- Email: [email protected]
A simplified semiquantitative determination of hepatitis C virus genome molecules by the end-point dilution method
Molecular and Cellular Probes (1996) 10, 477–480
Short Communication
A simplified semiquantitative determination of hepatitis C virus genome molecules by the end-point dilution method C. Bathelier,1 G. Mercier2 and G. Lucotte3∗ 1
LCL Laboratory, 37 rue Boulard, Paris 14, 2Burckel Laboratory, 27 rue Taine, Paris 12, and 3International Institute of Anthropology, 19 Avenue de la Marne, Asnie`res, France (Received 12 February 1996, Accepted 22 May 1996)
We describe a semiquantitative method to measure hepatitis C virus (HCV) viral particle numbers, by carrying out reverse-transcription polymerase chain reaction (RT-PCR) on serial dilutions of serum samples. The virus concentrations measured were 103–106 viral particles ml−1 of serum. The method described is relatively quick, and the only required manipulation is dilution of the serum. An optimal RT-PCR method is used for diluted and undiluted samples. 1996 Academic Press Limited
KEYWORDS: hepatitis C virus (HCV) RNA, reverse-transcription polymerase chain reaction (RTPCR), end-point dilution method, semiquantitative method, serum HCV RNA level estimation.
Hepatitis C virus (HCV) infection can induce chronic active hepatitis leading to liver fibrosis. HCV viraemia is correlated with viral replication, so quantification of the HCV RNA level in serum could provide a promising marker of replication and, therefore, of the efficiency of the antiviral treatment by a-interferon. HCV RNA particle numbers can be determined in serum samples with a new assay (Quantiplex HCVRNA assay; Chiron, Emeryville, CA, USA) based on the branched DNA (bDNA) signal amplification technique,1 but with a cutoff of 350 000 genome equivalents ml−1. Until now, the evaluation of HCV RNA in serum was performed by reverse-transcription polymerase chain reaction (RT-PCR).2–4 The most elegant method to quantify HCV RNA by RT-PCR is the competitive RT-PCR assay,5,6 which measures HCV RNA concentration by mixing complementary DNA of HCV RNA (cDNA) with 10-fold serial dilutions of
a competitor DNA or an RNA control that differ from the cDNA by having a small deletion or insertion. The usefulness of these two assays in clinical practice is however limited since they are expensive, time consuming and may require special skills. A simpler method to quantify HCV RNA levels is the limiting dilution PCR assay,7 which estimates the HCV RNA titre by carrying out serial 10-fold dilutions of extracted RNA prior to PCR. The titre is calculated from the highest dilution giving a positive signal. We report here the simplest method to estimate levels of HCV RNA determined by serial serum dilution endpoint titration, and RT-PCR detection,8 to obtain semiquantitative information about the virus load in infected individuals. Non-diluted and optimal reactions were performed using 100 ll of each serum sample, in order to obtain the 260-bp band following semiquantitative nested
∗ Author to whom correspondence should be addressed at: Regional Centre of Neurogenetics (Service of Neurology), Maison-Blanche Hospital, 45 rue Cognacq-Jay, 51092 Reims Cedex, France.
0890–8508/96/060477+04 $25.00/0
477
1996 Academic Press Limited
C. Bathelier et al.
478
(a)
(b)
(c)
50
25
13
6
3
1
1/10
1/100
Fig. 1. Serial dilutions (50, 25, 13, 6, 3, 1, 1/10, 1/100 ll of serum) for three samples [(a), (b) and (c)] and end-point (arrows) visualization (by electrophoresis on agarose and detection by ethidium bromide staining) of the 260-bp band.
