- Email: [email protected]
Getting IN on Viral RNA Condensation and Virion Maturation
Leading Edge
Previews Getting IN on Viral RNA Condensation and Virion Maturation Eric O. Freed1,* 1Virus-Cell Interaction Section, HIV Dynamics and Replication Program, Center for Cancer Research, National Cancer Institute, Frederick, MD 21702, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cell.2016.08.013
The retroviral enzyme integrase plays an essential role in the virus replication cycle by catalyzing the covalent insertion of newly synthesized viral DNA into the host cell chromosome early after infection. Now, Kessl et al. report a second function of integrase: binding to the viral RNA genome in virion particles late in the virus replication cycle to promote particle maturation. Retroviruses encode three essential enzymes: reverse transcriptase (RT), integrase (IN), and protease (PR). These enzymes are initially synthesized as part of a polyprotein precursor known as GagPol. During particle assembly, Gag-Pol is recruited into virions by the major retroviral structural polyprotein Gag. Each of the three viral enzymes performs a distinct and critical step in the retroviral replication cycle: RT converts the two single-stranded copies of genomic viral RNA that are carried in the virion into a doublestranded DNA copy early after infection; IN catalyzes the covalent insertion of the newly synthesized double-stranded viral cDNA into the infected cell genome, a process known as integration; and PR cleaves Gag and Gag-Pol concomitant with particle release from the infected cell. PR-mediated cleavage of Gag and Gag-Pol triggers a major morphological change in virion structure known as maturation. RT, IN, and PR are all targeted by antiretroviral therapy used to suppress HIV-1 replication in infected individuals. In this issue of Cell, Kessl et al. (2016) describe a previously unappreciated function of the HIV-1 IN protein in viral genomic RNA binding. Catalyzing the integration of viral DNA into the host cell chromosome is the primary function of IN; however, several lines of evidence hinted at the possibility that IN also performs another important task in the viral replication cycle. Although, as expected, many mutations in IN disrupt viral DNA integration, other mutations interfere with a late step in the virus replication cycle (Engelman, 1999). These so-called ‘‘class II’’ mutations disrupt particle
maturation, resulting in the formation of virions that contain an empty core with a ribonucleoprotein (RNP) complex located outside the core (Figure 1). More recently, a new class of IN-specific inhibitors referred to as allosteric IN inhibitors (ALLINIs) were shown to induce a similar phenotype (Balakrishnan et al., 2013; Desimmie et al., 2013; Fontana et al., 2015; Jurado et al., 2013). These results collectively suggested that IN not only directs viral DNA integration, but also functions in some undefined manner during the maturation step of the replication cycle. To address this possibility, Kvaratskhelia and colleagues used a cross-linking immunoprecipitation sequencing (CLIPseq) approach (Licatalosi et al., 2008) to examine whether HIV-1 IN binds directly to viral RNA in virions. Indeed, this turned out to be the case; the binding was found to be specific for viral RNA, as cellular RNAs were not significantly cross-linked to IN. The region of the viral RNA bound to IN was also specific; although the binding between IN and RNA occurred at multiple highly structured RNA elements rather than at one dominant site, the pattern for IN-RNA binding differed markedly from that of nucleocapsid (NC), another RNA-binding viral protein. To explore the consequences of INRNA binding, Kessl et al. then used atomic force microscopy to examine the structure of a fragment of HIV-1 RNA bound to IN. They observed that IN-RNA binding led to the formation of large RNP complexes in which IN appeared to bridge several RNA molecules. Interestingly, binding of IN to RNA stabilized low-order IN multimers and prevented
