Quantitative Detection of the M204V Hepatitis B Virus Minor Variants by Amplification Refractory Mutation System Real-Time PCR Combined with Molecular Beacon Technology

Experimental target design.Sixty-seven full-length HBV sequences representing all of the available human HBV genotypes were downloaded from the GenBank sequence database (http://www.ncbi.nlm.nih.gov ) in order to create a reference alignment for the design of target sequences; we included sequences from genotype A (M57663, AJ309370, AJ344115, AF297622, AF418677, AY161142, X70185, X51970), genotype B (D23678, D23679, D23677, AB010289, AB073838, AB010290, AB010291, AB010292, AB073854, AB073853), genotype C (D50489, D23680, M38636, AB026811, AB042282, AB033556, AB074047, V00867, Y18855), genotype D (AF418679, X59795, X72702, AY161150, AF121239, AF151735, AB090270, AF043593, AB033558), genotype E (AB032431, AB091255, AB091256, AB106564, X75664, X75657), genotype F (AB036906, AB036907, AB036908, AB036905, AB036909, AB036911, AB036910, AF223962, AF223964, AF223963), genotype G (AB064313, AB056514, AB064310, AB064311, AF160501, AF405706, AB056513, AB056515), and genotype H (AB064315, AY090460, AB059659, AB059660, AY090454, AY090457, AB064312). Sequences were aligned by using the CLUSTAL W program (version 1.81) (31) and manually corrected.

Based on the HBV reference sequence alignment, a single-stranded DNA target sequence containing 95 bases was designed so as to include codon 204 (MEGA version 3.1) (13). The nomenclature based on HBV reference sequence X02496 has been used in order to define the primers and molecular beacon position on the target sequences (30). Due to viral strain nucleotide polymorphisms, the target sequence was chosen to be designed at the least polymorphic site possible. Based on the design of the HBV WT target sequence containing the triplet ATG (encoding methionine) in codon 204, the Mut target sequence was constructed in such a way that the point mutational triplet GTG (encoding valine) would be included. Single-stranded target sequence synthesis and high-purity clearance were performed by Invitrogen Laboratories (www.invitrogen.com ).

Primer design.According to the ARMS technique, the reverse primer specifically designed for the WT and Mut target sequences were refractory to PCR on Mut and WT template DNA sequences, respectively. An additional mismatch at the n-1 position next to the 3′-end base of the Mut primer decreased the amplification efficiency of the WT target sequence to a greater extent than that of the Mut target sequence (20, 28). The primers used were 5′-CTCYTGGCTCAGTTTACTAGTGC-3′ (forward primer, positions 535 to 557), 5′-CCCAAWACCAVATMATCCAT-3′ (reverse WT primer, positions 610 to 629), and 5′-CCCAAWACCAVATMATCCGC-3′ (reverse Mut primer, positions 610 to 629). Due to polymorphisms in the viral strains, ambiguity codes had to be included in the primer design. Primers were designed according to the following standard requirements: (i) to avoid primer-dimers and (ii) to ensure that the melting temperature (T_m) of the primers would be similar to or at least 5 to 10°C lower than the T_m of the molecular beacon. The software tool used for assessment of the T_m calculations was from the Virtual Genome Center (http://www.yeastgenome.org/cgi-bin/web-primer ). Primers were synthesized by Invitrogen Laboratories.

Molecular beacon design.Molecular beacon, a single-stranded oligonucleotide hybridization probe that forms a stem-and-loop structure, was used for RT detection of amplification without interfering with the reaction (http://www.molecular-beacons.org/ ) (27, 29, 32). Molecular beacon consisted of 27 nucleotides and was 5′ labeled with 6-carboxyfluorescein (detected at 530 nm, channel 1 for the LightCycler and LightCycler 2.0 systems) and 3′ labeled with Black Hole Quencher 1. An additional CG terminal was added to both sides of the required sequence in order to form the stem. The molecular beacon used was 5′-CGCGGCGGAAAGCCCKRCGMACCACTGAACAAAGCCGCG-3′ (positions 560 to 586). Synthesis of the molecular beacon was performed by Biolegio Laboratories (www.biolegio.com ). Ambiguity codes were again included in the molecular beacon design due to polymorphisms in the viral strains.

Generation and quantification of the WT and Mut double-stranded DNA (dsDNA) target sequences.WT and Mut target sequences were amplified and converted to dsDNA forms by PCR. The cycling conditions used were 94°C for 2 min; 40 cycles of 94°C for 30 s, 58°C for 30 s, and 72°C for 30s; and a final extension of 72°C for 7 min. Thereafter, the Quant-iT PicoGreen dsDNA reagent, an ultrasensitive fluorescent nucleic acid stain for quantitating dsDNA in solution, was used to quantitate the PCR amplification procedure under RT conditions with the LightCycler 2.0 system (www.invitrogen.com , www.roche-applied-science.com ).

