Measuring the latent reservoir for HIV-1: Quantification bias in near full-length genome sequencing methods

Antiretroviral therapy (ART) effectively inhibits HIV-1 replication but is not curative due to the persistence of a latent viral reservoir in resting CD4(+) T cells. This reservoir is a major barrier to cure. Sequencing studies have revealed that the population of proviruses persisting in ART-treated individuals is dominated by defective proviruses that cannot give rise to viral rebound due to fatal defects including large deletions and APOBEC3-mediated hypermutation. Near full genome sequencing (nFGS) of individual proviruses is used in reservoir assays to provide an estimate of the fraction of proviruses that are intact. nFGS methods rely on a long-distance outer PCR capturing most (~9 kb) of the genome, followed by nested inner PCRs. The outer PCR is carried out at limit dilution, and interpretation of the results is based on the assumption that all proviruses are quantitatively captured. Here, we evaluate nFGS methods using the intact proviral DNA assay (IPDA), a multiplex digital droplet PCR assay that quantitates intact and defective proviruses with single molecule sensitivity using only short, highly efficient amplicons. We analyzed proviral templates of known sequence to avoid the additional complication of sequence polymorphism. With the IPDA, we quantitated molecular yields at each step of nFGS methods. We demonstrate that nFGS methods are inefficient and miss ~70% of full-length proviruses due to amplification failure at the initial outer PCR step. In contrast, proviruses with large internal deletions encompassing 70% of the genome can be quantitatively amplified under the same conditions. Accurate measurement of the latent reservoir of HIV-1 is essential for evaluating the efficacy of cure strategies, and the bias against full length proviruses in nFGS methods must be considered.