Automated Procedure for Improving the RNA Isolation Step in Viral ...

JOURNAL OF CLINICAL MICROBIOLOGY, Jan. 2004, p. 439–440 0095-1137/04/$08.00⫹0 DOI: 10.1128/JCM.42.1.439–440.2004

Vol. 42, No. 1

Automated Procedure for Improving the RNA Isolation Step in Viral Load Testing for Human Immunodeficiency Virus Min Xu, Yung Chan, Steven H. Fischer, and Alan T. Remaley* Department of Laboratory Medicine, Warren Grant Magnuson Clinical Center, National Institutes of Health, Bethesda, Maryland 20892-1508 Received 20 June 2003/Accepted 25 September 2003

An automated RNA isolation procedure on the Qiagen BioRobot is described for performing viral load tests for human immunodeficiency virus (HIV) with the Amplicor HIV type 1 test. The new procedure improves the precision of the assay and requires significantly less labor than the presently used manual RNA isolation procedure.

QS is used to calculate the number of HIV copies/ml of sample (3). When the recommended amount of QS for the manual RNA isolation procedure (6.6 ␮l/sample) was also used for the BioRobot, the QS absorbance values were approximately threefold higher and were above the linear range of the assay. In contrast, similar GAG absorbance results were obtained for the manual and automated RNA isolation methods, which suggests that the smaller QS RNA may not be as efficiently extracted as the longer viral RNA by the manual procedure. The correlation between the amount of QS added per sample and the calculated HIV viral load result was examined in Fig. 1. RNA was isolated from a plasma pool (17,000 HIV copies/ml) by the manual method (6.6 ␮l of QS/sample) and by the BioRobot with various amounts of QS. As expected, there was a negative correlation between the amount of QS added to the sample and the calculated HIV load result. Based on Fig.

Viral load testing for human immunodeficiency virus (HIV) is commonly performed by reverse transcriptase PCR, and except for the RNA isolation step, the assay has been fully automated in the Amplicor HIV type 1 (HIV-1) monitor test (3). Automation improves the analytical performance of the assay (1, 3) and makes it more feasible to implement in routine clinical laboratories. The isolation of HIV RNA by the recommended manual ethanol precipitation procedure is, however, a major impediment during the processing of a large number of samples and contributes to the imprecision of the assay (4). A procedure for automating the RNA isolation step on the Qiagen BioRobot is described in this report for the Amplicor HIV-1 monitor test. The Amplicor HIV-1 monitor test (1.5 version; Roche, Indianapolis, Ind.) was performed on RNA isolated from EDTAplasma by the manual procedure with the Cobas Amplicor (Roche), as described by the manufacturer. RNA was extracted from 220 ␮l of EDTA-plasma by anion-exchange chromatography with the BioRobot 9604 (2) and the QIAmp 96 virus kit (Qiagen, Valencia, Calif.). The same protocol was used for processing both centrifuged samples (500 ␮l spun at 23,600 ⫻ g for 1 h; 280 ␮l of supernatant removed) and noncentrifuged samples on the BioRobot (lysis buffer, 240 ␮l; proteinase K solution, 80 ␮l; and sample test volume, 50 ␮l), except that the noncentrifuged samples were eluted with 250 ␮l of elution buffer and centrifuged samples were eluted with 125 ␮l. The Amplicor HIV-1 monitor test uses an internal quantitation standard (QS). which is a short (155 bp) RNA oligonucleotide that contains on each end sequences complementary to the GAG primers that are also used to amplify the HIV GAG target gene. Once the QS and GAG genes are amplified, they are individually measured by using two different biotinylated capture probes that hybridize to specific regions in the middle of each amplicon. The amount of each amplicon is measured colorimetrically with a streptavidin-peroxidase conjugate, and the relative ratio of absorbance of GAG to that of

FIG. 1. The effect of QS concentration on HIV load test results. A plasma pool was processed with the BioRobot, with a different amount of QS per sample, as indicated on the x axis. The open circle represents the interpolated volume of QS needed to match the results of the automated RNA extraction procedure to those of the manual procedure. Results represent the mean of duplicates.

