Digital PCR as a tool to measure HIV persistence

Abstract

Although antiretroviral therapy is able to suppress HIV replication in infected patients, the virus persists and rebounds when treatment is stopped. In order to find a cure that can eradicate the latent reservoir, one must be able to quantify the persisting virus. Traditionally, HIV persistence studies have used real-time PCR (qPCR) to measure the viral reservoir represented by HIV DNA and RNA. Most recently, digital PCR is gaining popularity as a novel approach to nucleic acid quantification as it allows for absolute target quantification. Various commercial digital PCR platforms are nowadays available that implement the principle of digital PCR, of which Bio-Rad's QX200 ddPCR is currently the most used platform in HIV research. Quantification of HIV by digital PCR is proving to be a valuable improvement over qPCR as it is argued to have a higher robustness to mismatches between the primers-probe set and heterogeneous HIV, and forfeits the need for a standard curve, both of which are known to complicate reliable quantification. However, currently available digital PCR platforms occasionally struggle with unexplained false-positive partitions, and reliable segregation between positive and negative droplets remains disputed. Future developments and advancements of the digital PCR technology are promising to aid in the accurate quantification and characterization of the persistent HIV reservoir.

Development of sensitive ddPCR assays to reliably quantify the proviral DNA reservoir in all common circulating HIV subtypes and recombinant forms

“…One parameter by which to assess amplification reliability is precision, which is represented by CV (Figure A, Tables and ). The CVs reported here for LTR and JO are inversely correlated with target abundance and comparable to other reports . The precision of the GAG assay is very poor in the subtype reference plasmid dilution experiments as well as in the patient samples, and is clearly affected by undetectable samples in both experiments.…”

“…The false‐positive rates of 13% to 20% for PBMC DNA NTCs and 9% to 27% for water NTCs observed in this study (Figure ) are higher than reports that use a threshold above the highest droplet in NTCs or subtract the results found in NTCs from sample results . They are, however, similar to previous reports that used predetermined or algorithm‐driven thresholds yielding false‐positive rates of 8% , 13% , 15% , 21% , 25% , 40% and 100% , affirming the paradigm of false‐positive detection as the major drawback of digital PCR . The primers and probes used in this study can also be applied in qPCR, a platform that may be preferred to ddPCR to discriminate between the presence or absence of HIV DNA .…”

“…This cloud separation distance appears to affect quantification outcome, as the ratio between LTR‐ and JO‐positive cloud heights is significantly correlated with the ratio between LTR and JO quantification results (robust regression, p = 0.011) (Figure ). It has previously been suggested that digital PCR is tolerant to suboptimal PCR efficiency , but our observation demonstrates that factors additional to PCR efficiency determine cloud separation distance and thereby quantification outcome. Although the identity of these factors remains speculative (Data ), our observations point out the necessity of optimizing PCR efficiency as well as cloud separation distance of digital PCR assays in order to detect all available target molecules.…”

Quantification of HIV DNA Using Droplet Digital PCR Techniques

“…). The ‘rain’ phenomenon is not clearly understood (Kiselinova et al., ; Rutsaert et al., ; Trypsteen et al., ), but could be attributed to multiple droplets coalescing, leading to a higher background fluorescence and inefficient PCR amplification that could be due to sequence variation, or poor signal‐to‐noise separation that could be improved by optimizing the assay or by using dual‐quenched probes (such as ZEN‐quenched probes from IDT). Still, issues with ‘rain’ as well as false positive droplets (de Oliveira et al., ; Hindson et al., ; Kiselinova et al., ; Strain et al., ), a common problem across multiple digital PCR platforms, continues to complicate low‐level detection abilities.…”

“…This level of sensitivity has made ddPCR an attractive modality in detecting rare mutations associated with cancer phenotypes in liquid biopsies (Oellerich et al., ), as well as in low‐level pathogen detection (Mu, Yan, Tang, & Liao, ; Roberts et al., ; Sedlak, Cook, Cheng, Magaret, & Jerome, ; Srisutham et al., ; Strain et al., ; Wilson et al., ). Indeed, ddPCR has been used to detect HIV DNA, with single targets of 60 to 80 nucleotides [reviewed in Rutsaert, Bosman, Trypsteen, Nijhuis, and Vandekerckhove (); also see Trypsteen, Kiselinova, Vandekerckhove, and De Spiegelaere ()]. In particular, ddPCR has been implemented to quantify total HIV DNA in peripheral blood cells including PBMCs, CD4+ T cells (de Oliveira et al., ; Eriksson et al., ; Henrich, Gallien, Li, Pereyra, & Kuritzkes, ; Kiselinova et al., ; Strain et al., ), and regulatory T cells (Treg; Dunay et al., ), as well as in cells from the central nervous system (de Oliveira et al., ).…”