Imipramine impedes glioma progression by inhibiting YAP as a Hippo pathway independent manner and synergizes with temozolomide

Abstract Patients with malignant glioma often suffered from depression, which leads to an increased risk of detrimental outcomes. Imipramine, an FDA‐approved tricyclic antidepressant, has been commonly used to relieve depressive symptoms in the clinic. Recently, imipramine has been reported to participate in the suppression of tumour progression in several human cancers, including prostate cancer, colon cancer and lymphomas. However, the effect of imipramine on malignant glioma is largely unclear. Here, we show that imipramine significantly retarded proliferation of immortalized and primary glioma cells. Mechanistically, imipramine suppressed tumour proliferation by inhibiting yes‐associated protein (YAP), a recognized oncogene in glioma, independent of Hippo pathway. In addition to inhibiting YAP transcription, imipramine also promoted the subcellular translocation of YAP from nucleus into cytoplasm. Consistently, imipramine administration significantly reduced orthotopic tumour progression and prolonged survival of tumour‐bearing mice. Moreover, exogenous overexpression of YAP partially restored the inhibitory effect of imipramine on glioma progression. Most importantly, compared with imipramine or temozolomide (TMZ) monotherapy, combination therapy with imipramine and TMZ exhibited enhanced inhibitory effect on glioma growth both in vitro and in vivo, suggesting the synergism of both agents. In conclusion, we found that tricyclic antidepressant imipramine impedes glioma progression by inhibiting YAP. In addition, combination therapy with imipramine and TMZ may potentially serve as promising anti‐glioma regimens, thus predicting a broad prospect of clinical application.


| INTRODUC TI ON
Glioblastoma (GBM) is the most common and fatal primary brain tumour in adults, with a median survival of approximately 15 months. 1,2 The clinical first-line medication for GBM is temozolomide (TMZ), which have shown limited benefits mainly due to drug insensitivity or acquired resistance. 3 Therefore, it is urgent to explore new chemotherapeutic agents or repurpose the old clinical used drugs to treat GBM.
Nowadays, psychological depression is usually found to correlate with a poorer clinical outcome in cancer patients. Thus, antidepressants are widely used in tumour patients suffered with moderate to severe depression. 4,5 Emerging evidence indicates that certain types of antidepressants have anti-tumour properties in solid tumours apart from their intrinsic antidepressant effects. [6][7][8] Imipramine, a tricyclic antidepressant (TCA) that functions by inhibiting serotonin and norepinephrine reuptake in the central nervous system (CNS), 9,10 is currently considered to possess anti-tumour ability in various non-CNS tumours, such as small cell lung cancer, prostate cancer, colorectal cancer and lymphomas. [11][12][13] However, there are few studies to explore the influence of imipramine on glioma and the results provided by previous studies were inconsistent and controversial. For example, Levkovitz et al. reported that imipramine did not induce apoptosis, while Jeon et al. found that imipramine enhanced autophagic and apoptotic activities of glioma cells. 10,14,15 In addition, the specific role of imipramine, especially the mechanism that affects indefinite proliferation and invasion ability of glioma cells, has not been fully elucidated.
Yes-associated protein (YAP), the core effector of Hippo kinase cascade, usually plays a significant role in promoting cancer in most human tumours. 16,17 When Hippo kinase cascade is on, the upstream kinase LATS phosphorylates YAP, making it sequester in the cytoplasm and unable to enter the nucleus to initiate the transcription of downstream target genes. On the contrary, when Hippo kinase cascade is off, YAP enters the nucleus and interacts with the TEA domain (TEAD) family transcription factors, leading to expression of the downstream target genes and tumour growth. 18,19 Previous investigations performed by others and our group have found that YAP is closely related to the malignant progression of gliomas, and thus, targeting YAP may be helpful for the molecular therapy of glioma. [19][20][21][22] Notably, recent studies have shown that nortriptyline, another classical TCA, could increase the sphingolipid ceramide through inhibition of acid ceramidase (the enzyme responsible for ceramide metabolism), thus inhibiting YAP signalling in hepatic stellate cells. 23, 24 We therefore wonder whether imipramine will affect the activity of YAP.
In the present study, we investigated the effects of imipramine on glioma cell proliferation and glioma growth by conducting a serial in vitro and in vivo experiments. In addition, we elucidated the underlying molecular mechanism responsible for proliferation inhibition and evaluated the combinational effect of imipramine and TMZ. Our findings highlighted the importance of YAP as a potential molecular target and uncovered imipramine as a promising drug candidate for glioma therapy. Furthermore, imipramine may be a potential TMZ sensitizer and glioma patients may benefit from imipramine and TMZ combination therapy.
Primary glioma cell lines (GBM) were established by our laboratory as previously described. 25 All glioma cells were cultured in DMEM supplemented with 10% foetal bovine serum (FBS). Normal human astrocyte was cultured in astrocyte medium (ScienCell; Cat No.1801) supplemented with penicillin/streptomycin, 2% FBS and astrocyte growth supplement. All cells were maintained in humidified incubator with 5% CO2 at 37°C.

