Edited by: Gabriella Castoria, Second University of Naples, Italy
Reviewed by: Marzia Di Donato, University of Campania Luigi Vanvitelli, Italy; Erika Di Zazzo, University of Molise, Italy
This article was submitted to Cancer Endocrinology, a section of the journal Frontiers in Endocrinology
†These authors have contributed equally to this work
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
In our previous study, we have shown that CRLF1 can promote proliferation and metastasis of papillary thyroid carcinoma (PTC); however, the mechanism is unclear. Herein, we investigated whether the interaction of CRLF1 and MYH9 regulates proliferation and metastasis of PTC cells via the ERK/ETV4 axis. Immunohistochemistry (IHC), qPCR, and Western blotting assays were performed on PTC cells and normal thyroid cells to profile specific target genes.
Papillary thyroid carcinoma (PTC) contributes nearly 90% of all thyroid malignancies. It is therefore the most common endocrine cancer (
The Cytokine Receptor-Like Factor 1 (CRLF1) promotes proliferation and survival of normal neuron cells and B-cells by assembling with the cardiotrophin-like cytokine factor 1 (CLCF1) or p28. Its effects are mediated through the MAPK/ERK and PI3K/AKT pathways (
The
The expression of ETS family proteins is regulated by several transcription factors, some of which are substrates of ERK1/2, modulating the expressions of matrix metalloproteases (MMPs) and BCL2 family influences the migration and invasion, and survival of cells (
Herein, we identified that MYH9 directly binds with CRLF1 and increasing CRLF1 protein stability. Inhibition of MYH9 in PTC cells overexpressing CRLF1 significantly reversed malignant phenotypes, and CRLF1 overexpression activated ERK pathway. RNA-sequencing revealed that ETV4 is a downstream target gene of CRLF1, which was up-regulated following ERK activation. Moreover, ETV4 is highly expressed in PTC tissues and is associated with poor prognosis. Finally, the chromatin immunoprecipitation (ChIP) assays showed that ETV4 binds to the promoter of matrix metalloproteinase 1 (MMP1) on PTC cells. Based on these findings, our study demonstrates that CRLF1 interacts with MYH9, promoting cell proliferation and metastasis via the ERK/ETV4 axis in PTC.
In the previous study (
Interaction between CRLF1 and MYH9 in PTC cells.
Further, suppression of MYH9 markedly decreased the levels of the CRLF1 protein (
The cycloheximide (CHX) chasing assay revealed that MYH9 enhanced the half-life of the CRLF1 protein. The impact of MYH9 on CRLF1 stability was attenuated by MG132, the proteasome inhibitor (
Firstly, we examined MYH9-mRNA expression levels using the TCGA-database samples. Our analyses revealed that MYH9 expression levels were not significantly associated with PTC or the late clinical stage (
Further, rescue experiments were performed to determine the potential impact of MYH9 on the effects of CRLF1 on malignant phenotypes in IHH4 cells. We transfected two siRNAs-MYH9 into CRLF1-overexpressing and empty vector IHH4 cells. Consequently, the qPCR and Western blot assays validated that MYH9-mRNA and MYH9-protein levels declined (
Effects of MYH9 knockdown on CRLF1-induced growth of PTC cells.
Next, we evaluated whether MYH9 reverses CRLF1-induced tumor proliferation
Inhibited MYH9 in PTC cells transfected with empty vector had no impact on the tumor migration and invasion ability (
Impact of MYH9 knockdown on the effects of CRLF1 on the migration, invasion, and EMT process of PTC cells.
Epithelial–mesenchymal transition (EMT) is an essential biological process in cancer metastasis. Previously, we had shown that CRLF1 induced PTC cell metastasis by activating the EMT process (
Previous evidence revealed that CRLF1 activates the ERK pathway in PTC cells (
We performed RNA-Sequencing (RNA-Seq) on IHH4 and CRLF1-overexpressing TPC1 cells to determine the genes controlled simultaneously by CRLF1 (the heat maps are shown in
ETV4 is the downstream target gene of CRLF1 in PTC.
ETV4, an ETS family protein, is a transcription factor that functions as substrates for ERK1/2 (
In the mRNA expression profile, ETV4 was markedly higher in PTC tissues than in the normal thyroid tissue both in the TCGA and NFH cohort (
ETV4 transcriptionally up-regulates MMP1 expression and correlates with PTC progression.
The clinicopathological parameters in 100 PTC patients.
