric tumour-derived ANGPT2 regulation by DARPP-32 promotes angiogenesis
  1. Zheng Chen1,2,
  2. Shoumin Zhu1,
  3. Jun Hong1,
  4. Mohammed Soutto1,
  5. DunFa Peng1,
  6. Abbes Belkhiri1,
  7. Zekuan Xu2,
  8. Wael El-Rifai1,3,4
  1. 1Department of Surgery, Vanderbilt University Medical Center, Nashville, Tennessee, USA
  2. 2Department of General Surgery, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
  3. 3Department of Cancer Biology, Vanderbilt University Medical Center, Nashville, Tennessee, USA
  4. 4Department of Veterans Affairs, Tennessee Valley Healthcare System, Nashville, Tennessee, USA
  1. Correspondence to Professor Wael El-Rifai, Department of Surgery, Vanderbilt Ingram Cancer Center, Vanderbilt University Medical Center, 760 PRB, 2220 Pierce Avenue, Nashville, TN 37232-6308, USA; wael.el-rifai{at}vanderbilt.edu

Statistics from Altmetric.com

Significance of this study

What is already known on this subject?

  • Angiopoietin 2 plays a key role in angiogenesis.

  • DARPP-32 proteins overexpression was frequently observed along the stages of gastric carcinogenesis.

  • DARPP-32 proteins promote chemotherapeutic drug resistance and survival of cancer cells.

  • DARPP-32 promotes invasion of gastric cancer cells.

What are the new findings?

  • This study demonstrates that DARPP-32 proteins upregulate ANGPT2 mRNA and protein expression levels in gastric cancer cells.

  • DARPP-32 proteins regulate ANGPT2 expression and secretion through activation of STAT3 in gastric cancer cells.

  • ANGPT2 is induced in human gastric tumour epithelial cells, not in tumour associated vascular endothelial cells, by DARPP-32.

  • This study strongly suggests that the DARPP-32-STAT3-ANGPT2 axis regulates angiogenesis in gastric cancer.

How might it impact on clinical practice in the foreseeable future?

  • Understand the mechanisms of how DARPP-32 proteins enhance angiogenesis in gastric cancer.

  • DARPP-32-STAT3-ANGPT2 axis provides a new paradigm in gastric carcinogenesis.

  • The future development of potential inhibitors against DARPP-32 and ANGPT2 may offer a therapeutic window in gastric cancer.

Introduction

Almost one million cases of gastric cancer are diagnosed each year, establishing this disease as the fourth most common cancer and the second leading cause of cancer related deaths worldwide.1 ,2 In some regions of the world such as in Asia and Latin America, gastric carcinoma is the most common malignancy, and the incidence of gastric cancer is almost 10-fold higher than in the USA.3–5 Because of the lack of early specific symptoms, the diagnosis of gastric cancer is typically delayed in most patients until cancer has invaded the muscularis propria and patients present with advanced stages, becoming at higher risk of poor response to therapy and disease recurrence.6–8

Formation of tumour blood vessels is a crucial event during cancer formation and metastasis.9 Tumour neovascularisation depends on the production of specific angiogenic factors; either by host or tumour cells shifting the angiogenic balance towards a proangiogenic phenotype.10 Angiopoietin 2 (ANGPT2) is a family member of the human ANGPT-TIE system.11 It is primarily produced by endothelial cells12 and stored in Weibel-Palade bodies from where it can be rapidly released upon stimulation to act as an autocrine regulator of endothelial cell functions. Recent studies have shown that the overexpression of ANGPT2 correlates with poor prognosis in several cancers.13–16 ANGPT2 plays a key role in tumour initiation17 and increases the number of tumour vessel sprouts, possibly owing to the decreased pericyte coverage and more unstable vessels.18 ANGPT2 acts as a vessel destabilising agent that induces permeability and leads to dissociation of cell-cell contacts in cultured endothelial cells.19 The latest reports have suggested that ANGPT2 can be regulated by vascular endothelial growth factor A (VEGF-A), insulin-like growth factor 1, hypoxia inducible factors 1 and platelet-derived growth factor B.11 ,12 ,20 In addition, tumour necrosis factor-α (TNF-α)-induced activation of NF-κB can upregulate ANGPT2 expression in human umbilical vein endothelial cells (HUVECs).21 Consistent with the proposed link between inflammation and angiogenesis, activation of the Janus kinase 2 (JAK2)/signal transducer and activator of transcription 3 (STAT3) signalling pathway has been implicated in tumour angiogenesis in many types of human cancers.22 ,23 In fact, the inhibition or knockdown of STAT3 has been shown to suppress angiogenesis in different human cancers.24–26 However, the role of STAT3 in regulating the ANGPT-TIE system in cancer cells has not been explored.

