LINC01198 promotes proliferation and temozolomide resistance in a NEDD4-1-dependent manner, repressing PTEN expression in glioma

LINC01198 is upregulated in human glioma and is correlated with poor prognosis

Microarray data from GSE16011, Chinese Glioma Genome Atlas (CGGA) and the Repository for Molecular Brain Neoplasia Data (REMBRANDT) datasets show LINC01198 expression increased with tumor grade in glioma [8]. To examine whether LINC01198 plays an important role in glioma progression, we first analyzed the expression level of LINC01198 in human glioma and adjacent normal tissues using qRT-PCR, and we found that the LINC01198 expression level was significantly increased in glioma tissues compared with adjacent normal tissues (74/90) (Figure 1A). Next, we explored the relationship between clinicopathological characteristics and the LINC01198 expression in 90 glioma patients, as listed in Table 1. The results showed that glioma patients with LINC01198^High had a bigger tumor size (P = 0.001) and higher WHO grade (P = 0.006) (Figure 1B). Then, we investigated the prognostic implication of LINC01198 expression. Interestingly, our results showed that patients with LINC01198^High expression had a worse prognosis than those with LINC01198^Low expression (Figure 1C and 1D). Multivariate analysis identified LINC01198 expression as an independent predictor for prognosis of glioma patients (Tables 2 and 3). These results oracle that LINC01198 is likely participated in the progression of glioma.

Figure 1. LINC01198 is overexpressed in human glioma. (A) Expression level of LINC01198 was determined and normalized to GAPDH in 90 pairs of glioma tissues and their corresponding adjacent normal tissues by qRT-PCR. (B) The patients were divided into the tumor size ≤3 cm and tumor size >3 cm groups. The diagram shows the LINC01198 expression in each group. P = 0.001. (C) The patients were divided into the stage I-II and III-IV groups. The diagram shows the LINC01198 expression in each group. P = 0.006. (D and E) Glioma patients were divided into the LINC01198^High group and LINC01198^Low group according to the result of qRT-PCR. Survival and recurrence curves were constructed using the log-rank test. The results show the OS and recurrence for glioma patients with high LINC01198 expression and those with low LINC01198 expression.

LINC01198 enhances the proliferation and temozolomide resistance of glioma cells

The biologic function of LINC01198 in regulating the glioma progression remains unclear. Next, we carried out loss/gain-of-function studies in glioma cells. After detecting the expression levels of LINC01198 in six glioma cell lines, two normal brain tissues, and two glioma tissues, we performed knockdown of LINC01198 in U87 and LN229 cells that had high LINC01198 expression and upregulated its expression in U251 and SNB-19 cells that had low LINC01198 expression (Figure 2A–2C). CCK-8 assay result showed that knockdown of LINC01198 impaired glioma cell proliferation, whereas LINC01198 overexpression promoted cell proliferation in vitro (Figure 2D). Furthermore, CCK-8 assay was performed to investigate the role of LINC01198 in the sensitivity of glioma cells to temozolomide. The results showed that knockdown of LINC01198 dramatically increased the sensitivity of glioma cells to temozolomide, while LINC01198 overexpression reduced the sensitivity of glioma cells to temozolomide (Figure 2E).

Figure 2. LINC01198 promotes glioma proliferation and resistance to temozolomide. (A) LINC01198 expression was examined in several glioma cell lines, normal brain tissues, and glioma tissues using qRT-PCR. GAPDH was used as a control for loading. (B and C) LINC01198 expression in glioma cells was modified by shRNA interference and cDNA transfection. (D) Growth curve showing the proliferation activities of glioma cell transfectants in vitro. (E) IC50 value of temozolomide on glioma cells. Cells were treated with different concentrations of temozolomide for 72 hours. (F) Antitumor effect of temozolomide on glioma xenografts in an established model. (G) The data were expressed as percentage inhibition of tumor growth. (H) The survival time of mice bearing glioma subcutaneous xenograft received temozolomide therapy. Data are presented as the mean ± SD, n=3. *P < 0.05, **P < 0.01.

To further investigate the potential clinical relevance of LINC01198 in vivo, we subcutaneously injected U251 cells with or without stable LINC01198 overexpression into the dorsal flanks of 4-6 week BALB/c nude mice. Compared to Mock groups, temozolomide slightly inhibit tumor growth which derived from U251 cells overexpressing LINC01198 xenografts (Figure 2F and 2G). Moreover, the mice bearing subcutaneous xenograft U251 cells overexpressing LINC01198 tumors showed a short survival time compared with Mock groups tumors (Figure 2H).

