SUMO-specific protease 2-mediated deSUMOylation is required for NDRG2 stabilization in gastric cancer cells

Abstract

N-myc downstream regulated gene 2 (NDRG2) is frequently down-regulated in various cancers and functions as a candidate tumor suppressor gene. NDRG2 has been shown to be SUMOylated on the lysine 333 residue, which promoted its ubiquitination and sequentially degradation by the SUMO-targeted ubiquitin E3 ligase RNF4. However, how to regulated NDRG2 deSUMOylation process remains largely unknown. Here, we report that Sentrin/SUMO specific protease (SENP2) was down-regulated in clinic gastric cancer samples and possessed a tumor-suppressive role in gastric cancer. At the molecular level, we found that SENP2 interacts with NDRG2 and mediates the de-SUMOylation process of NDRG2. Overexpression of SENP2 stabilized NDRG2, whereas silencing SENP2 caused rapid NDRG2 SUMOylation and degradation, indicating SENP2 antagonizes NDRG2 ubiquitination and degradation, thereby promoting the stability and function of this protein. Thus, our study reveals that SENP2 acts as a tumor suppressor which is deregulated in gastric cancer and the specific de-SUMOylation activity of SENP2 for NDRG2 is critical for it stabilization as well as gastric cancer cells proliferation.

Keywords

SENP2 NDRG2 gastric cancer SUMOylation

1. Introduction

Gastric cancer (GC) is the third leading cause of cancer-related death worldwide and most gastric cancer patients are already diagnosed with late stage disease [1, 2, 3]. Despite research on GC has made some progress in recent years, the 5-year survival of patients with GC still not as satisfactory as expected [4, 5]. Thus, the exact molecular mechanisms still need further study.

SUMO is a small ubiquitin-like protein that can covalently modify a large number of proteins to alter their function, and localization and stabilization [6, 7]. SUMOylation is a highly dynamic and reversible process that is catalyzed by a activating (E1), conjugating (E2) and ligating (E3) enzyme cascade and reversed by a family of six SUMO-specific proteases (SENPs) [8, 9]. SENP2 is a nuclear envelope-associated protease that deconjugates a large number of SUMOylated proteins in vivo and has a broad de-SUMOylation activity in vitro [10]. Knockout of SENP2 in mice results in embryonic lethality, disturbing the p53-Mdm2 pathway, affecting the expansion of trophoblast progenitors and their maturation [11]. Further studies identified the pivotal roles of SENP2 in cell cycle and cell proliferation regulation [12, 13, 14, 15].

N-myc downstream-regulated gene 2 (NDRG2) is involved in multiple biological processes, including cell growth, differentiation, as well as apoptosis [16, 17]. NDRG2 functions as a tumor suppressor and its expression is down-regulated in various cancers, including breast cancer, colon cancer, melanoma, glio- blastoma, thyroid cancer and pancreatic cancer [16]. It has been reported that NDRG2 can be transcriptionally activated by p53 and involved in p53-mediated apoptosis pathway in certain tumor cells [18, 19]. Recently, NDRG2 is found to be a SUMOylated protein and modified by SUMO1 at K333 [20]. SUMOylated NDRG2 was rapidly underwent ubiquitination and sequentially degradation by the ubiquitin E3 ligase RNF4 [20, 21]. However, the specific de-SUMOylation protease that can reverse the SUMOylation of NDRG2-mediated processes is still unidentified.

Here, we show that SENP2 interacts with NDRG2 and mediates the de-SUMOylation process of NDRG2 to stabilize NDRG2. Moreover, SENP2 was down-regulated in gastric cancer samples which in turn caused NDRG2 destabilization in gastric cancer.

2. Materials and methods

Cell lines and clinic samples Seven human GC cell lines (NCI-N87, MKN74, AGS, NUGC-3, MKN45, MGC803, HGC-27) were obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). All cell lines were maintained in RPMI-1640 (Gibco BRL, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) except AGS in Ham’s F12 medium (Cellgro, Manassas, VA, USA) and incubated at an atmosphere containing 5% CO ${}_{2}$ at 37 ${}^{\circ}$ C.

