This article is mentioned in:

Introduction

Gastric cancer (GC) is a major cause of morbidity and mortality and has become a major public health concern worldwide (1). The early symptoms of GC are atypical; therefore, when patients develop apparent symptoms, the disease has already progressed to an advanced stage. Therefore, the opportunity for radical surgery for these patients is missed, and chemotherapy becomes the primary means of treatment.

RNA interference (RNAi) is the phenomenon of post-transcriptional gene silencing caused by double-stranded (ds)RNA (2). RNAi is considered a revolutionary novel post-gene transcriptional regulation technology due to its unique efficiency and specificity. This technology has become a powerful tool for the study of gene function and has provided novel techniques and application prospects for specific gene therapy. The RNAi technology is becoming known for gene function and regulation analysis in the post-genomic era.

Zinc finger proteins (ZNFs) are a class of proteins widely distributed in the human genome. ZNFs have been shown to regulate gene transcription, thus playing an important role in the occurrence and development of GC (3). It was previously demonstrated that the ZNF139 gene is closely associated with GC (4).

Proteomics relies on the bilateral fluorescence two-dimentional difference gel electrophoresis (2D-DIGE) to separate different proteins, which are then identified and analysed by mass spectrometry and bioinformatics (5). Proteomics technology is currently being used in various fields of biology due to its high sensitivity, accuracy and throughput. A previous study demonstrated that the ZNF139 gene is closely associated with GC (4). The aim of the present study was to further investigate the mechanisms underlying the role of ZNF139 in the occurrence, development and chemosensitivity of GC.

Based on the principle of RNAi, a ZNF139-targeted small interfering (si)RNA plasmid was developed in this study. The different proteins prior to and following siRNA-ZNF139 transfection were analyzed to provide a theoretical basis for the mechanisms underlying the role of ZNF139 in the occurrence, development and chemosensitivity of GC.

Materials and methods

Construction of siRNA-ZNF139

The 19-bp siRNA and ineffective interference control fragment were designed by the siRNA online software (http://i.cs.hku.hk/~sirna/software/sirna.php). The siRNA-ZNF139 template sequences were 5′-ACCTCGGAAGATTCAGCAT-3′ (sense strand) and 5′-ATGCTGAATCTTCCGAGGT-3′ (antisense strand). The ineffective interference control sequences were 5′-GACGAGTTGACTGCGATTG-3′ (sense strand) and 5′-CAATCGCAGTCAACTCGTC-3′ (antisense strand). The pSilencerTM 4.1-CMV neo-ZNF139 siRNA plasmid (Shanghai GenePharma Co., Ltd., Shanghai, China; Fig. 1A) and Escherichia coli were used to construct the engineered bacteria, then the endotoxin plasmid gross extraction kit was applied to extract the plasmid. Each recombinant plasmid was amplified and purified for the subsequent experiments.

Figure 1. Construction and transplantation. (A) Plasmid carrying the small interfering (si)RNA. (B) Subcutaneous tumor transplant. (C) Orthotopic tumor transplant.

RNAi-mediated inhibition of ZNF139 in vitro

Experiments were divided into the blank control, siRNA-ZNF139 and negative control groups. Based on the transfection reagent instructions of the X-tremeGENE HP DNA transfection reagent (Roche, Basel, Switzerland), the recommended amount of plasmid (0.8 µg/µl) and the appropriate transfection reagents were added for cell transfection. Transfected cells (~200 µl) were seeded in each well of the 96-well plate and then incubated in 5% CO2 at 37°C for 0, 24, 48 and 72 h post-transfection. Approximately 20 µl of MTT solution (5 mg/ml) was added into each well of the plate and incubated for another 4 h. The supernatant was carefully discarded and an absorbent paper absorbed the residual liquid. A total of 150 µl dimethyl sulfoxide (DMSO) was added to each well and agitated for 10 min. The optical density (OD) of each well was measured at 570 nm to illustrate the cell growth curve.

RNAi-mediated inhibition of ZNF139 in vivo

The in vitro cultured SGC7901 human GC cells were collected and subcutaneously injected into nude mice. In total, 45 BALB/C male nude mice (age, 4 weeks; weight, 17–23 g) were used, including 18 mice for inter-mouse passage subcutaneous tumor transplantation and 27 mice for orthotopic tumor transplantation. The mice were obtained from the Fourth Hospital of Hebei Medical University Animal Center specific-pathogen-free animal laboratory (Shijiazhuan, China). Inter-mouse passage of the subcutaneous tumor was performed (Fig. 1B) and orthotopic transplantation was conducted on the sixth generation of subcutaneous tumor using the OB biological glue method (Fig. 1C) (6). The incision and growth of the abdominal orthotopic tumor graft were observed daily. Interference was conducted when the orthotopic tumor graft grew to ≥10 mm in diameter. The orthotopic grafted nude mice were randomly divided into the aforementioned three groups when the grafted tumor was developed.