LightCycler RT-PCR optimization and amplification.The LightCycler system was used for the RT-PCR (www.roche-applied-science.com ). Multiple reactions with different concentrations of the master mix (MM) components were performed to define the optimum concentrations for an efficient and accurate amplification reaction. The MM final concentrations were set as follows: 10× LC FastStart DNA Master HybProbe, 30 μM each of the forward and reverse set of primers appropriate for the target sequence, 25 mM MgCl₂, 10 μM of the molecular beacon, and 1 μl of heat-labile uracil-DNA glycosylase. Twenty-microliter PCR capillary tubes were used for the mixture of 10.5 μl of MM with 10 μl of the HBV DNA target sequences. Each reaction mixture contained only one set of primers; for different targets, different reactions were needed (not a multiplex reaction). However, as the LightCycler has 32 possible positions, both WT and Mut targets were quantified within the same experiment.

Melting curves were determined to estimate the T_m of the molecular beacon. Different annealing temperatures were tested to optimize the reaction. Due to the difference between the annealing temperatures of the primers and the molecular beacon, we decided to enter a slope in the cycling protocol to ameliorate the assay; annealing was performed at 60°C, while fluorescence was measured at 52°C. The final cycling conditions used were an initial denaturation at 94°C for 10 min; 50 cycles of 94°C for 0 min and annealing at 60°C for 0.05 min; and a final extension of 72°C for 0.1 min.

Preparation of the DNA standard curve. (i) Pure targets.Serial dilutions were performed for WT and Mut target sequences, starting at the initially identified concentration and ending at a concentration of 10 copies per reaction mixture, in order to construct the DNA standard curve. The two different sets of primers (one specifically designed for the WT target and the other for the Mut target) were used with the different DNA concentrations of both targets for the evaluation of the ARMS RT-PCR assay with molecular beacon technology.

(ii) Mixtures.After evaluation of the assay with pure experimental targets, multiple reaction mixtures with both primer sets were prepared with both WT and Mut experimental target sequences, resembling a clinical sample. In the first mixture, a standard concentration of 10⁶ copies per reaction mixture was preserved for the WT target sequence, while a gradient of 10⁶ to 10 copies per reaction mixture was used for the Mut target. The second mixture contained the opposite ratio of WT and Mut target sequences.

Evaluation of the ARMS RT-PCR assay with molecular beacon technology with DNA extracted from clinical samples.Viral DNA was extracted from 500 μl of serum by using the QIAamp DNA Blood Mini Kit (www.qiagen.com ) according to the standard protocol. DNA was eluted in 60 μl (final volume) of DNase-free water.

Biological discriminatory ability (cutoff) estimation.Fifty-seven samples from treatment-naïve patients with viral loads of more than 10⁵copies/ml were randomly selected to be measured by the ARMS RT-PCR assay with molecular beacon technology. Patients were not exposed to any nucleoside/nucleotide analogue. Each sample was measured with both WT and Mut (forward-reverse) sets of primers. The results were elaborated to estimate the discriminatory ability (cutoff) of the assay with clinical samples.

Use with clinical samples.Samples from chronically infected HBV patients known to have failed antiviral treatment (mainly lamivudine and adefovir) were selected to be included for evaluation of the ARMS RT-PCR assay with molecular beacon technology. Samples from 12 patients who had developed the M204V mutation, as well as samples from 13 patients with no mutations at codon position 204, tested by sequencing, were used to evaluate the ARMS RT-PCR assay.

Bioethical approval.The Bioethical Committee of the Medical School, Athens University, approved this study (reference no. 37/26-4-06).

Statistical analysis.Pure target sequences were measured in 20 repeats with the ARMS RT-PCR assay with molecular beacon technology to evaluate the analytical sensitivity of the assay. The coefficient of variation (CV) was calculated for samples quantified in repeated measurements. Pearson's correlation coefficient was used to assess the strength of the linear association between the expected concentrations of the WT and Mut target sequence and those observed by the ARMS RT-PCR assay with molecular beacon technology. The fitted regression line of expected versus observed concentrations was compared with the line of equality by testing the two-tailed hypotheses of slope = 1 and intercept = 0.

To determine the biological cutoff of the assay, we obtained the distribution of Mut/WT ratios in 57 treatment-naïve patients. The biological cutoff was defined as the 95th percentile of that distribution, i.e., the ratio below which 95% of the observations were found. Our choice of the 95th percentile for determining the cutoff was based on the fact that in a normally distributed characteristic, the normal limits are usually chosen in such a way that 95% of the healthy population has values within these limits. Furthermore, taking into consideration publications reporting that the presence of the M204V mutation affects future treatment options, we would opt for a less relaxed cutoff value, i.e., 0.04% (95th percentile) rather than 0.22% (99th percentile), to avoid cross-resistance with other antivirals in patients falsely identified as not having the M204 variant population.

