Alternative ANKHD1 transcript promotes proliferation and inhibits migration in uterine corpus endometrial carcinoma

Data extraction and UpSet plot construction

Clinical information data and quantified gene expression transcriptome profiling of 390 primary UCEC samples were downloaded from the TCGA database (https://tcgadata.nci.nih.gov/tcga/), including 404 alternative SFs. In addition, missing percent spliced in (PSI) values of less than 25% were gathered from 527 primary UCEC samples in the TCGASpliceSeq database (https://bioinformatics.mdanderson.org/TCGASpliceSeq/)36. The AS patterns included AA, AD, AP, AT, ES, ME, and RI37. Each ASE is described with the gene name, TCGASpliceSeq database AS ID, and splicing pattern. An UpSet plot is shown to summarize the ASE profile in UCEC.

Construction UpSet plots of OS-SEs

Using the K-nearest neighbor algorithm, we inserted the missing expressed data in ASEs data. Data were filtered using three criteria, a mean PSI of less than 0.05, a standard derivation of less than 0.01 among all samples, and samples without follow-up recorded. Furthermore, ASEs combined with clinical information were analyzed with univariate Cox regression, and each prognostic value of ASEs was determined, and the OS-SEs were identified. UpSet plots were then set up to display OS-SEs. Afterward, z-score and −log10 (P value) were set as x-axis and y-axis, and volcano plots were used to illustrate the relationship between ASEs and prognosis. Finally, bubble plots were generated to display the top 20 significant OS-SEs in AA, AD, AP, AT, ES, ME, and RI.

Lasso regression and construction of a multivariate model with OS-SEs

The top 20 significant OS-SEs were integrated using Lasso regression to avoid the over-fitting of the multivariable model. Furthermore, a multivariate Cox regression model was set up based on the OS-SEs processed by Lasso regression, and a receiver operator characteristic (ROC) curve was applied to access its accuracy. We calculated the risk score using the following formula:

where n represents the number of OS-SEs selected by Lasso regression, and β each regression coefficient of OS-SEs. Afterward, samples were divided into high- and low-risk groups according to the median risk score. Kaplan–Meier curves were utilized to evaluate the relationship between risk score and survival probability. In addition, the PSI values of ASEs between high- and low-risk groups in the final model were illustrated using scatterplots and risk curves, respectively.

Independent prognostic analysis

Univariate and multivariate Cox analyses were applied to determine the independent prognostic value of the multivariate model risk score. Baseline information, including age, grade, and metastasis were integrated into the multivariate Cox analysis.

Co-expression and construction of a network between SF and OS-SEs

Based on the SpliceAid2 database, SF data were downloaded38. To determine the relationship between SF expression and PSI values, SFs and prognostic OS-SEs were co-expressed, and Pearson correlation was utilized to analyze the 390 SFs and prognostic OS-SEs based on expression levels. We filtered the regulation relationship by discarding interactions with P < 0.001 and an absolute value of correlation coefficient >0.400. A network between SF and OS-SEs was developed using Cytoscape (3.7.1)39. In our network, ellipses and arrows show OS-SEs and SF, in which the red and blue ellipses represent a high and low risk of OS-SEs, respectively, whereas red and green lines link SFs and OS-SEs represented positive and negative regulation, respectively.

Identification of OS-SEs related to metastasis

We applied the Kruskal–Wallis test and Mann–Whitney–Wilcoxon test to identify OS-SEs related to metastasis. In addition, according to the network analysis, these ASEs were also related to the regulatory network of SFs.

Co-expression analysis between ASEs and KEGG pathways

The prognostic signaling Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways estimated by gene set variation analysis (GSVA) were identified through univariate Cox analysis. A co-expression analysis between the prognostic pathways and metastasis-specific OS-SEs were conducted to identify potential downstream pathways.

Multidimensional validation

To minimize selection bias and the effects of limited data size, several databases were used to validate the expression of biomarkers we accessed at the cellular and tissue levels. Firstly, we used Genecard (https://www.genecards.org/) to find the top five genes related to pathways that we identified. In the top five genes, significant SF and OS-SEs were incorporated for further validation. Moreover, the Human Protein Atlas40, Genotype-Tissue Expression (GTEx)41, PROGgene Version242, Gene Expression Profiling Interactive Analysis (GEPIA)43, UCSC xena44, SurvExpress45, Ualcan46, Linkedomics47, cBioportal48, Expression atlas49, and Oncomine50 databases were utilized to validate the results in multiple dimensions. Furthermore, the Cancer Cell Line Encyclopedia (CCLE)51 was applied to validate the data at the cellular level. The String database52 was applied to construct a gene regulatory network at the molecular level.

