Knockdown of DIXDC1 Inhibits the Proliferation and Migration of Human Glioma Cells

Abstract

DIX domain containing 1 (DIXDC1), the human homolog of coiled-coil-DIX1 (Ccd1), is a positive regulator of Wnt signaling pathway. Recently, it was found to act as a candidate oncogene in colon cancer, non-small-cell lung cancer, and gastric cancer. In this study, we aimed to investigate the clinical significance of DIXDC1 expression in human glioma and its biological function in glioma cells. Western blot and immunohistochemistry analysis showed that DIXDC1 was overexpressed in glioma tissues and glioma cell lines. The expression level of DIXDC1 was evidently linked to glioma pathological grade and Ki-67 expression. Kaplan–Meier curve showed that high expression of DIXDC1 may lead to poor outcome of glioma patients. Serum starvation and refeeding assay indicated that the expression of DIXDC1 was associated with cell cycle. To determine whether DIXDC1 could regulate the proliferation and migration of glioma cells, we transfected glioma cells with interfering RNA-targeting DIXDC1; investigated cell proliferation with Cell Counting Kit (CCK)-8, flow cytometry assays, and colony formation analyses; and investigated cell migration with wound healing assays and transwell assays. According to our data, knockdown of DIXDC1 significantly inhibited proliferation and migration of glioma cells. These data implied that DIXDC1 might participate in the development of glioma, suggesting that DIXDC1 can become a potential therapeutic strategy for glioma.

Electronic supplementary material

The online version of this article (doi:10.1007/s10571-016-0433-5) contains supplementary material, which is available to authorized users.

Introduction

Glioma, which originated in neural epithelium, is one of the most common malignant tumors (Ozerlat 2012). Despite the development of microsurgical technique, conventional radiotherapy, stereotactic radiotherapy, new chemotherapy drugs and immunotherapy in recent 10 years, it is still hard to significantly prolong the survival of glioma, especially glioblastoma, whose survival is very short (Ferguson 2011; Robertson et al. 2011). The poor prognosis of these patients is due to the excessive proliferation and invasion biological characteristics of glioma cells (Brosicke et al. 2015; Denysenko et al. 2016; Mrugala 2013; Rolon-Reyes et al. 2015; Sigala et al. 2016; Tao et al. 2012). Thus, it is urgent to reveal the underlying molecular mechanisms of this malignancy and to seek the treatment-targeted point, so as to improve the prognosis of glioma patients.

DIX domain containing 1 (DIXDC1), a recently identified novel DIX domain containing protein, is the human homolog of coiled-coil-DIX1 (Ccd1). It is an important regulator of Wnt signaling pathway during zebrafish neural patterning (Shiomi et al. 2003, 2005; Wang et al. 2006). Containing a CH structure and MTH structure domains, DIXDC1 participates in the regulation of the cell shape and movement, and has been confirmed to belong to actin and tubulin cytoskeleton protein family (Wu et al. 2009). Studies have demonstrated that DIXDC1 may promote the nerve cell proliferation, and regulates the development and differentiation of the nervous system by adjusting GSK3-dependent β-catenin phosphorylation (Jing et al. 2009; Namba and Kaibuchi 2010). In the field of cancer research, it has recently been reported to be associated with the development of colorectal cancer. DIXDC1 regulates p21 and cyclin D1 through PI3K/AKT signaling pathway to promote the proliferation ability of colon cancer cells (Wang et al. 2009). Besides, DIXDC1 is an indicator of prognosis in non-small-cell lung cancer (NSCLC), and promotes tumor cell invasion and migration through PI3K-AKT/AP-1-dependent activation of metalloproteinases (Xu et al. 2014). According to the research of Tan et al., overexpression of DIXDC1 in gastric cancer activates the Wnt/β-catenin pathway, accelerating the invasion and metastasis of gastric cancer (Tan et al. 2015). Nevertheless, the function of DIXDC1 is less well understood in other tumors, especially in glioma.

In this study, we studied the expression of DIXDC1 in human glioma tissues and cell lines, and explored its role in regulating glioma cells proliferation and migration. We firstly recommend that DIXDC1 may be a useful molecular prognostic marker and provides a potentially useful target for glioma therapy.

