Reprogramming glioblastoma multiforme cells into neurons by protein kinase inhibitors

Background: Reprogramming of cancers into normal-like tissues is an innovative strategy for cancer treatment. Recent reports demonstrate that defined factors can reprogram cancer cells into pluripotent stem cells. Glioblastoma multiforme (GBM) is the most common and aggressive malignant brain tumor in humans. Despite multimodal therapy, the outcome for patients with GBM is still poor. Therefore, developing novel therapeutic strategy is a critical requirement. Methods: We have developed a novel reprogramming method that uses a conceptually unique strategy for GBM treatment. We screened a kinase inhibitor library to find which candidate inhibitors under reprogramming condition can reprogram GBM cells into neurons. The induced neurons are identified whether functional and loss of tumorigenicity. Results: We have found that mTOR and ROCK kinase inhibitors are sufficient to reprogram GBM cells into neural-like cells and “ normal ” neurons. The induced neurons expressed neuron-specific proteins, generated action potentials and neurotransmitter receptor-mediated currents. Genome-wide transcriptional analysis showed that the induced neurons had a profile different from GBM cells and were similar to that of control neurons induced by established methods. In vitro and in vivo tumorigenesis assays showed that induced neurons lost their proliferation ability and tumorigenicity. Moreover, reprogramming treatment with ROCK-mTOR inhibitors prevented GBM local recurrence in mice. Conclusion: This study indicates that ROCK and mTOR inhibitors-based reprogramming treatment prevents GBM local recurrence. Currently ROCK-mTOR inhibitors are used as anti-tumor drugs in patients, so this reprogramming strategy has significant potential to move rapidly toward clinical trials


Background
The ability to convert somatic cells into disease-relevant cell types through cellular reprogramming has opened new doors for basic research and cell replacement therapy [1]. Takahashi et al. demonstrated that defined factors could drive skin-derived fibroblasts to induced pluripotent stem (iPS) cells that could be further differentiated into the desired cell type. Direct somatic lineage reprogramming has attracted much attention as it bypasses conversion into iPS cells. Recently studies have demonstrated that a set of neural factors can directly convert fibroblasts into neurons [2][3][4][5][6][7][8][9][10][11]. Direct fibroblast-neuron reprogramming provides an alternative, potentially complementary tool to many of the proposed applications of iPS technology for both disease modeling and development of cell-based therapies.
Reprogramming healthy somatic cells with defined factors has been extensively investigated. However, reprogramming cancer cells has comparatively lagged behind [12][13][14]. Differentiation of tumors into post-mitotic cells is an ancient idea. For example, tretinoin (all-trans-retinoic acid) was used as differentiation therapy of acute

