Loss of NF2 (merlin) has been suggested as a genetic cause of neurofibromatosis type 2 and malignant peripheral nerve sheath tumor (MPNST). Previously, we demonstrated that NF2 sustained TGFβ receptor 2 (TβR2) expression and reduction or loss of NF2 activated non-canonical TGFβ signaling, which reduced Raf kinase inhibitor protein (RKIP) expression via TβR1 kinase activity. Here, we show that a selective RKIP inducer (novel chemical, Nf18001) inhibits tumor growth and promotes schwannoma cell differentiation into mature Schwann cells under NF2-deficient conditions. In addition, Nf18001 is not cytotoxic to cells expressing NF2 and is not disturb canonical TGFβ signaling. Moreover, the novel chemical induces expression of SOX10, a marker of differentiated Schwann cells, and promotes nuclear export and degradation of SOX2, a stem cell factor. Treatment with Nf18001 inhibited tumor growth in an allograft model with mouse schwannoma cells. These results strongly suggest that selective RKIP inducers could be useful for the treatment of neurofibromatosis type 2 as well as NF2-deficient MPNST.
This study identifies that a selective RKIP inducer inhibits tumor growth and promotes schwannoma cell differentiation under NF2-deficient conditions by reducing SOX2 and increasing SOX10 expression.
This article is featured in Highlights of This Issue, p. 355
Neurofibromatosis type 2 is a rare genetic disorder affecting the peripheral and central nervous systems (1, 2). Patients with neurofibromatosis type 2 are likely to develop multiple benign tumors such as schwannomas, meningiomas, and ependymomas (3). In particular, approximately 90% of patients with neurofibromatosis type 2 have vestibular schwannoma (4, 5); therefore, it is a hallmark of clinical diagnostics. Vestibular schwannoma can lead to deafness, imbalance, and life-threatening brain stem compression (6). The estimated incidence of neurofibromatosis type 2 is 1 in 33,000 people worldwide with a penetrance of 95%; however, surgical intervention is the only available treatment (7–9).
Neurofibromatosis type 2 is developed by an autosomal dominant inheritance pattern, and loss of function of the tumor-suppressor gene encoding neurofibromin 2 (also known as merlin or schwannomin; NF2) allows for development of the disease (10, 11). NF2 is involved in the Hippo pathway, regulation of the actin cytoskeletal organization, and inhibition of several signals, including Raf/MEK/ERK, PI3K/Akt, and mTOR signaling pathways (12, 13). Consistent with its reported roles, NF2 is critical for cell growth, proliferation, and tumorigenesis (13–16).
Similar to NF2, the tumor-suppressor Raf kinase inhibitor protein (RKIP) regulates the actin cytoskeleton and is a well-known inhibitor of the Raf/MEK/ERK, PI3Kinase/Akt, and mTOR signaling pathways (17–21). These functions are relevant to the suppression of cancer cell proliferation and metastasis (18, 22–25). The majority of cancers exhibit low or no level of RKIP, and recently, several studies have implicated the relationship between RKIP and cancer stem cells (26–28).
Previously, we reported the relevance of NF2 and RKIP in neurofibromatosis type 2 (29). Loss of the NF2 gene reduced TβR2 expression and caused an imbalance between TβR1 and TβR2 levels. TβR1 imbalance resulted in phosphorylation of RKIP, thereby promoting its destabilization. Indeed, activation of TβR1 and inactivation of TβR2 have been reported in various cancers (30–33). In fact, inhibition of TβR1 kinase activity via TEW7197, a general TGFβ inhibitor, showed prominent inhibition of schwannoma in neurofibromatosis type 2 mouse models (29). However, considering the physiopathological features of neurofibromatosis type 2, in which the average onset occurs at or before adolescence, it is classified as a pediatric genetic disease (11, 34, 35). The inhibition of TGFβ signaling by TEW7197 could have adverse effects because canonical TGFβ signaling is significant for normal homeostasis and development (36, 37).
In this study, we screened a new drug candidate that does not disrupt canonical TGFβ signaling but blocks TβR1-mediated reduction in RKIP levels. We found a novel chemical, Nf18001 that promotes the differentiation of neurofibromatosis type 2 cells and suppresses tumor growth in an allograft model without disruption of canonical TGFβ signaling.
Materials and Methods
All experimental procedures involving laboratory animals were approved by the Animal Care Committee of Pusan National University. FVB/NJ mice were obtained from The Jackson Laboratory. Before the experiment, all mice were maintained under temperature- and light-controlled conditions (20°C–23°C, 12/12 hours light/dark cycle) and provided with autoclaved food and water ad libitum.