PCR. Preliminary experiments8 with serial dilutions using solutions of known HCV concentrations have shown that the optimized RT-PCR assay described allows an end-point dilution positive limit corresponding to 1–10 targets to be obtained. For RNA extraction, 100 ll of serum was mixed with 500 ll of extraction solution [4 guanidium thiocyanate, 25 m sodium citrate (pH 7), 0·5% sarcosyl, 0·1 b-mercaptoethanol], and 10 lg of Escherichia coli-derived carrier tRNA was added to each tube. Thereafter, 100 ll of 2 sodium acetate (pH 4), 500 ll of water-saturated phenol, and 100 ll of chloroform-isoamylalcohol (49:1) were sequentially added to each tube. The tubes were shaken for 10 s, cooled on ice for 15 min, and the aqueous and phenol-chloroform phases separated by centrifugation for 20 min at 10 000 g and 4°C. The upper aqueous phase of each tube was transferred to fresh tubes and precipitated with 500 ll of isopropanol at −20°C for 1 h, followed by centrifugation for 20 min at 10 000 g and 4°C. The pellet was then dissolved in 150 ll of extraction solution and the RNA conserved at 0°C in 30 ll diethylpyrocarbonate-treated H2O with 60 U of RNAse inhibitor before freezing at −70°C. Reverse transcription and the first PCR were performed as follows. One aliquot of 10 ll for each sample was thawed on ice immediately before being added to 40 ll of fresh RNase-free reaction mix. The final concentrations of the buffer were: Tris-HCl,
20 m (pH 8·4); MgCl2, 2·5 m; KCl, 50 m; dNTP (Boehringer Mannheim), 0·25 m of each; RNAsin, 25 U/reaction; AMV reverse transcriptase (Promega), 5 U/reaction; Taq polymerase (Amplitaq, Cetus Perkin-Elmer), 0·5 U/reaction. Primers were all from the conserved 5′-non-coding region. The outer primers of the first PCR were 5′-CATGGTGCACGGTCTACGAGACC-3′ (antisense) and 5′-GGCGACACTCCACCATAGATC-3′ (sense), both with final concentrations of 0·5 m. After mixing the RNA with the reagents, mineral oil was added to prevent evaporation. Reverse transcription (RT) was performed at 50°C for 20 min (in an Hybaid Thermal Reactor) followed by 35 cycles of thermocycling. Melting was performed at 95°C for 1 min, annealing at 45°C for 2 min, and elongation at 72°C for 3 min. The second PCR was performed with 5 ll from the first reaction transferred to 45 ll of the same reaction buffer as above, excluding RT reagents and outer primers. The inner primers (0·5 l) were 5′-TCGCAAGCACCCTATCAGGCAG-3′ (antisense) and 5′-GGAACTACTGTCTTCACGCAGA-3′ (sense). Twenty five cycles of 95°C for 1 min and at 60°C for 1 min were then run. First and second round products were analysed by electrophoresis on a 1·5% agarose gel, stained by ethidium bromide and visualized by u.v. transillumination. A fragment of 260 bp was expected. Results of the semiquantification of HCV RNA in the serum of 48 infected patients by end-point titration and RT-PCR are presented in Table 1. Detection of
Semiquantitative determination of hepatitis C virus genome molecules by the end-point dilution method
479
Table 1. End-point results obtained by RT-PCR and limiting dilution assay for 48 HCV sera corresponding to infected patients ll of serum
Dilutions
No. of samples (%)
50 25 13 6 3 1 1/10
1/2 1/4 ≈1/9 1/16 1/33 1/102 1/103
1 (2) 2 (4·1) 17 (35·4) 11 (41·6) 9 6 (12·5) 2 (4·1)
the amplified 260 bp corresponding to the final RTPCR products was by ethidium bromide staining of gels (Fig. 1, for 3 samples). For each serum, we prepared several serial dilutions (1/2, 1/4, approx. 1/9, 1/16, 1/33, 1/102 and 1/103) of the initial 100 ll quantity. Figure 1 shows, for example, that for sample (a) no PCR signal was obtained when diluted 1:100, at 1:16 dilution for sample (b) and at approximately 1:9 dilution for sample (c). Table 1 summarizes the dilutions corresponding to detection limits for the 48 samples. Assuming a detection limit of 1–10 molecules of the optimal HCV RT-PCR used, and a quantitative recovery concerning reverse transcription of about 10% (a realistic estimation given by Promega), we can calculate that the level of virus concentration of these specimens is about 104–105 viral particles ml−1 of serum. In the first experiments, two replicates of the same serum sample (corresponding to final concentrations of 106, 105 and 103 HCV genomes ml−1) were tested in the assay; without differences shown for each replicate in detection for the dilution corresponding to the limit. Figure 2 shows that viraemia levels varied between 50 45
No. of sera (%)
40 35 30
}
patients from 103 HCV genomes ml−1 to 106 HCV genomes ml−1, with a modal value of 5×104 HCV genomes ml−1. An intrinsic effect of low dilutions was observed in our series8 and is probably responsible for part of the peak of values at 104–5×104. No correlation was observed in our (untreated) patients between HCV genome titre and serum alanine aminotransferase (ALT) levels. Semiquantitation of serum HCV RNA level was achieved by carrying out RT-PCR on serial dilutions from 100 ll of serum samples. This semiquantitative method was established using an end-point dilution method analogous to that used in assays for virus infectivity. This material provides a useful internal control for variables inherent in the assay, including RNA extraction, cDNA synthesis, and PCR itself. Samples positive by our method could be detected by ethidium bromide staining, making it unnecessary to increase the sensitivity of detection by Southern blotting. The method described here is relatively quick compared to other procedures1,5–6 as it requires few manipulations (two, or routinely one, series of seven dilutions, a positive control, a negative control) other than that of the optimal RT-PCR needed for the starting sample. A similar sort of semiquantitative level estimation was often used9–12 in routine experimentation procedures, and such an end-point dilution method can be accepted as a means to measure HCV particle levels when bDNA assay or competitive PCR are unavailable.
