Portal de Periódicos da CAPES

HIV-1 Reverse Transcriptase: A Metamorphic Protein with Three Stable States

2019; Elsevier BV; Volume: 27; Issue: 3 Linguagem: Inglês

10.1016/j.str.2018.11.011

1878-4186

Robert E. London,

Biochemical and Molecular Research

There has been a steadily increasing appreciation of the fact that the relationship between protein sequence and structure is often sufficiently ambiguous to allow a single sequence to adopt alternative, stable folds. Living organisms have been able to utilize such metamorphic proteins in remarkable and unanticipated ways. HIV-1 reverse transcriptase is among the earliest such proteins identified and remains a unique example in which a functional heterodimer contains two, alternatively folded polymerase domains. Structural characterization of the p66 precursor protein combined with NMR spectroscopic and molecular modeling studies have provided insights into the factors underlying the metamorphic transition and the subunit-specific programmed unfolding step required to expose the protease cleavage site within the ribonuclease H domain, supporting the conversion of the p66/p66′ precursor into the mature p66/p51 heterodimer. There has been a steadily increasing appreciation of the fact that the relationship between protein sequence and structure is often sufficiently ambiguous to allow a single sequence to adopt alternative, stable folds. Living organisms have been able to utilize such metamorphic proteins in remarkable and unanticipated ways. HIV-1 reverse transcriptase is among the earliest such proteins identified and remains a unique example in which a functional heterodimer contains two, alternatively folded polymerase domains. Structural characterization of the p66 precursor protein combined with NMR spectroscopic and molecular modeling studies have provided insights into the factors underlying the metamorphic transition and the subunit-specific programmed unfolding step required to expose the protease cleavage site within the ribonuclease H domain, supporting the conversion of the p66/p66′ precursor into the mature p66/p51 heterodimer. Metamorphic proteins, characterized by sufficient sequence-structure ambiguity to support at least two distinct, stable folds, are increasingly being identified and new functional roles discovered (Goodchild et al., 2011Goodchild S.C. Curmi P.M.G. Brown L.J. Structural gymnastics of multifunctional metamorphic proteins.Biophys. Rev. 2011; 3: 143Crossref PubMed Scopus (31) Google Scholar, Murzin, 2008Murzin A.G. Biochemistry. Metamorphic proteins.Science. 2008; 320: 1725-1726Crossref PubMed Scopus (174) Google Scholar). Examples include the cytokine lymphotactin (Volkman et al., 2009Volkman B.F. Liu T.Y. Peterson F.C. Chapter 3. Lymphotactin structural dynamics.Methods Enzymol. 2009; 461: 51-70Crossref PubMed Scopus (22) Google Scholar), redox-sensitive transcription factors such as OxyR (Choi et al., 2001Choi H. Kim S. Mukhopadhyay P. Cho S. Woo J. Storz G. Ryu S.E. Structural basis of the redox switch in the OxyR transcription factor.Cell. 2001; 105: 103-113Abstract Full Text Full Text PDF PubMed Scopus (434) Google Scholar), the DNA single-strand break repair scaffold XRCC1 (Cuneo and London, 2010Cuneo M.J. London R.E. Oxidation state of the XRCC1 N-terminal domain regulates DNA polymerase beta binding affinity.Proc. Natl. Acad. Sci. U S A. 2010; 107: 6805-6810Crossref PubMed Scopus (60) Google Scholar), the chloride ion channel CLIC1 (Littler et al., 2004Littler D.R. Harrop S.J. Fairlie W.D. Brown L.J. Pankhurst G.J. Pankhurst S. DeMaere M.Z. Campbell T.J. Bauskin A.R. Tonini R. Mazzanti M. et al.The intracellular chloride ion channel protein CLIC1 undergoes a redox-controlled structural transition.J. Biol. Chem. 2004; 279: 9298-9305Crossref PubMed Scopus (161) Google Scholar), the RNA polymerase (Tahirov et al., 2002Tahirov T.H. Temiakov D. Anikin M. Patlan V. McAllister W.T. Vassylyev D.G. Yokoyama S. Structure of a T7 RNA polymerase elongation complex at 2.9 A resolution.Nature. 2002; 420: 43-50Crossref PubMed Scopus (183) Google Scholar), the spindle checkpoint protein Mad2 (Skinner et al., 2008Skinner J.J. Wood S. Shorter J. Englander S.W. Black B.E. The Mad2 partial unfolding model: regulating mitosis through Mad2 conformational switching.J. Cell Biol. 2008; 183: 761-768Crossref PubMed Scopus (44) Google Scholar), the HIV-1 capsid maturation switches (Wagner et al., 2016Wagner J.M. Zadrozny K.K. Chrustowicz J. Purdy M.D. Yeager M. Ganser-Pornillos B.K. Pornillos O. Crystal structure of an HIV assembly and maturation switch.Elife. 2016; 5https://doi.org/10.7554/eLife.17063Crossref Scopus (93) Google Scholar), and the cyanobacterial circadian clock protein KaiB (Tseng et al., 2017Tseng R. Goularte N.F. Chavan A. Luu J. Cohen S.E. Chang Y.G. Heisler J. Li S. Michael A.K. Tripathi S. Golden S.S. et al.Structural basis of the day-night transition in a bacterial circadian clock.Science. 