R E T R A C T E D A R T I C L E
promyelocytic leukemia [15]. Recent established reprogramming method should significantly improve differentiation therapy.
Reprogramming and oncogenic transformation are stepwise processes that share many similarities. Overexpression of a set of oncogenes induces reprogramming and oncogenic transformation, while tumor suppressor gene such as p53 suppresses both processes. Although unidentified biological barriers may exist [14,16,17], reprogramming of both solid and liquid tumors to iPS cells has been reported by different groups [16,[18][19][20][21][22][23][24][25][26][27]. Loss of tumorigenicity by unknown mechanisms and induced dedifferentiation to pluriopotency seem to be common features of reprogrammed cells from different cancers. However, robust differentiation into specific lineages remains a stumbling block [28][29][30][31][32]. To overcome these roadblocks, therefore, we will detect whether cancer cells can be reprogrammed into normal-like cells under defined reprogramming conditions. Glioblastoma multiforme (GBM) is the most common and most aggressive malignant primary brain tumor in humans. Despite multimodal therapy [surgery with ionizing radiation and chemotherapy (temozolomide, TMZ)], the outcome for patients with GBM is still poor, with a median survival of only 14 months [33][34][35]. Therefore, it is imperative to develop novel therapeutic strategies for GBM treatment. Integration-free reprogramming of cancer cells is a safe and preferable strategy for clinical use. We screened a kinase inhibitor library and found that a combination of two inhibitors, Rho-associated protein kinase (ROCK) and mammalian target of rapamycin (mTOR), could substitute for transcription factors and convert human GBM cells into neural-like cells. The induced neural cells were subsequently trans-differentiated into functional neurons. The induced neurons lost their tumorigenicity and reverted to the 'normal state'. Importantly, ROCK-mTOR inhibitors coupled with reprogramming treatment prevented GBM local recurrence in mice, suggesting that our reprogramming treatment had a key role in preventing GBM recurrence.
Cortical neuron culture and co-culture with iNs Primary cortical neurons were isolated from P0 rats. Cortices were dissected and dissociated by trypsin digestion (0.25% Trypsin, 137 mM NaCl, 5 mM KCl, 7 mM Na 2 HPO 4 , 25 mM HEPES) and plated on poly-D-lysine coated glass coverslips. The neurons were maintained in growth medium consisting of MEM supplemented with B27, glutamine (all from Invitrogen), glucose, transferrin (Calbiochem), fetal bovine serum and Ara-C (both from Millipore-Sigma) for a week before co-culture with iNs. iNs were induced for 7 days by induction medium with P + Y and dissociated by trypsin (0.05% Trypsin). iNs were seeded onto a cortical neuron bed and maintained in neuronal mature medium.
Viral preparation, western blot and immunofluorescence ROCK1/2 and mTORC1 (Raptor) /C2 (Rictor) shRNAs were obtained from Millipore-Sigma. Western blotting analyses were performed to check the knockdown efficiency. Immunofluorescence staining was performed as follows: 5 × 10 4 modified human fibroblasts were planted on Matrigel-coated glass coverslips the day before induction. Cells were fixed for 20 min at room temperature in 4% paraformaldehyde in PBS, permeabilized for 30 min in PBS containing 0.2% Triton X-100 and 10% normal goat serum (NGS) and incubated overnight at 4°C in PBS containing 10% NGS and primary antibodies. Cells were washed three times with PBS and incubated for 2 h at room temperature with anti-rabbit or anti-mouse secondary antibodies conjugated to Alexa Fluor-488 or Alexa Fluor-594 (1:500, Invitrogen). Images were acquired on immunofluorescence microscope or Zeiss Trypan blue dye exclusion assays, qRT-PCR and TUNEL assays GBM, iP and iN cells were seeded at a density of 10,000 cells/well in a 12-well plate. Cells were counted with a ViCell cell viability analyzer (Beckman Coulter). Cell proliferation was normalized to the percentage of control cells. Three independent experiments were performed with triplicates for each cell line. For qRT-PCR assay, total RNA was extracted using Trizol isolation system (Invitrogen) according to the manufacturer's instructions. Total RNA was treated with DNase I (Qiagen) to prevent DNA contamination. 2 μg RNA was reverse transcribed using Transcriptor High Fidelity cDNA Synthesis Kit (Roche). Reverse transcription and real-time PCR were performed as a standard procedure. The terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labelling (TUNEL) assay was performed with TUNEL assay kit (Roche).

Electrophysiology
Coverslips plated with iNs were placed in the recording chamber mounted on an Olympus 1 × 81 microscope. All electrophysiology recordings were recorded at room temperature (22-25°C) via whole-cell recording with a patch electrode. Signals were amplified using Heka EPC-10 amplifier and filtered at 2 KHz via a Bessel low-pass filter. Data were sampled using Puls8.8 software, and analysed using Igor6.04.

Microarray analysis
Microarray analysis was performed at the Washington University Genome Center. Briefly, Illumina HumanHT-12 v4 Expression BeadChip was used and samples were labeled by biotin. The Direct Hybridization Assay was performed and the data were scanned on the BeadArray Reader. Scanned images were quantitated by Illumina Beadscan, v3. Quantitative data was imported into Illumina GenomeStudio software and normalized by Illumina's quantile method. The quantile-normalized background-subtracted data were calculated in excel. Data were filtered based on the average signal with baseline set as 50. Only the genes with division value > 3 were selected and considered as differentially expressed. All differentially expressed genes were clustered using MeV software.
Gene expression for each sample was analyzed to determine the relationship among GBM cells and iNs. Data were filtered based on average signal and baseline was set as 10 to increase the sensitivity. Differentially expressed genes were identified with a 3-fold change as the threshold. To examine the potential neuron differentiation pathways relevant to immortalization, all neuron differentiation genes from Gene Ontology Website (http://www.geneontology.org/, GO: 0030182) were compared with the microarray data. For gene enrichment analysis, we used web-based Gorilla program (http://cbl-gorilla.cs.technion.ac.il/). Thus, these results will greatly help for understanding mechanisms of GBM cell reprogramming.

