Samples
We tested for TP53 codon 72 polymorphism specimens from a sub-cohort of 978 Brazilian women enrolled in the Ludwig-McGill cohort study [13]. The Ludwig-McGill Cohort Study is an ongoing investigation of the natural history of HPV infection and the risk of cervical neoplasia. This study adhered to the tenets of the Declaration of Helsinki. All participants entered the study only after giving signed informed consent. All study procedures and the informed consent were approved by the institutional review boards and ethical committees of the Ludwig Institute for Cancer Research and the Maternidade Escola Vila Nova Cachoeirinha clinic, both in São Paulo, Brazil [13]. DNA was extracted from exfoliated cervical cells and purified by spin column chromatography. Four different methodologies were used to detect TP53 codon 72 polymorphism: DHPLC, RFLP using two different restriction enzymes - Bst UI and Bsa JI and Dot-blot hybridization. Because TP53 codon 72 polymorphism is located at TP53 exon 4, previous amplification of this region was performed. The resultant amplicon was then tested by the methodologies described below.
Amplification of TP53 exon 4
To establish and optimize TP53 codon 72 genotyping we performed a previous amplification of TP53 exon 4. Since this exon encodes a polyproline domain [14] GC-rich regions are frequently found. This condition favoured the generation of primer-dimers during amplification leading to low yield of PCR product. To avoid this problem we used hot start PCR to increase specificity [15]. In addition, because optimal primers annealing temperatures differ for more than 6°C we also used touchdown PCR to prevent non-specific extensions. The combination of both PCR strategies improved the yield of amplification and a single fragment of 279 base pairs was obtained (Figure 1A). The total reaction volume for DNA amplification was 50 μL. Each reaction was carried out using 0.5 μM of each of the two primers: TP53 +, 5'-TCCCCCTTGCCGTCCCAAG-3'; and TP53 -, 5'-CGTGCAAGTCACAGACTT-3'; in a mixture containing 10 mM Tris-HCl (pH 8.5), 50 mM KCl, 1.5 mM MgCl2, 200 mM of each deoxiribonucleotides, and 2 U of Taq Gold DNA Polymerase (Applied Biosystems -- Roche, New Jersey, NY). All other reagents were purchased from GIBCO BRL (Gaithersburg, MD). One microliter of DNA was added to the reaction. A negative control containing no DNA was also included with each series of reactions to check for contamination. For controls, we included 50 ng of DNA from SW756 cells and from a penile condyloma, known to be heterozygous and homozygous TP53 -Pro in codon 72, respectively. Reactions were conducted in a TC-341 thermal controller (Amersham Pharmacia Biotech, Buckinghamshire, England) with the following amplification profile: The following amplification profile was employed: 95°C for 10 min, 10 cycles of 95°C for 1 min, 58°C for 1 min and 72°C for 1 min followed by 15 cycles of 95°C for 1 min, 56°C for 1 min and 72°C for 1 min and an additional round of 15 cycles of 95°C for 1 min, 53°C for 1 min and 72°C for 1 min. A final extension step was added including 72°C for 3 min followed by 95°C for 10 min. Amplification products were run on 1.5% agarose gels containing 3 mg/ml of ethidium bromide at 90--100 V for 20--30 min. Bands were visualized and photographed on a UV transilluminator. Positive reactions were selected for TP53 codon 72 polymorphism detection.