A549, HEK293, and NCI-H28 cell lines were obtained from the ATCC. HCT116 cell lines were obtained from Dr. B. Vogelstein (Johns Hopkins University, Baltimore, MD), and the human schwannoma cell line Hei-193 (NF2 deficient) was provided by Dr. G. Zadeh (University Health Network, Toronto, ON). Mouse Schwann cells derived from NF2flox mice (38) were subjected to in vitro Cre-mediated deletion and were then transduced with pMSCV-hygro retroviral rescue constructs encoding either full-length merlin isoform 1 (MSchw-WT) or the empty vector (MSchw-KO). The cell lines were kindly provided by Dr. P. Greer (Queen's University, ON). Normal fibroblasts (GM00038, 9-year-old female N9) were obtained from Coriell Cell Repositories and maintained in Eagle's minimal essential medium supplemented with 15% FBS and 2 mmol/L glutamine without antibiotics. All cells were maintained in a 5% CO2 humidified incubator at 37°C. Hei-193, HEK293, and mouse Schwann cells were cultured in liquid DMEM medium supplemented with 10% FBS and 1% antibiotics. A549 and HCT116 cells were cultured in RPMI-1640 supplemented with 10% FBS and 1% antibiotics. The appearance and growth characteristics of all the cells used in this study were compared with published information to ensure their authenticity. Mycoplasma tests were performed every 6 months. Cells were passaged no longer than 2 months.
pCMV RKIP-HA was provided by G. Keum (David Geffen School of Medicine at the University of California, Los Angeles, CA). pCMV RKIP-T101A-HA and pCMV RKIP-T101D-HA were generated by Dr. J.H. Hwang (Pusan National University, Pusan, Korea). The pcDNA3 NF2-FLAG, pRK5 TGF-beta type 1 receptor-FLAG, and TGF-beta type 2 receptor–HA were obtained from Addgene. Transfection was performed using the Jetpei transfection agent (Polyplus) for mammalian expression of these vectors.
Cells were seeded at a density of 2 × 105 cells per well in 12-well plates and cultured overnight before transfection. The vector (1.5 μg) was mixed with 1.5 μL of Jetpei reagent in 150 mmol/L NaCl solution. The mixture was incubated for 15 minutes at room temperature (RT), after which it was added to the cells. After 3 hours, the serum-free medium was replaced with medium supplemented with 10% FBS.
For in vitro gene knockdown, siRNAs against target proteins were generated (Cosmo Genetech). The sequence of si-RKIP was CACCAGCATTTCGTGGGATGGTCTTTCAAGAGAAGACCATCCCACGAAATGCTGGTG and si-NF2 was CAGCCTGTCTTTCGACTTCAA. INTERFERin transfection reagent (Polyplus) was used for transfection. Briefly, cells were seeded at a density of 2 × 105 cells per well in 12-well plates and incubated overnight before transfection. A total of 1.5 pmol (21 ng) of siRNA duplexes were mixed with 4 μL of INTERFERin in 100 μL medium without serum. The mixture was incubated for 15 minutes at RT to allow the formation of the INTERFERin/siRNA complex. This mixture was then added to the cells and incubated for an additional 4 hours, after which the serum-free medium was replaced with medium supplemented with 10% FBS.
Antibodies against RKIP (1:2,000 for immunoblotting, ab76582), SOX10 (1:1,000 for immunoblotting and 1:200 for immunofluorescence, ab155279), myelin PLP (1:500 for immunoblotting and 1:300 for immunofluorescence, ab155279), MBP (1:500 for immunoblotting,1:300 for immunofluorescence, and 1:200 for FACS, ab62631), MPZ (for immunoblotting 1:500, ab31851), GFAP (for FACS 1:100 and 1:300 for immunofluorescence, ab270270), TβR1 (1:500 for immunoblotting, ab31013), and tenascin C (1:1,000 for immunoblotting, ac108930) were purchased from Abcam. Antibodies against SOX2 (1:1,000 for immunoblotting and 1:300 for immunofluorescence, 3579), OCT-4A (1:1,000 for immunoblotting, 2840), NANOG (1:1,000 for immunoblotting, 4903), c-MYC (1:1,000 for immunoblotting, 5605), p-SMAD2/3 (1:1,000 for immunoblotting, 8828), ERK (1:1,000 for immunoblotting, 9102), p-ERK (1:1,000 for immunoblotting, 9101), p-AKT S473 (1:1,000 for immunoblotting, 9271), GFAP (1:500 for immunoblotting and 1:300 for immunofluorescence, 3670), p-Elk-1 (1:1,000 for immunoblotting, 9181), c-Jun (1:1,000 for immunoblotting, 9165), and p-c-Jun (1:1,000 for immunoblotting, 9261) were purchased from Cell Signaling Technology. Anti–β-actin antibody (1:3,000 for immunoblotting, 66009–1-Ig) and Anti-HA antibody (1:1,000 for immunoblotting, 51064–2-AP) were purchased from Proteintech. Anti-FLAG (1:2,000, F1804) antibody was purchased from Sigma. Antibodies specific for GST (1:1,000 for immunoblotting, sc-138) were purchased from Santa Cruz Biotechnology. Anti-TβR2 antibody (1:500 for immunoblotting, bs-0117R) was purchased from Bioss.