25
ACKNOWLEDGEMENTS
20 15
This study was supported by grant no. R099.512B from Cis Bio-International.
10 5 0
3
3
4
4
5
6
10 5.10 10 5.10 10 10 –1 Log molecules ml
10
7
Fig. 2. HCV viraemia levels (frequency distribution), expressed in number of viral particles=molecules ml−1, in the 48 sera.
REFERENCES 1. Lau, J. Y. N., Davis, G. L. & Kniffen, J. (1993). Significance of serum hepatitis C virus RNA levels in chronic hepatitis C. Lancet 341, 1501–4. 2. Simmonds, P., Zhang, L. Q., Watson, H., et al. (1990).
480
3.
4.
5.
6.
7.
C. Bathelier et al. Hepatitis C quantification and sequencing in blood products, haemophiliacs and drug users. Lancet 336, 1469–72. Okamoto, H., Okada, S., Sugiyama, Y. et al. (1990). Detection of hepatitis C virus RNA by a two-stage polymerase chain reaction with two pairs of primers deduced from 5′-non coding region. Japanese Journal of Experimental Medicine 60, 215–22. Garson, J. A., Ring, C., Tuke, P. & Tedder, R. S. (1990). Enhanced detection by PCR of hepatitis C virus RNA. Lancet 336, 878–9. Kaneko, S., Murakami, S., Unoura, M. & Kobayashi, K. (1992). Quantification of hepatitis C virus viremia by reverse transcription-PCR. Journal of Medical Virology 37, 278–82. Clossais-Bernard, N. & Andre´, P. M. (1994). Automated quantitative determination of hepatitis C virus viremia by reverse transcription-PCR. Journal of Clinical Microbiology 32, 1887–93. Brillanti, S., Garson, J. A., Tuke, P. W. et al. (1991). Effect of a-interferon therapy on hepatitis C viraemia in community-acquired chronic non-A, non-B hepatitis. Journal of Medical Virology 34, 136–41.
8. Lucotte, G., Mura, C. Aouize´rate, A., Champenois, T. & Marchand, J. (1994). Confirmation of hepatitis C virus positive seras by polymerase chain reaction. In Methods in DNA Amplification (Rolfs A. et al., eds). Pp. 123–126. New York: Plenum Press. 9. Ulrich, P. P., Romeo J. M., Lane P. K., Kelly I., Daniel, L. J. & Vyas G. N. (1990). Detection, semiquantification, and genetic variation in hepatitis C virus sequences amplified from plasma of blood donors with elevated alanine aminotransferase. Journal of Clinical Investigations 86, 1609–14. 10. Cheng, R. C., Chan, R. T. & Lok, A. S. F. (1993). Longitudinal study of hepatitis C viremia in chronic hepatitis C. Journal of Medical Virology 41, 338–42. 11. Saldanka, J. & Minor, P. (1994). A sensitive PCR method for detecting HCV RNA in plasma pools, blood products, and single donations. Journal of Medical Virology 43, 72–6. 12. Puchlammer-Sto¨ckl, E., Mor, W., Kundi, M., Heinz, F. X., Hofmann, H. & Kunz, C. (1994). Prevalence of hepatitis C virus RNA in serum and throat washings of children with chronic hepatitis. Journal of Medical Virology 43, 143–7.