2017; 355: 1174-1180Crossref PubMed Scopus (89) Google Scholar). Recent analyses indicate that this class of proteins may be more common than generally appreciated (Porter and Looger, 2018Porter L.L. Looger L.L. Extant fold-switching proteins are widespread.Proc. Natl. Acad. Sci. U S A. 2018; 115: 5968-5973Crossref PubMed Scopus (70) Google Scholar). All of the above examples represent functional expansion of a single gene sequence at the level of folding, much as post-translational modifications can introduce new or modified protein functional characteristics (Jeffery, 2016Jeffery C.J. Protein species and moonlighting proteins: very small changes in a protein's covalent structure can change its biochemical function.J. Proteomics. 2016; 134: 19-24Crossref PubMed Scopus (49) Google Scholar). The driving force for the evolution of such structures is particularly strong for organisms that are subject to genome size constraints for which maximizing the information density of the coding sequence becomes critical. This constraint is most important for RNA viruses subject to both chemical and biochemical instability that limit the mean genome size to approximately 9,000 bases (Belshaw et al., 2007Belshaw R. Pybus O.G. Rambaut A. The evolution of genome compression and genomic novelty in RNA viruses.Genome Res. 2007; 17: 1496-1504Crossref PubMed Scopus (107) Google Scholar). Among the earliest metamorphic proteins characterized, HIV-1 reverse transcriptase (RT) remains unique in that two alternatively folded forms of the polymerase domain are both present in the p66/p51 RT heterodimer (Kohlstaedt et al., 1992Kohlstaedt L.A. Wang J. Friedman J.M. Rice P.A. Steitz T.A. Crystal-structure at 3.5 angstrom resolution of HIV-1 reverse-transcriptase complexed with an inhibitor.Science. 1992; 256: 1783-1790Crossref PubMed Scopus (1757) Google Scholar, Wang et al., 1994Wang J. Smerdon S.J. Jager J. Kohlstaedt L.A. Rice P.A. Friedman J.M. Steitz T.A. Structural basis of asymmetry in the human-immunodeficiency-virus type-1 reverse-transcriptase heterodimer.Proc. Natl. Acad. Sci. U S A. 1994; 91: 7242-7246Crossref PubMed Scopus (179) Google Scholar). Maturation from a p66 monomer to a p66/p51 heterodimer is a multi-step process, the characterization of which has required additional structural and spectroscopic investigations (London, 2016London R.E. Structural maturation of HIV-1 reverse transcriptase—a metamorphic solution to genomic instability.Viruses. 2016; 8https://doi.org/10.3390/v8100260Crossref PubMed Scopus (11) Google Scholar and references therein). The steps involved include an initial metamorphic transition followed by formation of an asymmetric p66/p66′ homodimer, additional conformational rearrangements, and a subsequent, subunit-specific programmed unfolding of the RH domain in one subunit that renders it proteolytically labile. These studies have revealed the remarkable behavior of an HIV pharmaceutical target of central importance, and elucidated maturation intermediates that may serve as alternative targets for drug interference while providing broader insights in the area of protein folding/unfolding. Each subunit of the mature p66/p51 RT heterodimer contains a polymerase domain. In p66, the domain adopts an active structure with a solvent-exposed catalytic site that is accessible to substrates, while in p51 the polymerase domain is more compact and the important catalytic residues are buried in the interior. These structural differences are most easily observed by color coding the polymerase subdomains: fingers (green), palm (blue), thumb (orange), and connection (magenta) (Figure 1). To simplify the presentation, we have not adhered to a rigorous distinction between domains and polymerase subdomains. In the p51 polymerase domain, the connection forms a large, non-covalent interface with the fingers/palm (use of the slash symbol for the fingers/palm refers to the discontinuity of the domain sequences so that they cannot be separated). Formation of this interface results in substantially greater buried hydrophobicity. This characteristic led Steitz and coworkers to conclude that this form of the polymerase is more stable than the extended form observed in the p66 subunit, resulting in the predictions that the monomeric forms of both p51 and p66 would adopt the more compact, p51-type structure observed in the p51 subunit of RT and that both of the homodimers formed would be asymmetric (Wang et al., 1994Wang J. Smerdon S.J. Jager J. Kohlstaedt L.A. Rice P.A. Friedman J.M. Steitz T.A. Structural basis of asymmetry in the human-immunodeficiency-virus type-1 reverse-transcriptase heterodimer.Proc. Natl. Acad. Sci. U S A. 1994; 91: 7242-7246Crossref PubMed Scopus (179) Google Scholar). This prediction of Wang et al., 1994Wang J. Smerdon S.J. Jager J. Kohlstaedt L.A. Rice P.A. Friedman J.M. Steitz T.A. Structural basis of asymmetry in the human-immunodeficiency-virus type-1 reverse-transcriptase heterodimer.Proc. Natl. Acad. Sci. U S A. 1994; 91: 7242-7246Crossref PubMed Scopus (179) Google Scholar was largely borne out in studies following a mutational