Cell culture and tissue collection

The human GC cell lines SGC7901, BGC823, MKN45, and AGS, and the immortalized gastric epithelial cell line GES were purchased from the Cell Resource Center of the Chinese Academy of Sciences, Shanghai, China. Cells were maintained in Dulbecco’s Modified Eagle’s Medium (Thermo Scientific HyClone, Beijing, China) supplemented with 10% fetal bovine serum (HyClone), 100 U/mL penicillin, and 100 U/mL streptomycin (HyClone) in a 37 °C humidified incubator with a mixture of 95% air and 5% CO₂.

A total of 20 fresh primary GC samples and matched adjacent non-cancerous tissues were obtained from patients undergoing surgery at Xijing Hospital, Xi’an, China. All samples were confirmed by the Department of Pathology at Xijing Hospital and stored in a liquid nitrogen canister. All patients provided informed consent for excess specimens to be used for research purposes and all protocols employed in the present study were approved by the Medical Ethics Committee of Xijing Hospital.

Mice

Female BALB/c nude mice were provided by the Experimental Animal Center of the Fourth Military Medical University and were housed in pathogen-free conditions. All animal studies complied with the Fourth Military Medical University animal use guidelines, and the protocol was approved by the Fourth Military Medical University Animal Care Committee.

Reagent and inhibitor

5-Fluorouracil was purchased from Sigma (Sigma-Aldrich Corporation, Los Angeles, CA, USA), and MAPK/ERK inhibitor (PD98059) and p53 inhibitor (A15201) were purchased from Invitrogen (Thermo Fisher Scientific, Cambridge, Massachusetts, USA); all were stored according to the manufacturer’s instructions.

RNA extraction and real-time polymerase chain reaction (PCR)

Total RNA was extracted from cell lines using the RNeasy Plus Universal Tissue Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The PCR primers for DDIT4 and ACTB were synthesized by TaKaRa (Dalian, China). The sequences were as follows: DDIT4, 5′-GGACCAAGTGTGTTTGTTGTTTG-3′ (Forward) and 5′-CACCCACCCCTT CCTACTCTT-3′ (Reverse); ACTB, 5′-TCATGAAGTGTGACGTTGACATCCGT- 3′ (Forward) and 5′-CCTAGAAGCATTTGCGGTGCACGATG-3′ (Reverse). cDNA was synthesized using the PrimeScript RT Reagent Kit (TaKaRa, Dalian, China). Real-time PCR was performed using SYBR Premix Ex Taq II (TaKaRa). Fluorescence was measured using a LightCycler 480 system (Roche, Basel, Switzerland). ACTB was used as an internal control for mRNA analysis. Each sample was run in triplicate.

Protein extraction and western blotting

Total proteins were prepared from fresh frozen tissue or cultured cells in radio immunoprecipitation assay (RIPA) lysis and extraction buffer (Beyotime Biotechnology, Shanghai, China) with protease and phosphatase inhibitors. Denatured proteins (20–50 mg) were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes. The following primary antibodies were used according to the manufacturer’s instructions: anti-DDIT4 (Dilution 1:500, Abcam, Cambridge, MA, USA) and anti-β-actin (Dilution 1:2000), anti-Ki67 (Dilution 1:1000), anti-p53 (Dilution 1:1000), anti-p-p53 (p-Ser6) (Dilution 1:1000), anti-p-p53 (p-Ser315) (Dilution 1:1000), anti-p21^Cip1 (Dilution 1:500), anti-p-p21^Cip1 (p-Thr145) (Dilution 1:500), anti-MEK1 (Dilution 1:1000), anti-p-MEK1 (p-Ser221) (Dilution 1:1000), anti-p42/44MAPK (Dilution 1:1000), and anti-p-p42/44MAPK (p-Thr202 and p-Tyr204) (Dilution 1:1000) (Cell Signaling Technology, Beverly, MA, USA). Densitometry of specific blotted bands was analyzed by ImageJ 1.48 software (Image-Processing and Analysis in Java; National Institutes of Health, Bethesda, MD, USA; http://imagej.nih.gov/), and the intensity values were normalized against the β-actin loading control.