Short Communication
A simplified semiquantitative determination of hepatitis C virus genome molecules by the end-point dilution method C. Bathelier,1 G. Mercier2 and G. Lucotte3∗ 1
LCL Laboratory, 37 rue Boulard, Paris 14, 2Burckel Laboratory, 27 rue Taine, Paris 12, and 3International Institute of Anthropology, 19 Avenue de la Marne, Asnie`res, France (Received 12 February 1996, Accepted 22 May 1996)
We describe a semiquantitative method to measure hepatitis C virus (HCV) viral particle numbers, by carrying out reverse-transcription polymerase chain reaction (RT-PCR) on serial dilutions of serum samples. The virus concentrations measured were 103–106 viral particles ml−1 of serum. The method described is relatively quick, and the only required manipulation is dilution of the serum. An optimal RT-PCR method is used for diluted and undiluted samples. 1996 Academic Press Limited
KEYWORDS: hepatitis C virus (HCV) RNA, reverse-transcription polymerase chain reaction (RTPCR), end-point dilution method, semiquantitative method, serum HCV RNA level estimation.
Hepatitis C virus (HCV) infection can induce chronic active hepatitis leading to liver fibrosis. HCV viraemia is correlated with viral replication, so quantification of the HCV RNA level in serum could provide a promising marker of replication and, therefore, of the efficiency of the antiviral treatment by a-interferon. HCV RNA particle numbers can be determined in serum samples with a new assay (Quantiplex HCVRNA assay; Chiron, Emeryville, CA, USA) based on the branched DNA (bDNA) signal amplification technique,1 but with a cutoff of 350 000 genome equivalents ml−1. Until now, the evaluation of HCV RNA in serum was performed by reverse-transcription polymerase chain reaction (RT-PCR).2–4 The most elegant method to quantify HCV RNA by RT-PCR is the competitive RT-PCR assay,5,6 which measures HCV RNA concentration by mixing complementary DNA of HCV RNA (cDNA) with 10-fold serial dilutions of
a competitor DNA or an RNA control that differ from the cDNA by having a small deletion or insertion. The usefulness of these two assays in clinical practice is however limited since they are expensive, time consuming and may require special skills. A simpler method to quantify HCV RNA levels is the limiting dilution PCR assay,7 which estimates the HCV RNA titre by carrying out serial 10-fold dilutions of extracted RNA prior to PCR. The titre is calculated from the highest dilution giving a positive signal. We report here the simplest method to estimate levels of HCV RNA determined by serial serum dilution endpoint titration, and RT-PCR detection,8 to obtain semiquantitative information about the virus load in infected individuals. Non-diluted and optimal reactions were performed using 100 ll of each serum sample, in order to obtain the 260-bp band following semiquantitative nested
∗ Author to whom correspondence should be addressed at: Regional Centre of Neurogenetics (Service of Neurology), Maison-Blanche Hospital, 45 rue Cognacq-Jay, 51092 Reims Cedex, France.
0890–8508/96/060477+04 $25.00/0
477
1996 Academic Press Limited
C. Bathelier et al.
478
(a)
(b)
(c)
50
25
13
6
3
1
1/10
1/100
Fig. 1. Serial dilutions (50, 25, 13, 6, 3, 1, 1/10, 1/100 ll of serum) for three samples [(a), (b) and (c)] and end-point (arrows) visualization (by electrophoresis on agarose and detection by ethidium bromide staining) of the 260-bp band.