Western blotting analysis
Cells were harvested and lysed using a RIPA buffer (50 mmol/L Tris-HCl, pH 7.5, 150 mmol/L NaCl, 1% NP-40, 0.1% SDS, and 10% sodium deoxycholate) to examine cellular signaling. Protein concentrations in the samples were measured using a Pierce BCA protein assay kit (Thermo Fisher Scientific) and a BSA standard. Samples (20-μg protein per lane) were separated by SDS-PAGE and transferred to Immobilon-P transfer membranes (Millipore Corp.), which were then incubated with 3% skim milk in TBS-T buffer (Tris-HCl–based buffer containing 20 mmol/L Tris pH 7.6, 150 mmol/L NaCl, and 0.05% Tween 20) for 1 hour at RT and then with primary antibody in TBS-T overnight at 4°C. The membranes were then washed and incubated with secondary monoclonal horseradish peroxidase (HRP)–conjugated goat anti-mouse, goat anti-rabbit, and mouse anti-goat IgG antibodies (Pierce, Thermo Fisher Scientific, Inc.) in TBS-T for 2 hours at RT. The blots were detected using a HRP-conjugated secondary antibody using ECL assay (Advansta).
Silica (S5631; silicon dioxide) and TβR1 kinase inhibitors (SB431542 and LY2157299) were purchased from Sigma-Aldrich. Porcine TGFβ1 was purchased from R&D Systems. TEW7197 was provided by Dr. S.J. Kim (CHA University, School of Medicine, Seongnam, Korea).
For RT-PCR, total cellular RNA was extracted using an RNA extraction kit (Qiagen). Gene expression studies were performed using cDNA synthesized from total RNA with MMLV RT (Invitrogen) and random hexamers. PCR from genomic DNA was performed using DiaStar Taq DNA polymerase (SolGent). Gene expression studies were performed using the following primers:
hSOX2 (Forward) 5′-TCGCAGACCTACATGAACGG-3′
hSOX2 (Reverse) 5′-ACATGTGAAGTCTGCTGGGG-3′
hSOX10 (Forward) 5′-CATGGAGACCTTTGATGTGGC-3′
hSOX10 (Reverse) 5′-TCAGAGTAGTCAAACTGGGGG-3′
hMBP (Forward) 5′-CAAGTACCATGGACCATGCC-3′
hMBP (Reverse) 5′-TTTATAGTCGGACGCTCTGCC-3′
hMPZ (Forward) 5′-CAACCCTACATTGACGAGGTG-3′
hMPZ (Reverse) 5′-CACTGACAGCTTTGGTGCTTC-3′
hPMP22 (Forward) 5′-GCAATGGACACGCAACTGATC-3′
hPMP22 (Reverse) 5′-CGAAACCGTAGGAGTAATCCG-3′
hPNLIPRP3 (Forward) 5′-CACTCCAAAGGAAGTCAGGCTAG-3′
hPNLIPRP3 (Reverse) 5′-ACATCCTGGCATGTGCTTCCCT-3′
hABCA1 (Forward) 5′-CAGGCTACTACCTGACCTTGGT-3′
hABCA1 (Reverse) 5′-CTGCTCTGAGAAACACTGTCCTC-3′
hDDIT3 (Forward) 5′-GAAGAGGAGGAAGACCAAGGG-3′
hDDIT3 (Reverse) 5′-CTCGATTTCCTGCTTGAGCCG-3′
hPDK4 (Forward) 5′-AGGTGGAGCATTTCTCGCGCTA-3′
hPDK4 (Reverse) 5′-GAATGTTGGCGAGTCTCACAGG-3′
hCENPE (Forward) 5′-GGAGAAAGATGACCTACAGAGGC-3′
hCENPE (Reverse) 5′-AGTTCCTCTTCAGTTTCCAGGTG-3′
hPLK1 (Forward) 5′-CTGCACAAGAGGAGGAAAGCC-3′
hPLK1 (Reverse) 5′-TGGTTGCCAGTCCAAAATCCC-3′
hTOP2A (Forward) 5′-GTGGCAAGGATTCTGCTAGTCC-3′
hTOP2A (Reverse) 5′-ACCATTCAGGCTCAACACGCTG-3′
hDLGAP5 (Forward) 5′-CTCGATCAGCTACTCAAGCAGC-3′
hDLGAP5 (Reverse) 5′-CAGGTCTTCCTTTACTTGGCACC-3′
mSOX2 (Forward) 5′-AGGATAAGTACACGCTTCCCG-3′
mSOX2 (Reverse) 5′-TAGGACATGCTGTAGGTGGG-3′
mSOX10 (Forward) 5′-ACTACAAGTACCAACCTCGGC-3′
mSOX10 (Reverse) 5′-GTTGGACATTACCTCGTGGC-3′
mMBP (Forward) 5-’TTCTTTAGCGGTGACAGGGG-3′
mMBP (Reverse) 5′-TAAATCTGCTGAGGGACAGGC-3′
mMPZ (Forward) 5′-CTGCTCCTTCTGGTCCAGTGAA-3′
mMPZ (Reverse) 5′-AGGTTGTCCCTTGGCATAGTGG-3′
mPMP22 (Forward) 5′-CGTCCAACACTGCTACTCCTCA-3′
mPMP22 (Reverse) 5′-GCCTTTGGTGAGAGTGAAGAGC-3′
Protein–protein interaction analyses
GST-pull down assay and immunoprecipitation assays were performed to evaluate protein–protein interactions. For GST pull-down assay, agarose-bead–conjugated GST–RKIP recombinant protein was incubated with FLAG-tagged TβR1-transfected HEK 293 cell lysates in PBS for 1 hour at 4°C. The immunoprecipitation assay was performed with FLAG-tagged TβR1- or HA-tagged RKIP-transfected Hei-193 cell lysates in PBS. Whole lysates were incubated with appropriate primary antibodies for 2 hours at 4°C and reacted with agarose-bead–conjugated protein A/G (Invitrogen) for 2 hours. After centrifugation, the precipitates were washed twice with RIPA buffer and subjected to SDS-PAGE and western blotting analysis.