Tissue microarray immunohistochemistry

GC tissue microarrays were purchased from Superchip (Shanghai, China). Each array contained 90 cases of paired adjacent gastric tissues and primary GC tissues. Immunohistochemical (IHC) staining was performed using anti-DDIT4 and anti-Ki67 antibodies as per the manufacturer’s instructions. IHC results were scored independently by two pathologists in a blinded manner. The scoring was based on the intensity and extent of staining. Staining intensity was graded as follows: 0, negative staining; 1, weak staining; 2, moderate staining; and 3, strong staining. The proportion of stained cells per specimen was determined semi-quantitatively as follows: 0, < 1%; 1, 1–25%; 2, 26–50%; 3, 51–75%; and 4, > 75%. The histological score (H-score) for each specimen was computed using the following formula: H-score = proportion score×intensity score. The samples were further characterized by H-score as negative (−, score: 0), weak (+, score: 1–4), moderate (++, score: 5–8), and strong (+++, score: 9–12). Samples with an H-score of > 4 were considered to exhibit high expression, and samples with an H-score of ≤ 4 were considered to exhibit low expression.

Immunofluorescence staining

Cells were plated onto glass coverslips, fixed with 4% paraformaldehyde for 20 min and permeabilized with 0.1% Triton X-100 in PBS for 15 min. Blocking solution was applied at room temperature for 1 h. Rabbit anti-human DDIT4 primary antibody (Abcam) was applied at 4 °C overnight. FITC-conjugated goat anti-rabbit and Cy3-conjugated goat anti-mouse secondary antibodies were incubated on the coverslips at room temperature for 2 h. Immunostaining signals and DAPI-stained nuclei were visualized at room temperature using a confocal microscope (FV10i; Olympus, Tokyo, Japan) equipped with a 10×/0.30 numerical aperture objective lens (Olympus) and Fluoview software (version 4.3; Olympus). No imaging medium was used. For better visualization, the images were adjusted using the level and brightness/contrast tools in Photoshop according to the guidelines for the presentation of digital data.

Lentivirus infection

DDIT4-overexpression or sh-DDIT4 lentivirus infection was conducted by GeneChem (Shanghai, China). Target cells (1 × 10⁵) were infected with 1 × 10⁷ lentivirus-transducing units in the presence of 5 mg/mL polybrene. An empty lentiviral vector was used as a negative control. After transfection and antibiotic selection for 6 weeks, cells were collected for further investigation.

High-content screening assay

Cells transfected with lentivirus stably expressing green fluorescent protein (GFP) were seeded into 96-well plates and treated with increasing doses of 5-FU (0, 10, 20 µg/mL for GC cells or 0, 2, 4 µg/mL for GES cells). Cells were imaged every 4 h for 48 h using the Thermo Scientific CellInsight CX7 High Content Analysis Platform (0, 10, 20 µg/mL for GC cells or 0, 2, 4 µg/mL for GES cells; Thermo Fisher Scientific, Cambridge, Massachusetts, USA). Proliferation curves were plotted and analyzed using HCS Studio Software (Thermo Fisher Scientific).

Colony formation assays

Transfected cells were seeded in six-well plates (1000 cells/well). After 14–18 days of incubation to establish stable clones, cells were fixed with 70% ethanol and stained with crystal violet solution. Colonies containing greater than 50 cells were counted. The experiment was conducted with three independent triplicates.

Cell cycle and apoptosis assays

For cell cycle analysis, target cells were selected with antibiotics (penicillin–streptomycin solution) for 48 h after transfection as indicated, fixed in 75% ethanol, and stained with propidium iodide supplemented with RNase A (Roche, Mannheim, Germany) for 30 min at 22 °C. The Annexin V-FITC Apoptosis Detection Kit (Cell Signaling Technology, Beverly, MA, USA) was used for apoptosis assays. Cells (1 × 10⁴) were stained according to the manufacturer’s protocol and sorted using a fluorescence-activated cell sorting sorter (BD Biosciences, La Jolla, CA, USA). Data were analyzed using ModFit software (BD Biosciences).