Materials and Methods

Patients and Specimens

All the human glioma specimens and the clinicopathological data used in this report were obtained from 98 patients newly diagnosed with gliomas who had not received any radiotherapy, chemotherapy, or hormone therapy before sample collection between 2002 and 2010 at the Department of Pathology, Affiliated Hospital of Nantong University. All fresh frozen human glioma tissue samples were snapped frozen in liquid nitrogen immediately after surgical removal and stored at −80 °C until use. The grade of the tissue samples was evaluated using hematoxylin- and eosin-stained sections, according to the WHO classification of tumors of the central nervous system. Written informed consent was obtained from each patient prior to tissue acquisition. Institutional approval was acquired from the Ethical Review Board of Affiliated Hospital of Nantong University prior to this study.

Immunohistochemistry (IHC)

Tumor tissue microarray blocks were dewaxed in xylene and rehydrated using graded alcohol. To retrieve the antigen, the sections were processed in 0.01 mol/L citrate buffer (pH 6.0) and heated to 121 °C in an autoclave for 3 min. Endogenous peroxidase activity was blocked by soaking in 3 % hydrogen peroxide. After rinsing in phosphate-buffered saline (PBS) (pH 7.2), the sections were applied 10 % goat serum for 1.5 h at room temperature in order to block non-specific reactions and were incubated with mouse anti-human Ki-67 antibody (Santa Cruz, diluted 1:500) and mouse anti-human DIXDC1 antibody (Abcam Biotechnology, diluted 1:100) at 4 °C overnight. The peroxidase–antiperoxidase method (Dako, Hamburg, Germany) was used to process all the slides. After being washed in PBS, the peroxidase reaction was visualized by incubating the sections with DAB (0.1 % phosphate buffer solution, 0.02 % diaminobenzidine tetrahydrochloride, and 3 % H₂O₂). Hematoxylin was used for nuclear counterstaining after the sections were being rinsed in water. Finally, the sections were dehydrated with graded alcohol and cover-slipped.

Immunohistochemical Analysis

Staining of DIXDC1 and Ki-67 was reviewed and scored independently by three pathologists blinded to the clinical and pathological parameters of the patients using a semiquantitative immunoreactivity score for both intensity of stain and percentage of positive malignant cells. Five high-power fields were selected randomly, and more than 300 cells were examined in each specimen. The intensity of immunostaining was coded as 0–3 (0, negative staining; 1, weak staining; 2, moderate staining; and 3, strong staining). The percentage of immunoreactive cells was documented as 0 (0–10 % tumor cells stained), 1 (11–25 % tumor cells stained), 2 (26–50 % tumor cells stained), 3 (51–75 % tumor cells stained), and 4 (76–100 % tumor cells stained). Then, we multiplied the two scores. Since there barely was a field reached 12 scores, the results were classified into two groups: low expression (≤4) and high expression (>4), cutting of at 4.

Western Blot Analysis

Western blot analysis was described as previously detailed (Wang et al. 2014). The primary antibodies used in this study were as follows: anti-DIXDC1 (1:500; Santa Cruz Biotechnology); anti-E-cadherin (1:1000; Santa Cruz Biotechnology); anti-N-cadherin (1:1000; Santa Cruz Biotechnology); anti-PCNA (1:1,000; Abcam Biotechnology); anti-human cyclin D1 (1:1,000; Santa Cruz Biotechnology); and anti-GAPDH (1:3,000; Santa Cruz Biotechnology). GAPDH was used as the loading control.

Cell Cultures and Transient Transfection

Human glioblastoma cell lines H4, U251MG, A172, and U87MG obtained from Cell Library, China Academy of Science, were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) (GibCo BRL, Grand Island, NY) containing 10 % fetal bovine serum, 100 U/ml penicillin–streptomycin mixture (GibCo BRL), and 2 mM l-glutamine at 37 °C with 5 % CO₂. All cells were grown in sterile culture dishes and passaged every 2 days using 0.25 % trypsin. The medium was changed every 2–3 days. DIXDC1-shRNAs were designed and synthesized by GeneChem (Shanghai, China). The sequences of the four shRNAs targeting on DIXDC1 are as follows: shDIXDC1-1: 5′-TATTGATCGGGAAGGAAAT-3′; shDIXDC1-2: 5′-ACTCAAGTAGGTAGTGAAT-3′; shDIXDC1-3: 5′-GCGAGAGCTAGAACACAAA-3′; and shDIXDC1-4: 5′-AGGGATCATTCTGGGTAAA-3′. U87MG cells were transfected with shRNAs using Lipofectamine 2000 (Invitrogen, Carlsbad, CA), following the manufacturer’s recommendation (Life Technologies). After being transfected for 48 h, cells were collected for the following assays.

