- Email: [email protected]
How a Virus Blocks a Cellular Emergency Access Lane to the Nucleus, STAT!
Cell Host & Microbe
Previews absence of HP1, the level of compaction typical for heterochromatin cannot be achieved, and the H3K9me3 mark will be lost rapidly. Heterochromatin has been linked to a G1/S phase transition checkpoint in other organisms. It is conceivable that a similar checkpoint is in place in the blood-stage malaria parasite, as it is crucial to closely monitor the integrity of its heterochromatic domains to prevent revelation of its entire antigenic repertoire simultaneously. The 50% replication defect in the Hda2 knockdown probably reflects the proportion of the parasite population not fit for S phase entry due to insufficient heterochromatin integrity. In the HP1 knockdown, which exhibits a complete arrest in the cycle following HP1 depletion, none of the asexual parasites fulfil the checkpoint criteria, exemplified by the complete breakdown of var gene silencing and the resulting hyperexpression of PfEMP1 on the surface of the host cell. Similar to the 50% replication defect in the Hda2 knockdown, half of the HP1 knockdown population commits to game-
tocytogenesis in the cycle of HP1 depletion. This probably reflects the proportion of the population expressing AP2-G to levels sufficient to activate early gametocyte genes. These two papers report the induction of gametocytogenesis by gene knockdown. Until now, the only way of inducing gametocytogenesis in vitro has been applying environmental stress. The question remains: what induces gametocytogenesis in vivo? Is there an external signal? Is it merely stochastic? Also, it remains to be explored experimentally whether there is a direct link between the frequency of var gene switching and the rate of gametocyte conversion. ACKNOWLEDGMENTS We would like to thank Dr. Kathrin Witmer (Imperial College) for critical reading of the manuscript and acknowledge financial support from the Wellcome Trust (Ref: WT094752). REFERENCES Brancucci, N.M.B., Bertschi, N.L., Zhu, L., Niederwieser, I., Chin, W.H., Wampfler, R., Freymond, C.,
Rottmann, M., Felger, I., Bozdech, Z., and Voss, T.S. (2014). Cell Host Microbe 16, this issue, 165–176. Canzio, D., Chang, E.Y., Shankar, S., Kuchenbecker, K.M., Simon, M.D., Madhani, H.D., Narlikar, G.J., and Al-Sady, B. (2011). Mol. Cell 41, 67–81. Coleman, B.I., Skillman, K.M., Jiang, R.H.Y., Childs, L.M., Altenhofen, L.M., Ganter, M., Leung, Y., Goldowitz, I., Kafsack, B.F.C., Marti, M., et al. (2014). Cell Host Microbe 16, this issue, 177–186. Flueck, C., Bartfai, R., Volz, J., Niederwieser, I., Salcedo-Amaya, A.M., Alako, B.T., Ehlgen, F., Ralph, S.A., Cowman, A.F., Bozdech, Z., et al. (2009). PLoS Pathog. 5, e1000569. Guizetti, J., and Scherf, A. (2013). Cell. Microbiol. 15, 718–726. Kafsack, B.F., Rovira-Graells, N., Clark, T.G., Bancells, C., Crowley, V.M., Campino, S.G., Williams, A.E., Drought, L.G., Kwiatkowski, D.P., Baker, D.A., et al. (2014). Nature 507, 248–252. Lopez-Rubio, J.J., Mancio-Silva, L., and Scherf, A. (2009). Cell Host Microbe 5, 179–190. Sinha, A., Hughes, K.R., Modrzynska, K.K., Otto, T.D., Pfander, C., Dickens, N.J., Religa, A.A., Bushell, E., Graham, A.L., Cameron, R., et al. (2014). Nature 507, 253–257. Tonkin, C.J., Carret, C.K., Duraisingh, M.T., Voss, T.S., Ralph, S.A., Hommel, M., Duffy, M.F., Silva, L.M., Scherf, A., Ivens, A., et al. (2009). PLoS Biol. 7, e84.