In vivo tumorigenicity

BGC823 cells (5 × 10⁵ cells in 0.2 mL of PBS) transfected with pCMV-DDIT4 or empty pCMV were injected subcutaneously into the dorsal flank of 5-week-old female Balb/c nude mice (five mice per group). Tumor diameter was measured every 3 days for 30 days. Tumor volume (mm³) was calculated based on the longest and shortest diameters as follows: volume = (shortest diameter)² × (longest diameter) × 0.5. Thirty days after injection, all mice were killed, and the tumor xenografts were isolated for further analysis. All experimental animals were supplied by the Experimental Animal Center of the Fourth Military Medical University. All protocols for animal studies were approved by the Fourth Military Medical University Animal Care Committee.

Cell counting kit-8 (CCK8) assay

For cell counting kit 8 assays, cells were seeded into 96-well plates at a density of 1000 cells in 100 μL of complete culture medium per well. At the indicated time points, the medium was replaced with a kit solution (TransDetect cell counting kit, Transgene, Beijing, China) and complete culture medium at a ratio of 1:9, and the samples were incubated for 2 h at 37  °C. The absorbance of each sample was analyzed at 450 nm using a microtiter plate reader (Tecan, Switzerland). The assay was repeated in triplicate.

Phospho-specific protein microarray analysis

Phospho-array detection was performed in cooperation with Wayen Biotechnology (Shanghai, China). At 48 h post-transfection, all treated cells were collected for protein extraction. Protein samples of 50 mg each were tagged with biotin reagent and hybridized on a Phosphorylation ProArray (Full Moon BioSystems, USA) using an Antibody Array Kit (FullMoon BioSystems, USA) for the detection of 248 site-specific cancer signaling phospho-antibody profiles. Finally, fluorescence intensity was scanned by GenePix 4000B (Axon Instruments, Houston, USA) using GenePix Pro 6.0. The raw data were manipulated using Grubbs’ method. The phosphorylation ratio was calculated as follows: phosphorylation ratio ¼ phospho value/unphospho value.

Statistical analyses

SPSS software (version 19.0, SPSS Inc., Chicago, IL, USA) was used for statistical analyses. Continuous data are presented as the mean ± standard deviation (SD), and Student’s unpaired t-test was utilized for comparisons between two groups. Frequencies of categorical variables were compared using the χ² test. A P value of less than 0.05 was considered significant.

DDIT4 was upregulated in GC tissues

To determine the expression pattern of DDIT4 in GC, we measured DDIT4 expression in 20 pairs of GC and adjacent normal tissue specimens using real-time PCR and western blotting. DDIT4 expression was upregulated in 13 of 20 GC tissues compared with matched adjacent normal tissues (Fig. 1a). Similar to DDIT4 mRNA levels, DDIT4 protein levels were increased in GC samples compared with their normal counterparts (Fig. 1b). Furthermore, we performed IHC using a GC tissue microarray containing 90 pairs of primary GC tissues and paired adjacent normal tissues. The IHC analysis revealed clear elevation of DDIT4 levels in GC tissues compared with the corresponding normal tissues, and DDIT4 expression was primarily located in the cytoplasm (Fig. 1c, d). Univariate survival analysis demonstrated that tumor size (P = 0.013), invasion depth (P = 0.015), lymphatic metastasis (P = 0.018), distant metastasis (P < 0.001), AJCC stage (P < 0.001), and pathological grade (P < 0.001) exhibit statistically significant associations with GC patient survival (Table 1). Taken together, DDIT4 expression was increased in GC tissues.

Variable	Univariate log rank survival analysis			Multivariate Cox survival analysis
	RR	95% CI	P value	RR	95% CI	P value
Age (years)
< 60	1.000
≥ 60	1.085	0.832–1.413	0.547
Gender
Male	1.000
Female	1.091	0.797–1.494	0.588
Tumor size (cm)
< 5	1.000
≥ 5	1.182	1.036–1.349	0.013	1.387	0.827–2.325	0.215
Invasion depth
T1/2	1.000
T3/4	1.610	1.098–2.360	0.015	1.496	0.696–3.216	0.302
Lymphatic metastasis
N0	1.000
N1–3	1.369	1.056–1.774	0.018	1.093	0.456–2.621	0.842
Distant metastasis
M0	1.000
M1	1.468	1.268–1.299	< 0.001	4.364	1.830–10.409	0.001
AJCC stage
I/II	1.000
III/IV	1.500	1.214–1.853	< 0.001	1.755	0.768–4.011	0.183
Pathological grade
I/II	1.000
III/IV	1.912	1.125–3.250	0.017	2.077	0.863–4.996	0.103
DDIT4 expression
Low	1.000
High	3.583	2.345–4.271	0.146

