Clinical specimens
Fresh-frozen primary tumors and paired adjacent nontumor tissues from nine BCa patients (Stage T1, n = 5; T2, n = 1; T3, n = 1; T4, n = 2) were collected at the time of surgery. All frozen tissue specimens were obtained from the Chungnam National University Hospital. Each tumor specimen was histologically verified by a board-certified pathologist and archived for further DNA study. A total of 51 voided urine samples, used for bisulfite-pyrosequencing verification, were freshly obtained from BCa patients with varying stages (n = 16), patients with benign urologic diseases (BUD) such as trigonitis, urinary stone, and benign prostate hyperplasia (n = 23), and healthy individuals (n = 12). In the clinical validation for the urine DNA-based methylation test, an independent set of fresh voided urine samples were obtained from patients with BCa (n = 55) at various stages (Ta – T4), patients with BUD (n = 25), and normal healthy subjects (n = 81) as shown in Table 1. Additionally, urine samples from patients with kidney (n = 6) or prostate cancers (n = 2) were also included. All voided urine samples from BCa patients were collected before definitive surgery. Normal healthy control samples were obtained from individuals without any history of genitourinary malignancy. Voided urine samples (40 mL each) were collected into 50 mL tubes containing preservative buffer (Genomictree, Inc. Daejeon, South Korea), and were then centrifuged at 3000 x g for 10 min. The pelleted urine sediment was stored at -20°C until DNA extraction. This study was approved by the Institutional Review Board of ChungNam National University Hospital, Daejeon, South Korea. Written informed consent was obtained from all study participants. This study adhered to local ethics guidelines.
Table 1
Clinicopathological features of subjects enrolled in this study
Characteristics | Tissues | Urine samples |
Healthy control | - | 93 |
Sex – no. (%) | | |
Male | - | 60 (64.5) |
Female | - | 33 (35.5) |
Age, mean (range) | - | 53.8 (26–85) |
BUD | - | 48a |
Sex – no. (%) | | |
Male | - | 29 (60.4) |
Female | - | 19 (39.6) |
Age, mean (range) | - | 52.5 (34–83) |
Bladder cancer (BCa) | 9 | 71 |
Sex – no. (%) | | |
Male | 7 (77.8) | 55 (77.5) |
Female | 2 (22.2) | 16 (22.5) |
Age, mean (range) | 74.3 (62–81) | 68.8 (33–85) |
Pathological stage – no. (%) | | |
Ta | - | 28 (39.4) |
T1 | 5 (55.6) | 35 (49.3) |
T2 | 1 (11.1) | 4 (5.6) |
T3 | 1 (11.1) | 2 (2.8) |
T4 | 2 (22.2) | 2 (2.8) |
Differentiation grade – no. (%) | | |
Low | 4 (44.4) | 35 (49.3) |
High | 5 (55.6) | 33 (46.5) |
Unknown | - | 3 (4.2) |
a Benign urologic diseases (BUD) included trigonitis, urinary stone, and benign prostate hyperplasia. |
Cpg Methylation Microarray Analysis
CpG methylation microarray analyses were performed using genomic DNA isolated from primary tumors and paired adjacent nontumor tissues from nine BCa patients with different stages (T1, n = 5; T2, n = 1; T3, n = 1; and T4, n = 2). CpG methylation microarray analysis was conducted as described previously [22] using human 244k CpG island microarrays containing 237,000 oligonucleotide probes covering 27,800 annotated CpG islands (Agilent Technologies, CA, USA) according to the manufacturer’s instructions. Raw methylation microarray data were submitted to Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo) with accession number GSE171369.
Methylation microarray data were analyzed using the Agilent Feature Extraction software version 9.3.2.1 and a GeneSpring software version 7.3.1 (Agilent, CA, USA). To determine differentially methylated targets between primary tumor and paired adjacent nontumor tissue samples, statistical analysis was performed using a parametric analysis of variance test with Benjamini and Hochberg multiple testing correction (P < 0.01), followed by fold change analysis. Next, multiple-probe enriched genes were further selected as methylation candidate genes if their probes yielded a positive call for methylation in the bladder primary tumor compared to non-cancerous tissues with at least two probes.