Seven human GC cell lines (NCI-N87, MKN74, AGS, NUGC-3, MKN45, MGC803, HGC-27) were obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). All cell lines were maintained in RPMI-1640 (Gibco BRL, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) except AGS in Ham’s F12 medium (Cellgro, Manassas, VA, USA) and incubated at an atmosphere containing 5% CO ${}_{2}$ at 37 ${}^{\circ}$ C.

Five human GC cell lines (AGS, BGC823, MGC803, NCI-N87, and SGC7901) were obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). BGC823, MGC803, NCI-N87, and SGC7901 cells were maintained in RPMI-1640 (Gibco, NY, USA) supplemented with 10% fetal bovine serum (FBS); AGS cells were maintained in F12 medium (Gibco, NY, USA) with 10% FBS; 293T cells were maintained in DMEM with 10% FBS. All cells were incubated at an atmosphere containing 5% CO ${}_{2}$ at 37 ${}^{\circ}$ C. Tissue samples were obtained from 20 GC patients who underwent surgical resection at our department between 2014 and 2016. All patients provided informed consent for collection of the tissue samples and publication of the related data and this study was approved by the Ethics Committee of our hospital.

2.1 Cell proliferation assay

The cell proliferation was detected by 3-(4, 5-dime- thylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay. AGS cells were seeded into 96-well plates at 2000 per well and incubated at 37 ${}^{\circ}$ C with 5% CO ${}_{2}$ for different time points. After that, MTT solution was added to each well and the plates were incubated for another 4 h. The formazan crystals were then dissolved with 100 $\mu$ L dimethyl sulfoxide per well. The absorbance of individual wells was read using a microplate reader (SpectraMax M3, Molecular Devices, Sunnyvale, CA, USA).

Figure 1.

SENP2 is down-regulated in clinic gastric cancer samples. A. The mRNA expression levels SENP2 in 20 paired GC tissues and adjacent normal tissues were normalised to the expression of beta-actin and calculated by the 2 ${}^{\rm-\Delta\Delta ct}$ method. ** $P<$ 0.05. B. The protein expression levels of SENP2 in representative samples of GC tissues (T) and paired adjacent non-cancerous tissues (N). C. The protein expression of SENP2 in different GC cell lines.

2.2 Quantitative PCR analysis

Total RNA was extracted using RNAeasy™ Animal Total RNA Isolation Kit with Spin Column (Beyotime, China) according to the manufacturer’s instructions. Reverse transcription PCR (RT-PCR) and quantitative PCR was carried out by using a SYBR Green Premix Ex Taq (Takara) on Light Cycler480 (Roche, Switzerland). Primers were selected from PrimerBank (https://pga.mgh.harvard.edu/primerbank/) and listed as follows: For SENP2: Forward Primer GGCTGGTTAGGATTCTCGGC; Reverse Primer GGCAGCATT GTAGAGACTGTTTT. For GAPDH: Forward Primer TGTGGGCATCAATGGATTTGG; Reverse Primer A CACCATGTATTCCGGGTCAAT. Relative quantitation analysis of gene expression data was conducted according to the 2 ${}^{\rm-\Delta\Delta Ct}$ method.

2.3 Immunoprecipitation (IP)

Cells were lysed in RIPA buffer for 30 min with gentle rocking at 4 ${}^{\circ}$ C. Lysates were subjected to IP with 100 $\mu$ L anti-FLAG M2 affinity resin (Sigma, USA) overnight at 4 ${}^{\circ}$ C. Resin-containing immune complexes was washed with ice cold RIPA buffer washes. Proteins were eluted with 2 $\times$ SDS loading buffer and boiled for 5 min.

2.4 Western blots

Cells were lysed by 2 $\times$ SDS loading buffer and the total protein was separated by 10% SDS-PAGE and transferred onto NC membranes which were blocked with 5% nonfat milk in PBST for 45 min before incubation with primary antibody overnight at 4 ${}^{\circ}$ C. Membranes were washed with PBST 5 min for three times and incubated with secondary antibody for 1 hr. The signals of the proteins were visualized on an ImageQuant LAS 4000 system (GE Healthcare). The following primary antibodies were used: SENP2 antibody (Abcam, Cambridge, UK), SUMO1 antibody (Cell Signaling Technology, Beverly, MA, USA), NDRG2 antibody (Abcam, Cambridge, UK), and $\beta$ -actin (Cell Signaling Technology, Beverly, MA, USA).

