Background
The Hedgehog (Hh) signaling pathway is an important regulator of embryonic development and homeostasis in metazoans [
1,
2]. In vertebrates, the main components include Hh ligand, the membrane receptor Patched (PTCH), signal transducer protein Smoothened (SMO), negative regulator suppressor of fused (SuFu), and glioma-associated oncogene (GLI) transcription factors including GLI1, GLI2, and GLI3 [
3]. In germ cells, dysregulation of Hh leads to various congenital abnormalities such as Greig cephalopolysyndactyly syndrome and Pallister-Hall syndrome caused by deleterious mutations in
Gli3 [
4,
5], as well as holoprosencephaly-like features and pituitary anomalies resulting from loss-of-function mutations in
GLI2 [
6]. Additionally, aberrant activation of Hh signaling in somatic cells has been implicated in human cancers [
7] including basal cell carcinoma [
8], medulloblastoma [
9], lung cancer [
10], breast cancer [
11], and glioma [
12]. Excess Hh ligand expressed by cancer or stromal cells, inactivating mutations in PTCH or SuFu, and activating mutations in SMO can all lead to derepression of GLI [
13] and inappropriate activation of target gene transcription [
14,
15]. These genes regulate cellular processes associated with tumorigenesis, including tumor cell survival/proliferation and metastasis and cancer stem cell self-renewal [
14,
15]. As such, various inhibitors of Hh signaling components have been developed for cancer therapy [
16‐
18].
Glioma arises from neurogliocytes and is a common type of central nervous system neoplasm. Around 54% of glioma cases are classified as glioblastoma (World Health Organization grade IV glioma) [
19,
20], which is difficult to treat; even with early diagnosis and aggressive surgery and radio−/chemotherapy, the median survival of these patients is 15 months [
21], with a 5-year survival of just 5% [
22,
23]. This is due to the malignant behaviors of glioma stem cells—including proliferation, angiogenesis, and invasiveness—that are modulated by Hh signaling [
12,
24]. Combined inhibition of Hh and Notch pathways sensitizes cluster of differentiation (CD) 133
+ glioma stem cells to chemotherapy [
25], while targeted inhibition of the Hh pathway improved the survival of glioma xenograft model mice [
26].
Rho GTPases modulate cell morphogenesis, proliferation, invasion, and survival through regulation of the actin cytoskeleton [
27,
28]. Most Rho GTPases identified to date (e.g., RhoA, RhoC, Rac1, and Cdc42) have oncogenic functions when abnormally activated. For example, loss of RhoC inhibited cancer cell metastasis in a RhoC
−/−; pyV-MT mouse model of mammary tumors [
29], and knocking out one allele of the
Rac1 gene impaired K-Ras-induced oral papilloma growth [
30]. The switch between GDP-bound inactive and GTP-bound active states of Rho proteins is mediated by GTPase-activating proteins (GAP) and guanine nucleotide exchange factors (GEFs) [
31]. GAPs accelerate GTP hydrolysis by Rho proteins; formation of GDP-bound Rho proteins block Rho GTPase signaling. On the other hand, GEFs facilitate the conversion of GDP-bound inactive Rho proteins to a GTP-bound active form by overriding the inhibitory effects of GDP dissociation inhibitors; thus, GEFs are generally considered to be pro-oncogenic. ARHGEF16 (also known as Ephexin4, GEF16, or NBR) is a GEF that can activate RhoG, Rac1, and Cdc42 proteins of the Rho GTPase family [
32‐
34] and thereby promote migration and resistance to apoptosis of breast cancer cells [
35] independent of Ephrin signaling. However, the mechanism underlying the functions of ARHGEF16 is not fully understood.
In this study, we identified ARHGEF16 as a target gene of GLI2 that interacts with cytoskeleton-associated protein 5 (CKAP5) to regulate glioma cell migration and proliferation, thus promoting glioma progression.