Dopamine and adenosine 3′,5′-cyclic monophosphate-regulated phosphoprotein Mr 32000 (DARPP-32), also known as protein phosphatase 1 regulatory (inhibitor) subunit 1B, is abundantly expressed in spiny neurons of the neostriatum.27 We have previously shown that the DARPP-32 gene also encodes alternatively spliced mRNA that generates an additional protein isoform known as truncated DARPP-32 (t-DARPP).28 DARPP-32 and t-DARPP are frequently overexpressed in gastric adenocarcinomas.28 ,29 DARPP-32 promotes activation of chemokine (C-X-C motif) receptor 4 and cancer cell invasion.30 In addition, overexpression of DARPP-32 proteins is associated with a potent antiapoptotic advantage for gastric cancer and breast cancer cells through a p53-independent mechanism that involves preservation of the mitochondrial membrane potential, activation of phosphoinositide-3-kinase (PI3K-AKT), and increased B cell chronic lymphocytic leukemia (CLL)/lymphoma 2 (BCL2) levels.29 ,31 ,32 Additionally, DARPP-32 proteins promote resistance of cancer cells to gefitinib and trastuzumab by promoting stabilisation and interaction of epidermal growth factor receptor (EGFR), erb-b2 receptor tyrosine kinase 2 (ERBB2) and erb-b2 receptor tyrosine kinase 3 (ERBB3) which lead to activation of PI3K-AKT signalling.33–36

The primary objective of this study was to investigate the role of DARPP-32 proteins in promoting angiogenesis in gastric cancer. We have uncovered that DARPP-32 proteins enhance angiogenesis through regulation of tumour-derived ANGPT2. We have demonstrated that DARPP-32 proteins induce ANGPT2 expression in gastric cancer cells through regulation of STAT3 activity. These novel findings underscore the importance of DARPP-32 proteins in regulating angiogenesis, which is a crucial step in promoting gastric tumorigenesis.

Materials and methods

Detailed protocols and standard procedures are described in the online supplementary methods.

Cell culture

Human gastric cancer cell lines were obtained from American Tissue Culture Collection (Manassas, Virginia, USA) and maintained in culture using F12 medium (GIBCO, Carlsbad, California, USA) (AGS and SNU-16) or Dulbecco's modified Eagle's medium (GIBCO) (MKN-28) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, California, USA) and 1% penicillin/streptomycin (GIBCO). HUVECs were purchased from PromoCell (Heidelberg, Germany), and maintained in culture using endothelial cell growth medium (Cedarlane, Burlington, North Carolina, USA).

DARPP-32 and t-DARPP overexpression

The expression plasmids for FLAG-DARPP-32 and FLAG-t-DARPP were generated by PCR amplification of the full-length DARPP-32 cDNA. Flag-tagged DARPP-32 or flag-tagged t-DARPP were cloned into the adenoviral (pACCMV) shuttle vector, a kind gift from Dr David Carbone (Ohio State University). The recombinant adenoviruses were generated by cotransfecting human embryonic kidney (HEK)-293 cells (American Tissue Culture Collection) with the shuttle and (pJM17) backbone adenoviral plasmids (a gift from Dr Carbone) using the calcium phosphate transfection kit (Applied Biological Materials, Richmond, British Columbia, Canada). Stably transfected AGS cells expressing DARPP-32 or t-DARPP were generated using pcDNA3 plasmids (Invitrogen) as described previously.32 ,36

Western blotting

Western blotting was conducted as previously described.36 Detailed antibody information is included in online supplementary methods.

Quantitative real-time RT-PCR

Sixty-three gastric cancer and 39 normal gastric tissue samples were collected from the National Cancer Institute Cooperative Human Tissue Network and the pathology archives at Vanderbilt University Medical Center (Nashville, Tennessee, USA). All tissue samples were collected, coded and deidentified in accordance with the Vanderbilt University Institutional Review Board-approved protocols. Quantitative real-time PCR (qRT-PCR) was conducted as previously described.36

HUVEC tube formation assay

HUVEC tube formation assay in AGS and MKN-28 cells was conducted following the protocol of In Vitro Angiogenesis Assay Tube Formation Kit (Trevigen, Gaithersburg, Maryland, USA). A detailed protocol is included in online supplementary methods.