strategy in which residues Lys219–Met230 were deleted (Zheng et al., 2014Zheng X. Pedersen L.C. Gabel S.A. Mueller G.A. Cuneo M.J. DeRose E.F. Krahn J.M. London R.E. Selective unfolding of one ribonuclease H domain of HIV reverse transcriptase is linked to homodimer formation.Nucleic Acids Res. 2014; 42: 5361-5377Crossref PubMed Scopus (22) Google Scholar). This segment forms a disordered loop in p51 but is structurally and functionally important in the extended polymerase domain present in the p66 subunit (Ding et al., 1998Ding J. Das K. Hsiou Y. Sarafianos S.G. Clark Jr., A.D. Jacobo-Molina A. Tantillo C. Hughes S.H. Arnold E. Structure and functional implications of the polymerase active site region in a complex of HIV-1 RT with a double-stranded DNA template-primer and an antibody Fab fragment at 2.8 A resolution.J. Mol. Biol. 1998; 284: 1095-1111Crossref PubMed Scopus (303) Google Scholar). The p51 construct lacking these palm loop residues, p51ΔPL, does not dimerize, allowing a crystal structure of the monomer to be determined (Zheng et al., 2014Zheng X. Pedersen L.C. Gabel S.A. Mueller G.A. Cuneo M.J. DeRose E.F. Krahn J.M. London R.E. Selective unfolding of one ribonuclease H domain of HIV reverse transcriptase is linked to homodimer formation.Nucleic Acids Res. 2014; 42: 5361-5377Crossref PubMed Scopus (22) Google Scholar, Zheng et al., 2015Zheng X. Perera L. Mueller G.A. DeRose E.F. London R.E. Asymmetric conformational maturation of HIV-1 reverse transcriptase.eLife. 2015; 4: e06359Crossref Scopus (17) Google Scholar). This result contrasts with many point mutations which in our experience reduce, but do not block, dimer formation. Although the structure of p51ΔPL is generally similar to that of the p51 subunit of RT, there are two major differences: (1) the residues linking the thumb to the palm and connection domains are disordered, so that the orientation of the thumb subdomain is unconstrained; and (2) the C-terminal helix of the connection domain, αM, which occupies the space between the connection and thumb, is unraveled. Additional NMR studies with p66ΔPL and with other p66 constructs under conditions favoring the monomer indicated that there are no interface contacts between the polymerase and RH domains in the monomer, leading to the p66M structure shown in Figure 1C. The figure also indicates where the deleted palm loop would have been located. As summarized below, each of the three different structures shown in Figure 1 plays a specific role in the RT maturation process, and in particular the structural difference between the p51 subunit and the p66 monomer plays a critical role in determining at what point in the maturation process the RH domain becomes susceptible to proteolytic cleavage. Figure 2 summarizes the basic steps involved in formation of the p66/p51 heterodimer from the p66 monomer. The process is initiated by dissociation of the connection domain from the fingers/palm in p66M (step 1). This dissociation releases multiple interdomain constraints that allow the remaining intradomain constraints to become more dominant, supporting subsequent maturation steps while reducing the rate of fingers/palm:connection reassociation. These domain rearrangements and interactions create a large, solvent-exposed hydrophobic surface that supports initial homodimer formation with a second monomer (step 2), so that the initially formed p66/p66′ homodimer exists as a structural heterodimer, sometimes referred to in the literature as an asymmetric homodimer. Note that subsequent to dimerization it becomes possible to distinguish the two subunits of the homodimer, and we introduce primes to denote the subunit formed directly from the monomer without initial domain rearrangements that is destined for proteolytic processing. The homodimer may dissociate, reversing the process, or undergo further adjustments that include more significant conformational changes in the palm and connection domains, and subsequent formation of the additional subunit interface between the p66 RH domain and the p66′ thumb′ domain (step 3) that is not present in the initially formed homodimer. The unraveled C-terminal segment of the connection domain in p66′ that connects the polymerase′ and RH′ domains then occupies the available channel that has been created in between these two domains in p66′ (step 4) and waits patiently for the rate-limiting unraveling of N-terminal residues in the supernumerary RH′ domain. Upon release of Tyr427′ and a few adjacent residues, helix αM′ in the p66′ subunit extends by incorporating these residues, and the stability of the resulting structure forces the truncated and destabilized RH′ domain to persist for a longer period of time with increasing likelihood of more extensive unfolding—a process likely supported by additional factors in the virion (Ilina et al., 2018Ilina T.V. Slack R.L. Elder J.H. Sarafianos S.G. Parniak M.A. Ishima R. Effect of tRNA on the maturation of HIV-1 reverse transcriptase.J. Mol. Biol. 2018; 430: 1891-1900Crossref PubMed Scopus (7) Google Scholar, London, 2016London R.E. Structural maturation of HIV-1 reverse transcriptase—a metamorphic solution to genomic instability.Viruses. 