Flow Cytometry Analysis of Cell Cycle

To synchronize the cell cultures, U87 cells were cultured with 10 % FBS-containing medium overnight. Then, the cultures were washed with PBS and changed to serum-free medium. After serum starvation for 72 h, the cells were released into cell cycle by refeeding with 10 % FBS-containing medium and this point was considered as 0 h. Then, cells were harvested at 0, 4, 8, 12, and 24 h. For flow cytometrical analysis of cell cycle distribution, the harvested cells were fixed with 70 % ethanol at 4 °C overnight. After being washed with PBS, the cells were incubated with 1 mg/ml RNase A for 20 min at 37 °C. Then, the cells were stained with propidium iodide (PI, 50 μg/ml) (50 μg/ml; Becton–Dickinson, San Jose, CA) in PBS and 1 % Triton X-100 at 4 °C for another 20 min. The data were acquired and analyzed by a Becton–Dickinson flow cytometer BD FACScan (San Jose, CA) as well as Cell Quest acquisition and analysis programs. All of the samples were assayed in triplicate.

Cell Proliferation Assays

In order to evaluate cell viability, U87MG cells transfected with control-shRNA and DIXDC1-shRNA were seeded onto 96-well cell culture cluster plates (Corning Inc., Corning NY, USA) at a concentration of 2 × 10⁴ cells/well in volumes of 100 μl and grown overnight. Cell Counting Kit-8 reagents (Dojindo, Kumamoto, Japan) were added to each well at the proper time points and then incubated for another 2 h at 37 °C under the conditions of protection from light. A microplate reader (Bio-Rad) was used to measure the absorbance of each well at 450 nm.

Colony Formation Assay

U87MG cells transfected with control-shRNA and DIXDC1-shRNA were seeded in 60 mM dishes at 2000 cells/dish and cultured in 5 ml of DMEM containing 10 % FBS for 14 days. After being washed with PBS, colonies were fixed with methanol for 30 min, and then stained with 0.5 % crystal violet (Solarbio, Beijing, China) for 30 min. Clearly visible colonies (>50 cells/colony) were counted as positive for growth.

Monolayer Wound Healing Assays

To assess the effects of DIXDC1 on glioma cell migration, monolayer wound healing assays were used. U87MG cells were seeded in 6-well plates and incubated until they reached nearly complete confluence. Thirty-six hours after transfection, cells were serum-starved for 12 h. Subsequently, a scratch was made on the cell surface with a sterile 10 µl pipette tip and this point was considered as 0 h. Cells were washed with PBS, cultured in 5 % fetal bovine serum-Dulbecco’s modified eagle medium (FBS-DMEM) at 5 % CO₂ and 37 °C. The width of the wounds was photographed under an inverted Leica phase-contrast microscope (Leica DFC 300 FX) at 0, 24, and 48 h time points.

Transwell Migration Assay

U87MG cells transfected with DIXDC1 shRNAs were starved overnight in DMEM media with 0.1 % FBS, and then were trypsinized and resuspended into DMEM containing 0.1 % bovine serum albumin. Into the transwell chambers (Corning, 8 lm pore size), 1 × 10⁵ cells were seeded in the upper chambers in serum-free medium, and DMEM with 10 % FBS was added to the bottom chambers. Twenty-four hours after incubation, top (non-migrated) cells were removed, and bottom (migrated) cells were fixed and stained with crystal violet to visualize nuclei. The number of migrating cells in five fields was counted under ×400 magnification, and the means for each chamber were determined.

Statistical Analysis

Statistical analysis was carried out by SPSS20.0 statistical package from SPSS Inc. (Chicago, USA). The Kaplan–Meier method was used to perform survival analysis, and the log-rank test was carried out. The statistical significance of the correlations between the DIXDC1 expression and the clinical features was analyzed by Chi-square (χ ²) or Fisher’s exact test. Univariate and multivariate analyses were carried out using Cox’s proportional hazards model, and the risk ratio and its 95 % confidence interval were recorded for every marker. Comparisons between groups were undertaken by Student’s t test. All data in this study represent the results of at least three independent experiments. P < 0.05 was considered to be significant.

