Preclinical investigation of ibrutinib, a Bruton's kinase tyrosine (Btk) inhibitor, in suppressing glioma tumorigenesis and stem cell phenotypes

Abstract

Standard interventions for glioma include surgery, radiation and chemotherapies but the prognosis for malignant cases such as glioblastoma multiforme remain grim. Even with targeted therapeutic agent, bevacitumab, malignant glioma often develops resistance and recurrence. Thus, developing alternative interventions (therapeutic targets, biomarkers) is urgently required. Bruton's tyrosine kinase (Btk) has been long implicated in B cell malignancies but surprisingly it has recently been shown to also play a tumorigenic role in solid tumors such as ovarian and prostate cancer. Bioinformatics data indicates that Btk is significantly higher in clinical glioma samples as compared to normal brain cells and Btk expression level is associated with stage progression. This prompts us to investigate the potential role of Btk as a therapeutic target for glioma. Here, we demonstrate Btk expression is associated with GBM tumorigenesis. Down-regulation of Btk in GBM cell lines showed a significantly reduced abilities in colony formation, migration and GBM sphere-forming potential. Mechanistically, Btk-silenced cells showed a concomitant reduction in the expression of CD133 and Akt/mTOR signaling. In parallel, Ibrutinib (a Btk inhibitor) treatment led to a similar anti-tumorigenic response. Using xenograft mouse model, tumorigenesis was significantly reduced in Btk-silenced or ibrutinib-treated mice as compared to control counterparts. Finally, our glioma tissue microarray analysis indicated a higher Btk staining in the malignant tumors than less malignant and normal brain tissues. Collectively, Btk may represent a novel therapeutic target for glioma and ibrunitib may be used as an adjuvant treatment for malignant GBM.

Keywords: Bruton’s tyrosine kinase; cancer stem cells; glioma; ibrutinib.

Figure 1. An elevated Btk expression is found in clinical GBM samples

A. Btk expression is elevated in advanced GBM clinical samples. (A) A GEO expression database used in this analysis contained different sample groups from primary tumor (subdivided into 3 categories, oligodendrioglioma, astrocytoma), secondary GBM and normal brain/astrocytes. Statistical analysis of Btk expression between tumor versus normal/astrocytes groups revealed that GBM and secondary GBM samples contain a significantly higher level of Btk mRNA as compared to normal brain and astrocyte samples (Student T-test, *P<0.05). GEO dataset used GDS4467. B. Yamanaka Brain dataset (GSE4381) analysis demonstrates the elevated Btk expression in different brain tumor samples. 1. Anaplastic Astrocytoma (N=4); 2. Anaplastic Oligoastrocytoma (N=1); 3. Anaplastic Oligodendroglioma (N=2); 4. Glioblastoma (N=22). In the 22 GBM patients, the higher the Btk level appears to be associated with death. The lower panel represents another dataset (Sun brain dataset, GSE4290) analysis also suggests an association between Btk mRNA level and GBM progression. 0. No grade (N=23); 1. Grade 2 (N=45); 2. Grade 3 (N=31); 3. Grade 4 (N=81). C. Btk protein expression was also found elevated in clinical tissue microarray, as indicated by proteinatlas database. 2 out of 9 samples showed high intensity while 9 out of 12 samples exhibited moderate staining. The staining of Btk was mainly identified in the cytoplasm and membrane. Representative and magnified micrograph of a low grade D. and high grade E. GBM patient sample respectively, demonstrating elevated Btk expression. Antibody used HPA002028; http://www.proteinatlas.org/ENSG00000010671-BTK/cancer/tissue/glioma.

Figure 2. Increased Btk, stemness marker expression and temozolomide resistance in GBM spheres

A. Identification of CD133+ population of GBM cells using flow cytometric analysis. CD133+ cell population was identified in both U87MG and DTBRG-05MG cell lines (approximately 4.4% and 9.2% respectively). B. CD133+ GBM cells when cultured under serum-deprived conditions showed an increased propensity for GBM sphere formation as compared to their CD133- parental counterparts. C. Comparative Western blots analysis between GBM parental cells and GBM spheres. In both cell lines, Btk and oncogenic Akt/mTOR signaling expression were both found higher in the GBM spheres than their parental counterparts. D. GBM GBM spheres exhibited enhanced temozolomide resistance. The IC50 value was found to be approximately 300 and 500 μM for U87MG and DBTRG-05 cells respectively. U87MG and DBTRG-05 GBM spheres remained approximately 85.7% and 67.8% viable at the IC₅₀ concentrations of their respective parental counterparts.