PCR. Preliminary experiments8 with serial dilutions using solutions of known HCV concentrations have shown that the optimized RT-PCR assay described allows an end-point dilution positive limit corresponding to 1–10 targets to be obtained. For RNA extraction, 100 ll of serum was mixed with 500 ll of extraction solution [4 guanidium thiocyanate, 25 m sodium citrate (pH 7), 0·5% sarcosyl, 0·1 b-mercaptoethanol], and 10 lg of Escherichia coli-derived carrier tRNA was added to each tube. Thereafter, 100 ll of 2 sodium acetate (pH 4), 500 ll of water-saturated phenol, and 100 ll of chloroform-isoamylalcohol (49:1) were sequentially added to each tube. The tubes were shaken for 10 s, cooled on ice for 15 min, and the aqueous and phenol-chloroform phases separated by centrifugation for 20 min at 10 000 g and 4°C. The upper aqueous phase of each tube was transferred to fresh tubes and precipitated with 500 ll of isopropanol at −20°C for 1 h, followed by centrifugation for 20 min at 10 000 g and 4°C. The pellet was then dissolved in 150 ll of extraction solution and the RNA conserved at 0°C in 30 ll diethylpyrocarbonate-treated H2O with 60 U of RNAse inhibitor before freezing at −70°C. Reverse transcription and the first PCR were performed as follows. One aliquot of 10 ll for each sample was thawed on ice immediately before being added to 40 ll of fresh RNase-free reaction mix. The final concentrations of the buffer were: Tris-HCl,
20 m (pH 8·4); MgCl2, 2·5 m; KCl, 50 m; dNTP (Boehringer Mannheim), 0·25 m of each; RNAsin, 25 U/reaction; AMV reverse transcriptase (Promega), 5 U/reaction; Taq polymerase (Amplitaq, Cetus Perkin-Elmer), 0·5 U/reaction. Primers were all from the conserved 5′-non-coding region. The outer primers of the first PCR were 5′-CATGGTGCACGGTCTACGAGACC-3′ (antisense) and 5′-GGCGACACTCCACCATAGATC-3′ (sense), both with final concentrations of 0·5 m. After mixing the RNA with the reagents, mineral oil was added to prevent evaporation. Reverse transcription (RT) was performed at 50°C for 20 min (in an Hybaid Thermal Reactor) followed by 35 cycles of thermocycling. Melting was performed at 95°C for 1 min, annealing at 45°C for 2 min, and elongation at 72°C for 3 min. The second PCR was performed with 5 ll from the first reaction transferred to 45 ll of the same reaction buffer as above, excluding RT reagents and outer primers. The inner primers (0·5 l) were 5′-TCGCAAGCACCCTATCAGGCAG-3′ (antisense) and 5′-GGAACTACTGTCTTCACGCAGA-3′ (sense). Twenty five cycles of 95°C for 1 min and at 60°C for 1 min were then run. First and second round products were analysed by electrophoresis on a 1·5% agarose gel, stained by ethidium bromide and visualized by u.v. transillumination. A fragment of 260 bp was expected. Results of the semiquantification of HCV RNA in the serum of 48 infected patients by end-point titration and RT-PCR are presented in Table 1. Detection of
Semiquantitative determination of hepatitis C virus genome molecules by the end-point dilution method
479
Table 1. End-point results obtained by RT-PCR and limiting dilution assay for 48 HCV sera corresponding to infected patients ll of serum
Dilutions
No. of samples (%)
50 25 13 6 3 1 1/10
1/2 1/4 ≈1/9 1/16 1/33 1/102 1/103
1 (2) 2 (4·1) 17 (35·4) 11 (41·6) 9 6 (12·5) 2 (4·1)
the amplified 260 bp corresponding to the final RTPCR products was by ethidium bromide staining of gels (Fig. 1, for 3 samples). For each serum, we prepared several serial dilutions (1/2, 1/4, approx. 1/9, 1/16, 1/33, 1/102 and 1/103) of the initial 100 ll quantity. Figure 1 shows, for example, that for sample (a) no PCR signal was obtained when diluted 1:100, at 1:16 dilution for sample (b) and at approximately 1:9 dilution for sample (c). Table 1 summarizes the dilutions corresponding to detection limits for the 48 samples. Assuming a detection limit of 1–10 molecules of the optimal HCV RT-PCR used, and a quantitative recovery concerning reverse transcription of about 10% (a realistic estimation given by Promega), we can calculate that the level of virus concentration of these specimens is about 104–105 viral particles ml−1 of serum. In the first experiments, two replicates of the same serum sample (corresponding to final concentrations of 106, 105 and 103 HCV genomes ml−1) were tested in the assay; without differences shown for each replicate in detection for the dilution corresponding to the limit. Figure 2 shows that viraemia levels varied between 50 45
No. of sera (%)
40 35 30
}
patients from 103 HCV genomes ml−1 to 106 HCV genomes ml−1, with a modal value of 5×104 HCV genomes ml−1. An intrinsic effect of low dilutions was observed in our series8 and is probably responsible for part of the peak of values at 104–5×104. No correlation was observed in our (untreated) patients between HCV genome titre and serum alanine aminotransferase (ALT) levels. Semiquantitation of serum HCV RNA level was achieved by carrying out RT-PCR on serial dilutions from 100 ll of serum samples. This semiquantitative method was established using an end-point dilution method analogous to that used in assays for virus infectivity. This material provides a useful internal control for variables inherent in the assay, including RNA extraction, cDNA synthesis, and PCR itself. Samples positive by our method could be detected by ethidium bromide staining, making it unnecessary to increase the sensitivity of detection by Southern blotting. The method described here is relatively quick compared to other procedures1,5–6 as it requires few manipulations (two, or routinely one, series of seven dilutions, a positive control, a negative control) other than that of the optimal RT-PCR needed for the starting sample. A similar sort of semiquantitative level estimation was often used9–12 in routine experimentation procedures, and such an end-point dilution method can be accepted as a means to measure HCV particle levels when bDNA assay or competitive PCR are unavailable.
