Since the mid 1990s when combination antiretroviral therapy (ART) became available, HIV-infected individuals were given hope and health and the means to prevent HIV-related disease. Current therapies in widespread use today are able to suppress virus replication to undetectable levels for extended intervals, based on the measurement of HIV-1 RNA in plasma. Despite the effectiveness of ART in suppressing virus replication, viremia is quickly re-established in the vast majority of individuals who interrupt therapy. It is clear from numerous clinical studies over the years that ART suppresses virus replication, but does not eliminate HIV from infected individuals. To define the reservoirs that persist despite highly suppressive ART and measure the effects of new therapeutic interventions, assays that provide quantitative and qualitative indicators of biologically relevant HIV reservoirs are needed. This review focuses on the detection and significance of unintegrated DNA as a biological marker of de novo infection in HIV-infected individuals on long-term highly suppressive ART.
HIV PERSISTENCE DESPITE ANTIRETROVIRAL THERAPY
Persistence of HIV infection in patients who respond favorably to antiviral treatment has been attributed to reservoirs that are established and maintained despite the dramatic reduction in detectable viremia in response to therapy. Even in a highly atypical individual with undetectable HIV DNA in blood and tissue, plasma virus levels rebounded upon the cessation of antiviral therapy . Thus, therapy alone can only suppress virus replication and does not eliminate HIV that persists in individuals, despite potent antiretroviral treatment.
There is considerable evidence that lymphocytes from aviremic patients harbor proviral genomes that are transcriptionally silent, but produce replication-competent virus upon reactivation . The reservoir of latently infected cells is established in all individuals during the early phase of infection. These quiescent, resting memory CD4+ T cells harboring latent HIV constitute a reservoir that is a major impediment to virus eradication. Because of the intrinsic stability of the latent reservoir, it has been concluded that the reservoir would ensure lifetime persistence in individuals on ART.
In addition to the reservoir of latently infected cells, there is evidence of ongoing virus replication in patients on effective ART in whom viremia is suppressed to levels below the limit of detection of 50 copies/ml [3–5]. Recently, more sensitive RNA assays have demonstrated that residual viremia persists in the majority of patients on long-term suppressive ART and that there is a correlation between the levels of residual viremia and the risk of virologic rebound [6▪]. Although the source of residual viremia needs to be further defined, the correlation between the levels of viremia and the likelihood of virologic rebound may be a consequence of higher levels of ongoing cycles of virus replication in different patients. In a similar analysis of a large cohort of patients on ART, a significant linear relationship between measurable low-level viremia (>3 copies/ml) and the probability of virologic failure was determined and may also be a consequence of ongoing replication in some patients [7▪]. Interestingly, a significant fraction of residual plasma viruses from a patient on ART were found to be replication-competent  when cloned and assessed in vitro in the absence of antiretroviral drugs. Again, the source of the residual viremia in this well suppressed patient was not identified, but one possibility is that virions containing replication-competent viral genomes are generated in a compartment in which new rounds of infection are occurring and are being released into the peripheral blood. In addition, there is a correlation between the levels of residual viremia and the size of the CD4+ T-cell reservoir, suggesting that productively infected cells contribute to residual plasma viremia [9▪].
Potential sites of cryptic replication in patients on ART include the central nervous system, gut-associated lymphoid tissue, bone marrow, and lymph nodes. Recent work evaluating the effects of antivirals on HIV burden in lymph node provided evidence that suppression of virus replication in this tissue is incomplete as compared to that in peripheral blood [10▪▪]. Measurement of antiviral drug levels in the cells of lymph node and gut revealed that therapeutic levels needed to effectively suppress virus replication in these tissues are not maintained. In addition, there are elevated levels of episomal viral DNA in these tissues, suggesting they are sites of continuous virus replication. Consistent with these findings, analysis of SIV-infected macaques that were treated with combination ART for 1 year revealed that virus in lymph node was the source of rebound upon treatment interruption [11▪]. These data provide evidence that virus replication continues in lymph node tissue despite highly suppressive antiviral therapy.
