Tumor specimens were derived from an 89-year-old Chinese woman treated in the absence of chemotherapy and radiotherapy, and had no family history. The patient had moderately/poorly differentiated right colon invasive ulcerative adenocarcinoma with liver metastasis as stage 4bN1bM1a, IV A. The right hemicolectomy of this patient was performed at the Department for Surgical Oncology of Nanjing First Hospital affiliated to Nanjing Medical University (Nanjing, China) in March 16, 2015. Following the surgical procedure, specimens were immediately obtained from primary tumor tissues, and prepared for cell culture and pathological analysis. Informed consent was obtained from the patient and the experiments were approved by the Ethics Committee of Nanjing First Hospital of Nanjing Medical University in accordance with generally accepted guidelines for the use of human material.

The tumor tissues were immersed in Dulbecco's PBS (DPBS; Gibco; Thermo Fisher Scientific, Inc.) with 1,000 U/ml penicillin G, 1,000 µg/ml streptomycin and 2.5 µg/ml amphotericin B (DPBS with 10X Antibiotic-Antimycotic). Subsequently, the tumor tissues were washed twice with transport medium and immersed in DPBS with 10X Antibiotic-Antimycotic for 1-2 h at 4°C. Following two washes with transport medium, the tissues were ground. The samples were plated in a sterile 6-cm culture dish and cultured in RPMI-1640 supplemented with 10% heat inactivated fetal bovine serum (both from Gibco; Thermo Fisher Scientific, Inc.) and 1X Antibiotic-Antimycotic (RPMI-1640 complete medium), and maintained in humidified incubators at 37°C in an atmosphere of 5% CO₂ and 95% air. After 2 weeks, the initial cell passage was performed when increased tumor cell growth was observed. Differential trypsinization was used to remove fibroblasts; a pure tumor cell population was obtained comprising the cc-006cp cell line (21).

Subsequently, the monoclonal cell subline derived from the cc-006cp cell line was plated onto semisolid medium. A total of 10³ viable cc-006cp cells were plated in 3 ml of RPMI-1640 complete medium containing 0.3% agarose layered over RPMI-1640 complete medium containing 0.5% agarose; the cells were incubated at 37°C with 5% CO₂. Colony formation was observed after 12-15 days, following which single colonies were selected and cultured with RPMI-1640 complete medium in an incubator at 37°C with 5% CO₂. Half of the media was replaced every 2-3 days depending on the pH. Using this method, a cell line of high purity and tumorigenicity was successfully derived from human moderately/poorly differentiated colon cancer; the cell line was stably passaged and was designated as cc-006cpm8, which is one of eight sublines of the cc-006cp cell line.

For electron microscopy, 1-5×10⁵ of the cc-006cpm8 cells plated in a 35-mm dish were fixed with 2.5% glutaraldehyde in 0.1 mol/l phosphate buffer for 1 h at room temperature, and post-fixed with 1% OsO₄ in the aforementioned buffer at 0°C for 30 min. The cells were rinsed with the same buffer, dehydrated with ethanol, immersed twice in absolute propylene oxide and embedded in Quetol 812. Sections were cut at a thickness of 100 nm with a diamond knife and mounted onto grids. Following staining with uranyl acetate and lead citrate, the cells were observed using a JEM-1011 electron microscope at 80 kV.

The cc-006cpm8 cells were treated with 0.075 mol/l KCl solution for 20 min at 37°C, and then fixed with a solution of methanol: Acetic acid (3:1) solution at 0°C for 10 min. Following fixation, the slides were place in trypsin solution (0.02% trypsin +3% Tris solution) in a water bath at 37°C for 3 min. The cells were then stained with 10% Giemsa solution for 4 min at room temperature. The karyotypes of 20 randomly selected cells were analyzed using the GSL 120 Imaging System equipped with Zeiss.

The chamber slides of the cc-006cpm8 cells were washed with PBS (pH 6.5), and then placed 3-5 drops of Wright-Giemsa Stain (code no. CO 135; Shanghai BOYAO Biotechnology Inc.) upon the slides for 1-2 min at room temperature, and added 6-10 drops of PBS (pH 6.8) to mix Wright-Giemasa stain for 6-8 min, and the stained slides were then rinsed with water and quickly air-dried.