25
ACKNOWLEDGEMENTS
20 15
This study was supported by grant no. R099.512B from Cis Bio-International.
10 5 0
3
3
4
4
5
6
10 5.10 10 5.10 10 10 –1 Log molecules ml
10
7
Fig. 2. HCV viraemia levels (frequency distribution), expressed in number of viral particles=molecules ml−1, in the 48 sera.
REFERENCES 1. Lau, J. Y. N., Davis, G. L. & Kniffen, J. (1993). Significance of serum hepatitis C virus RNA levels in chronic hepatitis C. Lancet 341, 1501–4. 2. Simmonds, P., Zhang, L. Q., Watson, H., et al. (1990).
480
3.
4.
5.
6.
7.
C. Bathelier et al. Hepatitis C quantification and sequencing in blood products, haemophiliacs and drug users. Lancet 336, 1469–72. Okamoto, H., Okada, S., Sugiyama, Y. et al. (1990). Detection of hepatitis C virus RNA by a two-stage polymerase chain reaction with two pairs of primers deduced from 5′-non coding region. Japanese Journal of Experimental Medicine 60, 215–22. Garson, J. A., Ring, C., Tuke, P. & Tedder, R. S. (1990). Enhanced detection by PCR of hepatitis C virus RNA. Lancet 336, 878–9. Kaneko, S., Murakami, S., Unoura, M. & Kobayashi, K. (1992). Quantification of hepatitis C virus viremia by reverse transcription-PCR. Journal of Medical Virology 37, 278–82. Clossais-Bernard, N. & Andre´, P. M. (1994). Automated quantitative determination of hepatitis C virus viremia by reverse transcription-PCR. Journal of Clinical Microbiology 32, 1887–93. Brillanti, S., Garson, J. A., Tuke, P. W. et al. (1991). Effect of a-interferon therapy on hepatitis C viraemia in community-acquired chronic non-A, non-B hepatitis. Journal of Medical Virology 34, 136–41.
8. Lucotte, G., Mura, C. Aouize´rate, A., Champenois, T. & Marchand, J. (1994). Confirmation of hepatitis C virus positive seras by polymerase chain reaction. In Methods in DNA Amplification (Rolfs A. et al., eds). Pp. 123–126. New York: Plenum Press. 9. Ulrich, P. P., Romeo J. M., Lane P. K., Kelly I., Daniel, L. J. & Vyas G. N. (1990). Detection, semiquantification, and genetic variation in hepatitis C virus sequences amplified from plasma of blood donors with elevated alanine aminotransferase. Journal of Clinical Investigations 86, 1609–14. 10. Cheng, R. C., Chan, R. T. & Lok, A. S. F. (1993). Longitudinal study of hepatitis C viremia in chronic hepatitis C. Journal of Medical Virology 41, 338–42. 11. Saldanka, J. & Minor, P. (1994). A sensitive PCR method for detecting HCV RNA in plasma pools, blood products, and single donations. Journal of Medical Virology 43, 72–6. 12. Puchlammer-Sto¨ckl, E., Mor, W., Kundi, M., Heinz, F. X., Hofmann, H. & Kunz, C. (1994). Prevalence of hepatitis C virus RNA in serum and throat washings of children with chronic hepatitis. Journal of Medical Virology 43, 143–7.