2.5 Colony formation

For colony formation assay, AGS cells were seeded in six-well plates contained a base layer consisting of 0.5 ml culture media (0.6% agar). After appropriate time of growth, the cells were washed with PBS, fixed with 4% paraformaldehyde, and stained with 0.1% crystal violet. Photographs were captured and cell clones were counted.

2.6 Statistical analysis

Values are shown as mean $\pm$ standard error of the mean (SEM). Statistical differences were determined by Student’s t test. A $p$ value $<$ 0.05 was considered significant.

Figure 2.

SENP2 inhibits gastric cancer cells proliferation. A. The mRNA levels of SENP2 were determined by Real-time PCR in AGS cells stable expressing the indicated shRNAs. B. AGS cells stable expressing the indicated shRNAs were harvested and subjected to western blot with anti-SENP2 antibody. C. The cell growth curve of AGS cells stable expressing the indicated shRNAs. D. The colony formation ability of AGS cells stable expressing the indicated shRNAs.

Figure 3.

SENP2 interacts with NDRG2 in gastric cancer cells. A. 293T cells were transfected with indicated plasmids for 48 hours. Cell lysate were subjected to Flag M2 beads IP followed by western blot with HA antibody. B. AGS cells lysate was subjected to IgG or NDRG2 IP followed by western blot with anti-SENP2 antibody. C. BL21 bacteria expressing His or His-NDRG2 were incubated with Flag-SENP2 cell lysate and then subjected to western blot with anti-Flag antibody.

Figure 4.

Overexpression of SENP2 increased NDRG2 abundance. A–F: AGS cells were transfected with Flag-SENP1(A), Flag-SENP2(B), Flag-SENP3(C), Flag-SENP5(D), Flag-SENP6(E), Flag-SENP7(F), respectively and subjected to western blot with anti-NDRG2 antibody.

Figure 5.

Silencing of SENP2 destabilized NDRG2. A. AGS cells with control or SENP2 knockdown were subjected to western blot with indicated antibodies. B. AGS cells with SENP2 knockdown were treated with proteasome inhibitor MG132 for 4 hours and subjected to western blot with indicated antibodies. C. AGS cells with control or SENP2 knockdown were treated with 50 ug/ml CHX for the indicated time and subjected to western blot with indicated antibodies. D. AGS cells lysate with control or SENP2 knockdown were subjected IP with anti-NDRG2 antibody followed by western blot with anti-ubiquitin antibody.

Figure 6.

SENP2 deSUMOylates NDRG2. A. 293T cells were transfected with indicated plasmids for 48 hours. Cell lysate were subjected to Flag M2 beads IP followed by western blot with Myc antibody. B. AGS cells lysate with control or SENP2 knockdown were subjected IP with anti-NDRG2 antibody followed by western blot with anti-SUMO1 antibody.

3. Results

3.1 SENP2 is down-regulated in clinic gastric cancer samples

The mRNA expression of SENP2 was determined by real-time PCR in 20 GC and paired adjacent normal tissues. We found that SENP2 was down-regulated in 20 GC samples compared with paired adjacent normal samples ( $p<$ 0.001, Fig. 1A). The protein expression level of SENP2 was detected by western blot analysis in GC and adjacent normal tissues and similar results were also obtained. Representative blots of six paired samples are shown in Fig. 1B. The protein levels of SENP2 were also assessed in five GC cell lines including AGS, BGC823, MGC803, NCI-N87, and SGC7901. We found that the protein levels of SENP2 were hard to detected in most GC cells and we chose AGS cells which contained the relative highest SENP2 proteins for further study (Fig. 1C). Taken together, these data indicate that SENP2 is down-regulated in GC sample and cell lines.