The cc-006cpm8 cell line, together with the positive and negative controls, were amplified using the GenePrint 10 system (Promega Corporation). The amplified products were then processed using the ABI3730×l Genetic Analyzer (Applied Biosystems; Thermo Fisher Scientific, Inc.). Data were analyzed using GeneMapper4.0 software (Applied Biosystems; Thermo Fisher Scientific, Inc.) and then compared with the American Type Culture Collection (ATCC), Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ) or Japanese Collection of Research Bioresources (JCRB) databases for reference matching. The STR assay was conducted at the China Center for Type Culture Collection (CCTCC).

For growth curve and population doubling time (PDT) analyses, 1×10³ cells/well were plated onto 24-well culture plates (Corning Incorporated) and cultured for 8 days in a CO₂ incubator. The cells were counted every hour in five wells using a JuLi Stage real-time cell counter. Alterations in cell confluence over time were presented as a growth curve. The PDT of cells in the logarithmic phase was calculated as follows: PDT = (t-t₀) [lg2/(lgNt-lgN₀)] (N₀ and Nt represent cell numbers at t₀ and t separately).

To determine the plating efficiency, the cc-006cpm8 cells were plated at various densities, ranging between 1×10² and 5×10² cells in three replicate 60-mm dishes (Corning^® Plastic Culture Dishes; Corning Incorporated). Colonies comprising >50 cells were counted 9 days later following staining with 0.5% crystal violet for 5 min at 25°C, and then counted under BX-4 Stereomicroscope (Shanghai Optical Instrument Factory).

cc-006cpm8 single cell suspensions (10³ cells in three replicate 60-mm dishes) with 0.3% agarose dissolved in RPMI-1640 complete medium were placed in tissue culture dishes (Corning^® Plastic Culture Dishes; Corning Incorporated) on a layer of 0.6% agarose (in the same medium), which had previously set. The dishes were then transferred to a 37°C incubator with 5% CO₂. After 2 weeks, colonies comprising >50 cells were counted and the number was divided by 1,000 to determine the colony-forming efficiency. Experiments were conducted in triplicate.

Tumorigenicity was investigated by inoculating 5×10⁶ cells/site cc-006cpm8 cells subcutaneously into the shoulder and contralateral hip of five BALB/c male nude mice, which were 4-6 weeks old and weighing 20-25 g, purchased from the Comparative Medical Center of Yangzhou University (Yangzhou, China). All mice were maintained in a HEPA-filtered environment at 24-25°C and humidity was maintained at 50-60%. Observations were conducted at least twice a week via caliper measurements to analyze progressive tumor growth. The study was performed in the Laboratory Animal Center of Southeast University (Nanjing, China), and followed the Constitution of the Laboratory Animal Ethics Committee of Southeast University. The animal experimental ethical inspection form was approved by the Animal Care and Welfare Committee of Southeast University.

Mycoplasma contamination was assayed using the Myco-PCR-Mix PCR Detection kit (Zhong Qiao). Cell lines should be pre-cultured in the absence of mycoplasma active antibiotics for several days to maximize test sensitivity. Samples should be derived from cultures that are at 90-100% confluence. Cell culture supernatants can be tested directly without prior preparation. According to the reaction ratio, the sample/Positive Control template/Negative Control template is added respectively to Myco-PCR-Mix in the PCR tubes. The thermal cycler program was as follows: the reaction is started with an initial denaturation for 5 min at 95°C; followed by 30 sec at 95°C, 45 sec at 55°C, 45 sec at 72°C for 36 cycles; and cooling down to 10°C. The PCR products were then electrophoresed on 1.5% agar gel. The results of electrophoresis can be observed directly under ultraviolet light after staining with ethidium bromide (EB). If there is a 250 bp band of sample same to positive control band size, it is indicated that the sample was infected by mycoplasma. This method has been reported to have high sensitivity in the detection of mycoplasma in cell cultures and other cell culture-derived biologicals.