Methods
Reagents, antibodies, and constructs
The GLI inhibitor GANT61 and protease inhibitor cocktail were purchased from Sigma-Aldrich (St. Louis, MO, USA). Puromycin was from Genechem (Shanghai, China) and Solarbio (Beijing, China), respectively. Lipofectamine 2000 transfection reagent (#11668019) and TRIzol reagent (#15596018) were from Thermo Fisher Scientific (Waltham, MA, USA). Protein A agarose beads (#11134515001) and Protein G agarose beads (#11243233001) were from Roche (Palo Alto, CA, USA), and Glutathione Sepharose 4B beads (#17–0756-01) were from GE Healthcare (Little Chalfont, UK). Antibodies against the following proteins were used for western blotting: ARHGEF16 (ab86068), GLI1 (ab49314), GLI2 (ab26056), SMO (ab38686), SuFu (ab52913), PTCH1 (ab55629), CKAP5 (ab86073), and normal rabbit IgG (ab171870) (all from Abcam, Cambridge, MA, USA); Forkhead box M1 (Abgent, San Diego, CA, USA; AT2097a); glyceraldehyde 3-phosphate dehydrogenase (Millipore, Billerica, MA, USA; MAB374); β-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA; sc-1616-R); and Flag (F3165) and c-Myc (M4439) (Sigma-Aldrich). Antibodies against GLI2 (sc-271786) used in the chromatin immunoprecipitation (ChIP) assay was purchased from Santa Cruz Biotechnology. All other chemicals used were of analytical grade and were purchased from Sigma-Aldrich. A reverse transcription kit (Takara Bio, Otsu, Japan; RR047A) and real-time quantitative (q)PCR assay kit (Takara Bio; RR820A) were used for mRNA quantitation. Cell-light™ EdU Apollo567 In Vitro Kit (Cat #: C10310–1) was purchased from Guangzhou RiboBio Co., LTD in China.
Luciferase reporter constructs used to examine transcriptional activation of ARHGEF16 by GLI2 with the dual-luciferase assay were generated by inserting
ARHGEF16 promoter sequences into the pGL3-basic vector. Primers used to generate the three reporter constructs are shown in Additional file
1: Table S1. pCMV6-Entry-Gli2-MYC/DDK (RC217291) containing human
GLI2 mRNA (NM_005270) was purchased from OriGene (Rockville, MD, USA). The first 984 bases of
GLI2 mRNA were deleted with a mutagenesis kit (Toyobo, Osaka, Japan; SMK-101) to generate a constitutively active form of GLI2 (GLI2A) lacking amino acids 1–328 [
36]. Human
ARHGEF16 mRNA (NM_014448) was inserted into pGBKT7 and pGEX-6p-1 plasmids using the In-Fusion Cloning kit (Clontech Laboratories, Mountain View, CA, USA; #639619) to generate pGBKT7-ARHGEF16 and pGEX-6p-1-ARHGEF16, respectively.
ARHGEF16 or
CKAP5 silencing constructs were generated with the BLOCK-iT Pol II miR RNAi Expression Vector kit (Invitrogen, Carlsbad, CA, USA; K4936–00); lentiviruses (LVs) expressing GLI2A or ARHGEF16 or for
ARHGEF16 knockdown were obtained from GeneChem (Shanghai, China). GV358 and GV307 LV vectors were used for overexpression or knockdown, respectively; the target sequences are shown in Additional file
1: Table S2. The target sequences for
GLI2 knockdown was TCCTGAACATGATGACCTA [
37].