2016; 8https://doi.org/10.3390/v8100260Crossref PubMed Scopus (11) Google Scholar). These transitions are discussed in greater detail below. As illustrated in Figure 1, the p66 monomer (p66M) contains a globular complex of the fingers/palm:connection subdomains, with the thumb (T) and RH domains connected by flexible segments. In the above, the colon represents the non-covalent interface between the fingers/palm and the connection subdomains. This structure is well supported by both crystallographic and NMR studies of Ile-labeled p66M (Zheng et al., 2014Zheng X. Pedersen L.C. Gabel S.A. Mueller G.A. Cuneo M.J. DeRose E.F. Krahn J.M. London R.E. Selective unfolding of one ribonuclease H domain of HIV reverse transcriptase is linked to homodimer formation.Nucleic Acids Res. 2014; 42: 5361-5377Crossref PubMed Scopus (22) Google Scholar). In contrast with the methyl labeled NMR studies, the size of RT makes 1H-15N TROSY (transverse relaxation optimized spectroscopy) spectra of U-[2H,15N]p66 considerably more difficult to interpret, and previous studies have led to divergent interpretations (Sharaf et al., 2014Sharaf N.G. Poliner E. Slack R.L. Christen M.T. Byeon I.J. Parniak M.A. Gronenborn A.M. Ishima R. The p66 immature precursor of HIV-1 reverse transcriptase.Proteins. 2014; 82: 2343-2352Crossref PubMed Scopus (24) Google Scholar, Zheng et al., 2017Zheng X. Mueller G.A. Kim K. Perera L. DeRose E.F. London R.E. Identification of drivers for the metamorphic transition of HIV-1 reverse transcriptase.Biochem. J. 2017; 474: 3321-3338Crossref PubMed Scopus (6) Google Scholar). However, Zheng et al., 2017Zheng X. Mueller G.A. Kim K. Perera L. DeRose E.F. London R.E. Identification of drivers for the metamorphic transition of HIV-1 reverse transcriptase.Biochem. J. 2017; 474: 3321-3338Crossref PubMed Scopus (6) Google Scholar have concluded that this approach also yields results consistent with the monomer structure shown in Figure 1C. Despite its thermodynamic stability, the fingers/palm:connection interface does not persist indefinitely. We have previously described this structure as "spring-loaded" because upon dissociation, removal of the interface constraints allows the component subdomains to rapidly express conformational preferences that differ from those observed in the complex. The difference is most immediately apparent for the fingers/palm, for which the two subdomains rotate away from each other, adopting a more expanded conformation similar to that in the p66 subunit of RT (Figure 3) (Unge et al., 1994Unge T. Knight S. Bhikhabhai R. Lövgren S. Dauter Z. Wilson K. Strandberg B. 2.2 A resolution structure of the amino-terminal half of HIV-1 reverse transcriptase (fingers and palm subdomains).Structure. 1994; 2: 953-961Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar, Zheng et al., 2015Zheng X. Perera L. Mueller G.A. DeRose E.F. London R.E. Asymmetric conformational maturation of HIV-1 reverse transcriptase.eLife. 2015; 4: e06359Crossref Scopus (17) Google Scholar). This expansion appears to result in part from the preference of α-helix E to eliminate the kink near Phe160, adopting a more standard helical geometry after the interdomain constraints have been removed (Zheng et al., 2017Zheng X. Mueller G.A. Kim K. Perera L. DeRose E.F. London R.E. Identification of drivers for the metamorphic transition of HIV-1 reverse transcriptase.Biochem. J. 2017; 474: 3321-3338Crossref PubMed Scopus (6) Google Scholar). Straightening of the helix may also contribute to dimerization, since αE is located at the subunit interface. In addition, a structural comparison of p66M and p66E indicates that the C-terminal segment of the palm subdomain undergoes a significant structural change, from an extended structure that runs along the surface of p66M and includes the disordered loop mentioned above to a more compact, three-stranded amphiphilic β-sheet structure in p66E. Circular dichroism analysis of the isolated segment (residues 226–241) indicates that it adopts a conformation more similar to that in the p66 compared with the p51 subunit, and the β-sheet structure is likely stabilized by hydrophobic interactions of the side chains (Zheng et al., 2017Zheng X. Mueller G.A. Kim K. Perera L. DeRose E.F. London R.E. Identification of drivers for the metamorphic transition of HIV-1 reverse transcriptase.Biochem. J. 2017; 474: 3321-3338Crossref PubMed Scopus (6) Google Scholar). This amphiphilic β sheet, which corresponds to β12-β13-β14 in p66E, forms a β sandwich with the hydrophobic surface of the β6-β9-β10 sheet in the palm. Occasional separation of these two β sheets, perhaps linked to motion of the thumb subdomain (Ivetac and McCammon, 2009Ivetac A. McCammon J.A. Elucidating the inhibition mechanism of HIV-1 non-nucleoside reverse transcriptase inhibitors through multicopy molecular dynamics simulations.J. Mol. Biol. 2009; 388: 644-658Crossref PubMed Scopus (71) Google Scholar) creates the non-nucleoside RT inhibitor binding pocket that surprisingly is not present in the unliganded enzyme (Hsiou et al., 1996Hsiou Y. Ding J. Das K. Clark A.D. Hughes S.H. Arnold E. Structure of unliganded HIV-1 reverse transcriptase at 2.7 angstrom resolution: implications of conformational changes for polymerization and inhibition mechanisms.Structure. 