Cells were cultured on coverslips, washed with PBS, fixed with 4% paraformaldehyde (PFA) for 30 minutes at RT and then permeabilized with 0.2% Triton X-100 at RT for 5 minutes. After treatment with blocking solution (3% goat serum diluted in PBS) for 1 hour, cells were incubated with antibodies in blocking buffer overnight at 4°C. Finally, cells were incubated with fluorescein isothiocyanate and rhodamine-conjugated secondary antibodies at 4°C for 7 hours. Nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI) at RT for 10 minutes. After the cells were washed three times with PBS, coverslips were mounted with mounting solution (H-5501; Vector Laboratories). Immunofluorescence signals were detected using a fluorescence microscope (Zeiss and Logos).
Cell viability was determined using the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. Cells were seeded in 96-well plates and treated with the indicated chemicals for specified times. After removing the medium, 200 μL (0.5 mg/mL) MTT solution in PBS was added to each well. Plates were then incubated at 37°C for 4 hours, the MTT solution was removed, and the cells were lysed using a solubilization solution (1:1 DMSO:ethanol). The amount of formazan dye produced was quantified by measuring the absorbance using an ELISA microplate reader (Thermo Fisher Scientific) at 560 nm.
Allograft tumor growth assay
For the allograft tumor growth assay, 1 × 107 mouse schwannoma cells were seeded subcutaneously on FVB/NJ mice (8-weeks-old). Tumor-bearing mice were administered the carrier (n = 8) or Nf18001 (20 mg/kg; n = 8) by intraperitoneal injection for 5 weeks (three times per week). Every week, the tumor volume and body weight were measured. After termination of the experiment in each group, mice were euthanized, dissected, and tumor tissues were isolated.
Tumorsphere formation assay
Hei-193 or mouse schwannoma cells (3 × 106 cells/plate) were incubated with DMEM/F12 medium (supplemented with 2% B27 and 40 ng/mL bFGF) added to the indicated chemicals on non-coated plates for the specified days.
Cells were seeded in 6-well plates and incubated with DMSO (control), Nf18001 (5 μmol/L), or TEW7197 (5 μmol/L). After 4 days, the cells were fixed with 70% ethanol and labeled with propidium iodide. The samples were analyzed by FACS for cell-cycle analysis. After 7 days, the cells were fixed with 4% PFA and stained with antibodies. A minimum of 10,000 cells were analyzed using an Attune NxT flow cytometer (Thermo Fisher Scientific).
To estimate the transcriptional activity of TGFβ signaling, 3TP-luciferase vectors were transfected into cells for 24 hours, and cells were exposed to the indicated chemicals for an appropriate duration. After washing with wash buffer (Promega), the cells were lysed using lysis buffer (Promega), and luciferase activity was determined using a luminometer (MicroDigital).
Growth curves were generated by seeding 2 × 104 cells into a 6-well plate. After the indicated day, the cells were trypsinized and dissociated into single cells. A minimum of 10,000 cells were counted using an Attune NxT flow cytometer (Thermo Fisher Scientific).
Total RNA (500 ng) was extracted using an RNAeasy Kit (Qiagen). RNA labeling, hybridization on Human Gene 2.0 ST Array (Affymetrix), and data analysis were performed using DNA Link. Genes showing at least 2-fold differences in either cell line were selected for further analysis. The full dataset is available in the NCBI's Gene Expression Omnibus (GEO) database (GEO GSE186107; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE186107).
Quantification and statistical analysis
Data are expressed as the mean ± SEM of the values from the independent experiments performed, as indicated in the corresponding figure legends. The numbers of biological replicates and their representation are indicated in each figure legend. Statistical analyses were performed using GraphPad Prism v7.00. Two-tailed Student t tests were used for single comparisons, and an ANOVA with Tukey's multiple comparisons test was used for multiple comparisons, unless otherwise specified. Statistical significance was set at P < 0.05.