The cell line and the original colon cancer tissues were analyzed by immunohistochemistry. The cc-006cpm8 cells were collected by centrifugation at 220 × g for 3 min; the pellets were fixed in 10% formaldehyde, embedded in paraffin and then sliced into 6-µm sections. The sections were stained using the EnVision system. In the original colon cancer tissues, IHC staining was performed on formalin-fixed paraffin-embedded tissues. The paraffin sections were dewaxed in xylene and rehydrated with distilled water. The slides were subsequently incubated with the following antibodies for 60 min at room temperature: Anti-vimentin (cat. no. MA3-745, Invitrogen; Thermo Fisher Scientific, Inc.; 1:20), anti-pan cytokeratin (cat. no. ab215838; 1:400), anti-epithelial membrane antigen (EMA; cat. no. ab140355; 1:150), anti-carcinoembryonic antigen (CEA; cat. no. ab4451; 1:100) (all from Abcam), anti-p53 (cat. no. sc-47698, Santa Cruz Biotechnology, Inc.; 1:250) and anti-Ki-67 (cat. no. ab8191; 1:50), anti-mucin 2 (cat. no. ab76774; 1:50), anti-caudal type homeobox transcription factor 2 (CDX2; cat. no. ab70458; 1:1,000), anti-mutL homolog 1 (MLH1; cat. no. ab92312; 1:500), anti-mutS protein homolog (MSH2; cat. no ab70270; 1:2,500), anti-mutS homolog 6 (MSH6; cat. no. ab92471; 1:500), anti-postmeiotic segregation 2 (PMS2; cat. no. ab110638; 1:200), anti-BRAF V600E (cat. no. ab200535; 5 µg/ml) (all from Abcam), anti-α-fetoprotein (AFP; cat. no. PA5-11480, Invitrogen; Thermo Fisher Scientific, Inc.; 1:1,0000), anti-glypican-3 (cat. no. ab207080; 1:500), anti-hepatocyte-specific antigen (cat. no. ab75677; 1/1) (both from Abcam). The reaction was performed using Autostainer Link 48 (Dako; Agilent Technologies, Inc.). Secondary antibodies (Dako EnVision+, HRP rab/mouse, K5007; Dako; Agilent Technologies, Inc.) were incubated for 20 min and 3,3′-diaminobenzidene substrate chromogen counterstaining for 4 min at room temperature.

For the cell line, the percentage of stained positive cells was scored via a semiquantitative method according to the following scheme: 0 (<5% of tumor cells); 1+ (5-25% of tumor cells); 2+ (26-50% of tumor cells); 3+ (51-75% of tumor cells) and 4+ (>75% of tumor cells). For the tumor tissues, the slides for IHC were observed independently of the diagnosis established from the HES-stained slides. For the slides of colon cancer tissue, the positive index of IHC was determined in regions of high proliferation. Automated detection was conducted using a Dako AutoStainer Link 48. Normal expression was defined as nuclear staining within tumor cells, using infiltrating lymphocytes as the positive internal control. Negative protein expression was defined as the complete absence of staining within tumor cells compared with the positive labeling of internal non-neoplastic tissues.

The cc-006cpm8 cells were cultured in a 100-mm plate and then separated by trypsin-EDTA when the cells attained 90% confluence. Subsequently, the cells were centrifuged at 300 × g for 10 min at room temperature and 10⁷ cells/100 µl were suspended (PBS pH 7.2, 2 mM EDTA and 0.5% bovine serum albumin; Sigma-Aldrich). The cells were then divided into two groups; one group was stained with human cluster of differentiation 44-fluorescein isothiocyanate (CD44-FITC; cat. no. 130-098-210) and isotype control antibody mouse IgG1-FITC (cat. no. 130-098-847) (both from Miltenyi Biotec) at 4°C for 10 min. The other group of cells was stained with human EpCAM-phycoerythrin (CD326-PE; cat. no. 130-098-115) and isotype control antibody mouse IgG1-PE (cat. no. 130-098-845) (both from Miltenyi Biotec) under the same conditions. Analysis was performed with a BD FACSCanto II flow cytometer (BD Biosciences).

Exome capture sequencing was performed using the Human Whole-Exon Capture (xGEN^® Exome Research Panel) kit of Integrated DNA Technologies and the Illumina Hiseq sequencing platform. The High-Throughput Information Acquisition Technique was used to collect the captured DNA fragments. Each sample produced 10 Gb clean data on average via the PE150 sequencing strategy. The obtained data were annotated and counted for SNVs and InDels.