GBM and iN cell transplantation and GBM recurrence treatment in mice
We implanted GBM cells and iNs in the corpus callosum of nude mice. On the day of transplantation, cells were harvested with accutase, counted, and re-suspended at concentration of 25,000 cells/μL. Four microliter cell suspension was injected into the corpus callosum of mice using the following stereotactic co-ordinates (in reference to Bregma point): 10.98 mm (anterio-posterior axis), 21.75 mm (latero-medial axis) and 22.25 mm (vertical axis). Control GBM cells were injected into left side and iNs were injected into right side of the same mouse. Mice with cell transplants were perfusion fixated at 2 weeks, 1 month and 4 months. Brains were excised and sectioned on a cryostat for IHC analysis of the cell implants. The primary antibodies used are described above. Subsequently, various secondary antibodies were used to visualize the specific primary immunoreaction product in single and double IHC staining. For GBM local recurrence experiments, we used a mouse model of local recurrence following surgical resection of orthotopic GBM tumors [37][38][39][40]. GBM tumors generated by GM97 cells in nude mice were surgically removed after 4 weeks when the primary tumors had reached~50 mm 3 . 95% of each tumor was excised and a small fragment of tumor, the largest being 2 mm 3 , was not removed to model postoperative residual GBM. Matrix-Driven Delivery (MDD) Pellet of ROCK-mTOR inhibitors with dried induction medium was prepared by Innovative Research of America. We implanted the MDD pellet including mTOR-ROCK inhibitors (2 μM P + Y, 5 μl per day for 21 days) with or without dried induction medium into the resection cavity of GBMs. MDD pellet with dried induction medium, pellet only and no treatment were controls. Induction media in each MDD pellet include 1 μM dexamethasone, 0.5 mM isobutylmethylxanthine and 200 μM indomethacin with 105 μl DMEM/F12 (2% FBS). Mouse brains were excised and sectioned on a cryostat for IHC and TUNEL assay analyses.

Screening a protein kinase inhibitor library to reprogram GBM cells
Reprogramming somatic cells are into iPS cells by transcription factors may cause genomic instability, which increases the risk of cancer-cell induction [41][42][43][44][45]. Therefore, we tried to develop a transgene-free method to efficiently reprogram GBM cells into normal-like cells. Cellular senescence has been shown to regulate reprogramming fibroblasts to iPS cells and fibroblast-neuron conversion [46][47][48][49]. Since many protein kinases are involved in senescence and proliferation processes, we screened a protein kinase inhibitor library (355 inhibitors, Calbiochem). We prepared a GBM cell line (U118) with expression of MAP2 promoter-RFP. Through phenotypic change screening, we found that candidate kinase inhibitors reprogrammed GBM cells into induced progenitor-like cells in induction medium (Fig. 1a, b). We observed that a subpopulation of cells was MAP2-RFP positive with marked morphological changes after 3-7 days in induction medium with candidate kinase inhibitor treatment. Cell morphology ranged from large nuclear and flat shaped cells (cancer cells) to small, bipolar cells, termed induced neural-like cells (Fig. 1a,b). Eight candidate small molecules/compounds were selected for further confirmation (Fig. 1c). Secondary screening confirmed that three compounds induced morphological changes with higher efficiency (Fig. 1c). Two candidate small molecules, namely rapamycin and palomid 529 (P529), are mTOR inhibitors and the third one (Y27632) is a ROCK inhibitor. To determine the combinatorial effects of these inhibitors on GBM-neural cell conversion, we found that using mTOR-ROCK inhibitors (P529/Y27632) together converted GBM cells into neural-like cells with the highest efficiency (~90% induced efficacy after 7 days induction) (Fig. 1c).
To investigate whether ROCK-mTOR inhibitors have a general effect on human GBM cells, another GBM cell line (GM97) and one patient-derived primary GBM cell (PDC1) were treated with ROCK-mTOR inhibitors. After induction for 3-7 days, about 70-90% of these GBM cells converted into neural-like cells (Additional file 1: Figure S1). Thus, reprogramming treatment of ROCK-mTOR inhibitors is able to convert multiple GBM cells into neural-like cells.