Chemical screening for RKIP inducers in the neurofibromatosis type 2 cell line
Because RKIP degradation in NF2 cells is mediated by non-canonical TGFβ signaling, inhibition of TβR1 kinase shows a tumor-suppressive effect on neurofibromatosis type 2 cells and mouse models (29). However, complete blocking of TGFβ signaling is not suitable for the treatment of neurofibromatosis type 2 because TGFβ signaling involves many biological processes such as early development, tumor suppression, and wound healing. Therefore, we aimed to identify a novel chemical capable of inducing RKIP expression without disrupting canonical TGFβ signaling. To this end, we measured RKIP expression in a human neurofibromatosis type 2 cell line (Hei-193) by treating the cells with 260 novel chemicals, designed on the basis of TEW7197 that did not interfere with TGFβ kinase activity. Among the libraries, we observed that 14 chemicals could induce RKIP expression in Hei-193 (Fig. 1A; Supplementary Fig. S1). Of these, four chemicals suppressed viability in both Hei-193 and neurofibromatosis type 2 mouse cells, similar to TEW7197 (Fig. 1B). Next, we assessed the effect of these chemicals on the binding between TβR1 and RKIP and found that SR08002 (Supplementary Fig. S2A) inhibited this binding (Fig. 1C).
Previously, we also revealed that mechanical stress, such as treatment with silica, could reduce RKIP expression in TβR2-deficient cells (29, 39). Thus, we examined the effect of SR08002 on silica-induced RKIP reduction. Reduced RKIP expression in HCT116 cells (TβR2-deficient colon cancer cell line; refs. 31, 32) in response to silica was abolished by Nf08001 (Fig. 1D; Supplementary Fig. S2B). Although SR08002 suppressed the viability of HCT116 cells, it also showed cytotoxicity in normal fibroblasts (Fig. 1E).
Optimization of Nf08001
Despite the induction of RKIP expression and anti-proliferating effect of SR08002 in neurofibromatosis type 2 cells, its optimization was required because of its toxicity to normal cells (Fig. 1E). Thus, we generated 39 derivatives from SR08002 (hereafter referred to as Nf08001) and examined their effect on the proliferation of neurofibromatosis type 2 cell lines and normal fibroblasts (Supplementary Fig. S3). On the basis of these results, we selected two chemicals (Nf18001 and Nf18011) that showed obvious antiproliferative effects in neurofibromatosis type 2 cell lines but not in normal fibroblasts (Fig. 2A); in addition, these chemicals induced RKIP expression and inhibited ERK activation (Supplementary Fig. S4A). Next, we examined the effect of these chemicals on the interaction between TβR1 and RKIP. As expected, Nf18001 and Nf18011 blocked the interaction between TβR1 and RKIP in a dose-dependent manner (Fig. 2B). Nf18001 blocked the interaction at low concentrations compared with the other chemicals (Supplementary Fig. S4B). In addition, Nf18001 induced RKIP expression in both human and mouse neurofibromatosis type 2 cell lines (Fig. 2C; Supplementary Fig. S4C) as well as reduced the phosphorylation of ERK and its downstream factors (Supplementary Fig. S4D).
Because TEW7197 was ruled out of our drug candidates, owing to TGFβ signaling inhibition, we analyzed the effect of all chemicals on TGFβ signaling through Smad2/3 phosphorylation and a 3TP-luciferase assay. Although TEW7197 strongly inhibited Smad2/3 phosphorylation, which was induced by TGFβ1 treatment, our newly identified chemicals did not inhibit Smad2/3 phosphorylation as much as TEW7197 did (Fig. 2D; Supplementary Fig. S5A). The luciferase activity assay showed that TEW7197, but not the novel chemicals, hampered TGFβ signaling in response to TGFβ1 treatment (Fig. 2E). Moreover, our chemicals did not alter other signaling cascades, such as IGF-1–induced AKT activation in Hei-193 (human neurofibromatosis type 2 cell line; Supplementary Fig. S5B) and A549 (human lung adenocarcinoma; Supplementary Fig. S5C) cells. Considering our chemical screening results, we chose Nf18001 as a candidate for further study (Supplementary Fig. S5D).
Selective effect of Nf18001 in NF2-deficient conditions
To confirm the effect of Nf18001, we examined the TβR1–RKIP interaction. Nf18001 interrupted binding in a dose-dependent manner (Fig. 3A; Supplementary Fig. S6A). Because the TβR1–RKIP interaction occurred under NF2- or TβR2-deficient conditions, we assumed that restoring NF2 might abolish the chemical effects such as RKIP induction, ERK inhibition, and antiproliferation. Indeed, RKIP induction and p-ERK suppression by Nf18001 were observed only in mouse schwannoma and HCT116 cells (TβR2-deficient cells) but not in normal fibroblasts (Supplementary Fig. S6B). We did not observe any chemical effects (antiproliferating effect, RKIP induction, and p-ERK suppression) in NF2-transfected Hei-193 cells (Fig. 3B and C) and mouse schwannoma cells (Supplementary Fig. S6C and S6D). NF2 transfection increased RKIP expression (Fig. 3C; Supplementary Fig. S6C), and knockdown of NF2 increased the sensitivity to Nf18001 in the NF2-expressing cell line (Supplementary Fig. S6E). In addition, Nf18001 did not inhibit TGFβ signaling, even at high concentrations (Fig. 3D).