1996; 4: 853-860Abstract Full Text Full Text PDF PubMed Scopus (278) Google Scholar). Formation of this β-sheet structure also creates a subdomain interface that facilitates positioning of the adjacent thumb so that the tips of the fingers and thumb make a small, hydrophobic contact that competes with reassociation of the connection subdomain and helps to position the domains optimally for dimer formation (Figure 3). The non-conservative L289K mutation in the thumbnail has been shown to inhibit dimerization (Goel et al., 1993Goel R. Beard W.A. Kumar A. Casas-Finet J.R. Strub M.P. Stahl S.J. Lewis M.S. Bebenek K. Becerra S.P. Kunkel T.A. et al.Structure/function studies of HIV-1(1) reverse transcriptase: dimerization-defective mutant L289K.Biochemistry. 1993; 32: 13012-13018Crossref PubMed Scopus (40) Google Scholar, Zheng et al., 2010Zheng X. Mueller G.A. Cuneo M.J. Derose E.F. London R.E. Homodimerization of the p51 subunit of HIV-1 reverse transcriptase.Biochemistry. 2010; 49: 2821-2833Crossref PubMed Scopus (16) Google Scholar), probably by interfering with the formation of the hydrophobic fingers:thumb interface. Hence, although the globular fingers/palm:connection structure shown in Figure 1C is more stable, the changes summarized in Figure 3 compete with reassociation of the fingers/palm and connection subdomains, retarding the reassociation rate. The effect is to extend the time period during which the structurally transformed monomer may associate with a second monomer that is in the ground state. The mechanism shown in Figure 2 hinges on the ability of the metamorphic structure to adopt a group of relatively unstable structures for a sufficient time period to allow a stable dimer to form; or, from a thermodynamic perspective, on an ideal balance between the thermodynamic forces favoring the monomer and the kinetic factors protracting the lifetime of the p66E structure. Stabilization of the transiently reorganized p66M structure is contingent on formation of an initial dimer that follows the creation of a large interface that can interact with p66M. The requirement for structural rearrangement prior to dimer formation is consistent with the inability of p66M lacking the palm loop noted above to dimerize (Zheng et al., 2015Zheng X. Perera L. Mueller G.A. DeRose E.F. London R.E. Asymmetric conformational maturation of HIV-1 reverse transcriptase.eLife. 2015; 4: e06359Crossref Scopus (17) Google Scholar), and corresponds to a conformational selection process whereby the predominant p66M selects a p66E-like structure for dimerization (Figure 2). This conclusion is also supported by kinetic studies of Barkley and coworkers (Braz et al., 2010Braz V.A. Holladay L.A. Barkley M.D. Efavirenz binding to HIV-1 reverse transcriptase monomers and dimers.Biochemistry. 2010; 49: 601-610Crossref PubMed Scopus (29) Google Scholar, Venezia et al., 2009Venezia C.F. Meany B.J. Braz V.A. Barkley M.D. Kinetics of association and dissociation of HIV-1 reverse transcriptase subunits.Biochemistry. 2009; 48: 9084-9093Crossref PubMed Scopus (24) Google Scholar). The metamorphic transition discussed above must create a sufficient interface to support initial dimerization with p66M. It is, however, likely that additional conformational changes, particularly involving segments near the domain and subunit interfaces, undergo subsequent adjustments from initial p66E-like structures to p66E. The exposed hydrophobic surfaces created by the fingers/palm:connection dissociation provide initial high-affinity, low-specificity surfaces that allow these rearrangements to occur, further enhancing dimer stability. A comparison of the domain/subdomain structures indicates that the fingers, thumb, and RH domains exhibit fairly minimal changes, while the palm and connection undergo more significant conformational alterations. Changes in the palm subdomain include formation of a short β sheet by the C-terminal segment discussed above, as well as adjustments at the interface involving residues 91–103, that are disordered in the RT216 structure (Unge et al., 1994Unge T. Knight S. Bhikhabhai R. Lövgren S. Dauter Z. Wilson K. Strandberg B. 2.2 A resolution structure of the amino-terminal half of HIV-1 reverse transcriptase (fingers and palm subdomains).Structure. 1994; 2: 953-961Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar). This segment interacts with residues in the p66′ β7-β8 loop upon formation of the dimer (Pandey et al., 2001Pandey P.K. Kaushik N. Talele T.T. Yadav P.N. Pandey V.N. The beta7-beta8 loop of the p51 subunit in the heterodimeric (p66/p51) human immunodeficiency virus type 1 reverse transcriptase is essential for the catalytic function of the p66 subunit.Biochemistry. 