With two catalytic activities and many substrates, how does HIV's reverse transcriptase enzyme know what to do to which substrate? Zooming in on the enzyme's molecular interactions provides tantalizing clues.
To replicate within their host cell, retroviruses such as HIV must make a double-stranded DNA copy of their single-stranded RNA genome. This elegant process, by which a viral enzyme — reverse transcriptase — synthesizes some 20,000 nucleotides, with the flow of genetic information moving in the opposite direction to normal (DNA to RNA), is nothing short of amazing. It also provides numerous opportunities for therapeutic intervention, as reverse transcriptase is the target of nearly half of the drugs approved for treating AIDS. But reverse transcription is not simple, for the enzyme comes into contact with diverse nucleic-acid substrates (DNA and/or RNA). On page 184 of this issue, Abbondanzieri et al.1 describe how this enzyme discriminates between substrates at intermediate steps of reverse transcription.
Reverse transcriptase has two distinct enzymatic activities: it is a DNA polymerase capable of copying either an RNA or a DNA template into a complementary DNA sequence; and it is an RNase H, capable of degrading the RNA strand of an RNA–DNA duplex into small pieces once it has been used as a template for the first DNA strand (called the 'minus' strand; Fig. 1). But the enzyme overlooks some RNA segments, called polypurine tracts (PPTs), sparing them from degradation; these segments have an unusual sequence (and so structure), which prevents the enzyme's RNase H domain from cutting them2.
Like other DNA polymerase enzymes, reverse transcriptases initiate DNA synthesis using short nucleic-acid segments called primers. The primer used by HIV reverse transcriptase to generate the minus-strand DNA is a host-cell transfer RNA sequence. For the formation of the second (plus) DNA strand, which is complementary to the minus strand, the enzyme uses PPT segments that it has generated by cleaving the template RNA on both sides of these fragments (Fig. 1b).
The molecular details of the interactions between HIV reverse transcriptase and its various substrates have been revealed through structural analyses of the enzyme when in complex with nucleic acids (DNA–DNA or RNA–DNA strands)2,3. They have also been studied by analyses of nucleic acids (DNA–DNA) in the presence of dNTPs (the incoming free nucleosides that join and extend a nucleic acid)4. Together with complementary biochemical studies, these structures showed that reverse transcriptase interacts with both nucleic acids and the incoming dNTPs when in the polymerization mode, with its polymerase active site aligned at one end (the 3′ end) of a primer. When in the RNA-cleavage mode, it interacts with RNA–DNA, with the active site of the RNase H domain aligned with the RNA strand2 (Fig. 1). A remaining puzzle has been how reverse transcriptase can distinguish PPT RNA, as an efficient primer for the initiation of second-strand-DNA synthesis, from the rest of the RNA sequence, which, having already been used as a template, is cleaved by RNase H.
Abbondanzieri et al.1 present fascinating clues about the interactions between HIV reverse transcriptase and various nucleic-acid substrates. To determine the orientation in which individual enzyme molecules bind to an RNA–DNA hybrid, the authors used the technique of fluorescence resonance energy transfer (FRET). Typical biochemical experiments involve analysing populations of molecules or complexes. By contrast, single-molecule spectroscopy techniques such as FRET allow direct, time-dependent observation of individual molecular events. For FRET analysis, two molecules of interest — in this case, reverse transcriptase and the nucleic acid — are labelled with fluorescent chemical groups, and the transfer of light energy from one molecule to the other, which depends on the distance between them, is used as a measure of their interactions.
The authors find that, when reverse transcriptase binds to a nucleic acid in which a relatively short RNA segment is hybridized to a longer DNA strand (mimicking the binding of a random primer to the minus-strand DNA), this enzyme binds almost exclusively in an RNase H cleavage mode, and the active site of the polymerase domain does not bind to the 3′ end of the RNA to extend this sequence. But if the RNA is the PPT primer, a portion of the reverse transcriptase binds to it in a polymerase-competent orientation, with the 3′ end of PPT available for extension by the polymerase active site (Fig. 2).
Abbondanzieri and colleagues also show that, in the presence of dNTPs, the number of reverse transcriptase molecules that bind to the PPT primer in the polymerase-competent orientation increases. Moreover, if a short DNA primer is bound to the longer DNA template, most reverse transcriptase enzymes bind in the polymerase-competent orientation. These results indicate that the PPT sequence 'design' is such that not only does it avoid degradation, but it is also an efficient primer for the initiation of second-strand DNA synthesis.
Remarkably, HIV reverse transcriptase can 'flip' its orientation on a nucleic-acid substrate without dissociating from it1. Such a transition between two binding orientations is unexpected, given the extensive contacts between this enzyme and its nucleic-acid substrate. The authors also find that an incoming dNTP substrate can bind to a reverse transcriptase in complex with a template–primer duplex (which lacks a hydroxyl group in the 3′ end of the primer to prevent dNTP incorporation), thereby stabilizing the polymerase-competent orientation of reverse transcriptase. Formation of such a 'dead-end' complex has previously been reported5 to prevent a reaction that unblocks chain-terminated primers and to increase the efficacy of some chain-terminating drugs that target HIV reverse transcriptase.
The authors have made another intriguing observation: the drug nevirapine — an inhibitor of HIV reverse transcriptase that binds in a pocket at the base of the flexible 'thumb' subdomain of the enzyme (Fig. 2a) — allows the enzyme to flip more rapidly. Nevirapine binding forces the thumb subdomain to undergo a conformational change that Abbondanzieri et al. propose relaxes the enzyme's grip on the nucleic-acid substrate, thus allowing it to switch orientations more rapidly. This observation implies that certain steps in the process of reverse transcription may be particularly susceptible to antiviral treatments — an idea that has also been proposed previously6,7.
Abbondanzieri and colleagues' work vividly illustrates the importance of understanding how the orientational dynamics of reverse transcriptase relate to the enzyme's activities. Their findings also raise the question of whether similar surprises are in store with regard to the enzyme's interactions with the host tRNA, during the initiation of minus-strand-DNA synthesis. Another question is whether other multifunctional enzymes indulge in similar molecular acrobatics. Examples of such enzymes are DNA polymerases, which, like reverse transcriptases, have separate non-polymerase activities, or enzymes that interact with nucleic acids, such as helicases, ligases, topoisomerases, integrases and restriction endonucleases. Whatever the answers, this work suggests that reverse transcriptase is even more agile in its handling of nucleic acids than we thought.
Abbondanzieri, E. A. et al. Nature 453, 184–189 (2008).
Sarafianos, S. G. et al. EMBO J. 20, 1449–1461 (2001).
Jacobo-Molina, A. et al. Proc. Natl Acad. Sci. USA 90, 6320–6324 (1993).
Huang, H., Chopra, R., Verdine, G. L. & Harrison, S. C. Science 282, 1669–1675 (1998).
Meyer, P. R., Matsuura, S. E., Mian, A. M., So, A. G. & Scott, W. A. Mol. Cell 4, 35–43 (1999).
Palaniappan, C., Kim, J. K., Wisniewski, M., Fay, P. J. & Bambara, R. A. J. Biol. Chem. 278, 3808–3816 (1998).
Grobler, J. A. et al. J. Biol. Chem. 282, 8005–8010 (2007).
About this article
Cite this article
Arnold, E., Sarafianos, S. An HIV secret uncovered. Nature 453, 169–170 (2008). https://doi.org/10.1038/453169b