TMED9 Expression Level as a Biomarker of Epithelial Ovarian Cancer Progression and Prognosis

Materials and Methods

Patients and tumor specimens. Tissue samples were obtained from patients who underwent primary cytoreductive surgery at the Gangnam Severance Hospital between 1996 and 2012 and the Korean Gynecologic Cancer Bank as part of the Bio and Medical Technology Development Program of the Ministry of the National Research Foundation. A total of 417 tissue samples (namely 212 EOCs, 52 borderline tumors, 83 benign tumors and 70 non-adjacent normal ovarian epithelia) were included in the study. The clinical information of the patients, including age, International Federation of Gynecology and Obstetrics (FIGO) stage, (13) histological classification of tumors based on the World Health Organization grading system (14), surgical procedure, response to platinum-based chemotherapy, examine medical records and pathology reports, data on serum levels of cancer antigen 125 (CA-125), survival duration, and survival status were gathered from electronic medical record. All patients underwent a maximum cytoreductive procedure, then received paclitaxel/carboplatin-based adjuvant chemotherapy. From the date of surgery to the time of relapse or the last follow-up consultation, disease free-survival (DFS) was assessed using the Response Evaluation Criteria in Solid Tumors (RECIST; version 1.1), which is based on the patient’s response to therapy determined through spiral computed tomography or positron emission tomography – computed tomography (15). Overall survival (OS) was assessed from the date of surgery to either the patient’s death or the date of last follow-up consultation with living patients. In compliance with the provisions of the Institutional Review Board (IRB) of Gangnam Severance Hospital, all tumor tissues were histologically investigated by a gynecologic pathologist, and all biological specimens were collected after securing informed consent from the participants (IRB No. 3-2020-0377).

Tissue microarray (TMA) and immunohistochemistry. A TMA was constructed with tissue cores of 1.0 mm in diameter containing a sufficient proportion of tumor cells punched from the donor formalin-fixed paraffin-embedded (FFPE) tissues or tissue blocks using a tissue arrayer (Beecher Instruments, Inc., Silver Spring, MD, USA). The TMA blocks were cut to 5-μm thickness using a rotary microtome. After sectioning, TMA sections were deparaffinized with xylene and rehydrated in serially graded ethanol to distilled water. Antigen retrieval was performed by incubating the TMA sections in a steam pressure cooker (Pascal; Dako, Carpinteria, CA, USA) containing heat-activated antigen retrieval buffer (Dako) at pH 6.0 for anti-TMED9 (rabbit polyclonal antibody, cat. #HPA014650, 1:100; Sigma-Aldrich Co. LLC, St. Louis, MO, USA), and a 3% H₂O₂ solution in methanol was used to treat the sections for 10 min to block endogenous peroxidase activity. The slides were stained for 30 min at room temperature with anti-TMED9 after being rinsed. Subsequently, the antigen-antibody reactions were visualized using Envision+ Dual Link System-HRP (Dako) and DAB+ (3,3’-diaminobenzidine; Dako) for 10 min. The stained slices were deposited in Faramount aqueous mounting solution after being dehydrated, counterstained with hematoxylin, (Dako). Appropriate negative and positive controls were also included.

Evaluation of IHC staining of TMA. A high-resolution optical was used to capture images of the stained TMA sections. (NanoZoomer 2.0 HT; Hamamatsu Photonics K.K., Hamamatsu City, Japan) at 20× objective magnification (0.5 μm resolution). Utilizing Visiopharm software, version 6.5.0.2303 (Visiopharm, Hrsholm, Denmark), the scanned sections were analyzed. On a scale of 0 to 3, the intensity of the brown staining was rated (0=negative, 1=weak, 2=moderate, and 3=strong) and the percentage of cytoplasm-stained tumor cells (range=0-100) was obtained using a predefined optimized algorithm. The total histoscore was determined by multiplying the percentage of positively stained cells by the intensity score (score range=0-300).