How a Virus Blocks a Cellular Emergency Access Lane to the Nucleus, STAT! Matthew D. Daugherty1,2,* and Harmit S. Malik1,2 1Division
of Basic Sciences Hughes Medical Institute Fred Hutchinson Cancer Research Center, Seattle, WA 98109, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.chom.2014.07.013 2Howard
Early in viral infection, the STAT1 transcription factor is rapidly transported into the nucleus using a nonconventional import mechanism to establish an antiviral state. In this issue, Xu et al. (2014) show how the Ebola virus VP24 protein precisely blocks specialized STAT1 import while leaving other cellular import processes intact. As anyone stuck in traffic watching an ambulance race past can appreciate, it is important to make sure that emergency responders have special access to roadways to ensure their rapid response. A similar emergency response is mounted in cells that sense viral infection via the
interferon (IFN) signaling cytokines, allowing the host to rapidly establish an antiviral state by upregulating the expression of hundreds of genes, known collectively as interferon-stimulated genes (ISGs). Key to this signaling cascade is the phosphorylation and nuclear import of STAT1
150 Cell Host & Microbe 16, August 13, 2014 ª2014 Elsevier Inc.
and STAT2 (signal transducer and activator of transcription) proteins, which are powerful transcription activators that switch on ISG expression. As part of the rapid immune response, STAT1 nuclear import is mediated in a noncanonical way, akin to an ‘‘emergency access
Cell Host & Microbe
Previews A
B
C
Figure 1. Ebola VP24 Specifically Blocks the Noncanonical Nuclear Import of STAT1 (A) Most cellular proteins rely on a classical nuclear localization signal (cNLS) for their import into the nucleus, via cNLS binding to armadillo (ARM) repeats 2–8 (green) of karyopherin-a (KPNA) proteins, including the NPI-1 subfamily of KPNAs. (B) Upon IFN binding to the IFN receptor on a cell surface, Janus kinases (JAK1 and TYK2) become activated and phosphorylate STAT1. Phosphorylated STAT1 dimerizes and then presents a nonclassical NLS (ncNLS), which is recognized by ARMs 8–10 of NPI-1 KPNAs (blue) for nuclear import. STAT1 binding to KPNAs does not interfere with cNLS binding, suggesting that both cargos may be imported simultaneously. Once nuclear, dimeric STAT1 activates expression of several hundred interferon stimulated genes (ISGs), which establish an antiviral state in the cell. (C) Upon Ebola infection, STAT1 is phosphorylated normally but is unable to bind KPNA due to competition with Ebola VP24 for specific binding to ARMs 8–10. This competitive mechanism of VP24 still allows cNLS-containing cargoes to bind KPNAs, thus specifically blocking only STAT1 import.
lane’’ to the nucleus. In this issue of Cell Host & Microbe, Xu et al. (Xu et al., 2014) use insights from structural biology and biochemistry to uncover how Ebola virus blocks this specialized access mechanism to disable the innate immune system with exquisite specificity. The import of most proteins into the nucleus depends on a combination of karyopherin-a proteins (KPNA, also known as importin-a) that directly bind cargo proteins and karyopherin-b proteins that carry the karyopherin-a-cargo complex into the nucleus via the nuclear pore. Typically, KPNA proteins recognize and bind a classical nuclear localization signal (cNLS) primarily composed of one or two short regions of highly basic residues on import cargoes (Figure 1A) (Cook et al., 2007). STAT1, however, lacks such a cNLS and instead relies on a
noclassical NLS (ncNLS) for import by only a subset of KPNAs, known as NPI-1 karyopherins (KPNA1, KPNA5, and KPNA6 or importin-a 5, 6 and 7). The STAT1 ncNLS is exclusively presented in the STAT1 dimer, allowing discrimination between the preactivation STAT1 monomer and the postactivation STAT1 dimer. Interestingly, cNLS cargos and the STAT1 ncNLS bind physically distinct armadillo (ARM) repeat regions of KPNAs such that binding of STAT1 does not compete with cNLS binding, suggesting that both cargos can be carried at the same time by the same KPNA (Reich and Liu, 2006) (Figure 1B). Though this specialized access to the nucleus by STAT1 has been known for years, the details of what comprises an ncNLS and how NPI-1 KPNAs recognize such an import signal have been elusive.