To examine the mechanism of action of Nf18001 on RKIP in detail, we monitored the effect of Nf18001 on RKIP mutants. Previously, we revealed that phosphorylation at the threonine 101 site by TβR1 kinase promotes RKIP destabilization (29). Although Nf18001 induced wild-type RKIP expression, it did not alter the expression of two kinds of RKIP mutants (RKIP T101A, stable form; RKIP T101D, unstable form; Fig. 3E). This result indicated that Nf18001 protected RKIP from TβR1-mediated destabilization.
Because RKIP reduction is responsible for activation of TβR1, we speculated that Nf18001 might show an antitumor effect on TβR2-deficient cancer. To address this issue, we treated HCT116 cells and examined the expression of RKIP and cell proliferation. As expected, Nf18001 suppressed cell viability as strongly as TEW7197 (Supplementary Fig. S7A) and induced RKIP expression (Fig. 3F). Restoration of TβR2 in HCT116 cells abolished the effect of the chemicals (Fig. 3F; Supplementary Fig. S7A), suggesting that the biological effect of Nf18001 is dependent on the status of RKIP reduction, which is caused by an imbalance in TβR1. In addition, we observed an increase in the level of TβR2 restored by Nf18001 (Fig. 3F). This result is consistent with that of our previous report that RKIP reduction downregulates TβR2 expression (29). To extend our hypothesis, we examined the effect of Nf18001 on a mesothelioma cell line, in which RKIP reduction was frequently detected (39). Treatment with Nf18001 induced RKIP expression and suppressed cell proliferation in a dose-dependent manner (Supplementary Fig. S7B and S7C). These results implied that Nf18001 could be used for the treatment of mesothelioma, TβR2-deficient cancer, and neurofibromatosis type 2.
Gene expression profile of Nf18001-treated neurofibromatosis type 2 cells
To investigate the effect of Nf18001 on gene expression profiles, we performed a microarray analysis using Nf18001-treated Hei-193 cells (Fig. 4A) and found that the expression of various genes was decreased (Cluster A) or increased (Cluster B) in Nf18001-treated cells. Gene ontology analysis classified the clustered genes into two categories: Cell-cycle arrest–related processes, including “cell division,” and lipid metabolism–related processes, including “cholesterol biosynthetic process” (Fig. 4B; Supplementary Fig. S8). Pathway analysis also suggested a similar result, cell cycle and steroid pathway (Fig. 4C; Supplementary Fig. S8). In addition, the cluster of genes that were strikingly upregulated by Nf18001 included lipid metabolism–related genes such as PNILPRP3, NR4A2, and ABCA1, whereas cell-cycle–related genes, such as PLK1 and CENPE, were included in most downregulated gene clusters (Fig. 4D and E). In addition, we observed considerable upregulation of tenascin C, the readout of canonical TGFβ signaling (40); this finding supported our preliminary result that Nf18001 facilitated TGFβ signal activation in Hei-193 cells (Supplementary Table S1).
Nf18001 inhibits the cell cycle and promotes differentiation into schwann cells
On the basis of results of gene expression profiling, we monitored the cell cycle after treatment with Nf18001 and revealed that Nf18001 induced cell-cycle arrest at the G1 phase (Fig. 5A; Supplementary Fig. S9A). Thus, Nf18001 treatment completely inhibited cell growth (Fig. 5B) and induced morphological changes similar to dendrite formation (Fig. 5C; Supplementary Fig. S9B). Because the morphology of Hei-193 cells treated with Nf18001 resembled that of TEW7197-treated or NF2-transfected cells, we assumed that Nf18001 could induce cell differentiation. Microarray analysis data showed that the expression of lipid synthesis–related genes was induced by TEW7197, which then induced cell differentiation into Schwann cells in our previous data (29). Thus, we examined the expression of Schwann cell markers and cell-cycle–related genes. Consistent with our microarray data, the expression of cell-cycle–related genes (CDK1, CDK4, CCNB1, and PLK4) was decreased by Nf18001 treatment (Fig. 5D). In contrast, expression of Schwann cell markers (PLP, MBP, GFAP MPZ, PMP22, and SOX10; ref. 41) was increased by Nf18001 treatment (Fig. 5D; Supplementary Fig. S9C). We also observed that the expression of Schwann cell makers and TβR2 increased, but that of SOX2, a stem cell marker, was reduced at the protein level (Fig. 5E). To determine whether Nf18001 induced differentiation into Schwann cells, we re-examined the expression of differentiation markers by flow cytometry analysis and immunofluorescence. Results of flow cytometry analysis demonstrated that Nf18001 induced Schwann cell marker expression in schwannoma cells (Fig. 5F); immunofluorescence staining for Nf18001 also showed similar results comparing with TEW7197 (Fig. 5G; Supplementary Fig. S9D). These results indicate that our novel chemical Nf18001 is an effective reagent for cell-cycle arrest and Schwann cell differentiation.