Reprogramming GBM cells into induced neurons
Next, we examined whether the GBM cells could be reprogrammed into another lineage (neuron, a lineage switch) in a defined neuronal medium. After 3 days in neuronal medium with mTOR-ROCK inhibitors, induced neural-like cells further differentiated into bi-or multi-polar neuron-like cells. In 2-3 weeks, 70-90% of GBM U118 cells converted to neurons with mature neuronal morphology that expressed the neuronal markers, Tuj1 and MAP2 (Fig. 2a & c). We calculated the iN yield as the percentage of MAP2+ cells in relation to the initial number of plated GBM cells. At 3 weeks after induction, we obtained yields of 90 ± 15% (Fig. 2a). In contrast, after 4 weeks in neuronal medium without kinase inhibitors, control GBM cells maintained GBM cell morphology with negative staining of MAP2 (Fig. 2b). Furthermore, we verify the neuron induction in two other GMB cells, GM97 and PDC1. Three weeks after induction,~75-90% of GBM cells converted into iNs (Additional file 1: Figures. S1-S2). Moreover, iNs did not express astrocyte (GFAP) and oligodendrocyte marker (O4) (Additional file 1: Figure S3). These results suggest that ROCK-mTOR inhibitor treatment led to conversion of GBM cells to neurons.
So far, we identified iNs based on their morphology and positive staining of several neuronal markers. Next, we further characterized the functional aspects of iNs induced by ROCK-mTOR inhibitor treatment from GBM U118 cells. We found that~80% of iNs showed Synapsin-positive puncta that label the vesicles in the presynaptic nerve terminal (Fig. 2c). Furthermore, iNs showed positive immunostainings with TBR1 (60%) or LHX6 (55%), indicating that iNs have properties of different neuronal populations present in the forebrain (Fig. 2c).
To determine whether iNs generate action potentials and neurotransmitter-induced currents, the basic functions of neurons, we cultured U118 cells (7 days post-induction) with a monolayer culture of primary rat neurons, which might provide a better environment for maturation [2,3]. Three weeks after co-culture, whole-cell voltage-clamp recordings of iNs showed that 70% of iNs exhibited inward fast activating and inactivating sodium currents and outward potassium currents during a ramp depolarization. This was also confirmed by applying a series of voltage steps to evoke inward sodium currents and outward potassium currents (Fig. 2d postsynaptic currents were recorded from induced neurons without co-culture, suggesting that iNs can form functional synapses by themselves (Fig. 2f, 65% positive efficacy, n = 30). These data indicate that iNs are functional neurons. We also tested the effects of reprogramming treatment on normal primary neurons and astrocytes. We prepared primary mouse neurons/astrocytes and found that the reprogramming treatment of mTOR-ROCK inhibitors did not affect neuron/astrocyte properties under drug concentration for reprogramming (Additional file 1: Figure S4). These data indicate that normal neurons/astrocytes are only marginally responsive to our treatment strategy.

Genome-wide transcriptional profiling of induced neurons
To analyze the similarities between iNs and parental GBM cells, we performed microarray-based comparative global gene expression profiling. Microarray data was quantile normalized and filtered based on the average signal, and differently expressed genes were selected for further analysis. Hierarchical cluster analysis revealed a significant difference between iN cells and their parental GBM U118 cells (Fig. 3a, b). The clustering analysis of global gene expression revealed that iNs from GBM cells by the reprogramming treatment of ROCK-mTOR inhibitors are particularly similar to control neurons, which are induced from fibroblasts by p53 knockdown (Fig. 3b) [49]. This indicates that iNs generated by different methods show similarity of gene expression profiling. Using the Ontology Data-base for enrichment analysis, we found that the key factors involved in neuron differentiation were significantly changed in iNs, compared with those in parental GBM cells. Moreover, we examined mRNA levels of key factors reported to be involved in neuron conversion. We found that, in iNs induced by the reprogramming treatment of ROCK-mTOR inhibitors, expressions of a set of neurogenic transcription factors were increased by > 10-fold (Fig. 3c), suggesting that these factors may be involved in GBM-neuron conversion. These findings indicate that genetic trans-differentiation erases majority of the expression hallmarks of the cell of origin, while specifically inducing the neuronal phenotype.