The orthotopic xenografted nude mice were randomly divided into three groups (n=9 per group). The grouping and intervention methods were as follows: In the blank control group, 50 µl of saline was injected at multiple points at the tumor spot every other day. In the siRNA-ZNF139 group, 50 µl of the intervention complex was administered by multipoint injection at the tumor spot every other day. The intervention complex included 20 µg of the siRNA-ZNF139 plasmid and 20 µl of the transfection reagent. The medium was added to a total volume of 50 µl. In the negative control group, 50 µl of the intervention complex was administered by multipoint injection at the tumor spot every other day. The complex included 20 µg of the negative-control plasmid and 20 µl of the transfection reagent. The medium was added to a total volume of 50 µl.

During the intervention period following formation of the orthotopic tumor graft, the long and short diameters of the tumor were measured once every other day. The orthotopic tumor volume was also calculated with the following equation: V=ab2/2 (where a is the long diameter and b is the short diameter of the tumor). Two weeks after the intervention treatment, the nude mice were sacrificed using the cervical dislocation method. The in situ tumor was removed and the specimens were stored at −80°C for future use. The experiment was performed in strict accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The animal use protocol was reviewed and approved by the Institutional Animal Care and Use Committee of Hebei Medical University (permit no. 20060828001).

MTT assay

The SGC7901 human GC cells from different experimental groups and nude mice with in situ grafted tumors were prepared. Approximately 180 µl of the cell suspension were seeded in each well of the 96-well plate. The three chemotherapeutic agents used were 5-fluorouracil (5-FU), cisplatin (CDDP) and mitomycin C (MMC). Each agent (~20 µl) was added into each well of the 96-well plate to a total volume of 200 µl per well. The procedure was repeated five times. The final concentrations of the chemotherapeutic agents were 25 µg/ml 5-FU, 4 µg/ml CDDP and 4 µg/ml MMC. The plate was incubated in 5% CO2 at 37°C for 48 h. Following incubation, 20 µl of MTT (5 mg/ml) was added to each well and incubated for another 4 h. The supernatant was carefully discarded and the residual liquid was absorbed with an absorbent paper. Following addition of 150 µl DMSO to each well and vortex mixing for 10 min, the OD value of each well was measured at 570 nm with a microplate reader (Anthos 2010; Anthos Labtec Instruments GmBH, Wals-Siezenheim, Austria). The mean cell growth inhibitory rate (IR) was calculated as follows: IR = (1 - mean ODdrug treatment well/mean ODcontrol well) × 100%.

Preparation of protein samples

The Bradford method (7) was used to detect the protein sample concentrations. Protein samples from the two experimental groups (~50 µg) were collected and labelled with CyDye5 and CyDye3. Approximately 50 µg of the protein samples in the control and experimental groups were replenished. Active hydration was conducted at 50 V for 10 h and then at 500 V for 1 h, 1,000 V for 1 h, 8,000 V for 1 h and 8,000 V for 3 h. The final voltage increased from 10,000 to 65,000 V. The strips were transferred to 10 ml of balance solution A [75 mmol/l Tris-HCl (pH 8.8), 6 mol/l urea, 30% glycerol, 2% sodium dodecyl sulfate (SDS) and 1% disulfide dithiothreitol (DTT); Sigma-Aldrich, St Louis, MO, USA] for 15 min and then to balance solution B (2.5% indole-3-acetic acid was used to replace the 1% DTT in solution A) for another 15 min. The 12.5% SDS-polyacrylamide gel (PAGE) electrophoresis was performed at 50 V for 2 h and then at 120 V until the frontier of bromophenol blue was ~1 cm from the lower end of the gel. Scanning was performed by Typhoon 9400 (GE Healthcare, Little Chalfont, UK), and image analysis was conducted by DeCyder difference analysis software (GE Healthcare).

Liquid chromatography-mass spectrometry (LC-MS) and database query

Following Deep Purple staining, an Ettan Spot Picker (GE Healthcare) was used to dig the spot. LTQXL™ (Thermo Fisher Scientific, Waltham, MA, USA) was used for the mass identification of the samples following in-gel digestion. BioWorks 3.3.1 software (Thermo Fisher Scientific) was used for database searching. The FASTA-form protein database was obtained from the National Center for Biotechnology Information human protein sequence database. Cysteine reductive alkylation and methionine oxidation were set as the fixed and variable modifiers, respectively.