Male | 29 | 12 (26.7%) | 17 (30.9%) | 0.64 |
Female | 71 | 33 (73.3%) | 38 (69.1%) | |
<55 | 78 | 39 (86.7%) | 39 (70.9%) | 0.09 |
≥55 | 22 | 6 (13.3%) | 16 (29.1%) | |
T1+T2 | 62 | 29 (64.4%) | 33 (60.0%) | 0.65 |
T3+T4 | 38 | 16 (35.5%) | 22 (40.0%) | |
Absent (N0) | 42 | 24 (53.3%) | 18 (32.7%) | |
Present (N1) | 58 | 21 (46.7%) | 37 (67.3%) | |
Absent (M0) | 83 | 42 (93.3%) | 41 (74.5%) | |
Present (M1) | 17 | 3 (6.7%) | 14 (25.5%) | |
I+II | 80 | 40 (88.9%) | 40 (72.8%) | |
III+IV | 20 | 5 (11.1%) | 15 (27.3%) | |
No | 60 | 34 (75.6%) | 26 (47.3%) | |
Yes | 40 | 11 (24.4%) | 29 (52.7%) |
Next, we explored the potential mechanism through which ETV4 modulates PTC proliferation and metastasis. In liver and gastric cancer types, ETV4 positively modulates MMP1 by binding the MMP1 promoter regions (
Collectively, based on our findings, we speculate a simple model to elucidate the molecular mechanism of CRLF1 interaction with MYH9, promoting progression in PTC (
Proposed model demonstrating interactions between CRLF1, MYH9, and activation of the ERK/ETV4 axis to regulate PTC progression.
Herein, we firstly demonstrated that CRLF1 interacts with MYH9 inducing PTC cell proliferation and metastasis through the ERK/ETV4 pathway, both
We used the FLAG pull-down assay and mass spectrometry to evaluate MYH9 as the prospective interacting protein with CRLF1 in PTC cells to elucidate further the detailed molecular mechanism of CRLF1 as a promoter of malignant phenotypes. We verified these findings using the Co-IP assays. Previous studies indicated that the MYH9 performs diverse functions in different cancer types (
Next, the rescue experiments confirmed that interfering with MYH9 significantly reduces tumor proliferation on CRLF1-overexpressing PTC cells but not in vector-expressing cells,
The ERK pathway plays a vital part in the pathogenesis, as well as the progression of PTC (
We additionally demonstrated that ETV4 is the downstream gene of CRLF1 in PTC. ETV4 is regulated by sumoylation, ubiquitination, and p300-mediated acetylation through ERK activation (
Altogether, our findings elucidate the molecular mechanism of CRLF in inducing malignant phenotypes in PTC. We demonstrated that CRLF1 interacts with MYH9, promoting tumor proliferation and metastasis by activating ERK/ETV4; therefore, we provide a prospective therapeutic target for the treatment of PTC.
We conducted all the animal experiments following a protocol approved by the Animal Care and Use Committee of the Southern Medical University. We obtained written consent from all patient-subjects for the use of clinical materials for research purposes. Besides, we obtained ethical approval of this study from the Ethics Committee of the Nanfang Hospital.
Prof. Haixia Guan (The First Affiliated Hospital of China Medical University, Shenyang, China) helped us with the human PTC IHH4, BCPAP, and normal thyroid epithelial Nthy-ori-3-1 cell lines. We purchased the human PTC cell line TPC-1 and the human embryonic kidney 293T (293T) cell lines from Nanjing Cobioer Company (Cobioer, China) and the American Type Culture Collection (ATCC, Manassas, VA, USA), respectively. We cultured the IHH4 cell line in RPMI-1640 and Dulbecco's modified Eagle's medium (DMEM; Invitrogen) enriched with 10% fetal bovine serum (FBS, Gibco, USA). Next, we cultured the Nthy-ori-3-1 and BCPAP cell lines in RPMI-1640 augmented with 10% FBS. We cultured the TPC-1 and 293T cell lines in DMEM supplemented with 10% FBS. Notably, we cultured all the cell lines with penicillin (100 U/ml) and streptomycin (100 U/ml) at 37°C in a humidified 5% CO2 incubator.
We obtained 20 PTC tissue samples and their paired normal tissue samples in August 2018 for the qRT-PCR assays. We collected a total of 100 paraffin-embedded PTC samples from patients who were first diagnosed between January 2005 and December 2009 at Nanfang Hospital, Southern Medical University, for the IHC assays (
We lysed the total protein in one sodium dodecyl sulfate (SDS) sample buffer and used BCA protein assays to determine protein concentrations. We used 8–12% SDS-polyacrylamide gels by electrophoresis to separate the protein extracts and transferred them to polyvinylidene fluoride membranes (Millipore, USA). Subsequently, we used 5% skim milk or bovine serum albumin to block the proteins for 1 h. Then, we incubated the membranes with primary antibodies at 4°C overnight and then with horseradish peroxidase-conjugated secondary antibodies (Pierce, USA) at room temperature for 1 h. Next, we visualized the bound antibodies through enhanced chemiluminescence and captured using a XAR film. We used β-actin as a loading control.