ROCK1/2 and mTORC1 shRNAs generate iNs
Both ROCK and mTOR are serine/threonine protein kinases. ROCK is a kinase belonging to the AGC (PKA/ PKG/PKC) family, including ROCK1 and ROCK2 [50][51][52][53]. mTOR interacts with other proteins and serves as a core component of two protein complexes, mTOR complex 1 and mTOR complex 2 [54][55][56][57][58][59][60]. To determine if ROCK and mTOR pathways are involved in GBM-neuron conversion, we examined the effects of ROCK and mTOR knockdown on GBM-neuron conversion (Fig. 4a). Lentiviral constructs expressing short hairpin RNA (shRNA) against ROCK1/2 and mTORC1 (Raptor)/C2 (Rictor) were transfected into GBM U118 cells. We found that knockdown of ROCK1, ROCK2 or mTORC1 (Raptor) could reprogram GBM cells into neural-like cells, while knockdown of mTORC2 did not (Fig. 4b). Combination treatment of ROCK1/2 and mTORC1 shRNAs converted GBM cells to neural-like cells with kinetics and conversion efficiency similar to ROCK-mTOR inhibitors. Moreover, the iNs showed typical neuronal morphology and positive immunostaining for neuronal markers TUJ1, MAP2 and Synapsin. Efficiency of neuronal conversion by shRNA treatment was similar to ROCK-mTOR inhibitors (Fig. 4c & d). Thus, we conclude that ROCK1/2 and mTORC1 kinases are involved in conversion of GBM to neurons.

Loss of proliferation and tumorigenicity in iNs
We performed proliferation assays to determine iN cell proliferation. Trypan blue exclusion assays were performed after treating cells with ROCK-mTOR inhibitors. We observed that ROCK-mTOR inhibitor treatment reduced cell proliferation in time-dependent manner in GBM cells, but not in iNs ( Fig. 5a and Additional file 1: Figures S5A & S6A). In addition, apoptosis increased in GBM cells upon ROCK-mTOR inhibitor treatment, but not in iNs ( Fig. 5b and Additional file 1: Figures S5B &  S6B). These data suggest that ROCK-mTOR inhibitor treatment does not affect iN cell proliferation and survival, but may mediate GBM cell proliferation by regulating apoptosis.
We further evaluated cell proliferation and invasion. The mature phenotype of iNs was accompanied by cessation of proliferation and invasion as measured by soft-agar proliferation and invasion assays (Figs. 5c-e (See figure on previous page.) Fig. 1 Protein kinase inhibitor screen for reprogramming GBM cells. a & b Screening design and primary screening results. Human GBM cells (U118) expressing MAP2-promoter-RFP were seeded in 96-well plates. Kinase inhibitors were added at a final concentration of 2 μM in the induction medium. The medium was changed every other day until day 7 when cells converted to RFP-positive cells. Images were taken on day 7 after inhibitor treatment. Positive induced neural-like cells were counted by RFP-positive staining and quantified on day 7. The histogram represents the ratio of normalized data for kinase inhibitor treatment as compared with DMSO. c Secondary screening results. U118 cells were treated with candidate kinase inhibitors. R + Y: Rapamycin+Y27632; P + Y: P529 + Y27632. Quantitative data are the mean ± SEM from three independent experiments

A R T I C L E
and Additional file 1: Figure S6C & S6D). To determine tumorigenic properties in vivo, we injected parental GBM cells and iNs (1.5 × 10 5 ) into the brain frontal lobe, 2 mm lateral to the bregma of nude mice (Fig. 6a) [49].
No tumors were formed following iN transplantation up to 16 weeks (Fig. 6b). Transplantation of GBM cells treated with induction medium without ROCK-mTOR inhibitors resulted in tumor formation in 10 of 10 mice