Cell culture and transfection
H4, U87, and U118 human glioma cell lines and 293 T human embryonic kidney cell line were purchased from the American Type Culture Collection (Manassas, VA, USA). Cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; Gibco), 100 U/ml penicillin, and 100 μg/ml streptomycin at 37 °C in a humidified incubator of 5% CO
2. Cells were transiently transfected with polyetherimide for 293 T cells to detect the efficiency of sh-ARHGEF16 or sh-CKAP5 constructs (Additional file
2: Figure S1A-C), or with Lipofectamine 2000 for glioma cell lines according to the manufacturer’s instructions. LV systems were used to establish stable glioma cell lines overexpressing GLI2A or ARHGEF16 or to knock down
GLI2 or
ARHGEF16. Puromycin (0.5 μg/ml) was added to cultures to maintain stable overexpression in the cell lines.
Microarray analysis
Microarray analysis was performed by Compass Biotechnology (Beijing, China). Briefly, total RNA was extracted from U87 cells stably overexpressing GLI2A (U87 GLI2A) and control cells (U87 Control) using TRIzol reagent, and then processed for hybridization to an HT-12 v4 Expression BeadChip (Illumina, San Diego, CA, USA). The array was washed and then scanned using an Illumina BeadArray Reader. Differentially expressed genes (DEGs) between LV-Control and LV-GLI2A U87 cells were identified from the data.
Western blotting and real-time qPCR
Total protein was extracted from cultured cells using lysis buffer (0.5% Lubrol-PX, 50 mM KCl, 2 mM CaCl
2, 20% glycerol, 50 mM Tris-HCl [pH 7.4], and 1% protease inhibitor cocktail), and the relative levels of target proteins were evaluated by immunoblotting. For qPCR, total RNA was extracted from cultured cells using TRIzol reagent and 1 μg of total RNA was used for reverse transcription and real-time qPCR on an ABI StepOne Plus detection system (Applied Biosystems, Foster City, CA, USA). Sequences of primers for detecting each target gene are shown in Additional file
1: Table S3.
Dual luciferase assay
U87 cells grown to 70% confluence in 24-well plates were transfected in triplicate with 0.75 μg of pGL3-basic-ARHGEF16 promoter-luciferase reporter and 0.25 μg of GLI expression plasmid or empty vector along with 0.025 μg of pRL-TK for normalization. After 48 h, luciferase activity was measured with a luminometer using the dual-luciferase assay kit (Promega, Madison, WI, USA; TM040) according to the manufacturer’s instructions. The activity of pGL3-basic-ARHGEF16 promoter-luciferase reporter normalized to that of pRL-TK Rluc reporter was compared between U87 cells transfected with GLI expression plasmid or empty vector.
ChIP assay
H4 cells were cross-linked with 1% (v/v) formaldehyde in phosphate-buffered saline for 10 min at 37 °C with gentle shaking. After adding 0.125 M glycine to terminate the reaction, the cells were lysed with lysis buffer on ice. Chromatin DNA was sheared by sonication to obtain ~ 500 bp fragments that were then mixed with anti-GLI2 antibody and protein G-agarose to enrich DNA fragments bound to GLI2 through immunoprecipitation. After decrosslinking, the precipitated DNA was analyzed by qPCR to assess the ARHGEF16 promoter regions containing putative GLI binding sites.
Cell migration and proliferation assays
The cell migration assay was performed using transwell plates (8 μm pore size, 6.5 mm diameter; Corning Life Sciences, Lowell, MA, USA). Briefly, 2 × 104 cells in 200 μl of 2% FBS DMEM were seeded into the upper chamber of the transwell insert, while the bottom chamber was filled with 800 μl of 10% FBS DMEM. After 24 h, cells on the upper surface of the membrane were removed and those on the lower surface of the membrane were fixed with 4% paraformaldehyde and stained with Crystal Violet. Cells were observed using an optical microscope after washing to remove excess dye and quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA). The cells over the whole filter were counted while one field of each filter was presented in the figures.