Western blot analysis

Approximately 50 µg of the prepared protein sample from each experimental group was separated by SDS-PAGE electrophoresis. The gel was sliced and transferred onto the polyvinylidene fluoride membrane, which was initially treated with the closure solution at room temperature and agitated for 1 h. The primary antibodies [rabbit anti-human polyclonal antibodies against ZNF139 (cat. no. ab126124), pyridoxal kinase (PDXK; cat. no. ab38208), annexin A2 (ANXA2; cat. no. ab41803) and fascin (cat. no. ab183891), and mouse anti-human monoclonal antibody against β-actin (cat. no. ab6276); Abcam, Cambridge, UK] were added and then incubated at 4°C overnight. The membrane was washed, followed by the addition of horseradish peroxidase enzyme-labelled goat anti-rabbit IgG secondary antibody (cat. no. ab6721; Abcam) and incubated at room temperature for 1 h. The membrane was washed again and stained by chemiluminescence. The Bio-Rad Image Analysis system (Bio-Rad Laboratories, Inc., Hercules, CA, USA) was used with β-actin as the internal reference. The experiment was performed thrice.

Statistical analysis

Measured data are expressed as the mean ± standard deviation and analysed by SPSS 13.0 statistical software (SPSS, Inc., Chicago, IL, USA). The intergroup comparison was conducted by analysis of variance.

Results

RNAi-mediated inhibition of ZNF139 in vitro

The growth of the SGC7901 human GC cells of the siRNA-ZNF139 group slowed down significantly compared with that of the control and negative control groups (P<0.05) (Fig. 2A).

Figure 2. (A) Cell growth of SGC7901 human GC cells at different time points and (B) tumor size changes of human in situ grafted GC tumors following RNA interference. GC, gastric cancer; OD, optical density; siRNA-ZNF139, small interfering RNA of zinc finger protein 139.

RNAi-mediated inhibition of ZNF139 in vivo

The size of the human in situ grafted GC tumor of the siRNA-ZNF139 group was significantly decreased compared with that of the control and negative control groups (P<0.05) (Fig. 2B).

Chemosensitivity changes

The average IR of the cells of the siRNA-ZNF139 group was significantly higher (P<0.05) compared with that of the control and negative control groups (Table I).

Table I. Alterations in cell chemosensitivity following RNA interference.

Table I.

Alterations in cell chemosensitivity following RNA interference.

	SGC7901 human GC cells			Human in situ grafted GC cells
Experimental groups	5-FU	CDDP	MMC	5-FU	CDDP	MMC
Blank control	29.18±3.01	30.67±1.55	19.58±2.17	31.25±2.86	32.64±1.69	21.29±2.23
siRNA-ZNF139	40.08±1.97a	39.76±1.24a	33.91±1.39a	42.28±2.13a	41.94±1.53a	35.70±1.66a
Negative control	28.98±3.21	30.50±1.84	19.52±1.77	31.47±2.75	32.48±1.91	21.48±1.94

{ label (or @symbol) needed for fn[@id='tfn1-ol-0-0-3421'] } Mean cell inhibition rate (IR) = (1 - mean OD_{drug treatment well}/mean OD_{control well}) × 100%. Data are presented as the mean percentage ± standard deviation.

a P<0.05 compared with the blank control group. GC, gastric cancer; siRNA-ZNF139, small interfering RNA of zinc finger protein 139; OD, optical density; 5-FU, 5-fluorouracil; CDDP, cisplatin; MMC, mitomycin C.

2D-DIGE

According to the 2D-DIGE spectrum of the SGC7901 human GC cells and in situ grafted GC tumors, the protein points were 1,958±67 with 90.5% matching rate and 5,227±59 with 88.7% matching rate, respectively. DeCyder difference analysis software was used to select the eight and six distinct protein points in the SGC7901 human GC cells and in situ grafted GC tumors, respectively. The experiment was performed thrice.

