TRIM21 overexpression promotes tumor progression by regulating cell proliferation, cell migration and cell senescence in human glioma

TRIM21 is elevated in human gliomas

To investigate the expression of TRIM21 in gliomas, the expression levels of TRIM21 protein were examined by immunohistochemical staining performed on the Chinese tissue microarray (TMA), containing 264 gliomas and 16 controls. As shown in Figure 1A, 1B and Table S1, the immunoreactivity of TRIM21 was significantly elevated in LGG (low grade glioma) and HGG (high grade glioma) samples, and increased according to WHO grade. Furthermore, similar results could be seen between normal brain tissues and gliomas of different WHO grade or histological grouping (Figure 1B). We further determined whether the TRIM21 gene was also increased at the transcriptional level. We examined its mRNA level in a cohort of 17 normal brain tissues and 41 glioma specimens. Consistently, the mRNA expression of TRIM21 was also significantly increased in different WHO grade gliomas (Figure 1C). Moreover, the result was further confirmed in TCGA, CGGA and Rembrandt cohorts (Figure 1D-F). Taken together, our findings suggested that TRIM21 is elevated in human gliomas.

TRIM21 serves as a prognostic factor

To explore the relationship between TRIM21 expression and the clinical prognosis of patients with glioma, we first examined the prognostic significance of TRIM21 protein expression in a Chinese TMA cohort. Kaplan-Meier analysis was performed, and the results showed that in all patients with glioma, those with high TRIM21 expression had a markedly shorter overall survival (OS) and progression-free survival (PFS) than those with low TRIM21 expression (Figure 2A). As higher tumor grade is an indubitable risk factor for glioma, we further conducted similar analysis in patients with HGG and GBM. As shown in Figure 2B and 2C, high TRIM21 expression could also predict significantly unfavorable OS and PFS in patients with HGG and GBM. Similar analysis performed in the CGGA, TCGA, and Rembrandt cohorts revealed that, higher TRIM21 mRNA expression could significantly predict worse OS and PFS for all gliomas (Figures 2E and S1), HGGs (Figure 2F) and GBMs (Figure 2D).

Univariate and multivariate Cox regression analyses were also performed to examine the independence of the prognostic value of TRIM21 protein expression. As shown in Table 1, after correction for clinical characteristics suggested to be significant prognostic factors in univariate Cox regression, high TRIM21 protein expression was an independent risk predictor of both OS and PFS for all gliomas, HGGs, and GBMs in the Chinese TMA cohort. Together, our findings indicate that TRIM21 serves as a prognostic factor in all gliomas, HGGs and GBMs.

TRIM21 promotes cell growth and migration in vitro

To further explore the biological significance of TRIM21 in glioma, TRIM21-overexpressed or depleted U87-MG and U251 cells were established by lentivirus infection with optimized lentivirus vectors (Figures 3A, S2A and S2B). As determined by CCK-8 assays, TRIM21-overexpressed U87-MG and U251 cells displayed a faster growth rate than the controls; targeting TRIM21 by shRNAs markedly inhibited cell proliferation, and similar results were obtained in U373 cells (Figures 3B and S2C). Consistent results in U251 cells were observed in colony formation assays (Figure S2D). We further examined whether the cell cycle was regulated by TRIM21 in glioma. Cell cycle analyses and EdU assays demonstrated that the percentage of G1 phase cells was significantly reduced with TRIM21 overexpression and elevated with TRIM21 depletion, on the contrary, the percentage of S phase cells was significantly elevated with TRIM21 overexpression and reduced with TRIM21 depletion (Figure 3C-E). Moreover, transwell and wound healing assays displayed that TRIM21 overexpression markedly promoted cell migration, whereas TRIM21 depletion depressed cell migration (Figure 4A-D). Taken together, our results indicated that TRIM21 regulates glioma cell growth and migration in vitro.

TRIM21 promotes glioma growth in vivo

Our data indicated that TRIM21 controls glioma cells survival in vitro and suggested that it might play an important role in the modulation of glioma progression. To verify the impact of TRIM21 in glioma, we conducted an in vivo experiment with intracranial xeno-transplanted tumor models. In intracranial xenograft mice, tumor progression was monitored by bioluminescence imaging technique on day 14, 20, and 26 after implantation. Xenografts carrying TRIM21-overexpressed U87-MG cells showed a significant increase in signal compared with the control group. On the contrary, mice bearing TRIM21-depleted U87-MG cells showed a drastic regression of tumor growth compared with the control group (Figure 5A). Survival rates of different groups of intracranial xenograft mice were described by the survival curve. As shown in Figure 5B, compared with the control group, mice injected with TRIM21-overexpressed U87-MG cells and TRIM21-depleted U87-MG cells exhibited shorter and longer OS time, respectively. For a pathological form of the intracranial tumor, representative paralleled HE-stained images from each group are shown in Figure 5C. Immunostaining analyses revealed an elevation of Ki-67 expression in tumors derived from TRIM21-overexpressed U87-MG cells (Figure 5D and 5E).

