The 38 women included in the study were randomly selected among San Luis Potosí City residents with cervical infection by HPV16 demonstrated by nested PCR amplification of the E6 ORF 16. Their age range was 22 to 45 years (mean ± SD = 33.2 ± 5.9 years). Twenty seven (71.1%) had low-grade squamous cervical intraepithelial lesions (LSIL), eight (21.1%) had high-grade squamous cervical intraepithelial lesions (HSIL), and three (7.9%) had invasive cervical cancer (ICC).

The 5'-termini of the deposited sequences were aligned to start at nucleotide 83 of the E-P genome, the first one of the ATG start codon of the E6 open reading frame (ORF), and their upstream portions were ignored. Lengths of the deposited sequences ranged from 337 to 576 bp (average = 541 bp). All sequences were longer than the 477 bp expected for the complete E6 ORF, except HPV16-27 (377 bp), HPV16-9 (436 bp), and HPV16-21 (465 bp). The GenBank/EMBL/DDBJ accession numbers of the 38 sequences of this study, EU880235 to EU880272, are depicted in Table 2.

HPV16 subtypes and variants were identified by comparing their sequences with that of the E6 ORF of the HPV16 E-P Ref prototype variant 7. The HPV16 subtypes identified are listed in Table 2. The predominant one is the European subtype (E) with 36 cases (94.7%); the two cases of the Asian-American (AA) subtype represent 5.3% of the total (Table 2).

The frequency of HPV16 variants identified is shown in Fig. 1. For the E subtype the most frequent was the E-P Ref variant with 27 cases (71.1%) followed by the E-T350G variant (seven cases, 18.4%) and the E-C188G variant (two cases, 5.3%). AA-a was the only AA variant (two cases, 5.3%).

In the E6-HPV16 ORF sequences, 24 single nucleotide changes that had not been described before were found (Table 3); 21 of them (13 substitutions, 6 insertions, 2 deletions) appeared once in nine sequences; the remaining three were observed in two sequences. The A404T change was observed in a single E-P variant.

In 13 sequences there was a deletion of two neighboring bases (AC) located in nucleotide positions 56 and 57, immediately before the ATG protein start codon. The A334G synonymous substitution was found in 22 sequences, located next to nucleotide 335, commonly used for subtype-variant identification.

Amino acid sequences of the E6 oncoprotein encoded by the samples analyzed were compared with the E-P Ref sequence to identify non synonymous mutations (Table 3). E-T350G variants had the expected substitution of leucine for valine at position 83 (L83V). The expected amino acid changes were also found in AA-a (Q14H; H78Y; L83V) and E-C188G variants (E20Q; L83V). Deletion of nucleotide 484, identified in one of the two AA-a variants implies the frameshift in the E6 ORF starting at amino acid 127. The A404T change identified in a single sample, produced the I101F change. The C37A change identified in an E-P variant produced the change Q91K. The C206T change present in an E-T350G variant generated a stop codon instead of amino acid 35 and T351A caused the L83E change in the same sample. The C173A substitution produced the H24N change in one sequence, and the A182T the I27L changes in another one.

Identical dendrograms were generated with the 38 E6-2 nucleotide sequences using the Phylip and MEGA programs. Comparing all sequences it was not possible to resolve the A404T non synonymous mutation as a group independent of the E-P Ref sequence. A tree was thus constructed with the six E6-HPV16 ORF sequences representing each of the variants identified, including those containing the single nucleotide changes A334G and A404T. The optimal tree with branch length = 0.01907 was drawn to scale with the same evolutionary distance units used to infer the phylogenetic tree (Fig. 2). In this way the AA-a variant and the three known E-P variants (E-P Ref, E-T350G and E-C188G) could be related, with two new branches of the E subtype corresponding to those carrying the A334G and A404T changes.

The presence of the A334G change in 19 of the 32 E-P sequences (70.4%) makes the corresponding HPV16 "variant" the most prevalent in San Luis Potosi City.

The proportion of HSIL and ICC lesions ("HSIL + ICC fraction") associated to the 19 A334G variants (0.21) was clearly higher than that associated to the eight E-P Ref variants (0.00). Although the HSIL + ICC fraction was even higher for the AA-a, E-T350G and E-C188G variants, the number of samples of these variants were too small to reach definitive conclusions on their oncogenicity (Table 4).

Randomly selected cervical lesion samples from 38 women residing in San Luis Potosí City, with HPV16 infection diagnosed in our laboratory 16, had been obtained by one of us (JOV) at the Colposcopy Clinic, Secretaría de Salud, San Luis Potosi City.

Each scraping was taken with an endocervical brush ('cytobrush') that was immediately inserted into a 5 mL polypropylene tube (Nalge Nunc, Rochester, NY) containing 1 mL phosphate buffered saline (PBS: 137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 2 mM KH₂HPO₄, pH 7.4) supplemented with sterile 25 mM disodium ethylene-diaminotetraacetate (EDTA), pH 8.0 (PBS-EDTA). Once detached from the cytobrush and suspended in the PBS-EDTA vehicle, each sample was fixed by addition of 1.5 mL 96% ethanol and processed to extract DNA on the same day or up to 30 days after being kept at room temperature. Reagents were purchased from J.T. Baker (Xalostoc, Mexico) unless other source is specified.