Reprogramming orthotopic intracranial GBM prevents local recurrence
Despite advances in varieties of treatment modalities for GBM, local relapse remains a clinically significant problem. We examined whether our reprogramming treatment could prevent GBM local recurrence using a mouse model [37][38][39][40]. GBM tumors were generated by transplanting GFP-GM97 cells in nude mice. After four weeks, when primary tumors had reached~50 mm 3 ,~95% of each tumor was excised except for a small fragment (2 mm 3 ). This fragment was not removed to model residual tumor cells after surgery. Matrix-Driven Delivery (MDD) Pellet of ROCK-mTOR inhibitors with dried induction medium was prepared by Innovative Research of America. The finished pellet had a biodegradable matrix that effectively and continuously releases the active product in the animal for three weeks. We implanted the mTOR-ROCK inhibitors with or without induction medium in the MDD pellet into the resection cavity of xenografts. Controls included no treatment, induction medium only and ROCK-mTOR inhibitors without induction medium in the MDD pellet. Tumor recurrence and mouse survival were measured (Fig. 6c, d, n = 10).
Tumor recurrence was not observed up to 12 weeks in mice treated with mTOR-ROCK inhibitors with induction medium treatment (0/10 mice, Fig. 6c). Local recurrence was observed within 3-4 weeks in 10 of 10 mice without treatment or with induction medium only (Fig. 6c). 8/10 mice displayed tumor recurrence in mTOR-ROCK inhibitors without induction medium, suggesting that there is only a weak effect on GBM when using mTOR-ROCK inhibitors alone (direct chemotherapy). Immunohistochemistry (IHC) analysis revealed that 90% ± 5 of GFP/MAPs2 and 70% ± 8 of GFP/synapsin positive cells were detected 2 weeks after mTOR-ROCK inhibitor treatment with induction medium (Fig. 6e). These results suggest that GFP-GBM cells converted into mature neurons in vivo. Recurred tumors were examined from mTOR-ROCK inhibitors without induction medium, medium only or no treatment. It was observed that GFP-positive cells did not express MAP2 (Additional file 1: Figure S7). Tumors from mTOR-ROCK inhibitor treatment without induction medium had fewer Ki67 positive cells and more apoptotic cells as compared with tumors from medium only or no treatment control (Fig. 6f-g). Taken together, these results indicate that that mTOR-ROCK inhibitors under the reprogramming