TRIM21 suppresses cell senescence through the p53-p21 pathway in glioma cells

To further elucidate the mechanism by which TRIM21 modulates glioma progression, RNA isolated from U87-MG cells infected with lentivirus expressing shLuc and shTRIM21 was subjected to transcriptional RNA sequencing profiling. A total of 116 million paired-end reads with average of 19 million reads per sample were generated from RNA sequencing, and a summary of their transcriptome statistics is shown in Table S2. Meanwhile, comparing shLuc with shTRIM21 U87-MG cells, a total of 1 601 differentially expressed genes (DEGs) were identified on the basis of p-value ≤0.05. Among all DEGs, 773 and 828 genes were downregulated and upregulated, respectively. Gene Ontology analysis showed that the identified upregulated genes were mainly involved in endoplasmic reticulum unfolded protein response, response to topologically incorrect protein, cellular response to unfolded protein, and cellular response to topologically incorrect protein biological processes (Figure 6A). These biological processes are related to the E3 ligase function of TRIM21. Moreover, a number of identified upregulated genes were involved in cellular response to stress, apoptotic signaling pathway, response to stress and response to oxidative stress biological processes (Figure 6A). In addition, downregulated genes were mainly involved in mitotic cell cycle, cell cycle process, cell cycle phase transition, and DNA repair (Figure 6A). These biological processes are closely related to cell senescence. More importantly, KEGG analysis revealed that the identified DEGs were significantly enriched in protein processing in endoplasmic reticulum pathway, p53 signaling pathway, and cell cycle pathway (Figure 6B).

Because p53-p21 is one of the major pathways that mediate aging, we quantified the mRNA and protein levels of TP53 and CDKN1A in the pathway. As shown in Figure 6C and 6D, both mRNA and protein levels of TP53 and CDKN1A were significantly reduced in TRIM21-overexpressed U87-MG cells, whereas they were remarkably increased in TRIM21-depleted U87-MG cells. Furthermore, as determined by CCK-8, cell cycle, EdU, and SA-β-Gal staining assays, TRIM21 knockdown promoted cell senescence in U87-MG cells (Figure 6E-I). On the contrary, TRIM21-overexpressed U87-MG cells displayed a lower cell senescence rate (Figure S3). Importantly, p53 knockdown could rescue the cell senescence effect of TRIM21 knockdown on glioma cells (Figure 6E-I). Taken together, our findings suggested that TRIM21 may modulate glioma progression at least in part, through cell senescence via the p53-p21 pathway.

TRIM21 increases glioma cell resistance to temozolomide (TMZ)

TMZ is a first-line chemotherapeutic drug for GBM. It exerts its effect by causing DNA damage and inhibiting cell division. Several studies have indicated that TMZ resistance is associated with p53 malfunction [26-29]. We performed analyses of TRIM21 expression stratified by the status of common genetic aberrations of GBM in TCGA cohort. As shown in Figure 7A and 7B, the mRNA expression of TRIM21 was significantly increased in GBM patients with IDH1 wild type or non-G-CIMP compared with those with IDH1 mutations or G-CIMP. Meanwhile, a similar trend could be seen between MGMT-unmethylated, and MGMT-methylated (Figure 7C). Studies have shown that GBM patients with IDH1 mutation, MGMT-methylated and G-CIMP have higher total survival and responsiveness to TMZ treatment [30,31]. We next examined whether TRIM21 played a critical role in resistance to TMZ therapy. As determined by CCK-8 assays, the overexpression of TRIM21 decreased the sensitivity of U87-MG cells to TMZ, but the depletion of TRIM21 sensitized U87-MG to TMZ (Figure 7D and 7E). Together, our findings indicated that TRIM21 increases the resistance of glioma cells to TMZ, which may be the underlying mechanism in the poor prognosis of glioma patients with high expression of TRIM21.

Ethics statement

This study complied with the tenets of the Helsinki Declaration and the National Guidelines for Animal Use in Research (China). This study was approved by the Specialty Committee on Ethics of Biomedicine Research of the Second Military Medical University (Shanghai, China, hereafter referred to as ‘SMMU’). All animal experiments were conducted with the approval of the Institutional Animal Care and Use Committee of Fudan University. All animal studies were performed according to the Ethics Committee guidelines of Fudan University.