Results

DIXDC1 was Overexpressed in Glioma Tissues

In order to determine whether DIXDC1 is closely related to the development of gliomas, we evaluated the expression of DIXDC1 in glioma tissues of different grades by Western blot. We found that the expression of the expression of DIXDC1 increased along with the increased malignancy of gliomas (P < 0.05, Fig. 1a, b). In order to further study the relationship between DIXDC1 and clinical progression of glioma, we detected the expression of DIXDC1 and Ki-67 in 98 cases of paraffin-embedded glioma samples by immunohistochemistry. Ki-67 is thought to be a marker of proliferation (Yamashita et al. 2010). It was found that DIXDC1 was mainly expressed in the cytoplasm of glioma cells, and its expression was associated with higher grades of glioma, which was consistent with the expression of Ki-67 (Fig. 1c). Besides, its expression was positively correlated with Ki-67 in the examined glioma tissues (Table 1).

Fig. 1.

DIXDC1 is overexpressed in human glioma tissues. a Total protein extracted from glioma tissues (grades II–IV) was analyzed by Western blot with anti-DIXDC1 antibody. GAPDH was used as a loading control. b The band intensity of DIXDC1 and GAPDH was quantified; the bar chart demonstrates the ratio of DIXDC1 protein to GAPDH for the above by densitometry. The data are mean ± SEM of three independent experiments. (*P < 0.05) c Paraffin-embedded glioma tissue sections (grades II–IV) were stained with antibodies against DIXDC1 and Ki-67 followed by counterstaining with hematoxylin (SP × 200, scale bar 101 μm). Comparisons between groups were undertaken by Student’s t test

Table 1.

DIXDC1 expression and clinicopathological variables in 80 glioma specimens

Characteristics	Total	DIXDC1 expression		P value
		Low score ≤ 4	High score > 4
Age
<40	14	10	4	0.077
≧40	66	30	36
Gender
Female	34	16	18	0.651
Male	46	24	22
Tumor location
Frontal	18	7	11	0.507
Parietal	14	7	7
Occipital	17	7	10
Temporal	20	13	7
Unknown	11	6	5
Surgery
Biopsy	14	9	5	0.489
Partial resection	31	15	16
Gross total resection	35	16	19
Vessel density
Normal	17	11	6	0.172
Increased	63	29	34
Tumor diameter (cm)
<4	33	17	16	0.82
≧4	47	23	24
Necrosis
Absence	37	21	16	0.262
Presence	43	19	24
WHO grade
II	27	20	7	0.001*
III	31	16	15
IV	22	4	18
Ki-67
Low expression	43	35	8	0.000*
High expression	37	5	32

Statistical analyses were performed by the Pearson χ ² test

* P < 0.05 was considered significant

Relationship Between the Expression of DIXDC1 and Clinicopathological Factors of Glioma

In order to further illuminate the clinical roles of DIXDC1 in glioma, we summarized the immunohistochemical results of 98 glioma specimens (Table 1). As shown in Table 1, upregulated expression of DIXDC1 was associated with clinicopathologic grades (*P = 0.001) and expression of Ki67 (*P < 0.001). However, there was no significant association among age, gender, tumor location, type of surgery, vessel density, tumor diameter, or necrosis in 98 glioma specimens. Besides, we showed that DIXDC1 and Ki67 expression, and WHO grade were independent prognostic indicators for patients’ overall survival (P < 0.05, Table 2). The Kaplan–Meier test revealed that DIXDC1 expression was significantly associated with the overall survival of glioma patients (*P < 0.01, Fig. 2). As compared with low DIXDC1 expression, high expression of DIXDC1 demonstrated significantly shortened overall survival (*P < 0.01). The above data highlighted that DIXDC1 expression was a strong predictor of the glioma patients’ survival.

Table 2.