Immunofluorescence and immunohistochemistry

Immunofluorescence and immunohistochemistry were performed as previously described.36 Detailed antibody information is included in online supplementary methods.

Luciferase reporter assay

Luciferase activity was measured using the Dual-Luciferase Reporter Assay kit (Promega, Madison, Wisconsin, USA) according to the manufacturer's instructions. Detailed protocols and plasmid information are described in online supplementary methods.

In vivo experiments

Five-week-old female NIHS-Lystbg Foxn1nu Btkxid nude mice were purchased from Harlan Laboratories (Frederick, Maryland, USA) and maintained under specific pathogen-free conditions. Tetracycline inducible (Tet-on system) AGS cells stably expressing DARPP-32 or empty vector32 were injected subcutaneously (2×106 cells per site) into the flanks. Fresh doxycycline (400 µg/mL) was added to the drinking water three times a week starting from the injection day. Tumour volume was calculated by the formula: tumour volume=1/2 (length× width2). All mice were sacrificed and tumours were collected when the tumour volume of control group reached 400 mm3. The Vanderbilt Institutional Animal Care and Use Committee approved all animal work.

Results

Overexpression of DARPP-32 proteins upregulate ANGPT2 mRNA and protein expression in gastric cancer cells

The western blot analysis and qRT-PCR results demonstrated that DARPP-32 overexpression leads to upregulation of ANGPT2 at the mRNA and protein levels in AGS and MKN-28 gastric cancer cell lines, which express low levels of endogenous DARPP-32 proteins (figure 1A, B, see online supplementary figure S1A,B). We next examined if the phosphorylation of threonine 34 residue (T34) of DARPP-32 is essential for the induction of ANGPT2. The results indicated that mutant DARPP-32 (T34A) can induce ANGPT2 mRNA and protein levels suggesting that it is not a critical site (figure 1A, see online supplementary figure S1A). These results were further confirmed using truncated DARPP-32, known as t-DARPP, which lacks the N-terminal domain and the T34 site of DARPP-32 (figure 1B, see online supplementary figure S1B). These findings ruled out the N-terminal T34-dependent protein phosphatase 1 (PP1) regulatory function of DARPP-3237 as a mechanism regulating ANGPT2 in cancer. Using an inducible Tet-on expression system, our data indicated that the induction of AGS-Tet-on-DARPP-32 or AGS-Tet-on-t-DARPP stable cells with doxycycline for 48 h markedly increased ANGPT2 protein levels (figure 1C). These results confirmed that the T34 phosphorylation site of DARPP-32 is not necessary for the induction of ANGPT2 and suggested that DARPP-32 and t-DARPP are potent inducers of ANGPT2.

Figure 1

DARPP-32 proteins upregulate ANGPT2 expression in gastric cancer cells without activating NF-κB signalling pathway. (A and B) The qRT-PCR and western blot analysis of ANGPT2 mRNA (left panels) and protein (right panels) levels in AGS cells transiently overexpressing DARPP-32 (DP32), DARPP-32T34A mutant (DP32T34A) or t-DARPP (TDP). Error bars indicate SD, *p<0.05, **p<0.01 compared with control (One-way ANOVA). (C) Western blot analysis of ANGPT2, DARPP-32 and t-DARPP in AGS/Tet-on cells stably expressing empty vector (Ctrl), DARPP-32 (DP32) and TDP. (D and E) The qRT-PCR and western blot analysis of ANGPT2 expression in AGS cells transiently overexpressing DARPP-32, t-DARPP or control with or without TNF-α treatment; *p<0.05, **p<0.01, ***p<0.001 compared with control (One-way ANOVA). F) Western blot analysis showing phospho-P65 (S536) (P-P65), P65, DARPP-32 and t-DARPP in AGS cells transiently overexpressing DARPP-32, t-DARPP or control with or without TNF-α treatment. G) NF-κB luciferase reporter analysis in AGS cell stably expressing pGL4.32 (luc2P/NF-κB-RE/Hygro) and transiently overexpressing DARPP-32, t-DARPP or control treated with or without TNF-α; NS, not significant.