| Establishment of YAP-overexpression stable cells
Stable YAP-overexpressing U251 and GBM cells were generated by a lentiviral-based approach, which have been described previously. 20

| Cell counting kit-8 assay
Cell viability and half-maximal inhibitory concentration (IC50) values were determined using cell counting kit-8 (CCK8) assay according to the manufacturer's protocol as previously described. 20

| Colony formation assay
The proliferation ability of cells was detected using six-well plates described previously. 20 Briefly, a total of 1000 cells were seeded into each well followed by imipramine treatment for 48 h and cultured at 37℃ for 2 weeks to form colonies. Cells were then gently washed with PBS, fixed with 4% paraformaldehyde and stained with 0.03% crystal violet. The plates were dried at room temperature, and the number of colonies was counted.

| EdU incorporation assay
Following the manufacturer's instructions, we used the Cell-Light EdU Cell Proliferation Detection Kit (Ruibo Biotech) for EdU incorporation assay as previously reported. 20

| Transwell invasion assay
Cell invasion assay was conducted using a transwell system that incorporated a polycarbonate filter membrane with a diameter of 6.5 mm and pore size of 8 μm (Corning, NY), according to the manufacturer's instruction and previous study. 27

| Subcellular fractionation and Western blotting
Cellular fractionation was performed using a Membrane and Cytosol Protein Extraction Kit (Beyotime) following manufacturer's instructions. Protein concentration was detected using a BCA protein assay kit (Beyotime). Thereafter, equal amounts of protein were separated by 10% sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to 0.45 μm pore size PVDF membrane (Roche). After blocking for 2 h with 3% BSA, membranes were incubated with primary antibodies at 4°C overnight, then with secondary antibodies at room temperature for 2 h. Finally, the immunoreactive proteins were visualized using enhanced chemiluminescence and protein bands were measured with ImageJ software.

| RNA extraction and qRT-PCR
Total RNA was extracted from cultured cell lines with Trizol (Invitrogen), followed by synthesis of first-strand cDNA using a reverse transcription kit (Tiangen). Quantitative RT-PCR was performed with SuperReal PreMix Plus (Tiangen) according to the manufacturer's instructions. The relative mRNA expression levels of target genes were normalized to the GAPDH internal control and calculated with the 2 − ΔΔC t method. All primer sequences were synthesized by Sangon Biotech Co. and listed in Table 1.

| Immunofluorescence
The subcellular localization of YAP in tumour cells was detected using immunofluorescence assay. All steps were performed as described previously. 20