3.2 SENP2 inhibits gastric cancer cells proliferation

To understand the biological role of SENP2 in GC cells, we silenced its expression in GC cell line AGS cells by two different shRNAs against human SENP2. The knockdown efficiency of SENP2 was determined by real-time PCR as well as western blot. We found that both shRNAs decreased the expression of SENP2 efficiently (Fig. 2A and B). Thus, both shRNAs were used to investigate the biological roles of SENP2 in GC cells. As shown in Fig. 2C, silencing of SENP2 significantly promoted cells growth. The colony formation ability of AGS cells were also enhanced by SENP2 knockdown (Fig. 2D). Taken together, these data indicated a critical role of SENP2 in the regulation of GC cells growth.

3.3 SENP2 interacts with NDRG2 in gastric cancer cells

By searching several large protein-protein interaction databases, we found that NDRG2 was one of the binding partners of SENP2 [22]. 293T cells were transfected with Flag-SENP2 or HA-NDRG2 or both for 48 hours. Cell lysate were subjected to Flag M2 beads immunoprecipitation followed by western blot with HA antibody. Indeed, we found that HA-NDRG2 was detected in Flag-SENP2 immunoprecipitant (Fig. 3A). Next, by using AGS cell lysate and NDRG2 antibody, we found that endogenous SENP2 was also presented in endogenous NDRG2 immunoprecipitant (Fig. 3B). To test whether NDRG2 interacted directly with SENP2, we further performed in vitro pull down assay. Protein complexes pulled down with nickel beads from BL21 bacteria expressing His or His-NDRG2 incubated with Flag-SENP2 were subjected to western blot. Flag-SENP2 was ready to be detected in His-NDRG2 but not His alone complexes (Fig. 3C). These data indicated that SENP2 interacted with NDRG2 directly.

3.4 Overexpression of SENP2 increased NDRG2 abundance

Because both SENP2 and NDRG2 exhibited tumor suppressor function, we then sought to identify the potential mechanisms. NDRG2 has been found to be an unstable protein in tumors, we then asked whether SENP2 could regulate the stability of NDRG2. To this end, we overexpressed all the six SENPs into AGS cells, respectively. We found that SENP2 was the only member that increased the protein amount of NDRG2 in GC cells when overexpressed (Fig. 4A–F). These data suggested a highly specific role of SENP2 in the regulation of NDRG2 in GC cells.

3.5 Silencing of SENP2 destabilized NDRG2

Next, we found that NDRG2 protein was hard to detect in AGS cells with SENP2 knockdown (Fig. 5A), but recurrent in the present of proteasome inhibitor MG132 (Fig. 5B), suggesting that SENP2 could prevent NDRG2 from ubiquitin-mediated destruction. Indeed, our cycloheximide (CHX) experiment showed that SENP2 knockdown facilitated NDRG2 protein degradation (Fig. 5C). In agreement, knockdown of SENP2 dramatically enhanced polyubiquitination of endogenous NDRG2 (Fig. 5D). Taken together, these data suggested that SENP2 plays a critical role in the regulation of NDRG2 stabilization.

3.6 SENP2 deSUMOylates NDRG2

Recently, NDRG2 was shown to be SUMOylated by RNF4 and we have observed that SENP2 could bind to and stabilize NDRG2. Therefore, we reasoned that SENP2 could be a deconjugation protease for SUMOylated NDRG2. To test this possibility, we performed a de-SUMOylation assay in the transfected 293T cells. As shown in Fig. 6A, the expression of SENP2 wild-type (WT) but not the SENP2 catalytic mutant (CA) significantly disappeared the SUMOylated form of NDRG2. Moreover, we could also observed the accumulation of the SUMOylated NDRG2 in SENP2 knock down (KD) AGS cells in the present of MG132 (Fig. 6B).