Dissociated cell suspensions from nine freshly resected colorectal carcinoma tissues of various histological grades were collected for the isolation of cancer cells as aforementioned; two of the specimens were severely contaminated with bacteria and were discarded. A total of five specimens were primary cancer cells derived from a finite population of cells (~5-6 passages) (22). Only two specimens of resected colorectal tumor were used to establish cell lines; one cell line was cc-006cpm8 and was established in the present study for cellular characterization. In addition, the patient was diagnosed with invasive and ulcerative moderately/poorly differentiated adenocarcinoma of the right colon according to postoperative pathology, as demonstrated by H&E staining of the primary tumor (Fig. 1A and B).

The initial isolation of the primary tumor cells and their propagation in vitro in RPMI-1640 complete medium for 8 days was conducted until those clones were observed. Numerous different cellular shapes and clone morphologies were reported: i) Carcinoma-associated fibroblasts and normal fibroblasts; ii) sparse large semispheroid clusters with dichotomized processes; and iii) small closely-packed clones (Fig. 2A and D). Following differential trypsinization, the cc-006cp cell line possessed an epithelial phenotype without contaminating fibroblasts; few clones were packed closely together, whereas others were not (Fig. 2B and E), and grew slowly compared with the cc-006cpm8 cell line (Fig. S1). From further subcloning, cc-006cpm8 cells were selected as the monoclonal cell subline from the cc-006cp cell line in semisolid medium (Fig. 2C and F).

In RPMI-1640 complete medium, the cc-006cpm8 cells were cultured as several layers with tight adhesion to the plastic surface and no contact suppression via microscopy (Fig. 3A). Following culture on chamber slides and Wright-Giemsa staining, large nuclei, a thin layer of cytoplasm and numerous nucleoli were observed in the cc-006cpm8 cells via microscopy (Fig. 3B). Under a transmission electron microscope, the cc-006cpm8 cells exhibited a prominent nucleolus, small irregular microvilli, notable mitochondria and endoplasmic reticulum. Few lipid droplets and vacuoles were observed (Fig. 3C and D).

The cc-006cpm8 cells were passaged continuously >100 times over a period of 1 year; the cells exhibited maintained vigorous growth. From the growth curve of the cc-006cpm8 cells, the average PDT was determined to be 27 h (Fig. 4A). To analyze the proliferation of this cell line, soft agar colony and clone formation assays were performed. The former revealed that the colony formation rate of the cc-006cpm8 cells was 73.2% (Fig. 4B) and that of clone formation was ~66.5%; the values of which were obtained from a small number of the 100 cells cultured in 6-mm dishes for 9 days. The resulting clones were stained with crystal violet and colonies constituting >50 cells were counted (Fig. 4C). The findings of these assays revealed the proliferative properties of cc-006cpm8 cells.

Athymic nude xenograft mice were used in the present study to investigate the tumorigenicity of cc-006cpm8 cells in vivo. Tumor formation was analyzed at least twice a week via caliper measurements; tumors were detected in all mice after 4 weeks (Fig. 5A and B). These xenograft tumors exhibited similar histology to the primary tumor samples obtained from the patient, as determined by H&E staining (Fig. 5C and D). The xenograft tumor was nodular and adhered to the skin.

To confirm the origin of the cc-006cpm8 cell line, flow cytometric analysis was performed. The results revealed the proportion of cancer stem cells (CSCs) in the subline. Compared with the control, there were ~99.9% of EpCAM(+) cells in the cc-006cpm8 population, which suggested that cc-006cpm8 cells were of epithelial origin; however, compared with the control, there were 59.6% of CD44(+) cells in the cc-006cpm8 population (Fig. 6). The results may indicate that heterogeneous expression was also observed within the same population of cc-006cpm8 cells.

