Urine collection and capture of exfoliated bladder cells with a membrane filter
Urine samples (100 ml) were collected (n= 26) from the first miction in the morning into a clean sterile container (120 ml, Thermo Scientific Samco). Collection was carried out from healthy volunteers with no known urological diseases. Urine samples were pooled and stored at 4 ° C for up to 72 hours prior analysis. Each donor gave consent before study participation.
100 ml of each pooled urine sample were poured using a 50 ml syringe and passed through a single-use filter cartridge (Filter), presenting a nylon disc filter of 11 µm porosity and 25mm diameter. The Filter was rinsed with 5 ml of 1X PBS (pH 7.4) before being disconnected from the syringe. To avoid saturation, urine sample was passed through the filter under gentle positive pressure.
Urine DNA extraction
Urine DNA isolation has been carried out directly from bladder cells captured on Filter using the QIAamp DNA Blood Mini Kit (Qiagen). If the Filter has been stored at -20°C, it has to be left 5-10 min at room temperature before DNA extraction. 442 µl of lysis buffer (220 µl 1X PBS, 22 µl Proteinase K and 220 µl AL buffer) were added on the filter. Bladder cells have been homogenized by passing 3 times through a syringe equipped with a 21 gauge needle (Terumo) to shear genomic DNA. The lysate was collected into a 2 ml tube and incubated at 56°C for 15 minutes and then centrifuge briefly to reduce foam. Subsequent processing was done according to the DNA purification protocol. All centrifugation steps were carried out in a benchtop microcentrifuge (14,000 RPM) at room temperature. DNA was eluted from the column into a clean 1.5 ml tube by adding 50 ml of AE buffer to the column, then incubated at room temperature for 5 minutes, and centrifuged for 1 minute. DNA concentration was determined with Qubit 4 fluorometer (Invitrogen) using the highly sensitive Qubit quantification assay. All genomic DNA samples were diluted or concentrated to obtain a final concentration of 1.25 ng/μl. We determined the optimal conditions for storage and shipping of the filters before DNA extraction. All samples were examined for DNA (5 ng) integrity via PCR amplification of the GLOBIN gene.
Reproducibility and stability study
We studied the reproducibility of the urine filtration (17 urine samples belonging to Pool 1 to 4). We also studied the stability of filters, recovered after filtration (9 filters belonging to Pool 5 to 7), according to temperature and storage time. This study requires the following steps: DNA extraction from filter, DNA quantification and PCR amplification of the GLOBIN gene.
Bisulfite DNA modification
30 ng of universal methylated human DNA standard (Zymo Research) were modified by sodium bisulfite using the EZ DNA Methylation kit (Zymo Research) according to the manufacturer's instructions. The PCR tubes (0.2 ml) were placed in a thermal cycler and we performed the followings steps: 37°C for 15 min followed by 50°C for 15h30 (overnight). For DNA purification, all centrifugation steps were carried out in a benchtop microcentrifuge (14,000 RPM) at room temperature. Bisulfite modified DNA was eluted from the column into a clean 1.5 ml tube by adding 10 ml buffer M-Elution buffer to the column, then incubated at room temperature for 5 minutes, and centrifuged for 1 minute.
Multiplex Real-time PCR
All PCR reactions were performed on the real-time PCR instrument StepOnePlusTM (Thermo Fischer Scientific) with a final volume of 20 µl. The PCR cycling parameters were: initial denaturation at 95°C for 5 min followed by 40 cycles of 15s at 95°C, 45 s at 60°C. The fluorescence data was acquired at the end of each cycle.
The MASO-PCR technology is presented in Figure 2a. Our technology was performed to simultaneously detect four mutations of the FGFR3 gene (FGFR3mut) with 6Fam-S249C and Vic-Y375C (MASO-PCR1) and 6Fam-R248C and Vic-G372C (MASO-PCR2). PCR was conducted using 5 ng of DNA template (1.25 ng/µl), 1X Quantifast Multiplex PCR (Qiagen), 500 nM of primers (Eurogentec) and 200 nM of probe (Thermo Fischer Scientific) in a final volume of 20 µl (16 µl reaction mix and 4 µl of DNA). The DNA integrity has been checked by amplification of the Ned-GLOBIN gene included as an internal control. The primers and probe sequences are listed in Table 1a.
We used the QM-MSPCR, a highly sensitive and specific PCR developed previously by our team [18]. The QM-MSP technology is presented in Figure 2b. We performed two QM-MSPCR for co-amplification of 6Fam-SEPTIN9 with Vic-ALBUMIN (QM-MSPCR1) and 6Fam-HS3ST2 with Vic-SLIT2 (QM-MSPCR2), respectively. QM-MSPCR reactions were performed with 4 µl of bisulfite-converted positive control DNA (100% methylated) and 16 µl of PCR mix containing 1x KAPA PROBE FAST qPCR Master Mix ABI Prism (KAPA Biosystems), 400 nM primers (Eurogentec) and 250 nM TaqMan-MGB probes (Thermo Fischer Scientific). ALBUMIN sequence has been designed without CpG site and used for normalizing the DNA amounts. All primers are presented in Table 1b.