DARPP-32 proteins enhance TNF-α induced ANGPT2 expression through activation of STAT3

Previous studies have shown that TNF-α can induce ANGPT2 levels in HUVECs.21 Therefore, to investigate whether the induction of ANGPT2 expression by DARPP-32 proteins in cancer cells is mediated by TNF-α, we treated AGS cells transiently expressing DARPP-32, t-DARPP or empty vector with TNF-α overnight and evaluated ANGPT2 mRNA and protein expressions. As expected, our results showed that TNF-α induced ANGPT2 mRNA and protein levels in AGS cells. However, DARPP-32 or t-DARPP induced equal or higher levels of ANGPT2 mRNA and protein than treatment with TNF-α alone (figure 1D, E). Interestingly, our data indicated that the combination of TNF-α treatment and overexpression of DARPP-32 or t-DARPP induced significantly higher levels of ANGPT2 mRNA and protein than DARPP-32 or t-DARPP overexpression alone (figure 1D, E). These results suggested that DARPP-32 proteins could synergistically enhance TNF-α mediated induction of ANGPT2 expression in gastric cancer cells. In an attempt to determine the mechanism by which DARPP-32 proteins induce ANGPT2 expression, we investigated the role of DARPP-32 proteins in activating NF-κB. Western blot data indicated that DARPP-32 or t-DARPP expression has no effect on phosphorylation of NF-κB-p65 (S536) protein in AGS cells with or without TNF-α treatment (figure 1F). In line with these results, the luciferase reporter assay data indicated no significant difference in NF-κB reporter activity between control and DARPP-32 proteins expressing AGS cells with or without TNF-α treatment (figure 1G). Collectively, these results demonstrated no activation of the NF-κB pathway by DARPP-32 and t-DARPP in gastric cancer cells, suggesting that induction of ANGPT2 by DARPP-32 proteins involves a different signalling mechanism. We have previously shown that overexpression of DARPP-32 proteins activates the AKT pathway, promoting gastric cancer cell survival and chemotherapeutic drug resistance.32 Accordingly, we investigated if the DARPP-32-AKT axis is involved in the regulation of ANGPT2. Our data indicated that the inhibition of AKT by MK2206 in AGS cells did not abrogate the induction of ANGPT2 expression by DARPP-32 proteins (figure 2A). Surprisingly, the ANGPT2 expression levels were increased following inhibition of AKT (figure 2A and see online supplementary figure S2). Based on these results, we concluded that the induction of ANGPT2 by DARPP-32 proteins is independent of the AKT pathway in gastric cancer cells. Confirming our results in figure 1, DARPP-32 and t-DARPP expression had no effect on NF-κB but surprisingly induced phosphorylation of STAT3 (Y705), which was further enhanced following inhibition of AKT (figure 2B). The substantial increase in the levels of phospho-STAT3 following AKT inhibition in DARPP-32 protein expressing cells suggests that the DARPP-32-STAT3 axis may be involved in resistance to AKT inhibitors. The immunofluorescence data confirmed activation of STAT3 by DARPP-32, as indicated by increased nuclear p-STAT3 (Y705) levels in AGS cells (figure 2C–E). In accordance with this result, the luciferase reporter assay data indicated that DARPP-32 expression significantly induced STAT3 transcriptional activity with and without interleukin 6 treatment (p<0.01, figure 3A). To investigate the role of STAT3 in the regulation of ANGPT2 by DARPP-32 proteins, we blocked STAT3 activity with a specific inhibitor (AZD1480) or siRNA knockdown in AGS cells transiently expressing DARPP-32, t-DARPP or empty vector. The suppression of STAT3 significantly decreased the induction of ANGPT2 mRNA (p<0.05, figure 3B) and protein (figure 3C) expression by DARPP-32 proteins in AGS cells. Notably, the decrease in ANGPT2 protein levels was observed in cell lysate and conditioned media in response to STAT3 knockdown (figure 3C). Together, these results indicated that DARPP-32 proteins induce ANGPT2 mRNA and protein expression through activation of STAT3.

Figure 2

DARPP-32 proteins upregulate ANGPT2 expression and phosphorylate STAT3. (A and B) Western blot analysis of ANGTP2, phospho-AKT (S473) (P-AKT), phospho-P65 (S536), phospho-STAT3 (Y705) (P-STAT3) in AGS cells transiently overexpressing control, DARPP-32 or t-DARPP treated with or without AKT inhibitor MK2206 (5 µM, overnight). (C) Immunofluorescence for DARPP-32 and phospho-STAT3 (Y705) in AGS cells transiently overexpressing DARPP-32 or control (200×). (D) Quantitative data for phospho-STAT3 in panel C, ***p<0.001 (t test). E) Western blots confirming immunofluorescence data shown in panel C.