Dna Extraction And Bisulfite Treatment
Genomic DNA from tissue specimens were extracted using a QIAamp DNA Mini kit (Qiagen, Hilden, Germany). Genomic DNA from urine sediments were isolated using a GT NUCLEIC ACID PREP kit (Genomictree, Inc., Daejeon, South Korea) according to the manufacturer’s instructions.
For methylation assessment, purified genomic DNA were first bisulfite-treated to convert unmethylated cytosine nucleotides into thymidine without changing methylated cytosines using an EZ DNA Methylation-Gold kit (Zymo Research, CA, USA) according to the manufacturer’s instructions. Briefly, DNA was chemically modified with sodium bisulfite at 64°C in the dark for 2.5 h and then, the bisulfite-modified DNA was purified and eluted in 10 µL of distilled water. Eluted DNA was either immediately used for methylation analysis or stored at -20oC until the analysis.
Methylation Assessment By Bisulfite-pyrosequencing
To assess methylation status of candidate genes, bisulfite-pyrosequencing was performed as previously described [23]. Bisulfite PCR and pyrosequencing primers were designed to amplify 3 to 5 CpG dinucleotides sites in target regions of genes using a PSQ Assay Design software (Qiagen, Hilden, Germany). Sequences of primers used in pyrosequencing are listed in Table 2. These primers were synthesized by Bioneer (Daejeon, South Korea). Genomic DNA was modified by sodium bisulfite using an EZ DNA Methylation-Gold kit (Zymo Research, CA, USA) according to the manufacturer’s instructions. Briefly, 20 ng of bisulfite-treated DNA was amplified in a 25 µl reaction with primer set and Taq polymerase (Enzynomics, Daejeon, South Korea). PCR amplification was run for 40 cycles with an optimal annealing temperature.
Table 2
Primer sequences for pyrosequencing and mePENK-qMSP
Pyrosequencing |
Gene | Sequences (5’ ◊ 3’)a | Amplicon Size (bp) |
CDX2 | F: TGGTGTTTGTGTTATTATTAATAG R: Biotin-CACCTCCTTCCCACTAAACTA S: ATTAATAGAGTTTTGTAAATAT | 129 |
CEI | F: TGGAAATGTAAGTAGTTTTAGTGTAT R: Biotin-AAATTTCTTAACCAAACTTCTCATAT S: TGTAAGTAGTTTTAGTGTATTAAAT | 152 |
DMC1 | F: GAGGGGGGTAAGTGGTAAAAA R: Biotin-TCCCTCAAAATCACTAAA ATTCCT S: GGGGTAAGTGGTAAAAA | 165 |
IMP-1 | F: GGATTTYGAAAYGTTATTATTTAATAG R: Biotin-AACTAAAAACRAAATATCCCAAT S: ATTTYGAAAYGTTATTATTTAATAG | 126 |
PDE3A | F: TGGGAATTTAGTGAAGAG R: Biotin-CCACTATAAACCAACTTATCCCTAACT S: GGGTATTTTATATTATGGTAGTG | 84 |
PENK | F: ATATTTTATTGTATGGGTTTTTTAATAG R: Biotin-ACAACCTCAACAAAAAATC S: GGGTGTTTTAGGTAGTT | 322 |
SIM2 | F: Biotin-GTGGATTTAGATTAGGATTTTGT R: CACCCTCCCCAAATTCTT S: CCTCCCCAAATTCTTC | 205 |
VSX1 | F: GGAGTGGGATTGAGGAGATTT R: Biotin-AGTAAGTTTATGGGAGGGGGATT S: TTTTTGAAATGTTGGTAAG | 89 |
ZNF312 | F: AAGAGGGATTTGGAGAGAGAA R: Biotin-TCTCAATACACCCAACCTACATAC S: GATTTGGAGAGAGAAGG | 140 |
mePENK-qMSP |
Gene | Sequences (5’ ◊ 3’)a | Concentration |
COL2A1 | F: GTAATGTTAGGAGTATTTTGTGGGTA R: CTACCCCAAAAAAACCCAATCCTA P: Cy5-AGAAGAAGGGAGGGGTGTTAGGAGAGG | 0.2 µM 0.2 µM 0.1 µM |
PENK | F: TCGGGTGTTTTAGGTAGTTTCGC R: ACGACTCAAATCGCCTCGCG P: Fam-TGGGGGCGATCGCGTTATTTCGG | 0.2 µM 0.2 µM 0.1 µM |
aF, R, S, and P represent forward, reverse PCR primers, sequencing primers, and PCR probe, respectively. Biotin, Cy5 or Fam indicates 5’ biotinylation, 5’Cy5 conjugation, and 5’Fam conjugation, respectively. |
Pyrosequencing was performed using a PyroGold kit and a PyroMarK ID Q96 instrument (Qiagen, Hilden, Germany) following the manufacturer’s instructions. Methylation index (MtI) of each gene in each sample was calculated as the average value of mC/(mC+C) for all examined CpGs in target regions. All pyrosequencing reactions included samples without any DNA template as negative controls. Methylation-positive was considered if MtI of primary tumor was greater than that of the corresponding nontumor tissue.