Denaturing HPLC Analysis (DHPLC)
PCR products were heated at 95°C for 10 min and allowed to cool down to room temperature for approximately 30 min. Five to 10 μL of each sample were run on a Wave DNA Fragment Analysis System (Transgenomic, Omaha, NE) using a DNASep column and the run was monitored by ultraviolet light (UV) (260 nm). Optimum DHPLC temperatures were determined by an incremental temperature scan, using the software-predicted melting profile as a starting point (Transgenomic, Omaha, NE). Samples were run at more than one temperature due to the heterogeneity in the distribution of GC-rich regions that results in a heterogeneous melting temperature distribution throughout the amplified DNA fragments to be analyzed. By using 59°C as a melting temperature we were able to identify the formation of heteroduplexes in a subset of samples submitted to the analysis. The heteroduplexes are frequently found in samples with differences in alleles. Thus, heterozygous samples were identified due to a small shoulder peak in the chromatogram. After distinguishing the heterozygous samples, the subset of homozygous samples were mixed in approximately equimolar proportions with a control sample with TP53 codon 72 polymorphism previously identified as homozygous proline. This mixture was heated at 95°C for 10 min and allowed to cool down to room temperature for approximately 30 min. Five to 10 μL of each sample were then submitted to a new DHPLC analysis. This allowed differentiating the two homozygous genotypes. Heteroduplexes were observed in samples with genotype homozygous arginine, since the PCR product added to the sample was known to be homozygous proline. The homoduplexes, or samples with a single peak, were observed in samples with genotype proline, since it was the same genotype presented by the control sample added to the mixture. As an additional control, 10% of the samples were randomly selected and mixed with the same equimolar proportions of a control sample previously identified as homozygous arginine. Results were compared and no discordant results were observed between the two pulled analyses.
Dot Blot Hybridization
After amplification of TP53 gene exon 4, 1.5 μL of PCR products were spotted on nylon membranes (Hybond N+ - Amersham Pharmacia Biotech, Buckinghamshire, England) and UV-cross-linked. Dot-blotted PCR products were hybridized with sequence-specific 3'; biotin-labelled oligonucleotide probes TP53 Arg 5'--GGGCCACGCGGGGAGCA-3' and TP53 Pro 5'--GGGCCACGGGGGGAGCA- 3'. Hybridization was performed in two different temperatures: 62°C for TP53 Arg and 64°C for TP53 Pro probe. Briefly, membranes were blocked with 15 ml 0.1 × SSPE (18 mM NaCl; 1 mM NaH2PO4-H2O; 0.55 mM NaOH; 0.1 mM EDTA; adjusted for pH 7,4), 0.5% SDS for 20 min and then pre-hybridized with 15 ml of 2 × SSPE (360 mM NaCl; 20 mM NaH2PO4-H2O; 11 mM NaOH; 2 mM EDTA; adjusted for pH 7,4), 0.1% SDS for 20 min. All steps above were performed in probe-specific hybridization temperature. Probes were added to pre-hybridization solution to a final concentration of 1.2 pmoles/mL. After overnight hybridization, membranes were washed 2 times of 15 min with 2 × SSPE, SDS 0.1% in hybridization temperature. Incubation with conjugated horseradish peroxidase-streptavidin (1 mg/mL; Vector Laboratories, Burlingame, CA, USA) was performed in room temperature. Membranes were washed 2 times with 2 × SSPE, SDS 0.1% for 10 min to eliminate background and minimize misclassification. Positive reactions were detected by enhanced chemiluminescence (ECL Kit, Amersham) through exposure to Kodak X-OMAT film (Kodak, Rochester, NY). If needed to obtain optimal signal intensities, the exposure times were sometimes varied, as judged by evaluation of the hybridization controls present on each membrane.
Restriction Fragment Length Polymorphism (RFLP)
After TP53 exon 4 amplification, 7 μL of the amplified reaction products were digested in a 10 μL volume of 1 × NEB buffer 2 and 5 U of Bsa JI or Bst UI endonucleases (New England BioLabs, Newton, MA) at 60°C for at least 2 hours. Products digested by Bsa JI were run in 12% polyacrylamide gel and products digested by Bst UI were run in 8% polyacrylamide gel. In both cases silver staining was used to identify band profile [16].
To ensure quality control of all genotyping methods, the samples were blindly read by two persons independently (TR, LLV). Despite the high percentage of agreement, around 95% for each method, the samples with discordant results were submitted to a new round of amplification and genotyping.
Statistical Analysis
Cohen's kappa coefficients of agreement were computed to evaluate the inter-methodology variability in genotyping.