Figure 3

DARPP-32 protein-induced upregulation of ANGPT2 expression is dependent on STAT3 pathway. (A) STAT3 reporter analysis in AGS cells stably expressing DARPP-32 (DP32) or control (pcDNA) with or without interleukin (IL) 6 treatment (200 ng/mL, 4 h), *p<0.05, ***p<0.001 compared with AGS cells stably expressing pcDNA without IL-6 treatment (One-way ANOVA). (B) ANGPT2 mRNA level in AGS cells transiently overexpressing DARPP-32 proteins or control with or without STAT3 inhibitor AZD1480 (300 nmol/L, overnight), ***p<0.001 compared with AGS control cells without AZD1480 treatment (One-way ANOVA). (C) Western blot analysis of ANGPT2 in cell lysates and conditioned media of AGS cells transiently overexpressing DARPP-32 proteins or control, with control or STAT3 siRNA knockdown.

DARPP-32 proteins promote angiogenesis through upregulation of ANGPT2 expression

Our data indicated that DARPP-32 proteins induce ANGPT2 expression. The secreted ANGPT2 protein is known to promote angiogenesis; therefore, we investigated the role of DARPP-32 proteins in angiogenesis in gastric cancer. We used the HUVEC tube formation in vitro assay as a measure of angiogenesis. Conditioned media from AGS and MKN-28 cells transiently overexpressing DARPP-32 or t-DARPP contained higher levels of the ANGPT2 protein (figure 4A and see online supplementary figure S3C), and induced significantly more HUVEC tube formation than control cells (p<0.05, figure 4B and see online supplementary figure S3A,B). The removal of ANGPT2 from conditioned media of DARPP-32-overexpressing AGS cells using a specific ANGPT2 antibody binding method significantly decreased the HUVEC tube formation (p<0.01, figure 4C, D). The conditioned media from AGS wild type cells induced very weak tube formation compared with ANGPT2 or VEGF-α (see online supplementary figure S5A,B). These results clearly indicated that DARPP-32 proteins promote angiogenesis by inducing ANGPT2 expression in gastric cancer cells.

Figure 4

DARPP-32 proteins enhance angiogenesis in gastric cancer in vitro. (A) Western blot data showing ANGPT2 protein levels in cell lysates or conditioned media from AGS cells transiently overexpressing DARPP-32 proteins or control. (B) Human umbilical vein endothelial cells (HUVECs) tube formation analysis (left panel, 100×) and quantification data (right panel) using conditioned media from AGS cells transiently overexpressing DARPP-32 proteins or control, ***p<0.001 compared with AGS control cells (one-way ANOVA). (C) HUVEC tube formation analysis (left panel, 100×) and quantification data (right panel, one-way ANOVA) using conditioned media from AGS cells transiently overexpressing DARPP-32 or control, with or without ANGPT2 binding antibody. (D) Western blot analysis of ANGPT2 in conditioned media from the same cells as panel C.

DARPP-32 proteins regulate tumour growth and angiogenesis in xenograft mouse models

We next investigated the role of DARPP-32 in xenograft gastric tumour growth and angiogenesis in vivo. We used the Tet-on inducible AGS cells stably expressing DARPP-32 or empty vector to inject nude mice. Tumours derived from AGS-DARPP-32 cells grew substantially faster than the AGS control tumours (p<0.05, figure 5A–C). Western blot and qRT-PCR analyses confirmed the higher expression levels of DARPP-32, ANGPT2 and phospho-STAT3 (Y705) in the DARPP-32 xenograft tumours than control tumours (p<0.01, figure 5D, E). In addition, immunohistochemistry analysis indicated that DARPP-32 expression was associated with high protein levels of ANGPT2, phospho-STAT3 (Y705) and von Willebrand factor; a marker of blood vessels, endothelial cells and indicative of angiogenesis in DARPP-32 xenografts (p<0.001, figure 5F, G). On the other hand, we examined ANGPT2 expression levels in SNU-16-DARPP-32 shRNA or SNU-16-Scramble shRNA xenografts. As expected, Western blot analysis data indicated that ANGPT2 and phospho-STAT3 (Y705) protein levels were significantly lower in SNU-16 cells stably expressing DARPP-32 shRNA than control cells (figure 6A). As we previously reported,36 we confirmed that knockdown of DARPP-32 expression decreased xenograft tumour growth compared with control (p<0.05, figure 6B). Consistent with the in vitro data, knockdown of DARPP-32 expression reduced phospho-STAT3 (Y705) and ANGPT2 protein levels in DARPP-32 shRNA xenograft tumour samples as compared with control samples (figure 6C). Accordingly, immunohistochemistry data demonstrated that ANGPT2, phospho-STAT3 (Y705) and the von Willebrand factor expression levels were significantly lower in DARPP-32 knockdown xenograft tumour samples than control samples (p<0.001, figure 6D, E). Collectively, our results strongly suggested that DARPP-32 proteins enhance angiogenesis and tumour growth in vivo by upregulation of ANGPT2 expression.