4. Discussion

The NDRG family has three members, including NDRG1, NDRG2, NDRG3 and NDRG4 [23]. This protein family is characterized by an esterase/lipase/ thioesterase active site serine and an $\alpha/\beta$ hydrolase fold [24]. NDRG proteins play important roles in cell proliferation and differentiation, and NDRG2 mainly exerts biological activity by modulating protein expression and phosphorylation [25]. NDRG2 has been reported to be a tumor suppressor and cell stress-related gene that is involved in cellular metabolic processes and in stress responses. The associations between NDRG2 and cancer have been fully reported. In various caner types, NDRG2 levels are positively correlated with tumor differentiation but negatively correlated with lymph node metastasis and TNM stage. Epigenetic silencing of the NDRG2 promotor such as DNA methylation has been found in most primary tumors [16]. Recently, the posttranscriptional modifications of NDRG2 by SUMO or ubiquitin have been reported and SUMOylation-dependent NDRG2 degradation has also been documented [20]. However, how to reverse this SUMOylation process and which SENP member involved this process deserved deep investigation.

In the present study, we found that SENP2 was associated with tumor suppressor NDRG2 and mediated NDRG2 deSUMOylation in GC cells. As RNF4-mediated NDRG2 is required for its sequential ubiquitination and degradation, our data suggest SENP2 could stabilize NDRG2 by removing its SUMO chain and preventing the attachment of ubiquitin chain. In consistent with it, the half-life of NDRG2 was largely extended and the protein content of NDRG2 was greatly increased in the present of SENP2. In contrast, NDRG2 protein could be barely detected in SENP2 KD cells and its SUMOylation form was easily showed up in the present of a proteosomal inhibitor. We also observed that SENP2 was down-regulated in GC samples and cell lines and SENP2 obviously exhibited a tumor suppressor role in GC cells.

Thus, our study reveals that SENP2 acts as a tumor suppressor and SENP2-mediated NDRG2 deSUMOylation is critical for NDRG2 stabilization as well as GC cells proliferation.

Footnotes

Conflict of interest

None.

References

Nie

Liang

Cai

Shang

et al., A global burden of gastric cancer: The major impact of China, Expert Rev Gastroenterol Hepatol 11(7) (2017), 651–661. doi: 10.1080/17474124.2017.1312342.

Rosa

Costamagna

Doglietto

G.B.

and Alfieri

, Classification of nodal stations in gastric cancer, Transl Gastroenterol Hepatol 2 (2017), 2. doi: 10.21037/tgh.2016.12.03.

Shimizu

Kanda

and Kodera

, Review of recent molecular landscape knowledge of gastric cancer, Histol Histopathol (2017), 11898. doi: 10.14670/HH-11-898.

Almhanna

, Targeted therapy for gastric adenocarcinoma, Adv Pharmacol 65 (2012), 437–470. doi: 10.1016/B978-0-12-397927-8.00014-2.

Kanda

and Kodera

, Molecular mechanisms of peritoneal dissemination in gastric cancer, World J Gastroenterol 22(30) (2016), 6829–6840. doi: 10.3748/wjg.v22.i30.6829.

Enserink

J.M.

, Regulation of cellular processes by SUMO: Understudied topics, Adv Exp Med Biol 963 (2017), 89–97. doi: 10.1007/978-3-319-50044-7_5.

Pichler

Fatouros

Lee

and Eisenhardt

, SUMO conjugation – a mechanistic view, Biomol Concepts 8(1) (2017), 13–36. doi: 10.1515/bmc-2016-0030.

Huang

C.J.

Khan

F.A.

and Huo

L.J.

, DeSUMOylation: An important therapeutic target and protein regulatory event, DNA Cell Biol 34(11) (2015), 652–660. doi: 10.1089/dna.2015.2933.

Kim

J.H.

and Baek

S.H.

, Emerging roles of desumoylating enzymes, Biochim Biophys Acta 1792(3) (2009), 155–162. doi: 10.1016/j.bbadis.2008.12.008.

10.

Mukhopadhyay

and Dasso

, Modification in reverse: The SUMO proteases, Trends Biochem Sci 32(6) (2007), 286–295. doi: 10.1016/j.tibs.2007.05.002.

11.

Chiu

S.Y.

Asai

Costantini

and Hsu

, SUMO-specific protease 2 is essential for modulating p53-Mdm2 in development of trophoblast stem cell niches and lineages, PLoS Biol 6(12) (2008), e310. doi: 10.1371/journal.pbio.0060310.