Contribution of various potential prognostic factors to survival by Cox’s regression analysis on 98 glioma specimens

Variables	Univariate analysis		Multivariate analysis
	Hazard ratio (95 % CI)	P	Hazard ratio (95 % CI)	P
Age (<40 vs. ≥40)	1.471 (0.678–3.189)	0.329
Gender (female vs. male)	1.612 (0.915–2.838)	0.098
Tumor location (frontal vs. parietal, occipital, temporal and unknown)	1.011 (0.814–1.254)	0.924
Tumor diameter (<4 vs. ≥4 cm)	0.969 (0.472–1.988)	0.931
Surgery (biopsy vs. partial resection and gross total resection)	0.865 (0.515–1.452)	0.582
Vessel density (normal vs. increased)	0.556 (0.247–1.250)	0.155
Necrosis (absence vs. presence)	1.634 (0.872–3.061)	0.126
WHO grade (II vs. III and IV)	3.020 (1.936–4.710)	0.000*	3.514 (2.399–5.147)	0.000*
DIXDC1 expression (low vs. high)	2.450 (1.032–5.815)	0.042*	6.158 (3.314–11.445)	0.000*
Ki-67 expression (low vs. high)	3.878 (1.523–9.874)	0.004*	8.926 (4.440–17.944)	0.000*

Statistical analyses were performed by the Cox’s regression analysis

CI confidence interval

* P < 0.05 was considered significant

Fig. 2.

DIXDC1 relates to the glioma outcome. Kaplan–Meier estimates of overall survival time in 98 patients who had a glioma with different DIXDC1 expressions. Patients in the high DIXDC1 expression group had significantly shorter overall survival (P < 0.01)

DIXDC1 was Upregulated in Proliferating Glioma Cells

Given the fact that DIXDC1 expression was mainly correlated with Ki67 expression and higher histological grade, we speculated that the expression of DIXDC1 might have a role in promoting the proliferation of glioma cells. The expression level of DIXDC1 was detected in four glioma cell lines (H4, U251MG, A172, and U87MG). Among all these cell lines, the expression of DIXDC1 was significantly higher in U87MG cells (Fig. 3a, b). So, we chose U87MG for the following experiment. After serum deprivation for 72 h, U87MG cells were arrested in G1 phase. When refed with 10 % FBS-containing medium, the cells were released from G1 phase and re-entered into S phase (Fig. 3c, d). Then, cells were harvested at 0, 4, 8, 12, and 24 h. Western blots showed that the expression of DIXDC1 was gradually increased after serum refeeding. Meanwhile, the expression of proliferation markers, cyclin D1 and PCNA, were also upregulated similarly (Fig. 3e, f). These results suggested that DIXDC1 could be identified as an important regulator of glioma cell proliferation.

Fig. 3.

The expression of DIXDC1 and cell cycle-related molecules in glioma cell lines. a and b Different expressions of DIXDC1 in four glioma cell lines (H4, A172, U87MG, and U251MG) were assessed by Western blot. The protein levels of DIXDC1 were quantified by correcting the average pixel intensity for each band with that of GAPDH as an internal loading control. The data are mean ± SEM of three independent experiments. (*P < 0.05, compared with H4) c and d After serum starvation for 72 h and addition of medium containing 10 % FBS for the indicated times, flow cytometry was performed to analyze cell cycle distribution of serum-starved and serum-released U87MG cells. Mean ± SEM of three independent experiments. (*, ^# P < 0.05, compared with control cells starved of serum for 72 h). e and f DIXDC1, cyclin D1, PCNA, and GAPDH (loading control) expression in serum-starved and -refed U87MG cells were analyzed by Western blot. The bar graph demonstrated the relative expression of DIXDC1, cyclin D1, and PCNA versus GAPDH at different time points in U87MG cells. Mean ± SEM of three independent experiments. (*^{, #, ^} P < 0.05, compared with cells starved of serum for 72 h). S serum starvation, R serum refeeding. Comparisons between groups were undertaken by Student’s t test

Downregulation of DIXDC1 Inhibited Proliferation of Glioma Cells

In order to further explore the role of DIXDC1 in glioma cells proliferation, U87MG cells were transfected with shRNAs directed against DIXDC1 expression or control-shRNA to knock down endogenous DIXDC1 in U87MG cells. Forty-eight hours after transfection, cell protein was collected for interference efficiencies analysis by Western blot. According to Fig. 4a–d, shDIXDC1-2 obviously downregulated the protein level of DIXDC1 in U87MG cells, accompanied by the downregulation of cyclin D1 and PCNA. Compared to the control-shRNA-transfected cells, cells transfected with shDIXDC1-2 were found to be more in G1 phase, and less in S phase (Fig. 4f), indicating that DIXDC1 may promote G1-S transition. Furthermore, Cell Counting Kit (CCK)-8 assay showed that knocking down of DIXDC1 in U87MG cells markedly reduced the cell proliferation rate (Fig. 4e), so did colony formation assay (Fig. 4g, h). As shown in Supplementary Fig.S1, the experimental results for U251MG cells confirmed above study. Taken together, these data suggested that DIXDC1 might play an important role in proliferation of glioma cells.