Figure 5

DARPP-32 expression enhances xenograft gastric tumour growth and angiogenesis in vivo. (A) Representative xenograft tumours of sacrificed mice at end of experiment (day 62). AGS cells stably expressing Tet-on inducible DARPP-32 or control were injected subcutaneously (2×106 cells per site) into nude mice. (B) Tumour volume was measured at the indicated times; each data point represents the mean±SD for eight xenografts. *p<0.05, **p<0.01 (t test). (C) Quantification of tumour weight at the end of experiment. The tumour weight is indicated by mean±SD, *p<0.05 (t test). D) qRT-PCR analysis of DARPP-32 (left panel) and ANGPT2 (right panel) in xenograft tumour samples, **p<0.01, p<0.001 (one-way ANOVA). (E) Western blot analysis of phosphor-STAT3 (Y705), ANGPT2 and DARPP-32 in duplicate xenograft tumour samples. (F) Immunohistochemical staining (left panel) and quantification data (right panels) of DARPP-32, ANGPT2, phosphor-STAT3 (Y705) and von Willebrand factor from xenograft tumours (100×), each sample was amplified to show the details. G) Quantification data for panel F, *p<0.05, **p<0.01, ***p<0.001 compared with Tet-on control cells (t test).

Figure 6

Knockdown of endogenous DARPP-32 expression decreases ANGPT2 protein level, xenograft tumour growth, and angiogenesis. (A) Western blot analysis of DARPP-32, ANGPT2, phospho-STAT3 (Y705) (P-STAT3) proteins in SNU-16 cells stably expressing DARPP-32 shRNA (DP32 shRNA) or scrambled shRNA (Scr shRNA). B) SNU-16 cells stably expressing DARPP-32 shRNA or scrambled shRNA were injected subcutaneously (2×106 cells per site) into nude mice. Tumour volume was measured at the indicated times; each data point represents the mean±SD for 10 xenografts. (C) Western blot analysis of DARPP-32 and ANGPT2 proteins in SNU-16 xenograft tumour samples. (D) Immunohistochemical staining of DARPP-32, ANGPT2, phosphor-STAT3 (Y705) and von Willebrand factor in SNU-16 xenograft tumours (100×), each sample was amplified to show the details. (E) Quantification data for (D), *p<0.05, **p<0.01, ***p<0.001 compared with SNU-16 scrambled shRNA cells (t test).

DARPP-32 proteins and ANGPT2 are frequently co-overexpressed in human gastric cancer samples

Because of the in vitro data, we investigated the correlation between DARPP-32, t-DARPP and ANGPT2 mRNA expression in deidentified human gastric cancer tissue samples. Our data indicated that DARPP-32, t-DARPP and ANGPT2 are significantly co-overexpressed in gastric cancer samples (p<0.05, figure 7A–C). Strong positive correlations were found between ANGPT2 and DARPP-32 (r2=0.4, p<0.0001) or t-DARPP and DARPP-32 (r2=0.4, p<0.0001) mRNA expression levels (figure 7D, E). These clinical data suggested a possible positive regulation of ANGPT2 expression by DARPP-32 proteins in human gastric cancer.