| Patient-derived xenograft and intracranial tumour mouse models
All the in vivo experiments performed in this study were approved by the Institutional Ethics Committee of Xuzhou Medical University.
For the YAP restoring experiment, luciferase-GFP-YAP or luciferase-GFP-vector GBM cells (5 × 10 5 ) were injected into BALB/c male nude mice (4 weeks, 20 g) intracranially following with imipramine (20 mg/kg) or vehicle 5 days on, 2 days off for 3 weeks. For the synergism experiment, BALB/c nude mice or C57BL/6 mice were intracranially injected with luciferase-GFP-GBM cells (5 × 10 5 ) or luciferase-mCherry-GL261 cells (2 × 10 5 ), respectively. After transplantation, the mice were administered with imipramine (20 mg/kg) or TMZ (7.5 mg/kg) intraperitoneally alone or together with imipramine (20 mg/kg) 5 days on, 2 days off for 3 weeks. Bioluminescence imaging was used to detect intracranial tumour growth on day 7, day 14 and day 21. After the tumour-bearing mice exhibited hemiplegia, listlessness, cachexia and other neurological symptoms, they were anesthetized with 5% isoflurane for 90 s and euthanized by cervical dislocation. Then, the main organs (including heart, liver, spleen, lung, kidney and glioma-bearing brain) were removed for subsequent experiments. Kaplan-Meier curves were calculated to estimate the overall survival.  Encyclopedia of Genes and Genomes (KEGG) pathway were applied for the analysis of DEGs using the Dr. Tom online software (BGI). Qvalue of the pathway shown in the figure was <0.05.

| Statistical analysis
All experiments were repeated at least three times and presented as the mean ± SEM. Significant differences within groups were analysed using Student's t test for single comparisons and one-way ANOVA followed by Tukey's test for multiple comparisons using GraphPad Prism 6 software. Overall survival curves were calculated using the Kaplan-Meier method and compared using the logrank test. p Values <.05 were considered be statistically significant (*p < .05, **p < .01, ***p < .001).

| Imipramine inhibits the proliferation and invasion of glioma cells
To evaluate the effect of imipramine on glioma cell viability, we Since invasiveness has always been considered as a vital biological characteristic of malignant glioma cells, we also investigated the effect of imipramine on glioma cells invasion. The results indicated that the invasive ability of U251 and GBM cells was significantly inhibited by imipramine ( Figure 1G). As shown in Figure 1H, cell invasion was, respectively, inhibited to 30.32% and 32.67% after 20 μM imipramine treatment in both cells. Taken together, these results indicate that imipramine significantly suppresses proliferation and invasion ability of glioma cells.

| Imipramine inhibits YAP activity as a Hippo pathway independent manner
To investigate the molecular mechanism by which imipramine inhib-

| Combining imipramine with TMZ attenuated glioma cell proliferation in vitro and the growth of glioma in vivo
It is well documented that chemotherapy is a critical process in the postsurgical treatment of glioma. 34 Since imipramine exhibits the significant inhibitory effect of glioma cell proliferation in this study and processes the ability to penetrate the blood-brain barrier, 35 we wonder whether it could synergize with TMZ, the first-line drug in treating malignant glioma. Firstly, U251 and GBM cells were treated respectively with TMZ for 48 h and IC50 was determined by CCK8 assay. As shown in Figure 4A, In addition, the combination of both agents exhibited a synergistic effect due to the combination index (CI) calculated by Compusyn. 36 As illustrated in Tables 2 and 3   TMZ. The mode of administration and dosing schedule is presented in Figure 5A. We found that either TMZ or imipramine alone significantly delayed tumour growth, while the combined effect of TMZ and imipramine exhibited the highest of tumour growth inhibition ( Figure 5B,C).

TA B L E 3 Combination index (CI) of IMIP and TMZ in GBM cells
Importantly, imipramine in combination with TMZ significantly extended the median overall survival of tumour-bearing nude mice compared to either agent alone, consistent with our in vitro findings that imipramine enhances the sensitivity of glioma cells to TMZ ( Figure 5D).
To rule out the possibility that these observations were due to a specific intracranial tumour mouse model, we conducted this in vivo experiment of combination therapy in a murine-derived tumour model. The results analysis of both bioluminescence imaging ( Figure 5E,F) and haematoxylin and eosin (H&E) staining ( Figure 5G) showed that the inhibition in glioma growth was considerably more pronounced in imipramine and TMZ combination treatment. Similarly, the combination therapy greatly extended the median survival time of the tumour-bearing C57BL/6 mice ( Figure 5H). These results indicate that imipramine may act as a potent sensitizer for TMZ chemotherapy in vivo.
Additionally, we evaluated the histology of major organs including brain, heart, liver, spleen, lung and kidney excised from mice to investigate systemic toxicity ( Figure 5I). The results of HE staining indicated that no histopathological changes were observed compared among the groups, suggesting that imipramine and TMZ effectively treats intracranial gliomas without conferring any apparent toxicity to normal tissues.