2001; 40: 9505-9512Crossref PubMed Scopus (34) Google Scholar). The ribbon diagram in Figure 2 (middle, right) illustrates how the p66 connection domain must adjust from its monomer conformation (pink) to the conformation observed in the mature heterodimer (magenta). The transition involves elimination of the bend in helix αL and large movements of the subsequent loop containing Trp406. Straightening of αL creates a binding site for Lys331′ in the p66′ subunit. The β18-αK loop at the RH-domain interface also shows significant variation. It is unclear to what extent these changes may precede, accompany, or follow dimerization. Formation of the RH:thumb′ interface present in mature RT occurs on a somewhat slower time scale. However, NMR observations, particularly for Ile434 located at this interface, indicate that it is formed within the first hour after dimerization occurs, and likely on a significantly shorter time scale (Zheng et al., 2015Zheng X. Perera L. Mueller G.A. DeRose E.F. London R.E. Asymmetric conformational maturation of HIV-1 reverse transcriptase.eLife. 2015; 4: e06359Crossref Scopus (17) Google Scholar). Formation of this interface is likely a cooperative process in which the connection:RH, connection′:thumb′, connection:connection′, and RH:thumb′ interfaces are mutually supportive. The cooperative nature of the connection′-thumb′ interactions is suggested by the monomer structure in which both the αM′ helix and thumb′ orientation are disordered (compare Figures 1B and 1C). Biochemical and structural studies of the isolated RT RH domain have demonstrated that the F440|Y441 cleavage site that serves as the HIV-1 protease substrate is located on the central β strand near the center of the domain (Davies et al., 1991Davies 2nd, J.F. Hostomska Z. Hostomsky Z. Jordan S.R. Matthews D.A. Crystal structure of the ribonuclease H domain of HIV-1 reverse transcriptase.Science. 1991; 252: 88-95Crossref PubMed Scopus (525) Google Scholar, Himmel et al., 2009Himmel D.M. Maegley K.A. Pauly T.A. Bauman J.D. Das K. Dharia C. Clark Jr., A.D. Ryan K. Hickey M.J. Love R.A. Hughes S.H. et al.Structure of HIV-1 reverse transcriptase with the inhibitor beta-thujaplicinol bound at the RNase H active site.Structure. 2009; 17: 1625-1635Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar, Himmel et al., 2014Himmel D.M. Myshakina N.S. Ilina T. Van Ry A. Ho W.C. Parniak M.A. Arnold E. Structure of a dihydroxycoumarin active-site inhibitor in complex with the RNase H domain of HIV-1 reverse transcriptase and structure-activity analysis of inhibitor analogs.J. Mol. Biol. 2014; 426: 2617-2631Crossref PubMed Scopus (35) Google Scholar, Hostomska et al., 1991Hostomska Z. Matthews D.A. Davies 2nd, J.F. Nodes B.R. Hostomsky Z. Proteolytic release and crystallization of the RNase H domain of human immunodeficiency virus type 1 reverse transcriptase.J. Biol. Chem. 1991; 266: 14697-14702PubMed Google Scholar, Kirschberg et al., 2009Kirschberg T.A. Balakrishnan M. Squires N.H. Barnes T. Brendza K.M. Chen X. Eisenberg E.J. Jin W. Kutty N. Leavitt S. Liclican A. Liu Q. et al.RNase H active site inhibitors of human immunodeficiency virus type 1 reverse transcriptase: design, biochemical activity, and structural information.J. Med. Chem. 2009; 52: 5781-5784Crossref PubMed Scopus (92) Google Scholar) (Figure 1). Formation of a stable RH domain requires an additional 14 residues beginning at Tyr427 rather than at Tyr441 (Becerra et al., 1990Becerra A.P. Clore G.M. Gronenborn A.M. Karlstrom A.R. Stahl S.J. Wilson S.H. Wingfield P.T. Purification and characterization of the RNase H domain of HIV-l reverse transcriptase expressed in recombinant Escherichia coli.FEBS Lett. 1990; 270: 76-80Crossref PubMed Scopus (42) Google Scholar, Davies et al., 1991Davies 2nd, J.F. Hostomska Z. Hostomsky Z. Jordan S.R. Matthews D.A. Crystal structure of the ribonuclease H domain of HIV-1 reverse transcriptase.Science. 1991; 252: 88-95Crossref PubMed Scopus (525) Google Scholar, Hostomska et al., 1991Hostomska Z. Matthews D.A. Davies 2nd, J.F. Nodes B.R. Hostomsky Z. Proteolytic release and crystallization of the RNase H domain of human immunodeficiency virus type 1 reverse transcriptase.J. Biol. Chem. 1991; 266: 14697-14702PubMed Google Scholar). Multiple crystal structures of the isolated RH domain all include Tyr427 (Davies et al., 1991Davies 2nd, J.F. Hostomska Z. Hostomsky Z. Jordan S.R. Matthews D.A. Crystal structure of the ribonuclease H domain of HIV-1 reverse transcriptase.Science. 1991; 252: 88-95Crossref PubMed Scopus (525) Google Scholar, Himmel et al., 2009Himmel D.M. Maegley K.A. Pauly T.A. Bauman J.D. Das K. Dharia C. Clark Jr., A.D. Ryan K. Hickey M.J. Love R.A. Hughes S.H. et al.Structure of HIV-1 reverse transcriptase with the inhibitor beta-thujaplicinol bound at the RNase H active site.Structure. 