1082 Cell 166, August 25, 2016 Published by Elsevier Inc.
the formation of large protein aggregates commonly observed in vitro with purified IN (Hickman et al., 1994). Having established that IN binds specifically to HIV-1 genomic RNA, the authors then mapped the residues in IN responsible for this activity. Three adjacent lysines were identified as the primary determinants of RNA binding. Mutation of two of these lysine residues resulted in the loss of IN-RNA binding in vitro without otherwise disrupting IN activity. Importantly, mutation of these lysine residues in the context of a full-length HIV-1 molecular clone resulted in the formation of virions with the ‘‘class II’’ phenotype described above (Figure 1) (Engelman, 1999): these virions were non-infectious and contained empty capsids and eccentric RNPs. Finally, bringing the story back to ALLINIs, Kessl et al. showed that treatment of virus-producing cells with an ALLINI blocked IN-RNA binding in virions but had no effect on NC-RNA binding. An ALLINI-resistant mutant showed levels of IN-RNA binding that were undiminished by treatment with the inhibitor. The ability of ALLINIs to block IN-RNA interaction may be linked to their previously reported activity in promoting IN multimerization (Jurado and Engelman, 2013), which could mask RNA-binding sites within IN. Collectively, the results presented by Kessl et al. demonstrate that HIV-1 IN binds RNA in virions during the particle maturation process, thereby promoting the condensation of the RNA within the capsid core, perhaps by bridging separate RNA molecules. In turn, the binding of IN to RNA may preserve IN as a functional, low-order multimer and prevent
Figure 1. Effect of Integrase Disruption on HIV-1 Maturation At the left is depicted the ultrastructure of the immature HIV-1 particle as a cartoon representation and as visualized by cryo-electron tomography (Fontana et al., 2015). In the cartoon, the capsid and nucleocapsid domains of Gag and the protease, reverse transcriptase, and integrase domains of Gag-Pol are indicated. In the mature particle (lower-right), the mature integrase protein is depicted binding to, and bridging, the two copies of the viral RNA. When IN-RNA binding is disrupted by mutations or allosteric IN inhibitors (ALLINIs) (upper-right), the viral RNA forms an aggregate outside the viral capsid core, resulting in the formation of empty cores. Cryo-electron tomography images reprinted from Fontana et al., 2015, with permission from the American Society for Microbiology.
the formation of non-functional, high-molecular-weight IN aggregates. Disruption of IN-RNA binding, either by mutation of IN or inhibitor treatment, results in the formation of morphologically aberrant particles in which the RNP complex fails to localize within the capsid core. It is well established that the highly ordered program of PR-mediated Gag processing plays a central role in HIV-1 maturation and in the formation of the condensed conical core that is required for the early stages of the virus replication cycle to proceed in a productive fashion (Freed, 2015). The findings of Kvaratskhelia and colleagues provide fresh insights into the process by which the genomic viral RNA is packaged into the capsid core. The results suggest that direct binding of IN to the viral RNA during particle maturation plays a heretofore unappreciated—and essential—role in ensuring
that the viral RNA ends up within, rather than outside, the core (Figure 1). Interesting questions raised by this study include whether IN-RNA binding is unique to HIV-1 or is a conserved feature of retroviral IN proteins, and the precise mechanism by which ALLINIs disrupt IN-RNA binding. The discovery of a new function for IN may lead to the development of novel inhibitors that, like the ALLINIs, block maturation by disrupting IN-RNA interactions. REFERENCES Balakrishnan, M., Yant, S.R., Tsai, L., O’Sullivan, C., Bam, R.A., Tsai, A., Niedziela-Majka, A., Stray, K.M., Sakowicz, R., and Cihlar, T. (2013). PLoS ONE 8, e74163. Desimmie, B.A., Schrijvers, R., Demeulemeester, J., Borrenberghs, D., Weydert, C., Thys, W., Vets, S., Van Remoortel, B., Hofkens, J., De Rijck, J., et al. (2013). Retrovirology 10, 57.
Engelman, A. (1999). Adv. Virus Res. 52, 411–426. Fontana, J., Jurado, K.A., Cheng, N., Ly, N.L., Fuchs, J.R., Gorelick, R.J., Engelman, A.N., and Steven, A.C. (2015). J. Virol. 89, 9765–9780. Freed, E.O. (2015). Nat. Rev. Microbiol. 13, 484–496. Hickman, A.B., Palmer, I., Engelman, A., Craigie, R., and Wingfield, P. (1994). J. Biol. Chem. 269, 29279–29287. Jurado, K.A., and Engelman, A. (2013). Expert Rev. Mol. Med. 15, e14. Jurado, K.A., Wang, H., Slaughter, A., Feng, L., Kessl, J.J., Koh, Y., Wang, W., Ballandras-Colas, A., Patel, P.A., Fuchs, J.R., et al. (2013). Proc. Natl. Acad. Sci. USA 110, 8690–8695. Kessl, J.J., Kutluay, S.B., Townsend, D., Rebenburg, S., Slaughter, A., Larue, R.C., Shkriabai, N., Bakouche, N., Fuchs, J.R., Bieniasz, P.D., et al. (2016). Cell 166, this issue, 1257–1268. Licatalosi, D.D., Mele, A., Fak, J.J., Ule, J., Kayikci, M., Chi, S.W., Clark, T.A., Schweitzer, A.C., Blume, J.E., Wang, X., et al. (2008). Nature 456, 464–469.