Pure target sequences.Multiple repeats of the experiments revealed that the ARMS RT-PCR assay with molecular beacon technology showed a sensitivity of 10 copies per reaction mixture obtained after measurements of pure viral WT or Mut target sequences with the relevant (forward-reverse) set of primers. Concentrations as low as 10 copies per reaction mixture for the WT and Mut target sequence were quantitatively detected by the ARMS RT-PCR assay at frequencies of 100% (20/20) and 95% (19/20), respectively, as shown in Table 1. The expected concentrations of the WT and Mut targets were highly correlated with those observed with the ARMS RT-PCR assay with molecular beacon technology (Pearson's r = 0.998 for both WT and Mut target sequences). The equality line of the expected concentrations versus those observed with the assay fell within the 95% confidence interval of the corresponding fitted regression line for both WT (Fig. 1A) and Mut (Fig. 1B) target sequences when detected by the relevant set of primers. The variability of the assay, as measured by the CV, increased as the expected concentration was decreasing (Table 1). The linearity of the ARMS RT-PCR assay with molecular beacon technology was sustained at higher concentrations of both WT and Mut target sequences.

• Open in new tab
• Download powerpoint

FIG. 1.

Fitted regression line of the concentrations obtained with the ARMS RT-PCR assay with molecular beacon technology for the Mut target sequence (A) and the WT target sequence (B) versus the expected concentrations (black line), the equality line (yellow line), and the 95% confidence interval of the fitted regression line (blue line).

Mixtures of both WT and Mut target sequences.From the experiments performed with mixtures of both target sequences with the appropriate set of primers, we have seen that the observed measured concentrations were reaching a threshold of quantification beyond which any further gradient of the minor target sequence resulted in the same value observed regardless of the concentration of the minor target sequence. Further analysis was performed by using mixtures with 10⁶ copies of the WT target sequence per reaction mixture and gradient concentrations of the Mut target sequence, with the addition of intermitted concentrations of 5 × 10³, 2.5 × 10³, and 1.25 × 10³ copies per reaction mixture detected with the relevant set of primers for the Mut target sequence.

The WT target sequence at a concentration of 10⁶ copies per reaction mixture was detected with both the relevant WT primers and the Mut primers to determine the discriminatory ability of the ARMS RT-PCR assay with molecular beacon technology. While the WT target was detected in the 23rd cycle with the relevant set of primers, it was detected in the 34th cycle with the Mut primers. This 11-cycle window was defined as the discriminatory ability of the assay, suggesting that reliable Mut target detection can be performed within this window. As shown in Fig. 2, the discriminatory ability of the ARMS RT-PCR assay combined with the molecular beacon was 0.25% in repeated measurements, regardless of the initial concentrations of the fixed targets. The only important factor was the ratio of the Mut and WT target concentrations. The same concentrations of WT and Mut targets were detected at the same cycle when the relevant set of primers was used.

• Open in new tab
• Download powerpoint

FIG. 2.

Discriminatory ability of the ARMS RT-PCR assay with molecular beacon technology. The concentrations shown are 10⁶ copies per reaction mixture of the WT target sequence determined with the set of primers (prim) specifically designed for detection of the WT (blue line), 10⁶ copies per reaction mixture of the WT target sequence determined with the set of primers specifically designed for detection of the Mut (pink line), 10⁵ copies per reaction mixture of the Mut target sequence determined with the set of primers specifically designed for detection of the Mut (yellow line), 10⁴ copies per reaction mixture of the Mut target sequence determined with the set of primers specifically designed for detection of the Mut (light blue line), 5 × 10³ copies per reaction mixture of the Mut target sequence determined with the set of primers specifically designed for detection of the Mut (purple line), and 2.5 × 10³ copies per reaction mixture of the Mut target sequence determined with the set of primers specifically designed for detection of the Mut (brown line). The two-headed arrow shows the limits of the discriminatory ability of the assay.

The variability of the ARMS RT-PCR assay with molecular beacon technology was evaluated by repeated measurements of the Mut target sequence in mixtures with known concentrations of WT and Mut targets in predetermined proportions with the relevant set of primers for Mut detection (Table 2). The same procedure was also used with the Mut target sequence as the major viral population (Table 2).

Biological discriminatory ability (cutoff) with clinical samples.All 57 samples from treatment-naïve patients were analyzed with the ARMS RT-PCR assay with molecular beacon technology. It was possible to determine a Mut/WT ratio for 36 (63.2%) of the 57 samples; in the remaining 21 samples, the Mut variant was not detected. The range of Mut/WT ratios in the 36 samples was 1.81 × 10⁻⁷ to 2.23 × 10⁻³. The median value of the Mut/WT ratio based on the samples tested positive for Mut and WT variants was 0.01% (5th and 95th percentiles = 0.0001 and 0.04%). The 95th percentile of the Mut/WT ratio, i.e., 0.04%, was defined as the biological cutoff of the assay.

Use with clinical samples.In all (12/12) of the samples from chronic HBV patients who had failed antiviral treatment (M204V mutation detectable by DNA sequencing), the Mut variant was quantified as the dominant viral population, representing 82.8 to 99.4% of the total viral population (Table 3). Interestingly, among 13 samples derived from chronic HBV patients who had failed antiviral treatment and tested WT at codon 204 by sequencing, 5 (38%) had a detectable M204V Mut variant when quantitated by the ARMS RT-PCR assay (Table 3). In five samples, the M204 Mut variant was detected at a ratio above the predetermined biological cutoff, ranging between 0.05 and 28%, as shown in Table 3.