* Corresponding author. Mailing address: National Institutes of Health, Bldg. 10, Rm. 2C-433, 10 Center Dr., Bethesda, MD 208921508. Phone: (301) 402-9796. Fax: (301) 402-1885. E-mail: aremaley @nih.gov. 439

440

NOTES

FIG. 2. Comparison plot of HIV load test result on samples extracted for RNA by the automated and manual procedures. Results for centrifuged (F) (n ⫽ 71) and noncentrifuged (E) (n ⫽ 26) samples were analyzed by Deming regression analysis.

J. CLIN. MICROBIOL.

and on five different dilutions of the sample, and it was found to be linear down to 50 copies/ml (y ⫽ 0.959x ⫹ 0.21, r2 ⫽ 0.985, P ⬍ 0.0001), which is comparable to results obtained with the manual method (3). The intra-assay precision of the two RNA isolation procedures was also determined by the replicate analysis (n ⫽ 10) of a plasma pool (50,000 copies/ml). The coefficient of variation of the QS absorbance, GAG absorbance, and the calculated HIV load test result were all found to be significantly better for the automated method (manual: QS, 38%; GAG, 37%; HIV copies/ml, 26%; and automated: QS, 20%; GAG, 12%; and HIV copies/ml, 13%). The improved precision of the automated RNA isolation method is, therefore, most likely due to more reproducible recovery for both the QS and HIV RNA. In summary, a new automated RNA extraction procedure on the BioRobot is described for viral load testing for HIV with the Roche Amplicor HIV-1 monitor test, which yields quantitative results comparable to those of the manual RNA extraction procedure. The new RNA isolation procedure improves the overall precision of the HIV reverse transcriptase PCR assay and should also be valuable for reducing the labor associated with processing a large number of samples. REFERENCES

1, it was estimated that 4 ␮l of QS per sample would be optimum for matching the results of the automated RNA extraction procedure with the manual procedure. A comparison study (Fig. 2) was performed on RNA samples extracted by the manual and the automated procedures (4 ␮l/sample), and a high correlation was observed between the two methods (y ⫽ 1.01x ⫺ 0.09, r2 ⫽ 0.937, n ⫽ 97). Furthermore, the automated RNA extraction procedure appeared to work equally well for both noncentrifuged samples and for samples with low HIV load that were first enriched for HIV by centrifugation. The linearity of the automated RNA isolation method was tested in duplicate on a high-viral-load sample (106 copies/ml)

1. DiDomenico, N., H. Link, R. Knobel, T. Caratsch, W. Weschler, Z. G. Loewy, and M. Rosentraus. 1996. COBAS AMPLICOR: fully automated RNA and DNA amplification and detection system for routine diagnostic PCR. Clin. Chem. 42:1915–1923. 2. Jongerius, J. M., M. Bovenhorst, C. L. van der Poel, J. A. van Hilten, A. C. M. Kroes, J. A. van der Does, E. F. van Leeuwen, and R. Schuurman. 2000. Evaluation of automated nucleic acid extraction devices for application in HCV NAT. Transfusion 40:871–874. 3. Triques, K., J. Coste, J. L. Perret, C. Segarra, E. Mpoudi, J. Reynes, E. Delaporte, A. Butcher, K. Dreyer, S. Herman, J. Spadoro, and M. Peeters. 1999. Efficiencies of four versions of the AMPLICOR HIV-1 MONITOR test for quantification of different subtypes of human immunodeficiency virus type 1. J. Clin. Microbiol. 37:110–116. 4. Xu, M., Y. Chan, S. H. Fisher, and A. T. Remaley. 2002. Improvement of AMPLICOR human immunodeficiency virus type 1 viral load test (version 1.5) by addition of a coprecipitant during the RNA isolation step. J. Clin. Microbiol. 40:2616–2617.