Figure 7

ANGPT2 and DARPP-32 proteins are upregulated in human gastric cancer tissue samples. (A–C) qRT-PCR analysis of DARPP-32, t-DARPP and ANGPT2 mRNA expression in normal gastric and gastric cancer tissue samples. Horizontal line indicated mean. Significance was evaluated by Mann–Whitney U test. (D and E) Correlation between ANGPT2 and DARPP-32 (r2=0.4, p<0.0001) or t-DARPP and DARPP-32 (r2=0.4, p<0.0001) mRNA levels in gastric cancer tissue samples (relative mRNA expression levels were converted to log10 values, linear regression). (F) Immunohistochemical staining of ANGPT2 and DARPP-32 in normal and gastric cancer tissue samples (100×). Glandular and blood vessel tissues in the normal and gastric cancer samples were amplified to show the details of ANGPT2 and DARPP-32 expression.

Cytosolic co-overexpression of DARPP-32 and ANGPT2 proteins in human gastric epithelial tumour cells

To investigate whether DARPP-32 regulates ANGPT2 through a paracrine or an autocrine mechanism with regard to the cellular origin of ANGPT2, we examined ANGPT2 and DARPP-32 expression levels in human normal and gastric cancer tissue samples by immunohistochemistry. Our results demonstrated that normal gastric glandular tissue expressed low (or undetectable) levels of DARPP-32 and ANGPT2 proteins, whereas gastric cancer epithelial cells showed high levels of DARPP-32 and ANGPT2 predominantly in the cytoplasm (figure 7F). Meanwhile, blood vessel cells in gastric cancer tissues expressed low levels of endogenous ANGPT2 but not DARPP-32. As expected, more blood vessels, indicative of increased angiogenesis, were observed in cancer tissues overexpressing DARPP-32 than normal samples (figure 7F). These data clearly demonstrated that DARPP-32 regulates epithelial tumour-derived ANGPT2 through an autocrine mechanism in gastric cancer.

Discussion

DARPP-32 proteins are frequently overexpressed and amplified in gastric cancer,28 ,29 ,38 promoting cancer cell growth, invasion and survival.29 ,30 ,36 Angiogenesis is an important biological activity that promotes tumour growth and metastasis.39 ANGPT2 has been recently reported to play a key role in angiogenesis in different types of cancer.11 ,40

In this study, we investigated the expression and regulation of ANGPT2 in gastric cancer. Our results demonstrate that DARPP-32 proteins promote angiogenesis through the regulation of ANGPT2 expression and secretion from gastric cancer cells. The T34, located in the N-terminal domain of DARPP-32, plays a critical role in regulating PP1 activity in the brain.41 ,42 Our findings suggest that mutations of the T34 in DARPP-32 as well as t-DARPP, lacking the N-terminal domain, are able to induce ANGPT2. These findings indicate that PP1 is not a critical factor in regulating ANGPT2 in gastric cancer. On the other hand, our results demonstrate that DARPP-32 regulates ANGPT2 by activating STAT3. This is supported by several lines of evidence: (1) We showed that DARPP-32 proteins upregulated ANGPT2 mRNA and protein levels in different gastric cancer cell models; (2) Knockdown of DARPP-32 expression decreased the ANGPT2 protein level in the SNU-16 gastric cancer cell model; (3) DARPP-32 protein-induced ANGPT2 expression promoted angiogenesis as indicated by enhanced tube formation of HUVECs in vitro, and increased xenograft tumour growth and vascularisation in vivo; (4) Knockdown of DARPP-32 expression downregulated ANGPT2, decreasing xenograft tumour growth and angiogenesis in vivo; (5) Using STAT3 luciferase reporter and immunofluorescence analysis, we found that DARPP-32 proteins activate STAT3, which was essential for the upregulation of ANGPT2; (6) Clinical data from human primary gastric tumours indicated that DARPP-32 and ANGPT2 were frequently co-overexpressed and strongly correlated in gastric cancer.

Several cytokines such as hypoxia inducible factor 1 and VEGF can regulate ANGPT2 expression in different physiological and pathological conditions, establishing a link between inflammation and cancer.43 Previous studies indicated that TNF-α induces ANGPT2 expression in HUVECs, human lymphatic endothelial cells or blood vascular endothelial cells.21 ,44 Therefore, we postulated that ANGPT2 regulation by TNF-α might be mediated by a possible DARPP-32-NF-κB axis in gastric cancer cells. Our data showed that DARPP-32 proteins enhanced ANGPT2 expression independent of NF-κB. Interestingly, our data indicated that DARPP-32 proteins, in combination with TNF-α, synergistically induced higher ANGPT2 levels than DARPP-32 proteins alone in gastric cancer cells. We have previously shown that the DARPP-32 proteins play an important role in the activation of PI3K-AKT pathway in upper GI cancer cells.32 ,35 In this study, we examined the potential implication of AKT pathway in the regulation of ANGPT2 by DARPP-32 proteins. Surprisingly, we found that pharmacological inhibition of AKT enhanced, not suppressing, DARPP-32 induced ANGPT2 expression. Therefore, our results suggested that the regulation of ANGPT2 by DARPP-32 proteins was not mediated by NF-κB or AKT signalling pathways.