Identification of LC-MS

The Ettan spot picker was used to assess the different protein spots. Following in-gel digestion, LC-MS was used for the analysis. Among the different proteins of the SGC7901 human GC cells, six were identified by LC-MS, namely PDXK, desmin, nuclear phosphoprotein (NPM), heat-shock protein 70 (HSP70), ANXA2 and fascin. In the siRNA-ZNF139-transfected SGC7901 cells, PDXK, desmin and NPM were upregulated, whereas HSP70, ANXA2 and fascin were downregulated. Five proteins were identified by LC-MS among the different proteins in the in situ grafted human GC tumors, namely ANXA1, PDXK, fascin, ANXA2 and heterogeneous nuclear ribonucleoprotein (hnRNP). Following transfection, ANXA1 and PDXK were upregulated, whereas fascin, ANXA2 and hnRNP were downregulated (Table II and Fig. 3).

Figure 3. Two-dimensional difference gel electrophoresis spectrum of human gastric cancer (A) SGC7901 cell line and (B) in situ grafted tumor.

Table II. Identification results of difference proteins.

Table II.

Identification results of difference proteins.

Number	Proteins	Version	MW	Score	Coverage	pI	Expression
1	PDXK	gi|4505701	35079.90	30.14	11.20	5.75	U
2	Desmin	gi|55749932	53503.23	30.17	6.60	5.21	U
3	NPM	gi|10835063	32554.90	10.18	3.70	4.64	U
4	HSP70	gi|5123454	69995.16	30.16	4.70	5.47	D
5	ANXA2	gi|50845386	38579.82	40.23	2.90	7.57	D
6	Fascin	gi|4507115	54496.09	10.21	2.80	6.84	D
7	ANXA1	gi|4502101	38690.00	60.20	17.30	6.57	U
8	hnRNP	gi|55956919	35945.22	40.21	12.70	6.49	D

[i] MW, molecular weight; pI, isoelectric point; U, upregulated following ZNF139 inhibition; D, downregulated following ZNF139 inhibition; ZNF139, zinc finger protein 139; PDXK, pyridoxal kinase; ANXA2, annexin A2; NPM, nuclear phosphoprotein; HSP70, heat-shock protein 70; hnRNP, heterogenous nuclear ribonucleoprotein.

Western blot analysis

The expression of ZNF139 was significantly decreased in the siRNA-ZNF139 group compared with that in the control and negative control groups. PDXK expression was significantly increased, whereas the expression of ANXA2 and fascin was significantly decreased, in accordance with the proteomics results (Fig. 4).

Figure 4. Western blot analysis of the expression of ZNF139, PDXK, ANXA2 and fascin in (A) SGC7901 cells and (B) transplanted tumors. 1, blank control group; 2, small interfering RNA-ZNF139 group; 3, negative control group. ZNF139, zinc finger protein 139; PDXK, pyridoxal kinase; ANXA2, annexin A2.

Discussion

A previous study (8) demonstrated that siRNA may guide the RNA-induced silencing complex (RISC) to cut the homologous single-stranded mRNA and function as the primer to bind the target RNA. A number of secondary dsRNAs are synthesized under RNA polymerase, and the newly synthesized dsRNA may be sliced by the Dicer enzyme to generate a large number of secondary siRNAs. Therefore, the RNAi effect increases further, resulting in the completion of the degradation of the target mRNA. The results of the present study demonstrated that siRNA-ZNF139 was able to inhibit the expression of ZNF139 and the growth of SGC7901 human GC cells and in situ grafted GC tumors in nude mice. These findings may provide the theoretical basis for the involvement of the ZNF139 gene in the occurrence, development and chemosensitivity of human GC.

The multidrug resistance (MDR) phenomenon of GC is a major problem that currently complicates clinical treatment, and one of the main reasons for the poor therapeutic effect of chemotherapy (9). MDR occurs when the tumor cells resist a specific anticancer drug and then exhibit cross-resistance to other, previously unencountered anticancer drugs, which may be structurally unrelated and have different targets and mechanisms of action (10). The fundamental solution for MDR is dependent on in-depth studies on its cause and underlying mechanism; such studies may provide a method for its complete reversion (11). Based on the present study, the application of siRNA to inhibit the expression of ZNF139 in SGC7901 human GC cells and in situ grafted GC tumors may increase the sensitivity of tumor cells to 5-FU, CDDP and MMC. The inhibition of ZNF139 expression in GC cells may improve the efficiency of chemotherapy in GC, thus providing a basis for future research on novel GC treatments.