12.

Cubenas-Potts

Goeres

J.D.

and Matunis

M.J.

, SENP1 and SENP2 affect spatial and temporal control of sumoylation in mitosis, Mol Biol Cell 24(22) (2013), 3483–3495. doi: 10.1091/mbc.E13-05-0230.

13.

Shen

H.J.

Zhu

H.Y.

Yang

and Ji

, SENP2 regulates hepatocellular carcinoma cell growth by modulating the stability of beta-catenin, Asian Pac J Cancer Prev 13(8) (2012), 3583–3587.

14.

Tan

M.Y.

X.Y.

Liu

Wang

Bao

E.D.

Qiu

J.X.

et al., SUMO-specific protease 2 suppresses cell migration and invasion through inhibiting the expression of MMP13 in bladder cancer cells, Cell Physiol Biochem 32(3) (2013), 542–548. doi: 10.1159/000354458.

15.

Zhang

X.D.

Goeres

Zhang

Yen

T.J.

Porter

A.C.

and Matunis

M.J.

, SUMO-2/3 modification and binding regulate the association of CENP-E with kinetochores and progression through mitosis, Mol Cell 29(6) (2008), 729–741. doi: 10.1016/j.molcel.2008.01.013.

16.

Fan

Jiang

Yan

et al., Emerging role of N-myc downstream-regulated gene 2 (NDRG2) in cancer, Oncotarget 7(1) (2016), 209–223. doi: 10.18632/oncotarget.6228.

17.

Yao

Zhang

and Liu

, NDRG2: A myc-repressed gene involved in cancer and cell stress, Acta Biochim Biophys Sin (Shanghai) 40(7) (2008), 625–635.

18.

Cai

Gao

Shen

Zhang

et al., NDRG2 is a novel p53-associated regulator of apoptosis in c6-originated astrocytes exposed to oxygen-glucose deprivation, PLoS One 8(2) (2013), e57130. doi: 10.1371/journal.pone.0057130.

19.

Liu

Wang

Yang

Liu

Zhang

et al., N-Myc downstream-regulated gene 2 is involved in p53-mediated apoptosis, Nucleic Acids Res 36(16) (2008), 5335–5349. doi: 10.1093/nar/gkn504.

20.

Tantai

Pan

and Hu

, RNF4-mediated SUMOylation is essential for NDRG2 suppression of lung adenocarcinoma, Oncotarget 7(18) (2016), 26837–26843. doi: 10.18632/oncotarget.8663.

21.

Mokhonova

E.I.

Avliyakulov

N.K.

Kramerova

Kudryashova

Haykinson

M.J.

and Spencer

M.J.

, The E3 ubiquitin ligase TRIM32 regulates myoblast proliferation by controlling turnover of NDRG2, Hum Mol Genet 24(10) (2015), 2873–2883. doi: 10.1093/hmg/ddv049.

22.

Huttlin

E.L.

Bruckner

R.J.

Paulo

J.A.

Cannon

J.R.

Ting

Baltier

et al., Architecture of the human interactome defines protein communities and disease networks, Nature 545(7655) (2017), 505–509. doi: 10.1038/nature22366.

23.

Lee

G.Y.

Chun

Y.S.

Shin

H.W.

and Park

J.W.

, Potential role of the n-MYC downstream-regulated gene family in reprogramming cancer metabolism under hypoxia, Oncotarget 7(35) (2016), 57442–57451. doi: 10.18632/oncotarget.10684.

24.

Melotte

Ongenaert

van Criekinge

de Bruine

A.P.

Baldwin

H.S.

et al., The N-myc downstream regulated gene (NDRG) family: diverse functions multiple applications, FASEB J 24(11) (2010), 4153–4166. doi: 10.1096/fj.09-151464.

25.

Yang

Fan

Deng

et al., Clinical and pathological significance of N-myc downstream-regulated gene 2 (NDRG2) in diverse human cancers, Apoptosis 21(6) (2016), 675–682. doi: 10.1007/s10495-016-1244-3.