Laser-capture microdissection and RNA extraction and quality control. Hematoxylin and eosin were used to stain all FFPE tissue slides, and an experienced gynecological pathologist reviewed the slides. A Leica AS LMD laser microdissection system (Leica Microsystems Inc., Buffalo Grove, IL, USA) was used according to the manufacturer’s instructions to perform a laser-capture microdissection where the desired lesions were cut out from the tissue. The sectioned FFPE tissues were placed on slides with a polyethylene terephthalate membrane (Leica Microsystems, Inc., Buffalo Grove, IL, USA). TRIzol reagent (Invitrogen, Gaithersburg, MD, USA) was used to isolate total RNA, and the quality of the RNA was measured with an Agilent 2100 bioanalyzer using an RNA 6000 Nano Chip (Agilent Technologies, Amstelveen, Netherlands). An ND-2000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA) was used for RNA quantification.

Library preparation and sequencing. For control and test RNAs, library construction was performed using the QuantSeq 3’ mRNA-Seq Library Prep Kit (Lexogen, Inc., Austria), according to the manufacturer’s instructions. Briefly, 500 ng of total RNA was prepared for the control and experimental RNAs. RNA was hybridized with an oligo-dT primer, including an Illumina-compatible sequence at its 5’ end. Reverse transcription was performed, and after the RNA template was degraded, a random primer with an Illumina-compatible linker sequence at its 5’ end was used for second-strand synthesis. Magnetic beads were used to purify the double-stranded library. To perform cluster generation, complete adapter sequences were added to this library, and polymerase chain reaction components were removed from the final library. NextSeq 500 (Illumina, San Diego, CA, USA) was used for single-end 75-base pair high-throughput sequencing.

RNA sequencing read mapping and gene expression analysis. Through sequencing, we obtained FASTQ raw data for eight independent libraries for both platinum-sensitive and platinum-resistant strains; in total, 16 samples were obtained. All FASTQ reads were trimmed and adapters were removed for quality control using BBduk (BBMap v36.59) (16) and FASTQC (v0.11.7) (17) for the sequencing data. We carried out mapping through the STAR-HTSeq workflow, that is, STAR (v2.7.3a) (v2.7.3a) (18) and HTSeq-count (v0.12.4) (19), where the reference genome (GRCh38) and annotations were aligned to the sequencing reads. After quantifying read counts, we performed normalization and differential expression analyses on the levels of gene expression using the DESeq2 v1.26.0 R package (20) where normalization method was applied as a variance-stabilizing transformation. We selected differentially expressed genes (DEGs) that satisfied two different criteria: i) Adjusted p<0.05 (i.e., Benjamini and Hochberg method) and ii) absolute log2 fold-change >2).

Public database. To validate our RNA-seq data, three datasets [GSE190688 (21), GSE54388 (22), GSE 40595 (23)] that analyzed DEGs between normal and high-grade serous ovarian cancer (HGSOC) samples were downloaded from the GEO Database (https://www.ncbinlm.nih.gov/geo/, accessed on December 20, 2021). An online public database, Kaplan–Meier Plotter (http://www.kmplot.com) which contains 10,293 lung, breast, gastric, and ovarian cancer samples to evaluate clinical outcomes of 54,675 genes were used to verify the prognostic significance of TMED9. A PostgreSQL serving running the web tool is able to combine clinical and gene-expression data and information (24).