To further explore the mechanism of AS, the Assay for Targeting Accessible Chromatin with high throughput sequencing (ATAC-seq) was utilized to validate chromatin accessibility53. Chromatin Immunoprecipitation sequencing (Chip-seq) was also applied to validate the binding domain of ASE and SF. However, the anti-HSPB1 antibody was absent, and thus, HSPB90B1 was applied to replace HSPB1 because HSPB1 is homologous to HSPB90B1 (GSE126151)54. The Cistrome data browser55 was utilized.

Patients and specimens

A total of 36 tumor specimens, 15 metastatic UCEC tissues, 15 primary UCEC tissues, and 6 additional UCEC tissues with paired non-tumor tissues were collected between January 2019 and January 2020 at the Tongji Hospital affiliated with the Tongji University School of Medicine. Written informed consent was obtained from all patients, and the procedures were approved by the Institutional Research Ethics Committee of Tongji Hospital, affiliated with Tongji University School of Medicine. Immunohistochemistry was used for the 15 primary and 15 metastatic primary UCEC tissues, reverse-transcription polymerase chain reaction (RT-PCR) was performed on 6 UCEC tissues and paired pericancerous tissues, and western blotting was performed on three of the six patient tumor tissues and paired pericancerous tissues.

Immunohistochemistry

Tissue blocks were cut into 4-µm-thick sections, deparaffinized, rehydrated, and stained overnight at 4 °C using an Ultrasensitive TM S-P system (KIT-9710; MaiXin, Fujian, China), and incubated with antibodies against ANKHD1 (1:100, cat. no. ab199164; Abcam, Cambridge, UK). Tissue sections were incubated with a secondary antibody labeled with biotin at 37 °C for 30 min (Ultrasensitive TM S-P, MaiXin). Diaminobenzidine tetrahydrochloride substrate (MaiXin) was used as the chromogen. The number of cells expressing ANKHD1 was classified into five grades: 0 points (no cell staining), 1 point (1–25% cell staining), 2 points (26–50% cell staining), 3 points (51–75% cell staining), and 4 points (76% + cell staining). Based on the intensity of cell staining, ANKHD1 expression was classified into four grades: 0 (no staining), 1 (light yellow particles), 2 (yellow particles), and 3 (dark yellow or tan particles). The final score of each section was the number of cells in the section multiplied by the staining score. A score less than 2 was considered negative, whereas a score greater than or equal to 2 was considered positive. Phosphate-buffered saline (PBS) and goat serum were used as negative controls.

Reverse-transcription quantitative PCR and agarose gel electrophoresis

Total tissue RNA was extracted using the RNA Fast 200 kit (Fastagen, Shanghai, China) and reverse-transcribed using the TB Green Premix Ex TaqTM Kit (Takara, Kyoto, Japan) according to the manufacturer's instructions. RNA quality check was performed using a spectrophotometer (acceptable A260/280 ratio between 1.8 and 2.0). cDNA was used as a template for RT-PCR at 95 °C for 5 min, denaturation for 5 s at 95 °C, and annealing at 60 °C for 30 s (40 cycles). The primer sequences were as follows: ANKHD1-BP3, forward CCAGATCCTGCTTGGAACCC, reverse TGTTTCCAATATGAGGTGCCCA; HSPB1, forward GCTTCACGCGGAAATACACG, reverse GTGATCTCGTTGGACTGCGT; and GAPDH, forward GGAGCGAGATCCCTCCAAAAT, reverse GGCTGTTGTCATACTTCTCATGG. qPCR reactions were performed in triplicate and the comparative CT method (2 − ΔΔCT method) was used to calculate the relative gene expression levels.

A 1% agarose gel was prepared using agarose powder (Biowest, Madrid, Spain) and Tris-acetic acid (TAE) (Beyotime, Shanghai, China), and the DNA amplification products were mixed with loading buffer (Beyotime). Samples were then loaded for electrophoresis. After electrophoresis, the gel was placed into a TAE solution (Beyotime) containing 0.5 μg/mL ethyl bromide for 30 min of staining. The bands were observed using a gel imager (Biorad, Hercules, CA, USA). The gel containing the target bands was then cut for DNA sequencing (Genewiz, Suzhou, China).