Fig. 4.

Knockdown of DIXDC1 decreases glioma cell proliferation. a and c 48 h after transiently transfected with DIXDC1-targeting shRNAs or control-shRNA, the knocking-down efficiencies were examined using Western blot analysis. The bar chart demonstrates the ratio of DIXDC1 protein to GAPDH by densitometry. Mean ± SEM (*P < 0.05, compared with the control) b and d Western blot analysis showing the effect of decreased DIXDC1 on the protein expression of PCNA, cyclin D1 in U87MG cells. The bar chart demonstrates the ratio of DIXDC1, PCNA, and cyclin D1 protein to GAPDH by densitometry. Mean ± SEM of three independent experiments. (*^{, #, ^} P < 0.05, compared with the control) e Cell vitality of U87MG cells transfected with the shDIXDC1-2 or control-shRNA was examined by CCK-8 assay at the indicated time. Mean ± SEM (*P < 0.05, compared with the control) f Flow cytometric analysis of cell cycle distribution 48 h later following control-shRNA and shDIXDC1-2 transfection. g and h Knocking down of DIXDC1 suppressed cell growth as determined by colony formation assays. The colonies (>50 cells/colony) were counted. Colony-formation ability was ratio of the number of colony to the number of cell plated. Mean ± SEM of three independent experiments. (*P < 0.05, compared with the control group) Comparisons between groups were undertaken by Student’s t test

DIXDC1 Regulated the Migration in Glioma Cells

Recent study showed that DIXDC1 participated in regulating the migration of non-small-cell lung cancer cells and gastric cancer cells (Tan et al. 2015). Therefore, we supposed that DIXDC1 might be associated with glioma cells migration. Monolayer wound healing assay and transwell assay were carried out to evaluate cellular migration ability. As shown in Fig. 5a, c, and Supplementary Fig. S2a, c, glioma cells transfected with shDIXDC1-2 migrated much slower than the cells transfected with control-shRNA at 24 h time point. We observed similar phenomenon at 48 h time point. The number of ShDIXDC1-2 transfected cells migrated to the bottom chamber was less than that of the control group (Fig. 5b, d, and Supplementary Fig. S2b, d). Based on the results of monolayer wound healing assay and transwell assay, we speculated that downregulating the expression of DIXDC1 might inhibit the migration ability of glioma cells. E-cadherin and N-cadherin are calcium-dependent cell adhesion molecules that mediate cell–cell adhesion and also regulate tumor cells migration and invasion (Hazan et al. 2000). The expression of E-cadherin and N-cadherin were examined and found that decreased DIXDC1 expression resulted in the upregulation of E-cadherin and downregulation of N-cadherin (Fig. 5e, f, and Supplementary Fig. S2e, f). Taken together, our results indicated that DIXDC1 was associated with the migration of glioma.

Fig. 5.

Knockdown of DIXDC1 will inhibit the migration of glioma cells. a and c Wound healing assays with control-shRNA and shDIXDC1-2 transfected U87MG cells. Migration of the cells to the wound was visualized at 0, 24, and 48 h with an inverted Leica phase-contrast microscope (×200 magnification). Each time point is derived from three independent experiments. (*P < 0.05, compared with the control-sh) b and d Crystal violet staining of glioma cells that crossed the polycarbonate membrane of the transwell chamber to detect the effect of DIXDC1 on migration of glioma cells. Number of cells that migrated through the member was counted by an inverted Leica phase-contrast microscope (×400 magnification) in five fields. Columns, mean of triplicate experiments (*P < 0.05, compared with the control-sh). e and f Western blot analysis of DIXDC1, E-cadherin, N-cadherin, and GAPDH in control-shRNA and shDIXDC1-2 cell lines. The bar graph demonstrated the relative expression of DIXDC1, N-cadherin, and E-cadherin versus GAPDH in U87MG cells transfected with control-shRNA or shDIXDC1-2. Mean ± SEM of three independent experiments. (*^{, #, ^} P < 0.05 compared with the control group). Comparisons between groups were undertaken by Student’s t test