It is well established that suppressive ART reduces immune activation, but individuals on long-term ART generally have higher levels than uninfected individuals . If residual virus replication is responsible for the aberrant state of immune activation in patients on ART, treatment that reduces residual replication could lead to normalization. In support of this, therapy intensification with raltegravir led to a normalization of immune activation markers  and therapy deintensification resulted in an increase in the levels of immune activation [13▪▪]. The ability of raltegravir intensification to modulate the levels of immune activation in these patients also indicates that low-level ongoing replication persists despite effective ART.
EPISOMAL DNA AND RECENT CELLULAR INFECTION
HIV infection of a susceptible cell requires that the incoming viral genome be converted to a double-stranded linear cDNA that is transported to the nucleus to establish a provirus by integration into the host chromosomal DNA as shown in Fig. 1. cDNAs that fail to integrate are degraded or converted to circular episomes containing one or two copies of the long terminal repeat (LTR) via recombination or nonhomologous end joining DNA repair, respectively. Thus, three forms of unintegrated DNA can be formed by recent cellular infection (Fig. 1). Circular genomes are not substrates for integration, but are useful markers of recent de novo infection. It should be noted that there is not an absolute consensus that episomal DNA is a valid indicator of recent cellular infection by HIV [14,15]. The major conclusion from these studies is that two LTR circles are stable and, as such, may not be a reliable marker reflecting the dynamics of HIV infection in individuals on ART. We have addressed the issue of two LTR circle stability and find that in vivo, the population of two LTR circles turns over in a matter of weeks . This issue is further supported by the observed rapid decline in unintegrated DNA in longitudinal samples from patients who initiate ART [17▪,18]. Despite the controversy over the use of episomes as a marker of recent infections, a number of investigators have relied upon them to probe the nature of virus persistence in patients on ART, with findings that are worth noting.
EPISOMAL DNA AS MARKERS OF ONGOING REPLICATION
To investigate whether low-level ongoing replication contributes to the maintenance of HIV infection, Buzón et al. determined the impact of therapy intensification using raltegravir in patients on suppressive ART. They hypothesized that treatment with an inhibitor of integration would result in a measurable increase in the levels of two LTR circular HIV if ongoing virus replication was occurring in this patient cohort. In 30% of well suppressed patients who intensified treatment with raltegravir, a transient increase in two LTR circle levels was detected approximately 2 weeks after baseline measurements. This subset of patients also had higher levels of immune activation at baseline that was subsequently normalized after raltegravir intensification. The most likely explanation for these observations is that ongoing replication occurs in at least some patients on ART and is revealed by the accumulation of two LTR circles in response to the inhibition of integration.
In a related study, sequence analysis of proviral and episomal reverse transcriptase genes from longitudinal samples of peripheral blood mononuclear cells isolated from two individuals undergoing treatment intensification with raltegravir were analyzed [19▪▪]. The authors found a statistically significant compartmentalization of genomic sequences, suggesting the two are genetically distinct and may be derived from different types of cells or anatomical sites. The lack of sequence similarity between episomes and proviruses suggest that the majority of proviruses in the peripheral blood are not representative of virus that persists during ART, but are more likely to be defective and archival in nature and of limited utility in viral reservoir analysis. This finding is consistent with the previous work that similarly concluded that episomal analysis has a greater potential to reveal the nature of persistent virus reservoirs than the proviral DNA in circulating peripheral blood mononuclear cells (PBMCs) of infected individuals on effective antiviral therapy [20▪▪].
Analysis of episomal cDNAs has revealed the nature of the reservoir that fuels viral rebound during treatment interruption and contributes to the treatment failure in patients on ART [20▪▪]. Envelope variable regions were specifically amplified from episomes and proviruses present in PBMCs just prior to rebound and phylogenetically compared to env sequences in the rebounding virus. While episomes were genetically similar to the rebounding virus, proviruses were not, suggesting that the resurgence of virus replication originates from cells or an anatomic compartment in which de novo virus replication continues despite ART. Sequences generated from episomes can also provide clinically relevant information about virus that persists in patients under different treatment conditions. As previously demonstrated and in this work [16,20▪▪], episomal sequence analysis can predict the emergence of antiviral drug resistance mutations and reveal coreceptor usage. Importantly, the specificity of episomal amplification allows for the analysis of ongoing replication in a background of mostly defective proviral genomes [21▪].
MOLECULAR ANALYSIS OF UNINTEGRATED HIV GENOMES
Although most assays to detect unintegrated DNA in patients on ART have targeted the two LTR circular forms of the genome, alternative or modified methods may increase the sensitivity and provide more convincing evidence of de novo infection.