| DISCUSS ION
Glioblastoma, the most malignant primary brain tumours, is characterized by the extensive proliferative ability and insensitive to chemotherapeutics, which results in poor clinical outcomes and short survival time. 39,40 Hence, searching novel and effective drugs for the treatment of GBM is extremely urgent. In this study, we found that imipramine significantly inhibited the proliferation and invasion of glioma cells, as well as the intracranial PDX. Mechanistically, imipramine impedes glioma growth by inhibiting YAP expression and translocation from cytoplasm into nucleus, without affecting Hippo signalling pathway as a prerequisite. More importantly, imipramine is found to be a potent sensitizer for TMZ chemotherapy as combination therapy with TMZ, which exhibits strongly tumour suppressing ability both in vitro and in vivo ( Figure 6).
More than 90% of GBM patients suffer from depressive disorders 41 and increasing epidemiological studies suggest that antidepressants reduce the cancer risk in glioma patients. 42 Imipramine, the first identified TCAs, is involved in reuptake of serotonin and norepinephrine, importance of which is increasingly recognized. 42 Studies have shown that imipramine exhibits inhibition role in several types of tumours although the specific mechanism remained unclear. 11 15 To the best of our knowledge, our present study is the first to explore molecular mechanisms underlying proliferation regulation of imipramine on glioma cells using systematic screening methods. Based on RNA-seq technology, we identified YAP, a classical oncogene closed to development and progression of tumours, as the direct effector involved in regulation of imipramine on glioma progression.
As the key downstream transcription co-activator of Hippo signalling pathway, YAP usually plays a role on promoting tumour progression. In the present study, we found that imipramine inhibits YAP expression and translocation from cytoplasm into nucleus and thus retards glioma progression. Interestingly, we discovered that, F I G U R E 6 Schematic model of how imipramine inhibits glioma cell proliferation. The results showed that imipramine impedes tumour progression through diminishing YAP activity and sensitives glioma cells to temozolomide MGMT as shown in Figure 2F,I,J, neither mRNA nor protein levels of MST/ LATS were altered, indicating that regulation of YAP levels by imipramine was largely independent of MST/LATS kinase activity. Given that imipramine functions by inhibiting serotonin and norepinephrine reuptake, we hypothesized that the mechanism of imipramine on YAP inhibition may possibly be associated with regulation of serotonin and norepinephrine. Dethlefsen and colleagues have comprehensively shown that norepinephrine leads sequestration of YAP to the cytosol and suppression of downstream genes in breast cancer cells. 45 In addition, Fang et al. found that serotonin-pERK-YAP axis mediates liver regeneration and speculated serotonin as an up-regulator of YAP. 46 A recent study has demonstrated that TCAs induce hepatic stellate cell inactivation through increasing the sphingolipid ceramide, which was regard as a potent inhibitor of YAP signaling. 23 We proposed that important neurotransmitters and enzymes may be involved in the biological modulation process of imipramine on YAP in glioma directly or indirectly. However, the accurate regulation mechanism, especially the regulation of YAP mRNA level by imipramine, remains largely unknown and needs to be further studied.
Being a key therapeutic agent widely utilized to treat GBM patients after surgery, TMZ results in a modest increase in overall survival of patients. It has been confirmed that drugs which prompts increased sensitivity to temozolomide could further improve the prognosis of glioma patients. 47 In summary, our results demonstrate that imipramine dampened glioma progression by inhibiting YAP activity independent of activating Hippo pathway. Furthermore, imipramine may serve as a potential TMZ sensitizer and glioma patients with high expression of YAP and obvious depression may benefit from combination therapy.