RR risk ratio, 95% CI 95% confidence interval, AJCC American Joint Committee on Cancer

Increased DDIT4 in GC cells promotes proliferation and colony formation

To validate the expression pattern of DDIT4, we detected DDIT4 protein and mRNA levels in four GC cell lines (MKN45, AGS, SGC7901, and BGC823) and an immortalized gastric epithelial cell line, GES. Similar to the GC tissues, DDIT4 levels were significantly increased at both the protein and mRNA level in GC cells compared with GES cells (Fig. 2a, b). Moreover, immunofluorescence revealed that DDIT4 was mainly localized to the cytoplasm (Fig. 2c). To analyze biological function, we silenced DDIT4 in SGC7901 and BGC823 cells with a lentiviral vector and upregulated DDIT4 levels in GES cells using a DDIT4-overexpressing lentiviral vector. After cell transfection and antibiotic screening for 6 weeks, the lentiviral transfection efficiency was confirmed by real-time PCR and western blotting. Among three shDDIT4 vectors, shDDIT4-2 was the most effective at silencing DDIT4 in SGC7901 and BGC823 cells (Fig. 2d, e). In contrast, transfection of the DDIT4-overexpressing lentiviral vector significantly upregulated DDIT4 levels in GES cells (Fig. 2f). Therefore, shDDIT4-2 and the DDIT4-overexpressing lentiviral vector were employed in the subsequent experiments.

Given that DDIT4 has been implicated in oncogene- and stress-induced DNA damage [4], we hypothesized that DDIT4 upregulation might be involved in the initiation and development of GC. To test this hypothesis, we conducted high-content screening assays and colony formation assays to determine whether DDIT4 regulates GC cell proliferation. The high-content screening assays revealed that cell proliferation was significantly inhibited by DDIT4 silencing in SGC7901 cells compared with the lentiviral control (Fig. 3a). Given that chemotherapeutics cause cytotoxic effects and DNA damage, we assessed whether DDIT4 was involved in the response of GC cells to these agents. Thus, we monitored SGC7901 cell proliferation after treatment with 5-FU (10 or 20 µg/mL) and found that DDIT4 downregulation suppressed proliferation in the presence of 5-FU (Fig. 3b, c). Consistently, colony formation assays revealed that DDIT4 downregulation inhibited SGC7901 colony formation with or without 5-FU (Fig. 3d). Similar to SGC7901 cells, DDIT4 downregulation in BGC823 cells reduced cell proliferation (Fig. 3e–g) and increased sensitivity to 5-FU (Fig. 3h). In contrast, ectopic expression of DDIT4 promoted GES cell proliferation and colony formation (Fig. 3i, l) and attenuated the sensitivity of GES cells to 5-FU (Fig. 3j–l). In addition, we performed Transwell assays to determine whether DDIT4 regulates GC cell migration and invasion, but no significant difference was observed in cells transfected with shDDIT4 lentivirus compared with the control group (Additional file 1: Figure S1). Taken together, these observations indicated that DDIT4 acts as an oncogene that promotes GC cell proliferation and reduces chemosensitivity.

