MALT1 is a potential therapeutic target in glioblastoma and plays a crucial role in EGFR‐induced NF‐κB activation

Abstract Glioblastoma multiforme (GBM) is the most common malignant tumour in the adult brain and hard to treat. Nuclear factor κB (NF‐κB) signalling has a crucial role in the tumorigenesis of GBM. EGFR signalling is an important driver of NF‐κB activation in GBM; however, the correlation between EGFR and the NF‐κB pathway remains unclear. In this study, we investigated the role of mucosa‐associated lymphoma antigen 1 (MALT1) in glioma progression and evaluated the anti‐tumour activity and effectiveness of MI‐2, a MALT1 inhibitor in a pre‐clinical GBM model. We identified a paracaspase MALT1 that is involved in EGFR‐induced NF‐kB activation in GBM. MALT1 deficiency or inhibition significantly affected the proliferation, survival, migration and invasion of GBM cells both in vitro and in vivo. Moreover, MALT1 inhibition caused G1 cell cycle arrest by regulating multiple cell cycle–associated proteins. Mechanistically, MALTI inhibition blocks the degradation of IκBα and prevents the nuclear accumulation of the NF‐κB p65 subunit in GBM cells. This study found that MALT1, a key signal transduction cascade, can mediate EGFR‐induced NF‐kB activation in GBM and may be potentially used as a novel therapeutic target for GBM.

unclear. In this study, we investigated the role of mucosa-associated lymphoma antigen 1 (MALT1) in glioma progression and evaluated the anti-tumour activity and effectiveness of MI-2, a MALT1 inhibitor in a pre-clinical GBM model. We identified a paracaspase MALT1 that is involved in EGFR-induced NF-kB activation in GBM.
MALT1 deficiency or inhibition significantly affected the proliferation, survival, migration and invasion of GBM cells both in vitro and in vivo. Moreover, MALT1 inhibition caused G1 cell cycle arrest by regulating multiple cell cycle-associated proteins.
Mechanistically, MALTI inhibition blocks the degradation of IκBα and prevents the nuclear accumulation of the NF-κB p65 subunit in GBM cells. This study found that MALT1, a key signal transduction cascade, can mediate EGFR-induced NF-kB activation in GBM and may be potentially used as a novel therapeutic target for GBM.

| INTRODUC TI ON
Glioblastoma multiforme (GBM) is the highest grade of malignant glioma and the most common and lethal type of brain tumour in adults. [1][2][3] Advances in neurosurgery, radiotherapy and chemotherapy have facilitated in the improvement of treatment schemes for GBM; however, the prognosis of patients with GBM remains dismal, with a median survival of 12 to 15 months after initial diagnosis. 4,5 Therefore, there is a need for a detailed investigation on the clinical pathogenesis of GBM to discover novel molecular targets for treatment and to improve prognosis.
Nuclear factor κB (NF-κB) activation commonly occurs in cancer, and substantial experimental evidence suggests that it is involved in both cancer development and resistance to treatment. [6][7][8] Previous studies have reported that NF-κB activation is a core driver for the malignant phenotype of GBM, which is correlated to a negative prognosis in patients with GBM. 9,10 Constitutive NF-κB activation promotes the expression of various proteins involved in the proliferation, survival, migration and epithelial-to-mesenchymal transition of GBM cells. 11 However, different types of cancers activate NF-κB using different mechanisms. Hence, uncovering signalling pathways that lead to activation of NF-κB in GBM will provide the therapeutic targets and thus benefit tumour treatment.
Epidermal growth factor receptor (EGFR) signalling activates NF-κB because EGFR gene amplification and mutation are common in GBM, and the aberrant EGFR signalling is likely to be a vital mechanism of NF-kB activation in GBM. [12][13][14] Although many reports confirm that EGFR signalling is an important driver of NF-κB activation in GBM, the underlying molecular mechanism remains largely unknown.
Elucidating the association of EGFR with the NF-κB pathway in GBM will thus provide novel drug targets for the inhibition of GBM progression. Therefore, the identification of a specific signalling component with therapeutic potential and high specificity in the tumour-promoting NF-κB pathway may help improve therapeutic efficacy.
Mucosa-associated lymphoma antigen 1 (MALT1) is a paracaspase that belongs to the caspase family of proteases and possesses arginine-specific cysteine protease activity. 15 MALT1 mediates by proteolytic cleavage that inactivates inhibitors of the NF-κB signalling pathway such as TNFAIP3/A20, and the proteins that favour NF-κB activity such as CYLD and RelB. 16 In addition, MALT1 serves as a scaffold to recruit additional signal transducers to activate NF-κB signalling. 17 Pan et al have reported that MALT1 is involved in EGFR-induced NF-κB activation and promotes EGFR-associated solid tumour progression. 18 Moreover, other researchers also found that miR-181d could attenuate the mesenchymal phenotype by directly repressing MALT1 in glioblastoma. 19 Thus, MALT1 may be considered as a potential target of solid tumour treatment. Recently, Fontan and co-authors discovered a novel molecule called MI-2 that could inhibit MALT1 by forming a covalent linkage in the active site. 20 Therefore, MI-2 can be used as a therapeutic agent for the treatment of tumours that are dependent on MALT1 signals.
In the present study, we identified MI-2 that effectively inhibited the activity of NF-κB in GBM by inhibitor screening. Then, we investigated the mechanism and therapeutic potential of MALT1 inhibition in the treatment of GBM. The results show that MALT1 is required for EGFR-induced NF-kB activation in GBM cells. Our study provides insights into the applicability of MALT1 as a potential therapeutic target for GBM and describes the basis for further clinical investigations of MI-2 in GBM.