Wasinger et al conducted a study applying proteomics (12); this technology has been widely used in various fields of biology due to its high sensitivity, accuracy and throughput. Franco et al (13) applied the combination of 2D-DIGE and LC-MS, and observed that the occurrence of GC was associated with the type of Helicobacter pylori strain. Xin et al (14) applied proteomics technology to compare the differences in SGC7901 human GC cells when cultured with and without methionine and observed that the lack of methionine may inhibit the proteins that are associated with the regulation of the apoptosis of GC cells. Yang et al (15) compared the proteome differences between the SGC7901 human GC cell line and a vincristine-resistant SGC7901 cell line, and identified a variety of MDR-related proteins. In the present study, siRNA-ZNF139 was applied to inhibit the expression levels of ZNF139 in SGC7901 human GC cells and in situ grafted GC tumors. Proteomics was used to identify the different proteins expressed prior to and following inhibition. Among the different identified proteins, PDXK, ANXA2 and fascin exhibited similar changes. Following transfection, PDXK was upregulated, whereas ANXA2 and fascin were downregulated. Therefore, ZNF139 may regulate the expression of PDXK, ANXA2 and fascin, thus participating in the occurrence, development and chemosensitivity of GC.

PDXK, a member of the ribose kinase superfamily, is widely present in biological organisms. PDXK is a key metabolic enzyme of vitamin B6 and catalyses its phosphorylation. The catalysed phosphorylation enables transfer into active pyridoxal-5′-phosphate (PLP), which has an important function in the synthesis and catabolism of amino acids. Matsubara et al (16) demonstrated that PLP is able to inhibit the proliferation of endothelial cells. Furthermore, Komatsu et al (17) demonstrated the inhibition of tumor cell growth by PLP.

ANXs are mainly expressed inside the nucleus and cytoplasm and under the cell membrane; they are a class of structurally-related calcium-dependent phospholipid-binding proteins divided into five subfamilies, namely A, B, C, D and E. The A family of ANXs includes 13 members (A1-13) found inside vertebrate cells (18). ANXA2 was first observed inside Rous sarcoma virus-transformed chicken embryo fibroblasts, which are located in the human chromosome 15q21-q22. ANXA2 is composed of 12 introns and 13 exons, with 339 amino acids and a molecular weight of 36 kDa (19). ANXA2 is mainly expressed inside endothelial cells, macrophages, mononuclear cells, nerve cells and certain tumor cells. This protein may be highly expressed in proliferating and transforming cells but its expression is low in terminally differentiated cells. It was recently demonstrated that ANXA2 is associated with various cytokines, including vascular endothelial growth factor (VEGF); this protein promotes VEGF expression on the tumor cell surface and the formation of new blood vessels, thus promoting the growth, invasion and metastasis of tumor cells (20). Zhang et al (21) demonstrated that ANXA2 is associated with MDR. Meng et al (22) applied proteomics technology to compare the parental human bladder cancer cell line (pumc-91) with the respective doxorubicin-resistant cell line (pumc-91/ADM) and observed that ANXA2 was highly expressed in pumc-91/ADM, which may indicate it is a resistance factor.

Fascin is an evolutionarily conserved cytoskeletal protein with a molecular weight of 55 kDa. The main function of fascin is to bind F-actin and assemble it in the parallel actin bundle. Fascin has an important function in cell migration, adhesion and exchange of intracellular information. During the occurrence and development of tumors, fascin protein expression increases, resulting in increased cell surface projections and decreased cell differentiation. Therefore, cell proliferation may be promoted, the actin cytoskeleton reconstructed and cell motility enhanced. Tumor cells exhibit enhanced invasiveness and migration ability, enabling them to penetrate the basement membrane and spread to the surrounding tissues (23).

In the present study, western blot analysis detected a significant increase in PDXK protein levels following siRNA-ZNF139 transfection. However, the expression levels of ANXA2 and fascin were significantly decreased. Therefore, ZNF139 may be involved in the occurrence, development and chemotherapeutic sensitivity of GC by promoting the expression of ANXA2 and inhibiting the expression of fascin and PDXK.

It was previously reported that the ZNF139 gene is closely associated with GC (4). With the aim to investigate novel GC treatments and governed by the principles of RNAi, we herein constructed an siRNA plasmid to the ZNF139 gene. Based on the analysis of the protein differences prior to and following siRNA-ZNF139 transfection, it was concluded that ZNF139 may promote the expression of ANXA2 and fascin, while suppressing the expression of PDXK. Therefore, ZNF139 may be significantly involved in the occurrence, development and chemosensitivity of GC. These findings are consistent with those of Komatsu et al (17), Zhang et al (21) and Meng et al (22). In conclusion, ZNF139 may inhibit the occurrence and development of GC, while increasing its sensitivity to chemotherapeutic agents, thus providing novel targets for the treatment of GC patients.

References