Proliferative capacity of U87 and U118 glioma cells was examined with the soft agar colony formation assay. A 400-μl volume of 0.5% agar solution containing 10% FBS DMEM was added each well of a 12-well plate, and 200 μl of 0.6% agar solution was mixed with 200 μl of 20% FBS DMEM containing 2 × 103 cells and added to the top of the solidified 0.5% agar. A 200-μl volume of 10% FBS DMEM was added to the wells, which was replenished every three days, and the cells were incubated for 2 weeks; colonies with a diameter larger than 50 μm were counted.
H4 cell proliferation was assessed with plate colony formation, where 1.5 × 103 H4 sh-Control or H4 sh-ARHGEF16 cells were seeded in one well of the 6-well plate and cultured with 10% FBS DMEM for about 2 weeks. Then cell colonies, fixed with 4% paraformaldehyde and stained with Crystal Violet, were examined and photographed under a phase-contrast microscope and quantified using ImageJ software. Furthermore, Cell-light™ EdU Apollo567 In Vitro Kit was used for cell proliferation analysis following the manufacturer’s instructions.
Subcutaneous xenograft assay
Female BALB/c nu/nu athymic nude mice (4 weeks old) were used for experiments according to Guidelines for the Welfare of Animals in Experimental Neoplasia published by The United Kingdom Coordinating Committee on Cancer Research. The experimental protocol was approved by the Institutional Animal Care and Use Committee of Nanchang University and regional authorities. A total of 5 × 10
6 Control+sh-Control, GLI2A + sh-Control or GLI2A + sh-ARHGEF16 U87 cells were injected into the flanks of each of five nude mice to examine the role of GLI2/ARHGEF16 signaling on glioma growth in vivo, and tumor formation was observed starting 6 days later. The size of glioma xenografts was measured every two to three days, and tumor volumes were calculated by the formula: 0.5 × length×width×(length + width) [
38]. Twenty-four days after injecting the U87 cells, mice were sacrificed under anesthesia and the tumor xenografts were harvested for immunohistochemical analysis.
Furthermore, to address the GLI2 inhibition effects in a pre-clinical context, GLI2A or Control U87 cells were inoculated into ten nude mice as above, and then the mice loaded with GLI2A and Control U87 xenografts were randomly divided into two groups and treated with only solvent (corn oil: ethanol, 4:1) or GANT61 (25 mg/kg) in solvent through intraperitoneal injection every second day for 4 weeks [
18,
39]. In addition, 5 × 10
6 sh-Control or sh-ARHGEF16 U118 stable cell lines were inoculated in five nude mice as above to examine the effects of ARHGEF16 knockdown on tumor growth.
Protein-protein interaction assay
Cells were solubilized in lysis buffer containing 0.5% Lubrol-PX, 50 mM KCl, 2 mM CaCl2, 20% glycerol, 50 mM Tris-HCl (pH 7.4), and protease inhibitors. Pre-cleared cell lysate containing Flag-ARHGEF16 was incubated with 30 μl of 1:1 slurry magnetic beads conjugated with anti-Flag antibody for 4 h at 4 °C. The beads were washed four times with lysis buffer before adding sodium dodecyl sulfate (SDS) sample buffer, and the samples were analyzed by western blotting. To detect the interaction between ARHGEF16 and CKAP5, glutathione S- transferase (GST)-ARHGEF16 fusion protein was expressed in BL21 bacterial cells, purified, and immobilized on Glutathione Sepharose 4B beads (Amersham Pharmacia, Piscataway, NJ, USA), and then incubated with lysate from U87 cells. Bead-associated proteins were subjected to SDS-polyacrylamide gel electrophoresis and immunoblotting. Experiments were repeated at least three times.
Statistical analysis
Quantitative data are presented as mean ± SD of at least three experiments. Differences between groups were assessed with the Student’s t-test or by one-way analysis of variance and were considered statistically significant at P < 0.05 and highly significant at P < 0.01. Data were analyzed using SPSS v.17.0 software (SPSS Inc., Chicago, IL, USA).