Discussion

Gliomas develop from the glial cells which form structures that surround and support neurons. According to the World Health Organization classification, gliomas are classified into four grades based on the histological appearance of the tumors (Kleihues et al. 1993; Louis et al. 2007). Despite of current advanced diagnostic methods and therapies, the prognosis is grave, with long-term survival rates of less than 5 % because of the high invasion and proliferation characteristic of gliomas (Mahaley et al. 1989). Therefore, exploring new molecular markers and relevant mechanisms of gliomas will help identify new therapeutic targets for the treatment of gliomas.

DIXDC1, a positive regulator of Wnt signaling, has been recognized as an Axin2 C-terminus interaction protein (Shiomi et al. 2005; Wang et al. 2006). The DIX domain is important for mediating multiple signal transductions (Luo et al. 2005; Wong et al. 2004). Up to now, there have been few studies on the biological function of human DIXDC1, especially in the field of cancer. It has been reported that DIXDC1 is associated with the progression of NSCLC (Xu et al. 2014), gastric cancer (Tan et al. 2015), and colon cancer (Wang et al. 2009) so far. However, the expression and significance of DIXDC1 in gliomas have not yet been elucidated. According to previous study, NSCLC expressed high level of DIXDC1 in cytoplasm, while normal lung tissues showed negative or weak staining (Xu et al. 2014). Our study confirmed that DIXDC1 was overexpressed in gliomas. The expression level of DIXDC1 increased along with the high grades of gliomas and these tissues displayed stronger staining, which was consistent with previous study. The result of immunohistochemical staining indicated that DIXDC1 may play an important role in gliomas development. In addition, we found that DIXDC1 expression was positively correlated with WHO grade and Ki67 expression. Multivariate analysis with the Cox proportional hazards model revealed that DIXDC1 could be an independent prognostic factor for the survival of glioma patients. Accordingly, Kaplan–Meier analysis indicated that overexpression of DIXDC1 predicted poor prognosis.

Studies showed that altered regulation of the cell cycle is closely related to abnormal cell proliferation, leading to tumorigenesis and development of tumors (Uchida et al. 2012). It has been reported that DIXDC1 promoted G0/G1 to S phase transition concomitantly with upregulation of cyclin D1 to upregulate colon cancer cell proliferation (Wang et al. 2009). To further explore whether DIXDC1 is involved in the proliferation of glioma cells, we investigated DIXDC1 expression during cell cycle progression in glioma cells and found that DIXDC1 level increased after serum stimulation in accordance with cyclin D1 and PCNA, and releasing from G1. Silencing of DIXDC1 in glioma cells markedly suppressed the rate of cell proliferation, increased G1-phase cell populations, and decreased S phase populations. These results indicated that DIXDC1 might have an important role in the progression of cell cycle and thus promote the proliferation of tumors.

E-cadherin and N-cadherin are calcium-dependent cell adhesion molecules that mediate cell–cell adhesion and also regulate tumor cells migration and invasion (Hazan et al. 2000). Downregulation of E-cadherin is correlated with invasion and metastasis of tumors including hepatocarcinoma, breast cancer, and glioma (van Roy and Berx 2008; Wang et al. 2013). Previous research has shown that the increased migration and invasion abilities of DIXDC1-overexpressing gastric cancer cells might be promoted by DIXDC1-mediated E-cadherin downregulation (Tan et al. 2015). In our report, with wound healing assay and transwell assay, we found that DIXDC1 was associated with the migration of glioma. Moreover, our study indicated that N-cadherin was downregulated and E-cadherin was upregulated in shDIXDC1-2-transfected glioma cells. Taken together, DIXDC1 might take part in regulating glioma cells migration through modulating N-cadherin and E-cadherin expression.

In summary, our study first identified that the expression of DIXDC1 was significantly increased in human glioma tissues and was significantly associated with WHO grade as well as poor prognosis. Moreover, knockdown of DIXDC1 expression by shRNA could inhibit the proliferation and migration of glioma, providing a novel therapeutic target for glioma. However, the complex molecular mechanisms of DIXDC1 in glioma development need to be further investigated.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Compliance with Ethical Standards

Conflicts of interest

All authors declare no conflict of interest.