2009; 17: 1625-1635Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar, Himmel et al., 2014Himmel D.M. Myshakina N.S. Ilina T. Van Ry A. Ho W.C. Parniak M.A. Arnold E. Structure of a dihydroxycoumarin active-site inhibitor in complex with the RNase H domain of HIV-1 reverse transcriptase and structure-activity analysis of inhibitor analogs.J. Mol. Biol. 2014; 426: 2617-2631Crossref PubMed Scopus (35) Google Scholar, Kirschberg et al., 2009Kirschberg T.A. Balakrishnan M. Squires N.H. Barnes T. Brendza K.M. Chen X. Eisenberg E.J. Jin W. Kutty N. Leavitt S. Liclican A. Liu Q. et al.RNase H active site inhibitors of human immunodeficiency virus type 1 reverse transcriptase: design, biochemical activity, and structural information.J. Med. Chem. 2009; 52: 5781-5784Crossref PubMed Scopus (92) Google Scholar). The stable, isolated RH domain that results is resistant to cleavage by HIV-1 protease (Hostomska et al., 1991Hostomska Z. Matthews D.A. Davies 2nd, J.F. Nodes B.R. Hostomsky Z. Proteolytic release and crystallization of the RNase H domain of human immunodeficiency virus type 1 reverse transcriptase.J. Biol. Chem. 1991; 266: 14697-14702PubMed Google Scholar, Tomasselli et al., 1993Tomasselli A.G. Sarcich J.L. Barrett L.J. Reardon I.M. Howe W.J. Evans D.B. Sharma S.K. Heinrikson R.L. Human immunodeficiency virus type-1 reverse transcriptase and ribonuclease H as substrates of the viral protease.Protein Sci. 1993; 2: 2167-2176Crossref PubMed Scopus (35) Google Scholar). In contrast, cleavage of the p66/p66′ homodimer at F440′|Y441′ does not produce a stable RH domain, but rather a group of peptides that subsequently undergoes further proteolysis (Tomasselli et al., 1993Tomasselli A.G. Sarcich J.L. Barrett L.J. Reardon I.M. Howe W.J. Evans D.B. Sharma S.K. Heinrikson R.L. Human immunodeficiency virus type-1 reverse transcriptase and ribonuclease H as substrates of the viral protease.Protein Sci. 1993; 2: 2167-2176Crossref PubMed Scopus (35) Google Scholar). Conversion of p66′ to p51 is controlled by the accessibility of the cleavage site, and must occur either prior to domain folding or subsequent to a specific unfolding step. Excessive processing of p66 can result from destabilizing mutations within the RNase H domain, including those located in the segment between Tyr427 and Tyr441 (Abram and Parniak, 2005Abram M.E. Parniak M.A. Virion instability of human immunodeficiency virus type 1 reverse transcriptase (RT) mutated in the protease cleavage site between RT p51 and the RT RNase H domain.J. Virol. 2005; 79: 11952-11961Crossref PubMed Scopus (44) Google Scholar, Hostomska et al., 1991Hostomska Z. Matthews D.A. Davies 2nd, J.F. Nodes B.R. Hostomsky Z. Proteolytic release and crystallization of the RNase H domain of human immunodeficiency virus type 1 reverse transcriptase.J. Biol. Chem. 1991; 266: 14697-14702PubMed Google Scholar, Navarro et al., 2001Navarro J.M. Damier L. Boretto J. Priet S. Canard B. Quérat G. Sire J. Glutamic residue 438 within the protease-sensitive subdomain of HIV-1 reverse transcriptase is critical for heterodimer processing in viral particles.Virology. 2001; 290: 300-308Crossref PubMed Scopus (9) Google Scholar, Slack et al., 2015Slack R.L. Spiriti J. Ahn J. Parniak M.A. Zuckerman D.M. Ishima R. Structural integrity of the ribonuclease H domain in HIV-1 reverse transcriptase.Proteins. 2015; 83: 1526-1538Crossref PubMed Scopus (6) Google Scholar), and seriously impairs viral viability by resulting in excessive formation of p51 relative to p66. Not only is excessive processing of p66 a potential problem for the virus, but processing must halt when 50% of p66 has been cleaved—a difficult constraint for an enzyme to achieve. The solution to the problem of p66 processing is intrinsic to the structures shown in Figure 1. The cleavage site within the RH domain lies buried in the folded RH domain in the p66 monomer (Figure 1C), preventing premature processing. The step 1 metamorphic transition does not alter this situation, so the cleavage site remains protected. Subsequent to dimerization, and maturation of the p66/p66′ interface, a channel between the connection′ and thumb′ domain in p66′ is created and the subunit moves toward completing the structural transition by forming helix αM′ within this channel (Figure 4). At this point, an apparent conflict arises since the helix seeks to incorporate several residues that also form the N terminus of the stable RH domain. No such conflict exists in the p66 subunit, in which the corresponding residues extend from the connection to the RH domain without conflict (Figure 4). This leads to a functional tug-of-war between the polymerase′ and the RH′ domains of p66′ for common residues, a conflict that is completely dependent on the alternative structures of the polymerase domain. In the monomer, the RH domain wins the competition, retaining residues beginning with Tyr427, while in the final stage of dimer maturation the polymerase′ wins the tug-of-war, resulting in RH′ destabilization. The structural conflict rests on the strong dependence of RH-domain stability on a few N-terminal