STAT3 regulates several critical functions such as proliferation and angiogenesis in human normal and malignant tissues.45–47 Herein, we demonstrated that transient overexpression of DARPP-32 proteins induced strong phosphorylation and nuclear localisation of STAT3 (Y705) in gastric cancer cell lines. Conversely, knockdown of DARPP-32 decreased the phospho-STAT3 (Y705) protein level in gastric cancer cells. Importantly, the induction of ANGPT2 by DARPP-32 proteins was sharply decreased by the specific STAT3 inhibitor (AZD1480) or siRNA knockdown in AGS cells. These results clearly demonstrated that the induction of ANGPT2 by DARPP-32 proteins is mediated by the activation of STAT3.

ANGPT2 is a protein secreted by endothelial cells under normal physiological conditions.12 We investigated the role of secreted DARPP-32 induced ANGPT2 protein in angiogenesis using the HUVEC tube formation assay. Our data showed that conditioned media from DARPP-32 protein overexpressing cells promoted tube formation in HUVECs. We confirmed that this induction of tube formation was dependent on ANGPT2, as the removal of ANGPT2 from the conditioned media abrogated the promotion of tube formation by DARPP-32. Our in vivo xenograft mouse model also demonstrated the proangiogenic function of DARPP-32. These results clearly indicated that DARPP-32 enhances angiogenesis through the regulation of ANGPT2 in gastric cancer. VEGF-α is an important cytokine that induces angiogenesis in cancer.48 ,49 We found that gastric cancer cells express endogenous levels of VEGF-α that are not modulated by DARPP-32. While our data demonstrate that DARPP-32 proteins induce angiogenesis, mainly through the regulation of ANGPT2, we cannot rule out the contribution of other existing factors such as VEGF-α in this process.

Our in vivo animal data demonstrated that DARPP-32 was overexpressed in xenograft tumours and this was accompanied with an increase in ANGPT2 expression, blood vessels and tumour growth; validating the data from the in vitro cell models. Our in vitro data also indicated that the proangiogenic function of DARPP-32 was dependent on the induction of ANGPT2 expression. In contrast, the knockdown of DARPP-32 in xenograft tumours significantly decreased ANGPT2 expression levels, angiogenesis and tumour growth. ANGPT2 plays an important role in blood vessel formation, especially tumour initiation.17 In fact, a higher level of ANGPT2 was correlated with poor prognosis in breast cancer,13 hepatocellular carcinoma14 and leukaemia.50 Notably, our data indicated higher levels of ANGPT2 expression in primary human gastric tumours and a strong correlation between ANGPT2 and DARPP-32. We could not confirm the correlation with clinical outcome due to our small sample size; analysis of additional clinical samples could establish this pattern. Our findings demonstrated that DARPP-32 and ANGPT2 were highly expressed in the cytoplasm in gastric cancer cells but only ANGPT2 was weakly expressed in tumour-associated endothelial cells in human gastric tumour samples. This is in line with the results from Moon and his colleagues’ study, showing the expression of ANGPT2 in hepatocellular cancer cells.51 Interestingly, tumour stromal tissue displayed even lower expression level of ANGPT2 than normal gastric connective tissue. These results suggested that DARPP-32-induced ANGPT2 in gastric cancer may function as a promoter of angiogenesis through a paracrine mode of action. Further studies will be required to investigate the contribution of endogenous ANGPT2 in gastric cancer epithelial cells to angiogenesis through regulation of blood vessel endothelial cells. Collectively, our data clearly indicated that the DARPP-32-STAT3 axis enhances gastric cancer angiogenesis through the upregulation of ANGPT2.

In conclusion, our findings demonstrate, for the first time, to our knowledge, that DARPP-32 proteins play a major role in promoting angiogenesis by regulating the expression and secretion of ANGPT2 in gastric cancer cells. The DARPP-32-dependent angiogenesis provides a new paradigm in gastric tumorigenesis which could have an impact on its treatment.

References

View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.