LINC01198 functions as a scaffold for NEDD4-1 and PTEN

To investigate the potential mechanisms of LINC01198 in glioma cells, we predicted the interactions between LINC01198 and RNA-binding proteins by an RNA-protein interaction prediction tool (http://pridb.gdcb.iastate.edu/RPISeq/), and we found that LINC01198 probably binds to NEDD4-1 and PTEN (Supplementary Table 3) (Figure 3A). Then, we carried out RIP assays and found that LINC01198 directly binds NEDD4-1 and PTEN in glioma cells (Figure 3B). In addition, RNA pulldown assays further confirmed that LINC01198 indeed binds with NEDD4-1 and PTEN in glioma cells (Figure 3C). Our results showed that LINC01198 could bind to both NEDD4-1 and PTEN, indicating that LINC01198 may function as a scaffold for NEDD4-1 and PTEN in glioma cells. To further confirm this hypothesis, the LINC01198 gene was cut into LINC01198-D1 (1-500 nt) and -D2 (601-955 nt), and their expression plasmids were constructed and transfected into U87 cells (Figure 3D). Our results showed that LINC01198-D1 mostly bound with PTEN, whereas LINC01198-D2 bound with NEDD4-1 (Figure 3E and 3F). These data showed that LINC01198 may function as a scaffold and bind with PTEN and NEDD4-1 at its 5′ region and 3′ region in glioma cells, respectively.

Figure 3. LINC01198 functions as a scaffold for NEDD4-1/PTEN to regulate PTEN expression in glioma cells. (A) Predicted interaction probabilities of LINC01198 and RNA-binding proteins via a RNA–protein interaction prediction tool. (B) RIP assays of LINC01198 binding to indicated proteins in glioma U87 cell extracts. (C) RNA pulldown was used to examine the association of LINC01198 and NEDD4-1/PTEN. (D) The LINC01198 gene was cut into LINC01198-D1(1-500 nt) and -D2(501-955 nt), and their overexpression vectors were constructed and transfected into U87 cells. (E) RIP assays of LINC01198-D1/D2 binding to indicated proteins in glioma U87 cell extracts. (F) Biotinylated LINC01198-D1/D2 RNAs were incubated with U873 cell lysates, and western blotting analysis was performed to evaluate the specific association between them and NEDD4-1/PTEN.

LINC01198 increased NEDD4-1-induced PTEN inhibition

Previous studies have demonstrated that NEDD4-1-mediated PTEN ubiquitination is crucial in the regulation of PI3K/AKT signaling in neurodegeneration [9]. We thus evaluated the effect of LINC01198 on NEDD4-1-induced PTEN inhibition in glioma cells. Overexpression of LINC01198 significantly inhibited the expression of PTEN protein but not mRNA in U257 and SNB-19 cells (Figure 4A, Supplementary Figure 1A). Conversely, LINC01198 knockdown increased PTEN protein but not mRNA expression in U87 and LN229 cells (Figure 4B, Supplementary Figure 1B). However, in vitro experiments showed that overexpression or knockdown of LINC01198 did not affect NEDD4-1 expression in glioma cells (Figure 4C and 4D). Moreover, overexpression or knockdown of LINC01198 did not affect PTEN expression in CRISPR/cas9-induced NEDD4-1-knockout glioma cells (Figure 4E and 4F; Supplementary Figure 2). Then, to investigate the relationship between LINC01198 and PTEN, we detected the expression of PTEN in 90 cases of glioma tissues using western blotting or qRT-PCR. Scatterplot assays results indicated a negative correlation between LINC01198 and PTEN at the protein level (r = 0.165; P = 0.004; Figure 4G), but this correlation was not shown at the mRNA level (r = 0.003; P > 0.05; Figure 4H). These results indicate that LINC01198 regulates PTEN expression through NEDD4-1.

Figure 4. LINC01198 increased NEDD4-1-induced PTEN inhibition in glioma cells. (A) Forced LINC01198 expression inhibited PTEN expression in glioma cells. (B) Protein expression levels of PTEN were increased in glioma cells treated with LINC01198 shRNA. (C) Forced LINC01198 expression did not affect NEDD4-1 expression in glioma cells. (D) Protein expression levels of NEDD4-1 did not change in glioma cells treated with LINC01198 shRNA. (E) Forced LINC01198 expression did not inhibit PTEN expression in CRISPR/cas9-induced NEDD4-1-knockout glioma cells. (F) Protein expression levels of PTEN were not increased in CRISPR/cas9-induced NEDD4-1-knockout glioma cells treated with LINC01198 shRNA. (G and H). A negative correlation between LINC01198 and PTEN was observed in tumor tissues at the protein level (r = 0.501; P = 0.013), but this correlation was not observed at the mRNA level (r = 0.296; P > 0.05).