residues, particularly Tyr427. Support for this conclusion comes from observations on destabilization of an N-terminally truncated RH domain, RHΔNT, in which the first two residues, Tyr427 and Gln428, were deleted, and Leu429 mutated to methionine (L429M) (Zheng et al., 2014Zheng X. Pedersen L.C. Gabel S.A. Mueller G.A. Cuneo M.J. DeRose E.F. Krahn J.M. London R.E. Selective unfolding of one ribonuclease H domain of HIV reverse transcriptase is linked to homodimer formation.Nucleic Acids Res. 2014; 42: 5361-5377Crossref PubMed Scopus (22) Google Scholar), and from studies of a domain-swapped RH dimer in which the monomer-dimer interconversion rate is greatly accelerated in constructs lacking Tyr427 (Zheng et al., 2016Zheng X. Pedersen L.C. Gabel S.A. Mueller G.A. DeRose E.F. London R.E. Unfolding the HIV-1 reverse transcriptase RNase H domain - how to lose a molecular tug-of-war.Nucleic Acids Res. 2016; 44: 1776-1788Crossref PubMed Scopus (10) Google Scholar). The interconversion rate is negligible for the RH domain at 25°C, but sufficiently rapid for RHΔNT to render isolation of either form in a pure state challenging. Studies of HIV-2 RT, including data related to its maturation, are limited, although the structure of the HIV-2 RT heterodimer has been reported (PDB: 1MU2 [Ren et al., 2002Ren J. Bird L.E. Chamberlain P.P. Stewart-Jones G.B. Stuart D.I. Stammers D.K. Structure of HIV-2 reverse transcriptase at 2.35-A resolution and the mechanism of resistance to non-nucleoside inhibitors.Proc. Natl. Acad. Sci. U S A. 2002; 99: 14410-14415Crossref PubMed Scopus (162) Google Scholar]). A structural comparison indicates that all of the critical elements involved in maturation of the HIV-1 enzyme are present in the HIV-2 enzyme as well (London, 2016London R.E. Structural maturation of HIV-1 reverse transcriptase—a metamorphic solution to genomic instability.Viruses. 2016; 8https://doi.org/10.3390/v8100260Crossref PubMed Scopus (11) Google Scholar). One interesting difference arises because the analogous F440|Y441 cleavage site in the HIV-2 enzyme is resistant to HIV-2 protease, so that the protein is cleaved at a different site, M484|A485 (Fan et al., 1995Fan N. Rank K.B. Leone J.W. Heinrikson R.L. Bannow C.A. Smith C.W. Evans D.B. Poppe S.M. Tarpley W.G. Rothrock D.J. et al.The differential processing of homodimers of reverse transcriptases from human immunodeficiency viruses type 1 and 2 is a consequence of the distinct specificities of the viral proteases.J. Biol. Chem. 1995; 270: 13573-13579PubMed Google Scholar). This result is consistent with the maturation steps outlined above in which the polymerase′-RH′ tug-of-war leads to RH′ domain unfolding in the HIV-2 RT homodimer, but the actual site of cleavage is determined by the selectivity of the corresponding protease once the domain unfolds. As is observed for HIV-1 RT (Tomasselli et al., 1993Tomasselli A.G. Sarcich J.L. Barrett L.J. Reardon I.M. Howe W.J. Evans D.B. Sharma S.K. Heinrikson R.L. Human immunodeficiency virus type-1 reverse transcriptase and ribonuclease H as substrates of the viral protease.Protein Sci. 1993; 2: 2167-2176Crossref PubMed Scopus (35) Google Scholar), maturation releases not an intact RH domain but fragments that are further proteolyzed. Remarkably, the HIV-2 RT p68/p68′ homodimer precursor can be cleaved at F440|Y441 by the HIV-1 protease (Fan et al., 1995Fan N. Rank K.B. Leone J.W. Heinrikson R.L. Bannow C.A. Smith C.W. Evans D.B. Poppe S.M. Tarpley W.G. Rothrock D.J. et al.The differential processing of homodimers of reverse transcriptases from human immunodeficiency viruses type 1 and 2 is a consequence of the distinct specificities of the viral proteases.J. Biol. Chem. 1995; 270: 13573-13579PubMed Google Scholar). The ability of the HIV-1 and HIV-2 proteases to cleave two sites separated by 44 residues, each of which is buried in the folded RH domain, provides strong support for an HIV-2 RT maturation pathway that also entails complete, subunit-selective unfolding of one RH domain. As summarized herein, RT maturation is a story of ambiguity and conflict. There is conflict between the interdomain structural restraints and the conformational preferences of the isolated subdomains that drive many of the rearrangements characterizing the initial metamorphic transition. The conflict between thermodynamic and kinetic factors supports the extended lifetime of the unstable, conformationally altered polymerase domain that allows dimer formation and subsequent conformational adjustments before the thermodynamic stability of the monomer is reasserted. There is intrinsic instability inherent in the RH domain resulting from linker strain and perhaps other factors that, when coupled with the additional competition between the polymerase and RH domain for common residues, is sufficient to result in RH-domain unfolding, exposing the initially buried cleavage site. Nevertheless, all of these highly unusual behavior patterns are merely atypical combinations of conventional protein-folding tendencies.