The primary antibodies used in this study included human anti-CRLF1 (1:400; ab56500), anti-MYH9 (1:1000; ab55456), anti-MMP1 (1:1000; ab215979), anti-fibronectin (1:1000; ab32419), anti-vimentin (1:1000; ab8978), and anti-snail (1:1000; ab216347) from Abcam, USA; anti-ETV4 (1:1000; SAB1403795) and β-actin (1:4000, A5541) from Sigma-Aldrich, USA; and anti-E-cadherin (1:1000; #14472), anti-ERK1/2 (1:1000; #4695), and anti-p-ERK1/2 (1:1000; #4370) from Cell Signaling Technology, USA. All the blot figures included the location of molecular size markers.
We used the TRIzol Reagent (Invitrogen, USA) in isolating total RNA from the PTC cells and clinical tissues following the manufacturer's protocol. After that, we reverse-transcribed 2 μg of RNA using the M-MLV Reverse Transcriptase kit (Promega) as per the manufacturer's instructions. We established the threshold cycle value for each sample by qPCR using SYBR Green (Invitrogen) and a CFX96 Touch sequence detection system (Bio-Rad, USA). We used the β-actin gene as the internal control for all genes. Subsequently, we calculated the relative gene expression levels using the comparative threshold cycle (2−ΔΔ
We bought the effective siRNA oligonucleotides targeting the CRLF1 from Guangzhou Ribobio Company (Guangzhou, China) and transfected using Lipofectamine RNAiMax (Invitrogen) according to the manufacturer's protocol. We purchased the lentiviral vector encoding FLAG-tagged CRLF1 (EX-N0027-Lv121), His-tagged MYH9 (EX-T1335-Lv128), the control vector (EX-EGFP-Lv105), and the packaging system (HIV) from GeneCopeia (USA). Subsequently, we verified all the plasmids via DNA sequencing. The siRNA sequences used are shown in
We seeded a total of 600–800 cells in 200 μl of medium per well in 96-well plates (five replicates of each sample). Next, we added 20 μl of MTT (5 mg/ml, BD Biosciences) to each well on the indicated day (days 1, 2, 3, 4, or 5) and incubated for 4 h at 37°C. Then, we discarded the supernatants and added 100 μl of dimethylsulfoxide (DMSO) to each well to dissolve the crystals. After that, we used the spectrophotometric plate reader (BioTek ELX 800, USA) to measure the absorbance at 490 nm. For the colony formation assays, we seeded 800 cells in 2 ml of the medium into each well of a six-well plate and cultured for 7–10 days. We subsequently fixed the colonies with methanol for 10 min and stained with 0.5% crystal violet for 15 min. ImageJ software was used to measure the cell counts. We performed the experiments in triplicate.
We used Transwell chambers (8 μm pores, Corning, USA) for the cell migration and invasion assays. We first pre-coated without (migration assay) or with (invasion assay) Matrigel (BD Biosciences). After that, we seeded 5 × 104-5 × 105 cells suspended in 200 μl of a serum-free medium in the upper chambers, and plated 500 μl of medium augmented with 10% FBS in the lower chambers. After 48 h of incubation, we fixed the cells on the upper surface of the membrane with methanol for 10 min and stained with 0.5% crystal violet for 10 min. Subsequently, we used an inverted microscope to count the cell numbers.
The Institutional Research Medical Ethics Committee of Nafang Hospital approved all the animal experiments. We purchased 4- to 6-week-old female BALB/c nude mice (
We embedded the clinical PTC tissue samples and tumors resected from mice in paraffin. Briefly, we cut 4-m-thick sections and baked them at 60°C for 2 h. Then, we deparaffinized the sections with xylene and rehydrated and blocked the endogenous peroxidase activity with 0.3% H2O2. Next, we processed the sections for high-temperature antigen retrieval with citrate (pH 6.0) and incubated them with 5% bovine serum albumin to block nonspecific binding. After that, we incubated the sections with diluted rabbit anti-ETV4 antibody (1:100; SAB1403795, Sigma-Aldrich, USA), p-ERK1/2 antibody (1:400; #4370, Cell Signaling Technology, USA), or MYH9 (1:100; ab55456, Abcam, USA) at 4°C overnight. Next, we washed the slides thrice with phosphate-buffered saline plus 1:1000 Tween-20 and incubated them with secondary antibodies (1:1000) for 30 min at 37°C. Then, we immersed slides in diaminobenzidine (Zhongshan Biological and Technical Company, Beijing, China) for 10 min, and then terminated reactions using distilled water. Next, we counterstained the slides with hematoxylin, dehydrated, and cover-slipped. Two experienced pathologists scored all sections. We calculated the staining index of ETV4 as follows: staining index = staining × intensity proportion of positive tumor cells. We defined the staining intensity as follows: 0 (no staining); 1 (weak, light yellow); 2 (moderate, yellow-brown); and 3 (strong, brown). We defined the percentage of positive cells as follows: 0 (no positive cells); 1 (<10% positive tumor cells); 2 (10–50% positive tumor cells); and 3 (>50% positive tumor cells). We determined the staining index cutoff value for ETV4 expression using its median value (3 points). We used a staining index score of >3 points and 3 points to define the tumors with high expression, low expression, respectively.