Cell 166, August 25, 2016 1083
Previews Getting IN on Viral RNA Condensation and Virion Maturation Eric O. Freed1,* 1Virus-Cell Interaction Section, HIV Dynamics and Replication Program, Center for Cancer Research, National Cancer Institute, Frederick, MD 21702, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cell.2016.08.013
The retroviral enzyme integrase plays an essential role in the virus replication cycle by catalyzing the covalent insertion of newly synthesized viral DNA into the host cell chromosome early after infection. Now, Kessl et al. report a second function of integrase: binding to the viral RNA genome in virion particles late in the virus replication cycle to promote particle maturation. Retroviruses encode three essential enzymes: reverse transcriptase (RT), integrase (IN), and protease (PR). These enzymes are initially synthesized as part of a polyprotein precursor known as GagPol. During particle assembly, Gag-Pol is recruited into virions by the major retroviral structural polyprotein Gag. Each of the three viral enzymes performs a distinct and critical step in the retroviral replication cycle: RT converts the two single-stranded copies of genomic viral RNA that are carried in the virion into a doublestranded DNA copy early after infection; IN catalyzes the covalent insertion of the newly synthesized double-stranded viral cDNA into the infected cell genome, a process known as integration; and PR cleaves Gag and Gag-Pol concomitant with particle release from the infected cell. PR-mediated cleavage of Gag and Gag-Pol triggers a major morphological change in virion structure known as maturation. RT, IN, and PR are all targeted by antiretroviral therapy used to suppress HIV-1 replication in infected individuals. In this issue of Cell, Kessl et al. (2016) describe a previously unappreciated function of the HIV-1 IN protein in viral genomic RNA binding. Catalyzing the integration of viral DNA into the host cell chromosome is the primary function of IN; however, several lines of evidence hinted at the possibility that IN also performs another important task in the viral replication cycle. Although, as expected, many mutations in IN disrupt viral DNA integration, other mutations interfere with a late step in the virus replication cycle (Engelman, 1999). These so-called ‘‘class II’’ mutations disrupt particle
maturation, resulting in the formation of virions that contain an empty core with a ribonucleoprotein (RNP) complex located outside the core (Figure 1). More recently, a new class of IN-specific inhibitors referred to as allosteric IN inhibitors (ALLINIs) were shown to induce a similar phenotype (Balakrishnan et al., 2013; Desimmie et al., 2013; Fontana et al., 2015; Jurado et al., 2013). These results collectively suggested that IN not only directs viral DNA integration, but also functions in some undefined manner during the maturation step of the replication cycle. To address this possibility, Kvaratskhelia and colleagues used a cross-linking immunoprecipitation sequencing (CLIPseq) approach (Licatalosi et al., 2008) to examine whether HIV-1 IN binds directly to viral RNA in virions. Indeed, this turned out to be the case; the binding was found to be specific for viral RNA, as cellular RNAs were not significantly cross-linked to IN. The region of the viral RNA bound to IN was also specific; although the binding between IN and RNA occurred at multiple highly structured RNA elements rather than at one dominant site, the pattern for IN-RNA binding differed markedly from that of nucleocapsid (NC), another RNA-binding viral protein. To explore the consequences of INRNA binding, Kessl et al. then used atomic force microscopy to examine the structure of a fragment of HIV-1 RNA bound to IN. They observed that IN-RNA binding led to the formation of large RNP complexes in which IN appeared to bridge several RNA molecules. Interestingly, binding of IN to RNA stabilized low-order IN multimers and prevented