| Cell lines and antibodies
Human GBM cell lines U251 and U87 were purchased from the Shanghai Cell Bank, Chinese Academy of Sciences for this study. These cell lines were cultured in DMEM supplemented with 10% FBS at 37°C in a humidified incubator with 5% CO 2 . Antibodies against MALT1, p-pRb, cyclin D1, CDK4, p27, p21, p65, IκB-α and lamin A/C were obtained from Cell Signaling Technology (CST, Beverly, MA, USA). Antibodies specific to β-actin were purchased from Abcam (Cambridge, MA, USA). DAPI was obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA).

| Construction and production of the lentivirus
The CRISPR/Cas9 system was used to knock out MALT1 in GBM cells. A non-targeting sgRNA was used as control. We used the two MALT1 sgRNAs targeting the human MALT1 gene to generate MALT1-knockout cells as previously described. 21 The sequences of the sgRNA used in this study are as follows: con-
Then, 10 μL CCK-8 solution was added to each well and cultured for 3 hours before measuring absorbance at a wavelength of 450 nm using a microplate reader.

| Cell cycle assay
To determine cell cycle distribution after MALT1 knockout or MI-2 treatment, cell cycle analysis was conducted as previously described. 22,23 Briefly, the cells were seeded into six-well plates at a density of 2 × 10 6 cells per well and treated with MI-2 (0-4 μmol/L) Lakes, NJ, USA).

| Wound-healing assay
Cells were obtained with a plastic pipette tip and washed with PBS to remove debris. Then, the cells were incubated and treated with serum-free media with or without MI-2. Three different time regimens (0, 24 and 48 hours) and five selected points of the lesions were assessed using a microscope (Olympus, Japan).

| Western blot analysis
The expression level of proteins was examined by Western blot analysis according to previous reports. 25,26 Total protein and cytoplasmic protein were extracted from the MALT1-knockout cells and MI-2-treated cells, and the concentrations of these proteins were measured using BCA Protein Assay Kit (Beyotime, China).
Electrophoresis (10% SDS-PAGE gel) was used to separate these proteins (50 μg/lane) in each sample and then transferred onto polyvinylidene difluoride (PVDF) membranes for further analysis.
The membrane was blocked with 5% non-fat milk and probed with specific primary antibodies at 4°C overnight, and secondary antibodies were added at room temperature for 2 hours. Then, these proteins were detected using an enhanced chemiluminescence detection system.

| In vivo experiments
MALT1-knockout U87 (5 × 10 5 cells per mouse) cells and control cells were stereotactically implanted into the right striata of nude mice as previously described. 27,28 The animals used were randomly assigned. Sixteen mice were assigned in each group. Five of the mice were used to evaluate tumour size, and another five were used to assess the survival analysis. To test the efficiency of MI-2 treatment, U87 cells (5 × 10 5 cells per mouse) were injected into the brain of nude mice. Seven days later, the treated mice (n = 42) were randomly divided into three different groups: MI-2 with 20 mg/kg (n = 14), MI-2 with 40 mg/kg (n = 14) and vehicle (n = 14). The drugs and vehicle were delivered daily via intraperitoneal injections. The mice were killed when cachexia occurred, and their brains were removed, fixed with 4% paraformaldehyde and dehydrated sequentially with 20% and 30% sucrose at 4°C until sinking. For Kaplan-Meier survival analysis, the mice (n = 9 per cohort) were killed when these exhibited neurological symptoms that included weight loss, mental apathy and haemiplegia.