Discussion
Glioblastoma is the most common type of malignant tumor in central nervous system and has a high recurrence rate owing to a strong capacity for migration and proliferation [
20,
21]. The Hh signaling pathway regulates tissue patterning during embryogenesis and contributes to the maintenance of adult tissues [
1,
2]. Numerous studies have shown that the Hh signaling pathway promotes gliomagenesis by maintaining the cancer stem cell pool [
42‐
44]. The oncogenic effects of Hh signaling are mediated by the target genes of GLI transcription factors [
15], suggesting that GLI inhibitors could possibly be used for cancer therapy [
18,
45].
In this study we identified
ARHGEF16 as a target gene of GLI2 in glioma cells. We found that GLI2 binds to the
ARHGEF16 promoter to activate gene transcription, suggesting that
ARHGEF16 is a novel target gene of GLI2. Among the three GLI transcription factors in mammals, GLI2 plays the most important role; in mice, GLI2 deficiency results in severe developmental defects and embryonic lethality [
46‐
49]. ARHGEF16 is a GEF of the Rho GTPase family [
50] whose members modulate cell morphogenesis, proliferation, invasion, and survival through regulation of the actin cytoskeleton [
27,
28]. Accordingly, we found that ARHGEF16 enhanced migration and proliferation in glioma cells. ARHGEF16, which is also known as Ephexin 4, can bind to the cytoplasmic region of Ephrin receptor [
50]. Ephrin signaling plays a key role in cellular repulsion, attraction, and migration by controlling local cytoskeletal dynamics through Ephexin proteins and Rho GTPases [
51‐
53]. Our finding that
ARHGEF16 is a target of GLI2 provides the first evidence of potential cross-talk between the Hh and Ephrin signaling pathways in glioma development.
Our study also identified CKAP5 as an ARHGEF16-interacting protein. CKAP5 is an evolutionarily conserved member of the XMAP215 family of microtubule-associated protein [
54‐
56] that is highly expressed in the mammalian brain [
57] but has also been found to be upregulated in colonic and hepatic tumors [
58]. It is also necessary for the survival of head and neck cancer as well as lung cancer cells [
59], and its expression level in liver cancer is an independent prognostic factor for both progression-free and overall survival, with a significant correlation between high CKAP5 level and poor prognosis [
41]. CKAP5 is required for the assembly and maintenance of the spindle apparatus during mitosis and meiosis and associated processes such as chromosome segregation and apoptosis [
40,
60], and its deletion leads to the formation of multipolar spindles and cell death [
61]. The identification of CKAP5 as an ARHGEF16-interacting protein in this study suggests that regulation of spindle integrity is important for glioma cell proliferation and migration. Further studies are needed to explore the mechanistic basis for the interaction between CKAP5 and ARHGEF16.
Given the oncogenic effects of some GEFs, it is in principle possible to target the oncogenic GEFs and CKAP5 for cancer therapy. Though no compound targeting ARHGEF16 or CKAP5 has been reported currently, some compounds targeting GEFs have been developed based on the comprehensive insights into the structural basis of GEFs and small G proteins interaction [
62,
63]. For instance, brefeldin A can inhibit the small ARF family members of the small G protein superfamily through stabilizing the ARF-GDP-GEF complex and thus trapping the GEF in an unproductive sate with its substrate [
64]; in addition, the compound NSC23766 can block the interaction between Rac and Tiam or Trio, both of which are the GEFs for Rho small G protein family, to inhibit Rac activity, thus presenting anticancer effects [
65]. The strategies to inhibit or stabilize the interaction of ARHGEF16 and its G proteins, along with GLI2 inhibition, can be used for cancer therapy.
Acknowledgements
This work was supported in part by grants from the National Natural Science Foundation of China (31460305, 31671476 to S.L.), the Natural Science Foundation of Jiangxi Province, China (20171ACB20028 to S.L.) and the Postgraduate innovation special fund project of Jiangxi Province (YC2015-B011).