In general, DNA fractionation and amplification of other forms of unintegrated HIV genomes would improve the detection and monitoring of ongoing replication.
The analysis of unintegrated HIV genomes is improved significantly when total cellular DNA is fractionated. Cell lysates of total DNA can, and have, been used to measure unintegrated HIV DNA, but the high concentration of chromosomal DNA limits the volume that can be used for PCR, so the cell equivalents of unintegrated DNA is quite low. Gel electrophoresis has been used to fractionate DNA to measure the relative amounts of the different molecular forms of HIV from infected cells , but this approach is not suitable for high-throughput processing and the risks of contaminating clinical samples with exogenous DNA is too great. An alternative protocol for the analysis of unintegrated HIV genomes relies on the fractionation using a modified plasmid isolation protocol [17▪,20▪▪,22–24]. The vast majority of chromosomal DNA can be separated from the unintegrated genomes with the use of this modified plasmid isolation protocol. After fractionation, the levels of proviral genomes in the low-molecular-weight fraction are negligible [20▪▪]. The resulting low-molecular-weight fraction is enriched for unintegrated genomes that can be amplified with greater sensitivity in the absence of excessive amounts of nontarget DNA. However, the fractionation involves an alkaline lysis step that denatures the DNA and likely decreases the efficiency of detection of partially degraded targets. Preparation of lysates under neutral conditions results in the isolation of unintegrated HIV DNA that is a better substrate for subsequent detection (M. Sharkey, unpublished observation).
Most studies characterizing ongoing virus replication have relied upon the two LTR circular forms of unintegrated DNA. Because of the unique sequence junction that is formed upon end-to-end joining of the linear genome, detection by PCR is exceptionally specific (Fig. 2). Although this approach provides specificity, two LTR episomes are the least abundant forms of unintegrated viral DNA in infected cells and the paucity of two LTR circles in some individuals can limit the usefulness of quantitative studies. One alternative is to measure the levels of episomes containing a single LTR. One LTR circles are approximately ten times higher than two LTR circles and can be measured with specificity. PCR primers designed to bind in nef and near the major splice donor site allow for quantitative real-time PCR of one LTR circles (Fig. 2).
Detection of the linear, unintegrated HIV genomes in patients on ART as a marker of ongoing infection would provide an additional labile DNA target, but has yet to be demonstrated as feasible with clinical samples. In-vitro studies have determined that linear, unintegrated genomes are more abundant and more labile than circular forms, so they may be a better surrogate marker of de novo infection in patients on ART. Specific amplification of linear, unintegrated genomes can be accomplished through the addition of a double-stranded linker to the 5′ end of the genome followed by quantitative PCR using an internal virus-specific primer and a primer complementary to the linker (Fig. 3). Essentially, this approach has been previously used to detect linear genomes in viremic individuals  and from in-vitro experiments  and a quantitative real-time assay could be developed with the incorporation of a few modifications.
Despite the technological challenges, DNA fractionation and inclusion of all forms of unintegrated HIV DNA in the analysis can significantly improve the means by which investigators can monitor ongoing replication and the impact of treatments designed to reduce virus replication. Refinements and modifications could increase the pool of target molecules, resulting in better sampling and improved data acquisition.
With the ultimate goal of eliminating HIV that is responsible for persistence during ART, there is a critical need for assays that can gauge the effectiveness of therapeutic interventions and monitor efforts at eradication. It is also important to couple the episomal analysis with measures of the latent reservoir to determine the degree to which ongoing replication contributes to the maintenance of latency. Several assays have been used to quantify HIV levels in patients on ART based on the detection of HIV RNA or DNA, or determination of the levels of infectious units by viral outgrowth in culture. Although the viral outgrowth assay has been critical in identifying and measuring the reservoir of latently infected resting T cells, it is time-consuming, laborious, and unlikely to be useful in tissue analyses. Measurements of plasma RNA and proviral DNA tend to be static as demonstrated by intensification studies [13▪▪,23,27,28] and probably do not reflect the impact treatments have on both the latently infected cells and ongoing cryptic replication. Measurement of cell-associated viral RNA is useful to identify the cells in which there is active transcription from provirus, including cells in which latent provirus has been activated through protocols designed to purge quiescent cells that harbor replication-competent HIV. The most useful assay would be the one that can broadly apply, is amenable to high-throughput processing, can be applied to tissue analysis, and provides a quantitative and qualitative measure of the impact of treatment on HIV persistence. The detection and measurement of unintegrated DNA has been demonstrated to fulfill these criteria and has great potential as a biological tool to monitor ongoing replication and the effects of eradication efforts. While the ultimate test of virus eradication will involve treatment interruption, methods such as episome analysis are essential to guide efforts to improve suppression and gauge the effects of protocols designed to eliminate virus reservoirs.