Western blotting

Western blot (WB) analysis was performed to evaluate ANKHD1 expression in both tissues and cells before and after transfection. Proteins were extracted from cells and the protein concentration was determined using a bicinchoninic acid (BCA) kit (Beyotime). After blocking, membranes (Millipore, Boston, MA,USA) were incubated with anti-ANKHD1 antibody (ab117788, Abcam; dilution 1:2000), anti-AKT antibody(AF1777, Biotime; dilution 1:1000), anti p-AKT antibody(AF1546, Biyotime; dilution 1:1000), anti-BAX antibody(ab32503, Abcam; dilution 1:2000), anti-Bcl-2 antibody(ab182858, Abcam; dilution 1:2000) and then incubated overnight at 4 °C. Subsequently, horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (ab205718, Abcam; dilution 1:5000) was added to the membranes and incubated for 1 h before detection. All blots derive from the same experiment and were processed in parallel.

Immunofluorescence staining

Cells were fixed for 30 min at 25 °C in 4% paraformaldehyde in PBS, permeabilized with Triton X-100 (Sangon Biotech, Shanghai, China), and then blocked with 1% bovine serum albumin for 1 h at 25 °C. Cells were incubated with antibody against ANKHD1 (BS-5831R, Bioss, Beijing, China; dilution 1:100). After washing three times (5 min per wash) with PBS, the cells were incubated with fluorescein isothiocyanate-conjugated antibody (ab150077, Abcam; dilution 1:500). Nuclei were counterstained at 25 °C for 10 min with DAPI (Beyotime).

Cell culture and transfection

Human EC cell lines Ishikawa(FH0305, FuHeng Biology, China) and HEC-1b(GDC0129, CDCC, China) were maintained in 89% DMEM media (GIBCO, Carlsbad, CA, USA) supplemented with 10% FBS (GIBCO) and 100 U/mL penicillin (Hyclone, Logan, UT, USA) in a humidified atmosphere containing 5% CO2 at 37 °C. A lentivirus transfection system was utilized to generate stable cell lines with specific gene knockdown or overexpression. Cells (1.5 × 105/well) were added to six-well plates 4 h before transfection. The sequence of ANKHD1-BP3 siRNAs (5′–3′) was GCGTCTGGAGGATATGTTAAT. Plasmid pSLenti-U6-shRNA-CMV-EGFP-F2A-Puro-WPRE was purchased from the Obio Company (Shanghai, China). Plasmid CMV-GFP-3FLAG-puro-HSPB1 was purchased from NOVOBIO (Shanghai, China).

Cell counting kit (CCK)-8 and colony formation assays

A commercial cell counting kit (CCK)-8 (Sigma Chemical, St. Louis, MO, USA) assay was used to evaluate cell proliferation. Cells were seeded onto 96-well plates at a density of 5 × 103 cells per well and cultured at 37 °C with 5% CO2. Absorbance was measured after an additional 3 h of incubation. A microplate reader (Thermo Fisher Scientific, Waltham, MA, USA) was used to detect the absorbance at a wavelength of 450 nm.

At 48 h from transfection, cells were plated in 6 cm cell culture dishes (1000 cells/dish) and incubated for 14 days. Cells were then stained for 20 min with crystal violet and the number of colonies (>50 cells) was determined.

Migration and invasion assays

For migration assays, cells were cultured for 24 h without Matrigel matrix. In the upper chamber, cells were cultured in a serum-free medium, and the lower chamber was filled with a 10% FBS medium. Cell invasion assay was performed using 6.5 mm transwells with 8.0 µm pore polycarbonate membrane inserts coated with a 0.5 mg/ml Matrigel matrix (BD, Franklin Lakes, NJ, USA) placed in a 24-well plate (Corning, Corning, NY, USA). Cells were seeded onto 24-well plates at a density of 1 × 105 cells per well and cultured at 37 °C with 5% CO2 for 48 h. Non-invading cells on the upper membrane surface were removed, and cells that passed through the filter were fixed with 4% paraformaldehyde for 15 min and stained for 10 min with hematoxylin at room temperature. The number of invading cells was counted in five randomly selected high-power fields (200× magnification) under an Olympus IX73 inverted microscope (Olympus Corporation, Tokyo, Japan). The data are representative of three individual wells.