To extract the DNA, each fixed sample was mixed by vortexing and 1 mL transferred to a 1.5 mL tube and spun in a Hettich Mikro 20 microcentrifuge (Cologne, Germany) for 5 min at 13,000 rpm (16,250 × g). The supernatant was discarded by decantation and to each pellet were added 500 μL of Tris-EDTA-saline (TES: 10 mM Tris-HCl; 2 mM disodium EDTA, 0.4 M NaCl, pH 8.0 at 25°C), 50 μL of 10% sodium dodecyl sulphate and 20 μL of proteinase K (20 mg/mL). Mixtures were incubated at 55°C for 3 h, at the end of which 150 μL of 5 M NaCl were added and centrifuged again for 15 min. Each supernatant was aspirated and transferred to a tube to which 577 μL cold isopropanol were immediately added and then left stand for 10 min at 4°C to precipitate the nucleic acids. The tubes were centrifuged again for 10 min and the supernatants discarded by decantation. Each pellet was washed by vortexing with 1 mL of 70% cold ethanol and centrifuged for 10 min at 10,000 rpm (9,615 × g) and room temperature. Supernatants were discarded by aspiration and the pellets dried out by inverting the tubes for 15 min on a paper towel. Each pellet was dissolved with 50 μL TE (10 mM Tris-HCl, 1 mM disodium EDTA, pH 8.0 at 25°C).

DNA quality was verified by electrophoretic analysis in 1% agarose gels with TAE buffer (40 mM Tris-acetate, 1 mM disodium EDTA, pH 8.2 at 25°C). Two-μL from each sample were applied to gels which were run at 60 V for 90 min. λ-phage DNA digested with Hind III (Sigma-Aldrich, Mexico) was used as marker. After staining for 20 min with ethidium bromide (1 μg/mL) gels were transilluminated with ultraviolet light and their fluorescence recorded with the Bio-Rad ChemiDoc EQ (Hercules, CA) photodocumenter.

DNA was quantified by fluorometry with the PicoGreen dsDNA Quantitation kit (Molecular Probes; Eugene, OR) by interpolation in a standard curve containing up to 50 ng of λ-phage DNA. To each well of a black FIA 96 well plate (Greiner Bio-One, Frickenhausen, Germany) 198 μL of the assay solution (PicoGreen diluted 1:400 in TE) and 2 μL of standard DNA or problem samples were added, and their fluorescence determined using a 485 nm excitation filter and a 535 emission filter in the GENios Pro fluorometer (Tecan Systems, San Jose, CA) with the Magellan 4 software.

Nested PCR was used to generate enough DNA to sequence the E6-HPV16 ORF 16. The E6-1 product (~650 bp) was preamplified with the LCRS/E7AS primer pair in the first reaction (PCR1), and the E6-2 product (~626 bp) with the E6F/PU-2R16 primer pair in the second reaction (PCR2) (Fig. 3, Table 1). PCR1 mixtures of 50 μL contained 2 mM MgCl₂, the four deoxinucleotide triphosphates (0.4 mM each), forward and reverse primers (0.6 μM each), 1.5 U of Taq DNA polymerase and 25 ng of cervical DNA in 200 mM Tris-HCl 500 mM KCl, pH 8.4. Mixtures were preamplified by incubation in a Touchgene Gradient (Techne) thermocycler with initial denaturation at 94°C by 4 min, 40 cycles of amplification (1 min denaturation at 94°C, 1 min annealing at 55°C 1 min extension at 72°C) and 10 min final extension at 72°C. To generate E6-2, to PCR2 mixtures (same composition as PCR1 mixtures except for primers and DNA) 1 μL of each preamplified PCR1 mixture was added as template. Amplification products were electrophoresed in high-resolution sodium borate (SB)1% agarose gels 30.

The intensity of E6-2 bands generated by nested PCR was highest in the presence of 2 mM MgCl2, which was used in all subsequent amplification experiments. In a pilot test E6-2 bands were amplified by direct PCR in positive controls (pHPV16 and pHPV18) and in 23 out of 28 (82.1%) cervical samples. In view of these results we decided to amplify all samples through nested PCR.

E6-1 was preamplified in PCR1 mixtures with the LCRS/E7AS primer pair and E6-2 in PCR2 mixtures with the internal E6F/PU-2R16 primer pair. In this way conspicuous E6-2 bands were obtained from all samples with yields sufficient (> 1500 ng) to sequence the amplicons (Fig. 4).

The pHPV16 and pHPV18 plasmids containing the complete genomes of HPV16 and HPV18 respectively, donated by Dr. Alejandro García Carrancá (Instituto Nacional de Cancerología, Mexico City), were used as positive controls. PCR mixtures without DNA were used as negative controls.

The DNA purified with the Wizard kit (PCR Preps DNA Purification Systems, Promega, Madison, WI) from 40 μL of PCR2 mixtures was used to sequence the positive and negative strands of E6-2 amplicons with the method of Sanger et al. 28 at the National Laboratory for Genomic Biodiversity (Guanajuato campus of CINVESTAV, Mexico) and compared with the E6 ORF of the major HPV branches (E, AA, Af1, Af2) by multiple alignment with the ClustalW v1.82 software 31. Viral subtypes and variants were identified by comparing the E6-HPV16 ORF nucleotide sequences published by Yamada et al. 7. Allocation of nucleotide and amino acid positions in the E6-HPV16 ORF was based on the nucleotide sequences 32. Amino acid sequences of the E6 oncoprotein were predicted with the Translate program tool of the ExPASy database 33.

Their evolutionary history was inferred using the UPGMA method 13. Evolutionary distances were computed using the 2-parameter method 34 whose units are the number of base substitutions per site. Codon positions included were first + second + third + noncoding. All positions containing gaps and missing data were eliminated from the dataset (complete deletion option); there were a total of 477 positions in the final dataset. Phylogenetic analyses were conducted in MEGA4 35.