Downregulation of DDIT4 increases 5-FU-induced apoptosis but decreases 5-FU-induced S phase arrest in GC cells

Beneath the complexity and idiopathy of cancer lies a limited number of critical events that propel the tumor cell and its progeny into uncontrolled expansion and invasion [13]. Two such events are deregulation of apoptosis and the cell cycle, which together with the obligatory compensatory dysregulation of proliferation provide a minimal “platform” necessary to support further neoplastic progression [13]. Thus, we performed flow cytometry to determine whether DDIT4 modulates apoptosis and the cell cycle, which contribute to gastric carcinogenesis. DDIT4 downregulation promoted apoptosis of SGC7901 cells treated with 0, 10 or 20 µg/mL 5-FU compared with the negative controls (Fig. 3m). In addition, DDIT4 silencing in SGC7901 cells attenuated gastric cancer cell S phase arrest (Fig. 3n). Similarly, in BGC823 cells, DDIT4 downregulation significantly increased GC cell apoptosis and reduced S phase arrest compared with the control (Fig. 3m, n). In contrast, DDIT4 overexpression in GES cells significantly reduced apoptosis in the absence and presence of 5-FU (10 or 20 µg/mL) (Fig. 3m). Moreover, as demonstrated by cell cycle analysis, ectopic expression of DDIT4 in GES cells drove the cell cycle into S phase and G₂/M phase and reduced the population of cells in G₁ phase compared with the control (Fig. 3n). Taken together, these findings indicated that DDIT4 is associated with 5-FU-induced apoptosis and cell cycle progression in GC cells in a dose-dependent manner.

DDIT4 silencing inhibits GC cell tumorigenesis in vivo

To investigate the effect of DDIT4 on GC tumorigenic behavior in vivo, we conducted tumorigenicity assays in nude mice by subcutaneously injecting BGC823 cells stably expressing shDDIT4 or scrambled control shRNA into the dorsal flank of several mice. DDIT4 depletion resulted in a significant reduction in tumor growth (Fig. 4a). DDIT4-knockdown tumors grew significantly slower (Fig. 4b) and weighed significantly less on average (Fig. 4c, d) compared with control tumors. Furthermore, we used western blotting and IHC to detect DDIT4 expression in xenografts and validated the lower DDIT4 levels in DDIT4-knockdown tumors compared with control tumors (Fig. 4e, f). Finally, we observed a reduced Ki67 (proliferation marker)-positivity rate in tumor tissues in the DDIT4-knockdown group compared with the control group (Fig. 4f). Taken together, these observations indicated that DDIT4 promotes GC growth in vivo and might function as an oncogene in gastric carcinogenesis.

DDIT4 downregulation inhibits GC cell proliferation through the MAPK and p53 pathways

To understand the molecular mechanism underlying GC growth regulation by DDIT4, we utilized phospho-array assays that detect 131 phosphorylation sites in 12 critical cancer signaling molecules (Additional file 2: Table S1). Given that DDIT4 acts as a negative regulator of mTOR, we first examined the activity of the phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K)/AKTkt/mTOR pathway in GC cells upon DDIT4 downregulation. Phospho-array assays revealed that the PI3K/AKT/mTOR pathway did not exhibit significant changes in DDIT4-knockdown cells (data not shown). Western blotting assays confirmed this observation (Fig. 5a), indicating that DDIT4 might activate other signaling pathways in GC cells. To uncover the molecular mechanism underlying GC growth regulation by DDIT4, we assessed MAPK and p53 signaling pathways in DDIT4-downregulated GC cells. In contrast to the PI3K/AKT/mTOR pathway, obvious alterations in the MAPK and p53 signaling pathways were observed (Fig. 5b). Anti-apoptotic BCL-2 levels were reduced, whereas pro-apoptotic BAD and proliferation-suppressive p21^Cip1 levels were increased in DDIT4-downregulated cells (Fig. 5b). To further examine the effect of blocking the MAPK and p53 signaling pathways on gastric cancer cell proliferation following DDIT4 knockdown, we performed rescue experiments by blocking the MAPK and p53 signaling pathways using a MAPK/ERK inhibitor (PD98059) and a p53 inhibitor (A15201) in DDIT4-knockdown cells. MAPK and p53 inhibition abolished the BCL-2 suppression and p21^Cip1 elevation that was induced by DDIT4 downregulation (Fig. 5c). Consistent with western blot assays, the CCK8 assays indicated that MAPK and p53 inhibition restored GC cell proliferation, which was suppressed by DDIT4 knockdown (Fig. 5d). Taken together, these findings indicated that the MAPK and p53 signaling pathways might play a critical role in DDIT4-mediated GC cell proliferation.