Discussion
Collectively, we screened a kinase inhibitor library and found that a combination of the ROCK-mTOR kinase inhibitors can substitute for all transcription factors to reprogram GBM cells. This conversion is highly efficient, fast, and yields mature neurons. Induced neurons lose their tumorigenicity and reprogramming treatment prevents GBM local recurrence in mice. Reprogramming GBM by kinase inhibitors is a transgene-free method We believe that our findings will assist in the development of GBM therapy. ROCK1 and ROCK2 have tissue-specific expression patterns. ROCK1 is mainly expressed in lung, liver, spleen, kidney and testis, while ROCK2 is distributed mostly in brain and heart. By acting on the cytoskeleton, ROCK regulates cell shape and movement, cellular immortalization and differentiation [50][51][52][53]. mTOR regulates cell growth, proliferation, protein synthesis, and  [54][55][56][57][58][59][60]. ROCK inhibitors are already used as anti-tumor drugs in patients, while mTOR inhibitors are used to prevent transplant rejection, promote longevity, and treat cancer in clinical trials. Thus, of ROCK-mTOR inhibitors-based reprogramming strategy is expected to have low risk in vivo and may rapidly move towards clinical trials. GBM cell lines used in this study also have p53 mutation or inactivation. In future studies, we will examine the effects of p53 on GBM reprogramming. Among the various proteins involved in the mTOR pathway, much interest in GBM has coated SGK1 kinase. Recently, a new SGK1 kinase inhibitor has shown good antiproliferative, proapoptotic capacity and synergistic potential with radiotherapy in GBM in vitro and in vivo. [61,62] It is interesting to determine whether SGK1 regulates GBM reprogramming.
One key issue with reprogramming therapy is that reprogrammed cells may be reverted back to GBM cells Fig. 6 Reprogramming orthotopic GBMs in mice. a Experimental design for tumor formation in mice. b Parental GBM cells (Control) and iNs were implanted into the same nude mice (left and right side each mouse, respectively) and the tumor formation was measured. c & d GBM tumors generated by GM97 cells in nude mice with postoperative recurrence model. ROCK-mTOR inhibitors (P529 and Y27632, P + Y) with induction media in the MDD pellet were administered directly into the resection cavity of GBMs. P + Y without induction medium, medium only (both in the MDD pellets) or no treatment were used as controls. Tumor recurrence (c) and mouse survival (d) were measured (n = 10). e Postoperative residual GFP-GBM were treated with ROCK-mTOR inhibitors (P + Y) with induction medium. GFP-positive cells expressed the mature neuronal markers MAP2 and Synapsin 2 weeks after transplantation. Data are presented as mean ± SEM from 5 mice (300 cells/section and 2 sections per mouse were counted). The scale bar represents 10 μm. f-g Ki67 (IHC) and apoptotic (TUNEL assay) cells were evaluated in postoperative brain tissues treated with P + Y and induction media. Normal brain tissue was as the control. Ki67 and apoptotic cells from recurred tumors treated with P + Y without induction media, media only or no treatment were also examined. Scale bar 10 μm and form a tumor in vivo. We found that induced neurons do not show signs of further proliferation following 4 weeks in replacement of differentiation media with maintenance media. Although we did not observe that the reprogrammed cells revert back, we could not rule out reversal of reprogrammed cells into tumors. Additional experiments with genetically different samples and long-term in vivo observation are needed to address this issue. It is known that GBM cell lines are highly passaged cells that may not recapitulate the properties of primary GBM. We reprogrammed patient-derived primary GBM cells in vitro and in vivo. In the future studies, we will work on designing reprogramming strategies for patient-derived primary GBMs with different genetic background.
Delineating the molecular mechanism behind neural conversion will aid further development of the method. Gene expression analysis suggests that ROCK-mTOR may act as the "master regulator" to coordinate a set of defined factors in blocking cell reprogramming in physiological conditions. Consistent with this view, ROCK-mTOR controls both iPS reprogramming and neural conversion, implicating a general mechanism of reprogramming, where loss of ROCK-mTOR may generate lineage progenitors that develop complete and functional lineages in various induction media. In previous studies, we found that cellular senescence is a key step in the conversion of human fibroblasts into functional neural cells, including three neural lineages (astrocytes, neurons, and oligodendrocytes) [48,49]. By genetically manipulating the senescence process, we have provided a new method for converting most human fibroblasts into neural cells in only 1-2 weeks. Both ROCK and mTOR are involved in senescence and proliferation processes, implicating that the reprogramming process may also be regulated through the kinase-senescence signaling pathways. We found that expressions of a set of neurogenic transcription factors, such as Neurod2, Ascl1 and Myt1l, were increased during GBM-neuron cell conversion. Neurod2 is significantly downregulated in GBM patients, which is involved in the p53 pathway [63]. Ascl1 and Myt1l suppress GBM tumorigenicity and are required for glioblastoma stem cells to undergo neuronal lineage differentiation [64][65][66]. Correlations of reprogramming factors and GBM tumorigenesis need further investigation using patient tissues.
The reprogramming-based therapy presents new avenues for the development of GBM therapeutics. ROCK-mTOR inhibitors are already used as anti-tumor drugs in patients with few or no side effects. We expect that neurons generated by ROCK-mTOR inhibitors are "safe" in vivo. However, it is been shown that aberrant synapse formation between implanted neurons of the wrong subtype and endogenous cells can lead to significant motor side effects in human clinical trials for Parkinson's disease. We will identify the subtypes of our induced neurons and evaluate their functional properties in vivo. Our results also show that the reprogramming treatment with kinase inhibitors does not affect normal neurons and astrocytes, supporting the safety of our reprogramming strategy. We believe that our strategy is a novel, low risk, and effective method that can be combined with conventional therapy for clinical use.

Conclusions
In this study, we have developed a novel cell reprogramming-based method for GBM treatment. We screened a kinase inhibitor library, and found mTOR and ROCK inhibitors are sufficient to reprogram GBM cells into "normal" neurons. In vitro and in vivo tumorigenesis assays showed that induced neurons lose tumorigenicity and this reprogramming treatment prevented GBM local recurrence in mice. The role of the ROCK-mTOR pathways in GBM reprogramming is entirely novel. ROCK-mTOR inhibitors are already used as anti-tumor drugs in patients and promise few or no side effects in reprogramming. Therefore, the reprogramming-based therapy will provide new avenues for development of therapeutics for GBM in patients.

Additional file
Additional file 1: Figure S1. ROCK-mTOR kinase inhibitors reprogram GBM cells into iNs. Figure S2. ROCK-mTOR kinase inhibitors reprogram GBM cells into iNs. Figure S3. iNs did not express astrocyte and oligodendrocyte markers. Figure S4. Effects of mTOR-ROCK inhibitor treatment on normal primary neurons/astrocytes. Figure S5. iNs lose abilities of proliferation. Figure S6. Proliferation and invasion ability of iNs. Figure S7.