Cell lines and culture. The American Type Culture Collection (Manassas, VA, USA) and the European Collection of Cell Cultures (Salisbury, UK) provided the primary ovarian cancer cell lines OVCAR3 and A2780, respectively. From the Korea Cell Line Bank, SNU840 cells were obtained (Seoul, Republic of Korea). OVCA433 cells were obtained from the Catholic University of Korea at St. Vincent Hospital. RMUG-S was obtained from the Japanese Collection of Research Bioresources Cell Bank (Osaka, Japan). In Dulbecco’s modified Eagle medium treated with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin, ovarian cancer cells were grown. Two immortalized HOSE cell lines (IHOSE4138 and 8695) were established previously (25) and kept in Dulbecco’s modified Eagle’s medium with 10% FBS and 50 μg/ml gentamicin as supplements. All cell lines were incubated at 37°C in a humidified incubator with 5% CO₂.

siRNA transfection. Bioneer supplied the targeted small interfering RNAs (siRNAs) against TMED9 and the reference siRNA (Daejeon, Republic of Korea). The following were the siRNA combinations: TMED9 #1, 5’-GUGAACAUGCCAAUGACUA-3’ (sense) and 5’-UAGUCAUUGGCAUGUUCAC-3’ (antisense); TMED9 #2, 5’-GUGCAAGUGACUCAGUCAU-3’ (sense) and 5’-AUGACUGA GUCACUUGCAC-3’ (antisense). Cells grown in 6-well plates were transfected with siRNA utilizing Lipofectamine^® RNAiMAX Reagent (Invitrogen, Gaithersburg, MD, USA) as directed by the manufacturer, at a final concentration of 100 pmol per well.

Western blot analysis. Phenylmethylsulphonyl fluoride was administered to a cell lysis buffer (Cell Signaling Technology, Danvers, MA, USA) before the samples were extracted and lysed. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis was used to separate the proteins from the cell lysates and deposit them onto a nitrocellulose membrane. Antibodies against TMED9 and α-actinin were purchased from Atlas Antibodies AB (Bromma, Sweden) and Santa Cruz Biotechnology (Santa Cruz, CA, USA), respectively. Enhanced chemiluminescence substances were used to see the immunoreactive bands (Thermo Fisher Scientific, Waltham, MA, USA).

Cell proliferation assay. Cell proliferation of each cell line was measured using an EZ-Cytox assay kit (Daeil Lab Service, Seoul, Republic of Korea), according to the manufacturer’s instructions. Using a microplate reader, the intensity for each well was determined at 450 nm, and the study was performed in triplicate (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

Colony-formation assay. OVCAR3 and OVCA433 cells were transfected with siRNA targeting TMED9. After 48-h transfection, 500 cells per well were planted into a 6-well plate and the cells were kept in an incubator at 37°C with 5% CO₂ for 2 weeks. The plates were stained with 0.5% crystal violet for 30 min after being fixed with methanol for 10 min. Colonies were counted under a microscope after being rinsed with distilled water. A colony was only counted only when a single clone contained more than 100 cells. The experiment was performed three times.

Boyden chamber assay. Microchemotaxis chambers (Neuro Probe, Gaithersburg, MD, USA) with 48 wells were utilized to assess cell invasion. The bottom cells were filled with culture medium containing 10% FBS before being overlaid with Matrigel-coated membranes (#PFB8; Neuro Probe) from BD Biosciences (San Jose, CA, USA). The upper cells were seeded with siRNA-transfected cells (1×10⁵ cells/50 μl of medium containing 0.05% FBS). The membranes were fixed and dyed with Diff-Quick solution after 24 h (Sysmex, Kobe, Japan). An Axio Imager M2 microscope was used to extract the cells from the membrane’s top surface and measure the invader cells in six randomly selected high-power field for each filter (Carl Zeiss, Thornwood, NY, USA). Three times the experiment was conducted.

Statistical analysis. TMED9 expression levels were evaluated using Mann–Whitney analysis or Kruskal–Wallis test, as necessary. OS and DFS were examined using the Kaplan–Meier technique. The hazard ratios (HR) and confidence intervals for the univariate and multivariate designs were also computed using the Cox proportional hazards approach. SPSS version 25.0 was used to conduct statistical analysis (IBM, Armonk, NY, USA). Statistical significance was defined as p<0.05 was considered statistically significant.