Nf18001 inhibits stemness of schwannoma cells by inducing RKIP expression
SOX2 is a well-known stem cell factor that maintains stemness under both physiological and pathological conditions (42, 43), whereas SOX10 orchestrates the differentiation of Schwann cells with SOX2 (44–46). Indeed, Schwann cells in the Nf2-null mouse model showed low levels of SOX10 (47). In a previous experiment, we also observed the induction of SOX10 and reduction of SOX2 expression by treatment with Nf18001 (Fig. 5D and E). Therefore, we examined the effect of Nf18001 on other stem cell factors, such as c-Myc, Nanog, and Oct4. However, our chemical did not alter the expression of these factors (Supplementary Fig. S10A). Next, to test the antitumor effect of Nf18001, we performed a tumorsphere formation assay. Hei-193 and mouse schwannoma cells easily formed tumorspheres. Nf18001 disrupted sphere formation; however, TEW7197 did not show an inhibitory effect (Fig. 6A; Supplementary Fig. S10B). Indeed, our chemical reduced the tumorsphere size as well as the number of spheres (Fig. 6B and C). Under these conditions, the expression of stem cell factors (SOX2, c-Myc, Nanog, and Oct4) was downregulated to undetectable levels by treatment with Nf18001, whereas SOX10 expression was considerably increased (Fig. 6D; Supplementary Fig. S10C). To investigate the molecular mechanism of reduction in SOX2 expression, we examined the localization of SOX2 through immunofluorescence and cell fractionation and revealed that nuclear SOX2 was moved to the cytosol in response to Nf18001 and TEW7197 treatment (Fig. 6E and F). A nuclear export inhibitor (leptomycin B; Supplementary Fig. S10D) or a proteasome inhibitor (MG312; Supplementary Fig. S10E) blocked the Nf18001/TEW7197-induced reduction in SOX2 expression, indicating that our chemicals promoted SOX2 export and degradation. Next, we examined the effect of RKIP on reduction in SOX2 and induction of SOX10 expression. Ectopic expression of wild-type and stabilized RKIP (T101A) reduced SOX2 expression and RKIP T101A induced SOX10 expression without chemical treatment (Fig. 6G). In contrast, RKIP T101D did not reduce SOX2 expression (Fig. 6G). Moreover, siRNA specific for RKIP (si-RKIP) abolished the chemical-induced reduction in SOX2 expression in the nucleus and inhibited its cytosolic localization (Fig. 6H). These results imply that RKIP is critical for Nf18001-induced reduction in SOX2 and induction of SOX10 expression. Because our purpose was to develop a new drug candidate for neurofibromatosis type 2 without disruption of TGFβ signaling, we monitored the effect of Nf18001 on the canonical TGFβ signaling cascade. In Hei-193 cells, TGFβ-induced Smad2/3 phosphorylation was prolonged by Nf18001 treatment (Supplementary Fig. S11A and S11B). In addition, we observed similar features in RKIP- or NF2-transfected Hei-193 cells (Supplementary Fig. S11C and S11D). These results indicate that the TGFβ signaling feature of schwannoma cells, imparted by Nf18001, was very similar to that of schwannoma cells in which NF2 expression was restored.
Nf18001 has an antitumor effect in vivo
To examine the in vivo effect of Nf18001, we allografted the neurofibromatosis type 2 mouse model-derived schwannoma cell line to 8-week-old mice and administered Nf18001 by intraperitoneal injection (20 mg/kg, 3 times per week). Treatment with Nf18001 inhibited tumor growth (Fig. 7A; Supplementary Fig. S12A); in addition, tumor size and weight were suppressed by Nf18001 treatment (Fig. 7B and C). Injection of Nf18001 at a dose of 200 mg/kg (10 times higher than the therapeutic dose) into normal mice did not change the body weight (Supplementary Fig. S12B), indicating that Nf18001 did not cause severe toxicity. Next, we analyzed the expression of RKIP and related genes in tumors. The Nf18001-treated tumor tissues showed induction of expression of RKIP, Schwann cell markers (PLP, MBP, and MPZ), and SOX10 at the protein level (Fig. 7D). RT-PCR analysis also showed similar results, that is, the induction of expression of Schwann cell markers and SOX10 and the reduction in SOX2 expression (Fig. 7E). Considering our in vitro and in vivo results, Nf18001 is a potential drug candidate for neurofibromatosis type 2–derived schwannoma.
In this study, we evaluated the efficacy of Nf18001 as a potential therapeutic agent for neurofibromatosis type 2 and confirmed our therapeutic strategy to target the TβR1–RKIP network. Majority of patients with NF2 syndrome have NF2 gene mutation. Loss of NF2 causes imbalance of TGFβ receptors and predominated TβR1 induced phosphorylation-mediated degradation of RKIP (29). Consequentially, it promotes tumor formation via activation of MAPK and snail (39). In a previous study, we suggested TEW7197 as a novel agent for anti-neurofibromatosis type 2 therapy (29). TEW7197 showed an inhibitory effect on canonical TGFβ signaling, resulting in adverse effects on normal homeostasis (Fig. 2D and E). TGFβ signaling plays crucial roles in normal homeostasis, including adult stem cell differentiation, immune regulation, and wound healing (37, 48, 49). Thus, we designed a new chemical to specifically inhibit only the RKIP–TβR1 interaction without disturbing other interactions of TβR1, such as that with SMAD2/3. Our results show that Nf18001 disrupts the RKIP–TβR1 interaction more than TEW7197, without affecting canonical TGFβ signaling (Fig. 3A; Supplementary Fig. S4B). Moreover, TEW7197 inhibits several signaling pathways, including ERK, AKT, and TGFβ signaling pathways, whereas Nf18001 specifically inhibits ERK phosphorylation. Inactivation of ERK apparently occurs when RKIP expression is induced in NF2-deficient schwannoma cells and TbR2-deficient HCT116 cells (Fig. 3C; Supplementary Fig. S5 and S6). In this context, Nf18001 works only in abnormal NF2 or TβR2-deficient status, and therefore it causes reduction of cell viability in relevant cell line except normal fibroblast in vitro and tumor formation without lowering body weight in allograft assay.