Patient samples

The Chinese tissue microarray (TMA) cohort of human glioma tissue specimens from different WHO grades patients and normal brain tissue samples from trauma patients were obtained from the Department of Neurosurgery in Changzheng Hospital, Second Military Medical University (SMMU) (Shanghai, China), between January 2000 and December 2010. The human gliomas cohort of fresh-frozen tissues from 41 glioma patients and 17 normal controls were collected from the Department of Neurosurgery in Changzheng Hospital, SMMU, between June 2008 and July 2010. Informed consent was provided by each patient. The Specialty Committee on Ethics of Biomedicine Research of SMMU approved the patient sample acquisition.

Antibodies and chemicals

The following antibodies were used: TRIM21 (SC-25351; Santa Cruz), GAPDH (ab9385; Abcam), Ki-67 (9449; Cell Signaling Technology), p53 (9282; Cell Signaling Technology), p21 (ab109520; Abcam). The TRIzol Reagent was purchased from the Invitrogen (Thermo Fisher Scientific, USA). The Cell Counting Kit-8 was purchased from Dojindo Laboratories (Japan). Puromycin was purchased from Sigma. The protease inhibitors cocktail was purchased from the Sigma-Aldrich (USA). The D-luciferin was purchased from Promega (USA). The Lipofectamine 2000 was purchased from Thermo Fisher Scientific (USA).

Cell culture and lentivirus infection

293T and human malignant glioma cell lines (U87-MG, U251, U373) were obtained from the American Type Culture Collection (ATCC). All cells were maintained in DMEM with 10% (v/v) FBS, and were grown at 37°C with 5% CO₂. 293T cells were transiently transfected using Lipofectamine 2000 according to the manufacturer’s instructions. pCD513B-1-based gene-overexpression or pll3.7-based gene-knockdown lentiviral vectors and packing constructs were transfected into 293T cells. Virus-containing supernatant was collected 48 hr. after transfection. U87-MG, U251 and U373 cells were infected with virus-containing supernatant over 48 hr.

Plasmids constructions

TRIM21 cDNA was subcloned into pCD513B-1 lentiviral vector to generate the TRIM21 overexpression plasmid. pll3.7-shRNA plasmids generated by methods described previous [37]. The shRNA sequence of shTRIM21#2: 5’-UGGCAUGGUCUCCUUCUACAA-3’; shTRIM21#3: 5’-CUGCCUUCUUUAUGGGACUUA-3’; shTP-53#1: 5’-GUCCAGAUGAAGCUCCCAGAA-3’; shTP53#2: 5’-CGGCGCACAGAGGAAGAGAAU-3’; shLuc: 5’-CAAAUCACAGAAUCGUCGUAU-3’.

Immunohistochemistry assays

The immunohistochemistry assays were performed on TMA and nude mice xenograft tissue to detect TRIM21 and Ki-67 expression by methods described previously. The immunohistochemical staining were scored by two independent pathologists in a blinded manner. The evaluation of the staining density, intensity and the calculation of total immunoreactivity score were described previously. TRIM21 expression was denoted as low (total score ≤6) or high (total score >6) to divide the glioma patients into two groups.

Cell proliferation assays

Cell proliferation rate was determined using Cell Counting Kit-8 (CCK-8) according to the manufacturer’s protocol (Dojindo Laboratories, Japan). Briefly, the cells were seeded onto 96-well plates at a density of 1 200 cells per well. During a 4 to 7-d culture periods, 10 μL of the CCK-8 solution was added to cell culture, and incubated for 1 hr. The resulting color was assayed at 450 nm using a microplate absorbance reader (Bio-Rad). Each assay was carried out in triplicate.

Cell cycle analysis

For cell cycle analysis, cells were harvested and washed 48 hr post-treatment with PBS, followed by propidium iodide (50 μg/mL) staining in the presence of RNase (10 μg/mL) for 30 minutes at 4°C in the dark. The fraction of viable cells in G0/G1, S and G2/M phases of cell cycle were determined using a FACs flow cytometer and Cell Quest FACS system (Becton-Dickinson).

EdU assays

EdU assay was performed using the EdU Apollo 567 Cell Tracking Kit (Rib-bio; Guangzhou, China). Treated and control cells (5 × 10³/well) were seeded onto 96-well plates and incubated with 5-ethynyl-20-deoxyuridine (EdU; 200 μM) for 2 h at 37°C. Cells were fixed with 4% paraformaldehyde for 20 min, treated with 0.5% TritonX-100 for 10 min, washed with PBS three times, and incubated with 100 μL of Apollo reagent for 30 min. Nuclei were labeled with Hoechst 33342. The percentage of EdU-positive cells was calculated by image J software. Data reported represent the average of three independent experiments.