To confirm the epithelial origin of the cc-006cpm8 cell line associated with the original tumor sections, the expression of keratin and vimentin were investigated via immunocytochemistry. These results demonstrated the positive expression of pan-keratin (Fig. 7A) and negative expression of vimentin (Fig. 7B). These findings suggested the epithelial origin of the cc-006cpm8 cell line. On further observation of the cytokeratin (CK)7/CK20 profile, CK7⁺/CK20⁻ expression was present in this cell line and the corresponding original tumor (Fig. S2). Additionally, the expression of certain colon cancer markers in cc-006cpm8 cells and the original tumor sections were determined. Notable positive expression of CDX2 and CEA (Fig. 7C and D), and mucin 2 (Fig. E) was observed. Furthermore 50% of the cells were positively stained for Ki-67 (Fig. 7F); however, negative staining for p53 in the cell line and the tumor sections from the patient was reported (Fig. 7G). In addition, positive staining for EMA was detected in the original tumor sections from the patient. By contrast, the cc-006cpm8 cell line exhibited 1% positive staining (Fig. 7H). The results further suggested the biological characteristics and cancer heterogeneity in this cell line. The expression of DNA MMR proteins was also determined by immunocytochemistry in the cell line and the patient tumor sections, respectively. The results revealed the negative expression of MLH1 and PMS2 and notable positive expression of MSH2 and MSH6 (Fig. 8). Within MSI-H colorectal cancer, the BRAF V600E mutation is associated with sporadic tumorigenesis (23). IHC analysis of BRAF V600E in the cell line and the patient tumor sections was performed, and positive expression was reported in the two sample types (Fig. 8). As the specimens were derived from a patient with right colon adenocarcinoma and liver metastasis, the expression of AFP, HepPar-1 and glypican-3 were determined via IHC in the cc-006cpm8 cell and the tumor sections; the antibodies used are those used in the diagnosis of hepatocellular carcinoma. Therefore, this may eliminate the possibility that the cc-006cpm8 cells and patient tumor sections were of hepatic origin as negative expression was reported (Fig. S3).

Chromosomal analysis of the cc-006cpm8 cell line was performed at passages 5, 10 and 25, in which 20 karyotypes were investigated using Giemsa banding. The results revealed that the modal number of chromosomes in three different passages was similar, indicating chromosomal stability. The modal number of chromosomes in this cell line was 50, which accounted for 84.2% of cells; the number of chromosomes ranged between 47 and 53. Therefore, the cell line was reported as hyper-diploid. In cells of passage 10, the karyotypes of the clone with 50 chromosomes were xx, +7, +9, +9, +12, −15 and +18 (Fig. 9A and B). Other karyotypes were also observed, including +8, +10, +12, +13, +14, +16, +19, +21 and +x. Collectively, the frequencies of +7, +8, +9, +12 and +18 were higher. The results suggested that this cell line was characteristic of malignant tumor cells.

The cc-006cpm8 cell line was subjected to STR profiling by the CCTCC following the ANSI/ATCC ASN-0002-2011, Authentication of Human Cell Lines: Standardization of STR Profiling. The STR analysis of this cell line resulted in unique human profiles, which did not completely match the profiles in the ATCC, DSMZ or JCRB databases, or the National Experimental Resources Cell Sharing Platform.

Exome capture DNA sequencing was performed on the cell line; SNVs and InDels were counted from the data. As presented in Table I, 27,134 total SNVs and 34,860 total InDels were reported in the cc-006cpm8 cell line. Among the mutated genes, 1,510 novel SNVs and 32,658 novel InDels were reported. In addition, 20 hypermutated exons containing SNPs and InDels were detected in this cell line, including SNVs of MUC19, MUC16, MUC12, FLG and AHNAK2, and InDels of DEFB126, MIR3665, WNK lysine deficient protein kinase 1 (WNK1) and SLAIN motif-containing 1 (SLAIN1) (Fig. 10A and B). Additionally, 24 commonly mutated genes associated with CRC were reported; these mutations may occur in the coding sequences of genes in this cell line(Table II) (20). The genes associated with the DNA MMR pathway were analyzed. MLH1 and MSH6 revealed no mutations; however, MLH3, MSH2, MSH3 and PMS2 exhibited missense mutations (24-26). Furthermore, BRAF, APC, TP53, PIK3CA, ERBB2 and TCF7L2 possessed missense mutations comprising SNPs, and InDels of TGFBR2, ATM and APC, which occurred via frameshift mutations. By contrast, KRAS, NRAS, SMAD4, SMAD2, ACVR2A, FBXW7, CTNNB1, ARID1A, FAM123B and SOX9 exhibited no missense or frameshift mutations in the coding regions.