TUNEL assay

TUNEL assays were performed using an Elabscience® Tunel cell apoptosis detection kit (Elabscience, Wuhan, China). Cell smears were fixed with 4% paraformaldehyde for 10 min and washed three times with PBS. Protease K (Elabsciece) and 0.1% Triton X-100 (Sangon Biotech) were added and the mixture was incubated for 5 min. After washing with PBS, the mixture was incubated with balance solution (Elabsciece) for 30 min, and then the prepared TdT working solution (Elabsciece) was added and the mixture was incubated for 60 min. DAPI (Beyotime) was used for staining in the dark.

Subcutaneous tumor formation and peritoneal tumor metastasis assays

Fifteen BALB/C nude mice (Viton Lever, China; 4 weeks old, female, 10–12 g) were raised in a semi-barrier system with constant temperature and humidity, and their drinking water and feed were strictly sterilized. The study was approved by the Animals Ethics Committee of Tongji Hospital, affiliated with Tongji University School of Medicine. The mice were divided into three cell-based groups (five/group) as follows: Ishikawa, Ishikawa shRNA-ANKHD1-BP3, and Ishikawa NC. The cell density in each group was adjusted to 5 × 106 cells/mL, and 0.2 mL was injected subcutaneously into the back of the nude mice. The continuous observation was conducted; the animals were euthanized after 4 weeks, and the tumor was measured.

Another 15 BALB/C nude mice (Viton Lever; 4 weeks old, female, 10–12 g) were divided into three cell-based groups (five/group) as follows: Ishikawa, shRNA-ANKHD1-BP3, and SPC. The cell density was 1 × 106 cells/mL, and 0.2 mL was administered by direct intraperitoneal puncture into the abdominal cavity of the nude mice.

mRNA library construction and sequencing

Total RNA from HEC-1b and HSPB1-HEC-1b cell lines was extracted using Trizol reagent (Invitrogen, CA, USA) following the manufacturer’s procedure. RNA quantity and purity were obtained with a Bioanalyzer 2100 and an RNA 6000 Nano LabChip Kit (Agilent, Santa Clara, CA, USA) with RIN number >7.0. Poly(A) RNA was obtained from total RNA (5 µg) using poly-T oligo-attached magnetic beads using two rounds of purification. Following purification, the mRNA was fragmented into small pieces using divalent cations under elevated temperatures. Then the cleaved RNA fragments were reverse-transcribed to create the final cDNA library in accordance with the protocol for the TruSeq RNA Sample Preparation v2 (Cat. RS-122-2001, RS-122-2002) (Illumina, San Diego, CA, USA). The average insert size for the paired-end libraries was 300 bp (±50 bp). Paired-end sequencing was performed on an Illumina Hiseq 4000 at LC Sciences (Illumina), following the vendor’s recommended protocol.

Chromatin immunoprecipitation sequencing

We used the ChIp Assay Kit (Biyotime) to perform the ChIp Assay in accordance with the manufacturer’s instructions. Cells were cross-linked with 1% formaldehyde for 10 min at 37 °C and quenched with 125 mM glycine for 5 min. The DNA fragments with 200–1000 bp were prepared and then immunoprecipitated with Protein A + G Magnetic beads coupled with anti-HSPB1 (Cell Signaling Technology, Boston, MA, USA) antibodies. After reverse crosslinking, ChIP and input DNA fragments were end-repaired and A-tailed using the NEBNext End Repair/dA-Tailing Module (E7442, NEB) followed by adapter ligation with the NEBNext Ultra Ligation Module (E7445, NEB). The DNA library was amplified in 15 cycles and sequenced using the Illumina HiSeq (Illumina) with 2 × 150 pairs.

Statistical analysis

Statistical analysis was performed using R version 3.5.1 (Institute for Statistics and Mathematics, Vienna, Austria; www.r-project.org) (packages: impute, UpSetR, ggplot2, rms, glmnet, preprocessCore, forestplot, survminer, survivalROC, and beeswarm). SPSS Version 22.0 (IBM, Armonk, NY, USA) was used for experimental analyses. Differences between two groups were assessed with Student’s t-test, while variance was used for three groups. P values were two-sided, and we defined P < 0.05 as statistically significant.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

留言 (0)

沒有登入
gif