The Role of Unintegrated HIV DNA: An interview of Mark Wainberg, McGill University AIDS Centre, Jewish General Hospital, Montreal, QC, Canada
Dr. Mark A. Wainberg is head of AIDS research at the Lady Davis Institute for Medical Research, Director of the McGill University AIDS Centre located at the LDI, and Professor of Medicine and of Microbiology and Immunology at McGill University in Montreal. He is an internationally recognized scientist in the field of HIV/AIDS who served as President of the International AIDS Society between 1998-2000. He is well known for his initial identification of 3TC as an anti-viral drug, as well as for multiple contributions to the field of HIV drug resistance. His laboratory continues to work on drug resistance and drug development, and, as well, Dr. Wainberg has turned his attention to novel concepts in prevention of HIV infection. Dr. Wainberg’s laboratory has recently released a scientific paper discussing the role of unintegrated viral DNA forms in HIV pathogenesis, and as a viral reservoir (1). He kindly agreed to answer a couple of questions about this topic.
Alain Lafeuillade: What are the different forms of unintegrated HIV DNA found during HIV infection, and what is their significance?
Mark Wainberg: Unintegrated HIV DNA can exist in at least 3 forms.
First, linear proviral DNA is the product that is transcribed from viral genomic RNA by the viral reverse transcriptase enzyme in the cytoplasm of infected cells. This is the viral DNA that is directly responsible for establishing infection via integration into the host cell genome. This DNA is therefore able to traverse from the cytoplasm where it is made to the nucleus, where integration is mediated by the viral Integrase enzyme. The process of such transition is itself complex and involves a pre-integartion complex (PIC) that involves RT, Integrase, and the viral DNA; this complex is able to penetrate through nuclear pores to gain entry into the cytoplasm.
The other forms of unintegrated viral DNA that are found in infected cells are circular forms that contain either one or two copies of the viral LTR. These forms of viral DNA do not become integrated into the host genome and are found exclusively in the nucleus where they are generated from linear DNA. Despite the fact that they cannot integrate into cellular DNA, they can nonetheless be transcribed to yield gene products. As explained below, the latter products can have pathophysiological significance, as can the circular viral DNA forms themselves.
AL: Where are these unintegrated DNA forms located, and what is their relation with proviral HIV DNA?
MW: All unintegrated DNA forms derive from linear cDNA, which is itself unintegrated. Linear cDNA is the product of reverse transcription of the viral RNA genome and the precursor to integrated provirus. Thus linear unintegrated cDNA will be found in both the cell cytoplasm and nucleus. While HIV 1-LTR and 2-LTR circles are found exclusively in the nucleus, 2-LTR circle formation is entirely dependent on host non-homologous end joining (NHEJ) DNA repair factors. 1-LTR circle formation may arise through recombination, but host DNA repair factors have also been implicated. This may explain why circles are found only in cell nuclei. Only linear cDNA can integrate to become provirus. The circular products are incapable of sustaining replication on their own, but can express a limited set of genes.
AL: What interactions do they develop with cellular host factors?
MW: Linear cDNA is recognized by the cell as DNA damage, or possibly as a pathogen. 1-LTR circles can arise through RAD50/MRE11/NBS1 nuclease-mediated effects on linear cDNA. 2-LTR circles arise through the effects of the NHEJ components Ku, ligase 4 or XRCC4 on linear cDNA. Possibly the DNA-dependent protein kinase catalytic subunit (DNA-PKcs) also has a modest effect in this regard. There are good preliminary data that XPB and XPD, which are cellular DNA helicases that are components of the TFIIB basal transcription complex that plays a role in DNA nucleotide excision repair, may directly degrade linear cDNA rather than circularize it. Similarly, the homologous recombination (HR) DNA repair protein Rad52 has been reported to possibly directly degrade linear cDNA. Further, the host restriction factor APOBEC3G has been shown to modify linear cDNA, rendering it unsuitable for integration. There is also evidence that gene expression from unintegrated DNA is regulated via epigenetic modification of viral unintegrated DNA by the host, but the precise mechanisms are unclear. Overall, the relationship between pathogen sensing, DNA repair, and innate immunity remains an interesting and relatively unexplored area of HIV.
AL: What is their role in the natural history of HIV disease?
MW: The presence of unintegrated DNA prior to integration has been linked to activation and likelihood of productive infection in resting T-cells via expression of Tat and Nef from unintegrated DNA. Many studies use blockage of integration to mimic the preintegration state; in these circumstances, it can be seen that CD4, CCR5 and CXCR4 are downregulated by Nef, leading to a restriction of superinfection events, which can often hinder infection. We have also seen that that HLA molecules that are recognized by cytotoxic T-cells (CTLs) are downregulated by Nef whereas those that are recognized by natural killer cells (NKs) are not affected by Nef. Thus, Nef expression prior to integration should aid in immune evasion. For cells with slower viral replication kinetics, such as macrophages and quiescent T-cells, preintegration translation of viral early genes might lead to an extended period of benefit for the virus.
AL: What is their fate during effective antiretroviral therapy?
MW: Integrase inhibitors, such as Raltegravir, lead to a proliferation of unintegrated DNA (about 1.4 to 2 fold in our hands), due to greater substrate availability for the NHEJ pathways when integration is blocked. But unintegrated DNA forms do not possess an origin of replication, and thereby become dissipated in the overall population with each cell division. With rapid lymphocyte turnover in PBMCs, the 2-LTR surge is only visible for a few weeks during raltegravir therapy. However, if the unintegrated products are in cells which divide very slowly or do not divide, they may persist.
AL: What evidence do we have that they could participate to the HIV reservoir?
MW: There is very good evidence that unintegrated HIV DNA can persist and express genes in macrophages for up to 30 days (at which point experiments were stopped). Non-integrating lentiviral gene therapy vectors have been found in non-dividing cells of experimentally infected animals for up to one year post infection. Viruses in this circumstance are not infectious, as only early genes seem to be expressed (Tat, Nef, Rev), but such viruses can be rescued by incoming viruses which can complement the unintegrated infection. Further, if the preintegration complex (PIC) is still intact, stimulation of a resting T-cell might rescue infection from a preintegration blockage, a phenomenon known as preintegration latency. However, all of these states are relatively minor contributors when compared to bona fide latent reservoirs that are established during infection of CD4+ T-cells that are transitioning to a resting state.
(1) Sloan RD, Wainberg MA. Rhe role of unintegrated DNA in HIV infection. Retrovirology 2011; 8: 52. doi: 10.1186/1742-4690-8-52
Key words: HIV DNA, HIV reservoir