The author would like to thank Dunja Babic for critical reading and comments on this article.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
▪ of special interest
▪▪ of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (p. 156).
1. Chun TW, Justement JS, Murray D, et al. Rebound of plasma viremia following cessation of antiretroviral therapy despite profoundly low levels of HIV reservoir: implications for eradication. AIDS 2010; 24:2803–2808.
2. Palmer S, Josefsson L, Coffin JM. HIV reservoirs and the possibility of a cure for HIV infection. J Intern Med 2011; 270:550–560.doi: 10.1111/j.1365-2796.2011.02457.x..
3. Buzón MJ, Massanella M, Llibre JM, et al. HIV-1 replication and immune dynamics are affected by raltegravir intensification of HAART-suppressed subjects. Nat Med 2010; 16:460–465.
4. Sharkey ME, Teo I, Greenough T, et al. Persistence of episomal HIV-1 infection intermediates in patients on highly active antiretroviral therapy. Nat Med 2000; 6:76–81.
5. Agosto LM, Liszewski MK, Mexas A, et al. Patients on HAART often have an excess of unintegrated HIV DNA: implications for monitoring reservoirs. Virology 2011; 409:46–53.
6▪. Doyle T, Smith C, Vitiello P, Cambiano V, et al. Plasma HIV-1 RNA detection below 50 copies/ml and risk of virologic rebound in patients receiving highly active antiretroviral therapy. Clin Infect Dis 2012; 54:724–732.
A clinical study of a large group of patients on ART, in which residual viremia predicted rebound of limited virus replication, suggesting current antiviral regimens are not completely suppressive.
7▪. Maggiolo F, Callegaro A, Cologni G, et al. Ultrasensitive assessment of residual low-level HIV viremia in HAART-treated patients and risk of virological failure. J Acquir Immune Defic Syndr 2012; 60:473–482.
Another large clinical study in which low-level viremia is predictive of virological failure in patients on ART. The authors conclude that ongoing virus replication may contribute to residual viremia, at least in a subset of patients.
8. Sahu GK, Sarria JC, Cloyd MW. Recovery of replication-competent residual HIV-1 from plasma of a patient receiving prolonged, suppressive highly active antiretroviral therapy. J Virol 2010; 84:8348–8352.
9▪. Chun TW, Murray D, Justement JS, et al. Relationship between residual plasma viremia and the size of HIV proviral DNA reservoirs in infected individuals receiving effective antiretroviral therapy. J Infect Dis 2011; 204:135–138.
The study that demonstrates a correlation between the levels of residual viremia and the frequency of CD4+ T cells harboring HIV DNA in patients on ART, suggesting that productively infected cells contribute to residual viremia.
10▪▪. Cohen J. HIV/AIDS research. Tissue says blood is misleading, confusing HIV cure efforts. Science 2011; 334:1614.
An article summarizing the recent unpublished work concluding that antiretroviral drug concentrations in lymph node are suboptimal, resulting in de novo infection in tissue.
11▪. Horiike M, Iwami S, Kodama M, et al. Lymph nodes harbor viral reservoirs that cause rebound of plasma viremia in SIV-infected macaques upon cessation of combined antiretroviral therapy. Virology 2012; 423:107–118.
Using a SIV-infected macaque model, the authors find that high levels of viral RNA are detected in the lymphatic tissue of animals on ART and this compartment is a likely source of rebound virus upon treatment cessation.
12. Hunt PW, Landay AL, Sinclair E, et al. A low T regulatory cell response may contribute to both viral control and generalized immune activation in HIV controllers. PLoS One 2011; 6:e15924.