| Histopathology and immunofluorescence staining
The flash-frozen brains were serially cut at 12 μm thicknesses, and brain section that contained the largest tumour area was stained with haematoxylin and eosin (H&E) as previously described. 28 Proliferative and apoptotic indices of the tumours were assessed by immunofluorescence staining of anti-Ki67 and anti-cleaved caspase-3. Briefly, the brain sections with the tumour were incubated in 0.1% Triton X-100 and blocked with 1% BSA in PBS for 1 hour.
Subsequently, primary and secondary antibodies were applied to the brain slices and incubated. Cellular DNA was stained with DAPI for 15 minutes. All brain slices were examined and photographed with a microscope with an attached fluorescence detector. We used three tumour tissue sections for Ki67 and cleaved caspase-3 quantitative analysis. Ki67-and cleaved caspase-3-positive cells were counted in five randomly selected areas. Signal intensity was measured with ImageJ software.

| Statistical analyses
The data obtained in this study were processed with GraphPad Prism 6 software. One-way ANOVA and Student's t test were applied to assess statistical differences among groups, and survival analysis was evaluated by Kaplan-Meier survival curve and the log-rank test. P-values were marked as *P < 0.05, **P < 0.01 and ***P < 0.001.

| MALT1 contributes to GBM cell proliferation and colony formation
Western blot analysis was used to test the efficiency of U87 and U251 cell lines with MALT1 knocked out. Figure 1A clearly illustrates that the MALT1-knockout group did not express MALT1 compared with the sgRNA-negative control group, which means the MALT1 was successfully knocked out by MALT1-sgRNA. The sequencing results are shown in Figure S1 of Supplemental materials. CCK-8, EdU and colony formation assays were performed to assess the effects of MALT1 on cell proliferation. Figure

| MI-2 inhibits the migration and invasion of GBM cells
Wound-healing assay was performed to assess the role of MALT1 in the migration of GBM cells. The migration of U87 and U251 cells into scratch wounds is shown in Figure 3A,C, which indicate that the number of cells that had migrated into the wound decreased via a concentration-dependent way. Figure 3B illustrates that the number of U87 cells that migrated, respectively, decreased to 71.

| Knocking out MALT1 or inhibition by MI-2 suppresses intracranial GBM growth in vivo
To examine whether MALT1 is important to GBM growth in vivo, MALT1-knockout U87 cells were transplanted into the right striatum of nude mice using a stereotactic technique. H&E staining ( Figure 4A) showed that tumours derived from MALT1-knockout U87 cells were significantly smaller than those of the control group.

| MALT1 is required for EGF-induced NF-κB activation in GBM cells
To test whether MALT1 is involved in EGF-induced NF-κB activation, we examined the effects of MALT1 deficiency or inhibition on  Figure 5A,B). In addition, knocking out MALT1 induced a significant decrease in nuclear p65 expression, whereas an increase in expression of cytoplasmic p65 was observed ( Figure 5C,D). To test this finding, we applied EGF to stimulate the U87 and U251 cells and found that IκB-α was degraded effectively. In addition, IκB-α degradation was partially   nalling. 34 Therefore, whether MALT1 is also involved in regulating non-canonical NF-κB signalling or other signalling pathways in glioma cells should be confirmed in future studies.

| D ISCUSS I ON
This is the first systematic study on the role of MALT1 in GBM progression. We found that MALT1 plays critical roles in cell proliferation in vitro and in vivo by inducing G1 cell cycle arrest. Most importantly, we further confirmed that EGFR-induced NF-κB activation can be significantly blocked by MALT1 inhibition and further regulate GBM cell proliferation and survival. These findings suggest that MALT1 is a potential theoretical target for GBMs and provide a hint for clinical trials to evaluate the therapeutic potential of MALT1 inhibitor in human GBMs.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

AUTH O R CO NTR I B UTI O N S
MN and RY designed this study. XL, CY and LS performed the main experimental procedures. GL, QC, QS, YW, XC and HL carried out partial experiments. JW and SG performed the statistical analysis.
XL and MN wrote this manuscript. All the authors read and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets supporting the conclusions of this article are included in the article.