PENK methylation assessment in urine DNA by real time PCR
To measure PENK methylation quantitatively in DNA of urine sediment, total genomic DNA obtained from the sediments of voided urine was used for bisulfite treatment. All purified bisulfite-treated DNA was subsequently subjected to real time PCR-based methylation assessment for PENK (named as mePENK-qMSP). Primers and probes were designed to amplify the target region of PENK covering CpG targets (72 bp; +524 to + 595 bp) and were synthesized by Integrated DNA Technologies (IDT) (IA, USA). mePENK-qMSP assay have been established with a modified protocol based on the report described by Eads et al. [21] in which the fluorescence-based qMSP quantitated the original methylation level of the interested gene locus using the bisulfite-converted sequences-specific primers and probes. Region of COL2A1 DNA having no CpG sites was used for methylation-independent amplification as a control to determine the presence of bisulfite-treated DNA [24].
PCR reaction mixture contained 5x AptaTaq PCR master mix (Roche Diagnostics, Mannheim, Germany), PENK methylation-specific primers and probes, and COL2A1-specific primers and probes (Table 2). mePENK-qMSP assay was performed on a 7500 Fast System Real-Time PCR (Thermo Fisher Scientific, MA, USA). Real time PCR was performed with the following thermal cycling conditions: activation at 95°C for 5 min, 40 cycles of 95°C for 15 sec, and 60°C for 1 min. Heating and cooling rates were set to ≥ 4°C per sec and ≥ 3.5°C per sec, respectively. For each experiment, BCa cell (RT4) genomic DNA containing fully methylated PENK gene and whole genome amplified genomic DNA of human lymphocyte containing unmethylated PENK gene [25] were used as controls to validate mePENK-qMSP adequacy of each sample batch. Non-template controls were also included for each experiment to detect cross-contamination. Cycle threshold (CT) values were analyzed using the 7500 software (Thermo Fisher Scientific, MA, USA). For urine DNA testing, CT values for each experimental set were determined using a manually configured cutoff value. This cutoff value was established using the unmethylated DNA fluorescence level as a baseline. mePENK-qMSP was performed at one time for each sample. The relative level of methylated PENK gene in each sample was determined as 40-△CT [CT of amplified PENK gene – CT of COL2A1 (reference gene)] [26]. Higher values of 40-△CT represented higher levels of PENK methylation. If CT of PENK was undetected, the value was considered to be 25, the nearest value to the lowest of 40-△CT for test results of all samples.
Statistical analysis
All statistical analyses were performed using MedCalc software, version 9.3.2.0 (Basel, Belgium). A P value of less than 0.05 was considered statistically significant. Receiver operating characteristic (ROC), area under ROC (AUC), and 95% confidence intervals (CI) were calculated to confirm the accuracy of diagnosis, sensitivity, and specificity. Samples were categorized as negative or positive based on the cutoff value determined by the ROC curve analysis of the assay results.