Notably, Nf18001-induced RKIP expression reduced SOX2 expression (Fig. 6; Supplementary Fig. S10). Recently, mounting evidence has shown a relationship between RKIP and cancer stem cell factors (26, 27). In particular, our data show that RKIP induces nuclear export of SOX2, which leads to reduction in SOX2 expression (Fig. 6E and F). It is assumed that these are indirectly regulated through ERK inhibition by RKIP (26). Interestingly, Nf18001-induced RKIP expression also regulates expression of SOX10, which plays a crucial role as a transcription factor in the induction of Schwann cell differentiation (41, 45). Our microarray and tumorsphere formation assay data serve as evidence that RKIP and Nf18001 regulate both transcription factors (SOX2 and SOX10), inhibit cancer stemness, and induce the redifferentiation of schwannoma in neurofibromatosis type 2 (Figs. 4 and 6). The response of SOX2 and SOX10 to TGFβ may be attributed to the recovery of canonical TGFβ signaling by Nf18001 and RKIP. The highly upregulated tenascin C expression, revealed through microarray analysis, indicated that Nf18001 could induce recovery of canonical TGFβ signaling. Indeed, Nf18001 and TGFβ restored TβR2 expression in Hei-193 cells (Supplementary Fig. S6F and S11). TGFβ signaling is vital for Schwann cell differentiation and myelination (36, 50), and NF2-null schwannoma cells lack SOX10 function; hence, they exhibit abnormal de-differentiation and proliferation as well as development of schwannoma (47). These data and our results support the fact that Nf18001 induces RKIP expression and recovery of canonical TGFβ signaling, leading to the redifferentiation of schwannoma cells by reduction in SOX2 and induction of SOX10 expression in neurofibromatosis type 2. However, understanding the detailed mechanism of RKIP-dependent SOX2 and SOX10 regulation requires further investigation, and we plan to undertake such studies in the future.
In an allograft study, Nf18001 delayed the tumor growth in mouse schwannoma cells derived from neurofibromatosis type 2 mouse models without reducing body weight (Fig. 7A–C; Supplementary Fig. S12). In the allografted tumors, expression of RKIP, Schwann cell makers (PLP, MBP, and MPZ) and SOX10 was upregulated at the protein and mRNA levels upon treatment with Nf18001 (Fig. 7D and E) These data support our results; however, to ensure that the results are valid, further in vivo studies using transgenic mouse models over an extended period of time are needed.
Furthermore, Nf18001 therapy would be effective in diseases other than neurofibromatosis type 2, such as mesothelioma and TβR2-deficient cancers. In fact, the NF2 gene is highly mutated in mesothelioma (50%; refs. 51–53). In a previous study, we showed that mesothelioma and neurofibromatosis type 2 share a common tumorigenesis mechanism (29). Data from this study (Supplementary Fig. S6B and S7) support the fact that Nf18001 could be a potential therapeutic agent for these diseases.
In conclusion, our study shows that Nf18001, which can target the RKIP–oncogenic TβR1 network, may be a novel drug candidate for diseases such as neurofibromatosis type 2.
No disclosures were reported.
J.-H. Cho: Conceptualization, data curation, software, formal analysis, supervision, validation, investigation, visualization, methodology, writing–original draft. S. Park: Data curation, formal analysis, validation, visualization, writing–original draft, writing–review and editing. S. Kim: Resources. S.-M. Kang: Validation, visualization, methodology. T.G. Woo: Validation, visualization, methodology. M.H. Yoon: Validation, visualization, methodology. H. Lee: Resources. M. Jeong: Resources. Y.H. Park: Resources. H. Kim: Resources. Y.T. Han: Resources. Y.G. Suh: Validation, methodology. B.H. Kim: Validation, writing–review and editing. Y. Kwon: Validation, methodology. H. Yun: Resources, validation, methodology. B.J. Park: Conceptualization, resources, data curation, supervision, funding acquisition, validation, investigation, methodology, writing–original draft, project administration.
This work was supported by National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT; NRF-2020R1A4A1019322; to B.J. Park) and the Bio and Medical Technology Development Program of the National Research Foundation (NRF) funded by the Korean government (MSIT; NRF-2021M3E5E6038102).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.