1082 Cell 166, August 25, 2016 Published by Elsevier Inc.
the formation of large protein aggregates commonly observed in vitro with purified IN (Hickman et al., 1994). Having established that IN binds specifically to HIV-1 genomic RNA, the authors then mapped the residues in IN responsible for this activity. Three adjacent lysines were identified as the primary determinants of RNA binding. Mutation of two of these lysine residues resulted in the loss of IN-RNA binding in vitro without otherwise disrupting IN activity. Importantly, mutation of these lysine residues in the context of a full-length HIV-1 molecular clone resulted in the formation of virions with the ‘‘class II’’ phenotype described above (Figure 1) (Engelman, 1999): these virions were non-infectious and contained empty capsids and eccentric RNPs. Finally, bringing the story back to ALLINIs, Kessl et al. showed that treatment of virus-producing cells with an ALLINI blocked IN-RNA binding in virions but had no effect on NC-RNA binding. An ALLINI-resistant mutant showed levels of IN-RNA binding that were undiminished by treatment with the inhibitor. The ability of ALLINIs to block IN-RNA interaction may be linked to their previously reported activity in promoting IN multimerization (Jurado and Engelman, 2013), which could mask RNA-binding sites within IN. Collectively, the results presented by Kessl et al. demonstrate that HIV-1 IN binds RNA in virions during the particle maturation process, thereby promoting the condensation of the RNA within the capsid core, perhaps by bridging separate RNA molecules. In turn, the binding of IN to RNA may preserve IN as a functional, low-order multimer and prevent
Figure 1. Effect of Integrase Disruption on HIV-1 Maturation At the left is depicted the ultrastructure of the immature HIV-1 particle as a cartoon representation and as visualized by cryo-electron tomography (Fontana et al., 2015). In the cartoon, the capsid and nucleocapsid domains of Gag and the protease, reverse transcriptase, and integrase domains of Gag-Pol are indicated. In the mature particle (lower-right), the mature integrase protein is depicted binding to, and bridging, the two copies of the viral RNA. When IN-RNA binding is disrupted by mutations or allosteric IN inhibitors (ALLINIs) (upper-right), the viral RNA forms an aggregate outside the viral capsid core, resulting in the formation of empty cores. Cryo-electron tomography images reprinted from Fontana et al., 2015, with permission from the American Society for Microbiology.
the formation of non-functional, high-molecular-weight IN aggregates. Disruption of IN-RNA binding, either by mutation of IN or inhibitor treatment, results in the formation of morphologically aberrant particles in which the RNP complex fails to localize within the capsid core. It is well established that the highly ordered program of PR-mediated Gag processing plays a central role in HIV-1 maturation and in the formation of the condensed conical core that is required for the early stages of the virus replication cycle to proceed in a productive fashion (Freed, 2015). The findings of Kvaratskhelia and colleagues provide fresh insights into the process by which the genomic viral RNA is packaged into the capsid core. The results suggest that direct binding of IN to the viral RNA during particle maturation plays a heretofore unappreciated—and essential—role in ensuring
that the viral RNA ends up within, rather than outside, the core (Figure 1). Interesting questions raised by this study include whether IN-RNA binding is unique to HIV-1 or is a conserved feature of retroviral IN proteins, and the precise mechanism by which ALLINIs disrupt IN-RNA binding. The discovery of a new function for IN may lead to the development of novel inhibitors that, like the ALLINIs, block maturation by disrupting IN-RNA interactions. REFERENCES Balakrishnan, M., Yant, S.R., Tsai, L., O’Sullivan, C., Bam, R.A., Tsai, A., Niedziela-Majka, A., Stray, K.M., Sakowicz, R., and Cihlar, T. (2013). PLoS ONE 8, e74163. Desimmie, B.A., Schrijvers, R., Demeulemeester, J., Borrenberghs, D., Weydert, C., Thys, W., Vets, S., Van Remoortel, B., Hofkens, J., De Rijck, J., et al. (2013). Retrovirology 10, 57.
Engelman, A. (1999). Adv. Virus Res. 52, 411–426. Fontana, J., Jurado, K.A., Cheng, N., Ly, N.L., Fuchs, J.R., Gorelick, R.J., Engelman, A.N., and Steven, A.C. (2015). J. Virol. 89, 9765–9780. Freed, E.O. (2015). Nat. Rev. Microbiol. 13, 484–496. Hickman, A.B., Palmer, I., Engelman, A., Craigie, R., and Wingfield, P. (1994). J. Biol. Chem. 269, 29279–29287. Jurado, K.A., and Engelman, A. (2013). Expert Rev. Mol. Med. 15, e14. Jurado, K.A., Wang, H., Slaughter, A., Feng, L., Kessl, J.J., Koh, Y., Wang, W., Ballandras-Colas, A., Patel, P.A., Fuchs, J.R., et al. (2013). Proc. Natl. Acad. Sci. USA 110, 8690–8695. Kessl, J.J., Kutluay, S.B., Townsend, D., Rebenburg, S., Slaughter, A., Larue, R.C., Shkriabai, N., Bakouche, N., Fuchs, J.R., Bieniasz, P.D., et al. (2016). Cell 166, this issue, 1257–1268. Licatalosi, D.D., Mele, A., Fak, J.J., Ule, J., Kayikci, M., Chi, S.W., Clark, T.A., Schweitzer, A.C., Blume, J.E., Wang, X., et al. (2008). Nature 456, 464–469.
Cell 166, August 25, 2016 1083