Detection of FGFR3 mutations using MASO-PCR in patients with NMIBC
We developed a sensitive Mutated Allele Specific Oligonucleotide–PCR (MASO-PCR) assay to detect FGFR3 mutations using multiplex real-time PCR. We selected 263 urine DNA samples, including 176 FGFR3 wild-type (wt) and 87 FGFR3-mutated (mut) previously validated by AS-PCR, from NMIBC patients (AUVES cohort, project reference RECF0998-PHRC 2003) [11]. For initial diagnosis the distribution of patients (n= 57) among low/intermediate and, high-risk NMIBC was 51%/23% and 26%. For follow-up the distribution of patients (n= 30) among low/intermediate and, high-risk NMIBC was 56%/17% and 27%.
The distribution of FGFR3 mutations was:
For initial diagnosis (n= 107): FGFR3mut (n= 57): S249C (n=31), Y375C (n=14), R248C (n=7), G372C (n=3) and R248C/S249C (n=2), and FGFR3wt (n=50).
For follow-up (n= 156): FGFR3mut (n= 30): S249C (n=20), Y375C (n=4), R248C (n=3), G372C (n=2) and R248C/S249C (n=1), and FGFR3wt (n= 126).
MASO-PCR: FGFR3 positive control, primer specififcity, and determination of limit of detection (LoD)
- Construction of the control plasmids containing FGFR3 mutations
Positive control plasmids were designed to incorporate the region of the FGFR3 mutation. The synthetic mutated FGFR3 sequences (FGFR3mut) were synthesized and inserted into pMA-T vector using GeneArt (ThermoFisher Scientific). Each positive control plasmid was confirmed by sequencing before use.
-
FGFR3mut plasmid n°1 (2571 bp): Plasmid pMAT (2374 bp) + FGFR3 sequence with S249C and Y375C mutations (197 bp) (Additional file 1: Figure S1a)
-
FGFR3mut plasmid n°2 (2560 bp): Plasmid pMA-T (2374 bp) + FGFR3 Sequence with R248C and G372C mutations (186 bp) (Additional file 1: Figure S1b)
- Primer pair specificity
We used the same FGFR3 primer pairs (Table 1a) to amplify FGFR3 mutations with the Fast SYBR Green PCR master mix (SG-PCR, ThermoFisher Scientific). PCR reactions were performed in duplicate onto two separated runs. In each 20 µL reaction, G372C, R248C, S249C, and Y375C were amplified with a 1X SG (10 µl), 200 nM of primers and FGFR3mut plasmid (4µl) corresponding in a final volume of 20 µl. The thermal cycling conditions included an initial denaturation at 95°C for 3 min followed by 40 cycles: 95°C for 3 s and 60°C for 20 sec. At the end of PCR reactions, a melting curve analysis was carried out to check the specificity of the primers. Each melting curve was determined by heating the PCR product from 70°C to 95°C and monitoring the fluorescence at a transition rate of +0.3°C. The melting temperature (Tm) was calculated using the StepOnePlus software (Life Technologies) and also estimated by Howley's formula: [67.5 + (0.41* %G-C) - (395/length of amplicon)].
- LoD for analysis of FGFR3 mutations
The diploid human genome comprises about 6.109 base pairs (bp). Plasmids (2.50 ng/µl) were diluted at 2.106 (6.109 divided by 2.6.103) in the standard human DNA (FGFR3wt, 2.50 ng/µl), leading to a 1:1 ratio (FGFR3mut/FGFR3wt) and dilutions were used as FGFR3 positive controls. To determine the LoD, a serial dilution series of the each FGFR3 positive control (see above) was produced at 50%, 10%, 5%, and 1% with FGFR3wt (1.25 ng/µl). 4 µl of each dilution (5 ng) were amplified by MASO-PCR and Cts were analysed by applying a predefined threshold (DRn) at 0.15 for GLOBIN, S249C, Y375C, G372C and 0.24 for R248C, with Cts values above the threshold as positive and below the threshold as negative. All dilutions were amplified and then analyzed in duplicate on the same plate to PCR in two independent runs.
QM-MSP: Positive control and determination of the limit of quantification (LoQ)
The LoQ of each gene was determined on each duplex QM-MSP1 and QM-MSP2 by performing a dilution range with 10, 1, 0.1 and 0.01 ng of bisulfite-converted positive control DNA (100% methylated). The Cts were analysed using a threshold value (DRn) of 0.10 for determining the limit of DNA quantity and amplification efficiency (E). Each dilution was done in duplicate on the same PCR plate in two independent runs.
Stability and reproducibility study of all-in-one PCR master mixes
The all-in-one PCR master mixes were prepared in a single reaction mixture including all the necessary components (PCR, primers and TaqMan-mgb probes) for mutation and methylation assays using MASO-PCR and QM-MSPCR technology, respectively. We studied the stability of each all-in-one PCR master mix (MASO-PCR1,2 and QM-MSPCR1,2) from aliquots that were run in triplicate and stored at -20°C for 0, 1, 2, 3, 4, 6, 9 and 12 months.