Quantification of human immunodeficiency virus type 1 proviral load by a TaqMan real-time PCR assay

Abstract

Proviral human immunodeficiency virus type 1 (HIV-1) DNA could be a useful marker for exploring viral reservoirs and monitoring antiretroviral treatment, particularly when HIV-1 RNA is undetectable in plasma. A new technique was developed to quantify proviral HIV-1 using a TaqMan real-time PCR assay. One copy of proviral HIV-1 DNA could be detected with 100% sensitivity for five copies and the assay had a range of 6 log(10). Reproducibility was evaluated in intra- and interassays using independent extractions of the 8E5 cell line harboring the HIV-1 proviral genome (coefficients of variation [CV], 13 and 27%, respectively) and peripheral blood mononuclear cells (PBMC) from a patient with a mean proviral load of 26 copies per 10(6) PBMC (CV, 46 and 56%, respectively). The median PBMC proviral load of 21 patients, measured in a cross-sectional study, was determined to be 215 copies per 10(6) PBMC (range, <10 to 8,381). In a longitudinal study, the proviral load of 15 out of 16 patients with primary infection fell significantly during 1 year of antiretroviral therapy (P = 0.004). In the remaining patient, proviral HIV-1 DNA was detectable but not quantifiable due to a point mutation at the 5' end of the TaqMan probe. No correlation was observed between proviral load and levels of CD4(+) cells or HIV-1 RNA in plasma. TaqMan PCR is sensitive and adaptable to a large series of samples. The full interest of monitoring proviral HIV-1 DNA can now be ascertained by its application to the routine monitoring of patients.

FIG. 1

Sequence of the region amplified for HIV-1 quantification. In the B subtype consensus region from the Los Alamos data bank, the lowercase letters indicate the variable positions identified. Boxes 1, 2, and 3 correspond, respectively, to the target sequences of the forward primer P1, the probe, and the reverse primer P2 used for HIV-1 quantification. The target region was sequenced as described in Materials and Methods, and for each patient only the mutations relative to the consensus sequence are indicated. For patients 2, 7, and 13, the sequences were not obtained for the primer target regions.