NEDD4-1 turnover LINC01198-induced glioma progression

To further determine the biological roles of LINC01198 in glioma, CRISPR/cas9-induced NEDD4-1-knockout glioma cells were used in an experiment. A CCK-8 assay showed that forced LINC01198 expression neither increased cell proliferation nor promoted glioma cell resistance to temozolomide in NEDD4-1-knockout glioma cells (Figure 5A and 5B). Additionally, reduced LINC01198 expression neither inhibited cell proliferation nor increased glioma cell resistance to temozolomide in NEDD4-1 knockout glioma cells (Figure 5C and 5D). To further understand the mechanism of LINC01198 in glioma, we recovered PTEN expression in LINC01198-overexpressing cells (Figure 5E). The results showed that recovered PTEN expression reversed LINC01198-induced tumor growth (Figure 5F).

Figure 5. NEDD4-1 reversed LINC01198-induced glioma proliferation and resistance to temozolomide. (A) Forced LINC01198 expression did not increase cell proliferation in NEDD4-1-knockout glioma cells. (B) Forced LINC01198 expression did not promote resistance to temozolomide in NEDD4-1-knockout glioma cells. (C) Reduced LINC01198 expression did not inhibit cell proliferation in NEDD4-1-knockout glioma cells. (D) Reduced LINC01198 expression did not inhibit resistance to temozolomide in NEDD4-1-knockout glioma cells. (E) PTEN expression in LINC01198-overexpressing glioma cells was modified by cDNA transfection. (F) recovered PTEN expression reversed LINC01198-induced tumor growth in glioma. Data are presented as the mean±S.D, n=3.

LINC01198 enhances AKT activity through regulation of the NEDD4-1/PTEN axis

To explore whether LINC01198 influenced AKT signaling pathway activity, we analyzed the p-AKT expression in glioma cells. As shown in Figure 6A, LINC01198 overexpression increased p-AKT expression. Conversely, knockout of LINC01198 decreased the expression of p-AKT (Figure 6B). In particular, LINC01198 knockdown inhibited the interplay between NEDD4-1 and PTEN (Figure 6C; Supplementary Figure 3). Finally, the LINC01198-induced higher AKT activity was abrogated when knockout of NEDD4-1 was performed (Figure 6D). Collectively, these observations suggest that LINC01198 enhances AKT activity through NEDD4-1-induced PTEN ubiquitin degradation in glioma cells.

Figure 6. LINC01198 enhances AKT activity through regulation of the NEDD4-1/PTEN axis in glioma cells. (A) Forced LINC01198 expression upregulated p-AKT expression in glioma cells. (B) Protein expression levels of p-AKT were reduced in glioma cells treated with LINC01198 shRNA. (C) LINC01198 knockdown inhibited the interplay between NEDD4-1 and PTEN in glioma cells. (D) Forced LINC01198 expression did not upregulate p-AKT expression in CRISPR/cas9-induced NEDD4-1-knockout glioma cells.

LINC01198 expression is correlated with poor prognosis in glioma patients treated with temozolomide

We then analyzed retrospective data from 40 recurrent glioma patients after primary tumor resection receiving temozolomide therapy; the clinicopathological characteristics and OS for these patients were recorded in Table 4. Their LINC01198 expression levels were then detected (Figure 7A), and Kaplan–Meier survival analysis showed that the OS for the LINC01198^high group was significantly lower than LINC01198^low group (Figure 7B). The median OS was 4.7 months in the LINC01198^high group, and it was 9.6 months in the LINC01198^high group; therefore, we concluded that high levels of LINC01198 led to the temozolomide resistance in glioma patients (Figure 8).

Figure 7. LINC01198 induces temozolomide resistance in glioma patients. (A) Recurrent glioma patients were divided into the LINC01198^High group and LINC01198^Low group according to the result of qRT-PCR. (B) Comparison of overall survival curves for temozolomide-treated patients with high LINC01198 expression and those with low LINC01198 expression.

Figure 8. Working model: LINC01198 overexpression promotes the proliferation and temozolomide resistance of glioma cells.

Cells and tissue samples

The human glioma cell lines U87, U251, SHG-44, SNB-19, LN229 and T98G were purchased from ATCC (LGC Standards S.r.l.). Cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco) supplemented with 10% foetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin (Gibco) in 5% CO₂ at 37 °C.