SA-β-Gal activity assays

SA-β-Gal activity was determined using Sigma SA-β-Gal staining kit according to the manufacturer’s protocol. Briefly, the cells were seeded onto 24-well plate 24 or 48 hr before staining. After washing cells with PBS for triple, cells were fixed with the fixation buffer, followed by three washed with PBS and then incubated with SA-β-Gal staining solution at 37°C separated from CO₂ overnight. Cells were visualized and imaged using a microscope (Leica).

Migration assays

Cell migration was determined by Transwell (Costa) migration assay. U87-MG cells were precultured in serum-free medium for 48 hr. 1-3 × 10⁴ cells were seeded in serum-free medium in the upper chamber, and the lower chamber was filled with DMEM containing 10% FBS. After 48 hr, the non-migrating cells on the upper chambers were carefully removed with a cotton swab, and the migrated cells underside of the filter stained and counted in nine different fields.

Western blotting

Cell lysates or immunoprecipitates were subjected to SDS-PAGE, and proteins were transferred to nitrocellulose membranes (GE Healthcare Sciences). Membranes were blocked in Tris-buffered saline (TBS, pH7.4) containing 5% nonfat milk and 0.1% Tween-20, washed twice in TBS containing 0.1% Tween-20 and incubated with primary antibody overnight at 4°C followed by secondary antibody for 1 hr at room temperature. Proteins of interest were visualized using the Enhanced Chemiluminescence (ECL) system (Santa Cruz Biotechnology). Densitometry analysis of protein bands was performed on image J software.

qRT-PCR

Total RNA was isolated from cells or tissues using TRIzol reagent (Thermo Fisher Scientific), and cDNA was reverse-transcribed using the Superscript RT kit (Toyobo), according to the manufacturer’s instructions. Quantitative real-time PCR amplification was performed using the THUNDERBIRD SYBR qPCR Mix (Toyobo) on ABI PRISM 7900HT instruments (Applied Biosystems). The amplification was done in a total volume of 10 μL with the following steps: an initial denaturation step at 95°C for 5 min, followed by 40 cycles for denaturation at 95°C for 15 sec and elongation at 60°C for 45 sec. A melting curve analysis of each sample was used to check the specificity of amplification, and each sample was assayed in triplicate. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the endogenous control, and the 2^-ΔΔCt method was used as relative quantification measure of differential expression. The primer sequences for qRT-PCR used are as follows: TRIM21-F: 5’-CGTTGAGTCCCCTGTAAAGC-3’; TRIM21-R: 5’-CAGGCAGATAGGGCATGTG-3’; TP53-F: 5’-GGAGCCGCAGTCAGATCCTAG-3’; TP53-R: 5’-CAAGGGGGACAGAACGTTG-3’; CDKN1A-F: 5’-TAGCAGCGGAACAAGGAG-3’; CDKN1A-R: 5’-AAACGGGAACCAGGACAC-3’; GAPDH-F: 5’-GAAGGTGAAGGTCGGAGT-3’; GAPDH-R: 5’-GAAGATGGTGATGGGATTTC-3’.

Generation of glioma xenografts in mice

For generation of orthotopic xenografts, 6-week-old female nude mice (SLAC laboratory animal Center, Shanghai, China) were maintained in a barrier facility on high-efficiency particulate air (HEPA)-filtered racks and randomly divided into different experimental groups. U87-MG cells with luciferase expression were infected with TRIM21 overexpression or deletion lentiviruses and their corresponding control lentiviruses. Cells (1 × 10⁶) were surgically implanted into left amygdala of mice brains using a stereotactic apparatus (KDS310, KD Scientific; Holliston, MA, USA). For in vivo BLI, luciferase signals were detected 14, 20, 26 days after transplantation. Mice were intraperitoneally injected with D-luciferin (150 mg/kg) (E1603, Promega) and anesthetized with pentobarbital sodium. After 10 min of substrate administration, images were acquired through the In Vivo Imaging System (IVIS, PerkinElmer, Waltham, MA), and luminescence flux was recorded to assess tumor growth in a blinded fashion. After transplantation, animals were closely followed and euthanized by cervical dislocation when they exhibited central nervous system symptoms, or drastic loss of body weight. Tumors were excised, formalin-fixed, paraffin-embedded, and sectioned for hematoxylin and eosin (HE) staining and IHC. All animal experiments were conducted with the approval of the Institutional Animal Care and Use Committee of Fudan University. All animal studies were performed according to the Ethics Committee guidelines of Fudan University.

Statistics

All data are shown as mean values ± SD/SEM for experiments performed with at least three replicates. The difference between two groups was analyzed using two-tailed Student’s t-test unless otherwise specified. Analysis of survival was conducted by Kaplan-Meier survival and Cox regression. * represents P<0.05; ** represents P<0.01; *** represents P<0.001.