We transfected 293FT cells with an Lv-105 empty vector or the Lv121-CRLF1-flag plasmid. Thirty-six hours later, we treated the cells with IP lysis buffer (150 mM NaCl; 1% NP-40) on ice for 30 min. We then collected the proteins and centrifuged them to discard any precipitates. Then, we added 30 μl of FLAG beads (Sigma) to each sample and mixed them at 4°C overnight. The next day, we centrifuged and discarded the supernatant. We washed the beads five times with the IP buffer. After that, we added 30 μl of FLAG peptide (200 μg/ml, Sigma), and the mixture was shaken for 2 h. Next, we centrifuged the supernatant, followed by denaturing, and then performed SDS-PAGE. After electrophoresis, the gel was released and fixed for an hour and then washed with 30% ethanol for 10 min followed by a sensitizer for 2 min. After washing the gel with ultrapure water twice, we added a silver solution for 10 min. Subsequently, we performed de-staining and terminated the reaction when the expected bands appeared. The protein bands present were subsequently used for liquid chromatography–mass spectrometry (LC-MS) analysis.
We lysed the cell lines in a protein lysis buffer [20 mM Tris–HCl (pH 7.5), 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1% Nonidet P-40, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na3VO4, and 1 μg/ml leupeptin with protease inhibitor cocktail and phosphatase inhibitors] at 4°C for 30 min and centrifuged at 12,000 rpm (10 min, 4°C) to remove cell debris. For immunoprecipitation, we added the antibodies against FLAG or His to the lysates and incubated them overnight at 4°C, with rabbit IgG (1:100) serving as a control antibody. Then, we added the protein A/G agarose beads and incubated them with the lysates for 1 h at 4°C. After that, we washed the beads with the protein lysis buffer five times and any remaining bound proteins were eluted in protein loading buffer and analyzed by immunoblotting.
We plated the cells on coverslips in 48-well plates and cultured them overnight to allow for cell adherence. After fixation with 4% paraformaldehyde and permeabilization with 0.2% Triton X-100, we incubated the cells with antibodies. IgG was added as negative control. After that, we counterstained the cells with 0.2 mg/ml DAPI (Sigma-Aldrich, USA) and visualized them with a fluorescent confocal microscope (Carl Zeiss LS800, Germany). We used primary antibodies, including MYH9 (1:50; ab55456) and CRLF1 (1:50; ab56500) from Abcam, USA. Colocalization analysis was performed in single cells with JACoP plugin in ImageJ. Pearson's P colocalization index was calculated.
RNA was quantified in a Nanodrop (Thermo Scientific, Waltham, MA, USA) and quality checked using a Bioanalyzer-RNA 6000 nano kit (Agilent Technologies, Santa Clara, CA, USA). We prepared the libraries from 1 μg of RNA using the TruSeq Stranded mRNA kit (Illumina, San Diego, CA, USA). We performed Next-generation sequencing a NextSeq 500 platform (Illumina, San Diego, CA, USA) and a minimum of 30 million reads for each replicate were generated. The Cufflink RNA-Seq workflow was employed to perform bioinformatics analysis. We calculated differential gene expression as a log2 fold change (vector/CRLF1). The optimized FDR approach was employed to adjust
The ChIP assay was carried out using a ChIP assay kit (Thermo Scientific, Waltham, MA, USA). All procedures were performed according to the manufacturer's protocol. Briefly, chromatins were crosslinked, isolated, and digested with Micrococcal Nuclease to obtain DNA fragments. The samples were treated with the antibody or IgG for immunoprecipitation. After elution and purification, the recovered DNA fragments were subjected to qPCR and PCR assays. IgG served as a negative control.
Cells were transfected with siNC or siMYH9 and were then incubated with 20 μmol/L MG132 for 0–12 h or left untreated. The cells were then treated with 50 μg/ml CHX and incubated for different periods. Cells sampled were harvested and prepared for Western blot assays.
Data analysis was performed using SPSS Ver.22.0 (IBM Corporation, USA) and GraphPad Prism Ver.7.0 (GraphPad Software, San Diego, CA, USA). All data are shown as the mean ± SD and were obtained from three independent experiments. Categorical variables were analyzed using Fisher's exact tests. The mean values of two groups were compared with Student's
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below:
RNA-SEQ:
LC/MS data:
S-TY, J-NG, and B-HS performed
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at:
papillary thyroid carcinoma
chromatin immunoprecipitation
cycloheximide
co-immunoprecipitation
immunohistochemistry.