Figure 3. Btk-silencing resulted in decreased malignant GBM phenotypes and stemness

A. Btk knocked down U87MG and DBTRG-05MG cells exhibited a significantly decreased colony-forming ability. B. Btk-silenced U87MG and DBTRG-05MG cells demonstrated a decreased migratory ability as compared to their counterparts. C. Down-regulation of Btk resulted in a significantly reduced sphere-forming ability in both cell lines. D. Flow cytometric analysis showed that Btk-silencing was associated with reduced the percentage of CD133⁺ U87MG and DBTRG-05MG (approximately 4.4% and 0.9% reduction respectively). E. Comparative Western blots between parental and Btk-knocked down cells. Btk-downregulation was associated with reduced expression level of stemness genes including c-Myc, Nestin, as well as the oncogenic Akt/mTOR signaling and the EMT marker, Vimentin.

Figure 4. Ibrutinib suppressed GBM tumorigenic ability resembling Btk-gene silenced GBM cells

A. The colony forming ability of both U87MG and DTBRG-05MG cells were significantly suppressed at low concentration of Ib (0.5 uM at IC10 for U87MG while 4 uM at IC₂₀ for DTBRG-05MG cells). B. Ibrutinib treatment significantly decreased the migratory ability of both GBM cell lines (at 0.25uM and 4uM for U87MG and DTBRG-05MG respectively, these concentrations represent the IC₂₀ values). C. Western blots of total protein lysates collected from both U87MG and DTBRG-05MG cells treated with ibrutinib (Ib), temozolomide (TMZ), and two drugs combined (Combo). The treatment conditions were as the follows: U87MG (Ib,10μM; TMZ300 μM); DBTRG-05MG (Ib, 30μM; TMZ 500 μM). The expression level of Btk, p-Akt and p-mTOR was significantly reduced by Ib treatment while TMZ only treatment moderately reduced Btk expression. The combination treatment led to the most potent suppressive effect in Btk, p-Akt, mTOR and Nestin expression. D. The GBM sphere-generating ability was significantly affected by different treatments. The most effective sphere-inhibitory effect was by the combined treatment followed by Ib alone and TMZ. (10x magnification, scale bar = 100μm). The lower panels depict the percentage survival of GBM spheres under different treatment conditions. E. Western blots analysis of total cell lysates collected from GBM spheres after different treatments. Similarly, Btk, p-Akt, p-mTOR, Nestin and Vimentin expression level was suppressed by Ib treatment, while TMZ slightly increased p-Akt expression. The combination of Ib and TMZ (combo) exerted the most inhibitory effect. Treatment conditions: U87MG (Ib, 10 μM; TMZ 300 μM), DBTRG-05MG (Ib,30 μM, TMZ 500 μM).

Figure 5. Btk-silencing led to decreased GBM tumorigenesis in vivo

A. Representative bioluminescence imaging of wild-type Btk U87MG and Btk-silenced U87MG orthotopic xenograft mouse models. Btk-knockdown (KD) group clearly showed a lower bioluminescent signal over time indicating the delayed tumorigenesis. B. Semi-quantitative analysis of tumor growth represented by the fold change in bioluminescence intensity (BLI) over time. By week four post tumor implantation, it was clear that the control (CTRL) mice exhibited a significantly higher fold change in BLI as compared to those Btk-KD counterparts. C. Tumor biopsies were collected from both groups and Btk-KD samples demonstrated a significantly lower ability to generate GBM spheres as compared to those of CTRL counterparts. D. Comparative flow cytometry analysis showed that Btk-KD sample contained a significantly lower percentage of CD133+ cells as compared to its wild-type counterpart.

Figure 6. Evaluation of ibrutinib and temozolomide treatments using patient-derived xenograft mouse model

A. Patient-derived GBM sample was injected subcutaneously into the right flank of NOD/SCID mice for evaluating the anti-GBM effect of ibrutinib (Ib) and temozolomide (TMZ). The fold change in tumor size was plotted against time. The tumor suppressive effect was found the most pronounced in the Ib+TMZ group followed by Ib alone, TMZ alone and vehicle control. The insert represents the photographs of tumor biopsies from different groups. B. The body weight of the mice was measured and tracked over time, demonstrating the treatments did not exert apparent toxicity in all groups. C. The photographic representation of tumor biopsies. D. Western blot analysis of the total lysates obtained from the tumor biopsies demonstrated the expression of Btk, members of mTOR signaling and stem marker, Nestin was suppressed by both Ib treatment alone and TMZ/Ib combination.