13▪▪. Llibre JM, Buzón MJ, Massanella M, et al. Treatment intensification with raltegravir in subjects with sustained HIV-1 viraemia suppression: a randomized 48-week study. Antivir Ther 2012; 17:355–364.doi: 10.3851/IMP1917.
Results of a raltegravir intensification study, in which the authors report a normalization of immune system activation markers and an impact on the levels of residual HIV replication.
14. Butler SL, Johnson EP, Bushman FD. Human immunodeficiency virus cDNA metabolism: notable stability of two-long terminal repeat circles. J Virol 2002; 76:3739–3747.
15. Pierson TC, Kieffer TL, Ruff CT, et al. Intrinsic stability of episomal circles formed during human immunodeficiency virus type 1 replication. J Virol 2002; 76:4138–4144.
16. Sharkey M, Triques K, Kuritzkes DR, Stevenson M. In vivo evidence for instability of episomal human immunodeficiency virus type 1 cDNA. J Virol 2005; 79:5203–5210.
17▪. Mexas AM, Graf EH, Pace MJ, et al.
Concurrent measures Of total and integrated HIV DNA monitor reservoirs and ongoing replication in eradication trials. AIDS 2012; 26:2295–2306.
A report in which measurements of HIV DNA are used to define reservoirs and gauge the impact of treatments on different patient cohorts.
18. Koelsch KK, Liu L, Haubrich R, et al. Dynamics of total, linear nonintegrated, and integrated HIV-1 DNA in vivo and in vitro. J Infect Dis 2008; 197:411–419.
19▪▪. Buzón MJ, Codoñer FM, Frost SD, et al. Deep molecular characterization of HIV-1 dynamics under suppressive HAART. PLoS Pathog 2011; 7:e1002314.
This study genetically compared episomal and proviral genomes from PBMCs of patients undergoing treatment intensification with raltegravir. Episomal sequences were found to be distinct from proviral and likely are generated in a reservoir in which limited de novo infection is maintained despite ART.
20▪▪. Sharkey M, Babic DZ, Greenough T, et al. Episomal viral cDNAs identify a reservoir that fuels viral rebound after treatment interruption and that contributes to treatment failure. PloS Pathog 2011; 7:e1001303doi:10.1371/journal.ppat.1001303..
A study demonstrating that analysis of episomal HIV DNA provides biologically relevant and clinically significant data on viral persistence in patients on ART.
21▪. Fourati S, Lambert-Niclot S, Soulie C, et al. HIV-1 genome is often defective in PBMCs and rectal tissues after long-term HAART as a result of APOBEC3 editing and correlates with the size of reservoirs. J Antimicrob Chemother 2012; 67:2323–2326.
A study of virus reservoirs in drug-naïve versus long-term treatment patients demonstrating the accumulation of defective genomes in ART patients.
22. Vallejo A, Gutierrez C, Hernandez-Novoa B, et al. The effect of intensification with raltegravir on the HIV-1 reservoir of latently infected memory CD4 T cells in suppressed patients. AIDS 2012; 26:1885–1894.
23. Gutiérrez C, Díaz L, Vallejo A, et al. Intensification of antiretroviral therapy with a CCR5 antagonist in patients with chronic HIV-1 infection: effect on T cells latently infected. PLoS One 2011; 6:e27864.
24. Gandhi RT, Coombs RW, Chan ES, et al. No effect of raltegravir intensification on viral replication markers in the blood of HIV-1-infected patients receiving antiretroviral therapy. J Acquir Immune Defic Syndr 2012; 59:229–235.
25. Bukrinsky MI, Stanwick TL, Dempsey MP, Stevenson M. Quiescent T lymphocytes as an inducible virus reservoir in HIV-1 infection. Science 1991; 254:423–427.
26. Pierson TC, Zhou Y, Kieffer TL, et al. Molecular characterization of preintegration latency in human immunodeficiency virus type 1 infection. J Virol 2002; 76:8518–8531.
27. Chege D, Kovacs C, la Porte C, et al. Effect of raltegravir intensification on HIV proviral DNA in the blood and gut mucosa of men on long-term therapy: a randomized controlled trial. AIDS 2012; 26:167–174.
28. Lam YM, McBride KL, Amin J, et al. Switching virally suppressed, treatment-experienced patients to a raltegravir-containing regimen does not alter levels of HIV-1 DNA. PLoS One 2012; 7:e31990.