Glioma tissues and adjacent tissues were collected from patients with glioma at Renji Hospital from Jan 2008 to Jan 2012. Forty-eight of the tumor samples were pathologically graded as low-grade tumors (stage I and stage II), and 42 were graded as high-grade tumors (stage III and stage IV) according to the WHO criteria. All tissues were preserved in liquid nitrogen. In this study, patient consent was received for all investigations and experiments, which have been approved by the ethics committee for clinical research of Renji Hospital.

Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted using TRIzol Regentin in accordance with the manufacturer’s instructions (Invitrogen, Carlsbad, CA, USA). Reverse transcription was carried out using the extracted RNA and the M-MLV reverse transcriptase in accordance with the manufacturer’s instructions (Invitrogen, USA). Quantitative real-time polymerase chain reaction (qRT-PCR) analyses were performed using SYBR Premix Ex Taq II (Takara) on an ABI StepOnePlus real-time PCR system in accordance with the manufacturers’ instructions (Applied Biosystems, Foster City, CA, USA). The quantification of the expression of RNA was normalized to the expression of GAPDH. The 2^-∆∆CT method was used to evaluate the relative expression. The primers in this study used are listed in Supplementary Table 1.

RNA immunoprecipitation

The EZMagna RNA immunoprecipitation (RIP) kit (Millipore) was used according to the manufacturer's protocol. U87 cells were lysed in complete RIP lysis buffer, and the cell extract was incubated overnight at 4°C with magnetic beads conjugated with specific antibodies or control IgG. The beads were washed in PBS three times and incubated with proteinase K to remove proteins. Finally, purified RNA was used to perform the qRT-PCR analysis.

RNA pulldown assay

The RNA pulldown was performed as previously described [32]. RNAs (LINC01198, LINC01198D1, LINC01198D2) were transcribed using T7 RNA polymerase in vitro, purified using the RNeasy Plus mini kit, and treated with RNase-free DNase I. Transcribed RNAs were biotin-labeled with the Biotin RNA labeling mix (Ambio Life). Positive, negative, and biotinylated RNAs were mixed and incubated with U87MG cell lysates. Magnetic beads were added to each binding reaction, and this was followed by incubation of the samples at room temperature for 2 hours. Then, the beads were washed with washing buffer, and the eluted proteins were examined by western blotting.

Western blotting

Western blotting was performed according to standard protocols as described in reference [33]. The antibodies used in this study has added in Supplementary Table 2.

Proliferation and drug sensitivity analyses

For the proliferation analysis, the effect of the test agents on cell viability was assessed with the CCK-8 assay as described in reference [34]. Three independent experiments were performed. For the drug sensitivity analysis, the cells were seeded into 96-well plates at an initial density of 2 × 10³ cells per well. After 24 hours of incubation, the cells were incubated in fresh culture medium containing different concentrations of temozolomide for 72 hours (4μM, 20μM, 100μM, 500μM, 2,500μM). The CCK-8 assay was performed to analyze cell viability.

Plasmid construction and cell transfection

The full-length complementary cDNA strands of human LINC01198, LINC01198-D1, and LINC01198-D2 were synthesized by Invitrogen and cloned into the expression vector pCDNA3.1. The small hairpin RNA (shRNA) of LINC01198 was provided by Genepharm. Co. Plasmid vectors for transfection were prepared using DNA Midiprep kits and transfected into glioma cells using Lipofectamine 2000 according to the manufacturer's instructions.

Creation of CRISPR/Cas9-mediated knockout cell lines

Knockout of NEDD4-1 was performed by CRISPR/Cas9 technology, using a previously published protocol as reference [35].

Subcutaneous xenograft tumor models

Glioma cells (2 × 10⁵) were subcutaneously injected into 4-6 weeks old nude mice. When the tumor volumes reached an average of approximately 100 mm³, mice were randomly assigned to treatment groups and vehicle solution group. For treatment groups, temozolomide 25 mg/kg i.p. every 2 days for 4 weeks. The tumor volumes were measured twice every week in two dimensions with Vernier calipers. The tumor volumes were calculated using the following formula: length × width² × 0.5. The mice are considered dead when the tumor volumes reached 2, 000 mm³. Mice were housed according to protocols approved by the Medical Experimental Animal Care Commission of Renji hospital.

Statistical analysis

All analyses were performed using SPSS software (version 16.0). For continuous variables, the results were shown as mean ± SD. Student’s t-test was used to compare the difference between two groups. Differentially expressed LINC01198 between glioma and normal tissues was also evaluated by Student’s t-test. Kaplan-Meier curve and log-rank test were used to evaluate the effect of LINC01198 on survival of glioma patients. A two-sided P value < 0.05 was considered as statistically significant.