Resolution of this conundrum has now emerged from an unexpected source: Ebola virus. Ebola virus and the related Marburg virus are members of the filoviridae, enveloped negative-sense RNA viruses that can cause viral hemorrhagic fevers that can reach fatality rates near 90%. One characteristic of Ebola virus infections is the lack of IFN response in infected cells, essentially nullifying the innate immune response to viral infection. Such dysregulation of the innate immune system is likely responsible for the unchecked replication of the virus and rapid dissemination throughout the body (Basler and Amarasinghe, 2009). Previous work by the Basler group and colleagues showed that the VP24 protein from Ebola, a minor matrix protein present in incoming viral particles, was singlehandedly able to disrupt signaling by
Cell Host & Microbe 16, August 13, 2014 ª2014 Elsevier Inc. 151
Cell Host & Microbe
Previews IFN-a/IFN-b and IFN-g by preventing phosphorylated STAT1 from entering the nucleus (Reid et al., 2006). Additional biochemical mapping using truncations and individual mutations revealed that VP24 required the same regions on KPNA1, KPNA5, and KPNA6 to bind to these karyopherins as phosphorylated STAT1 required for binding, suggesting that VP24, like STAT1, contains a ncNLS (Reid et al., 2007). In the current study, Xu et al. (Xu et al., 2014) reveal mechanistic details of Ebola virus inhibition of STAT signaling while simultaneously providing molecular insight into STAT1 nuclear import. By solving the crystal structure of VP24 bound to KPNA5, the authors showed that VP24 tightly binds KPNA5 using an extended surface to interact with the three C-terminal ARMs of KPNA5. Using these structural data, the authors showed that the same residues required for KPNA50 s interaction with VP24 were also required for interaction with STAT1, thus providing the first structural and biochemical glimpse of how KPNA can recognize an ncNLS. These data also elegantly explain the specificity of VP24 and STAT1 binding by only a subset of KPNAs, as many of the critical binding residues in KPNA5 are well conserved among NPI-1 KPNAs, but not well conserved among non-STAT1-binding KPNAs. Expanding from KPNA5 binding of an ncNLS to the in vivo function of VP24, the authors showed that residues in VP24 that make direct contacts to KPNA5 are required to block nuclear import of STAT1 and its transcriptional functions. Thus, consistent with a previously proposed model (Reid et al., 2007), VP24 binding to KPNA5 serves as a competitive inhibitor of phosphorylatedSTAT1 binding. Importantly, although VP24 directly interfered with nuclear import of STAT1 via its ncNLS, it did not block binding of a cNLS-containing cargo protein by KPNA5. Thus, Ebola virus does not inhibit all nuclear import via NPI-1 KPNA proteins; this would presumably be highly deleterious for the virus. Instead, Ebola VP24 protein specif-
ically blocks STAT1’s ‘‘emergency access lane’’ to the nucleus with surgical precision (Figure 1C). Why would Ebola virus choose this strategy to antagonize the host immune system? The finding that Ebola virus disrupts STAT1 function is not unexpected, given that IFN signaling is such an important axis of the innate immune system. Indeed, numerous viruses antagonize STAT signaling by mechanisms such as STAT1 and STAT2 protein degradation or dephosphorylation of activated STAT1 (Najjar and Fagard, 2010), and these would appear to be at least as effective a strategy. However, encoding a competitive inhibitor of STAT1-KPNA interaction might present some evolutionary advantages to Ebola. This would especially be true if the constraint to maintain STAT1KPNA interactions imposes greater difficulty for hosts to evolve away from antagonism via ‘‘mimicry’’ by Ebola virus VP24 (Elde and Malik, 2009). In contrast, changes in STAT proteins at viral protein binding interfaces might be sufficient to escape antagonism by other mechanisms, and therefore easier to evolve in principle (Daugherty and Malik, 2012). Interestingly, Ebola virus may not be unique in employing this strategy; measles virus and rotavirus may also block STAT1 nuclear import, although it remains unclear whether they also employ a competitive inhibitor (Yarbrough et al., 2014). By establishing a clear mechanism for Ebola VP24 blockage of IFN signaling in infected cells and structurally defining the interaction between KPNAs and ncNLSs, this work continues the long tradition of using viruses to molecularly dissect cellular pathways that are often highly pleiotropic, especially in the field of nucleocytoplasmic transport (Yarbrough et al., 2014). For instance, one of the important implications of this work is a deeper understanding of the evolution and functions of the ‘‘emergency access lane’’ for nuclear import. Although we do not know whether an ncNLS-mediated STAT1 import is a superior strategy compared to import using a cNLS, such
152 Cell Host & Microbe 16, August 13, 2014 ª2014 Elsevier Inc.
a specialized system may have evolved so that STAT1 does not disrupt, or is not disrupted by, the import of normal cellular cargos upon a cellular emergency such as IFN induction (Reich and Liu, 2006). By using Ebola virus VP24 as a tool, and specific mutations of the newly defined ncNLS-binding residues on KPNAs, future work will be able to dissect which additional cargoes use this specialized system for nuclear import and with what consequences. One thing is for certain: with the ‘‘emergency access lane’’ in full service, it allows pathogens like Ebola virus to specifically block it without paying any of the pleiotropic costs of blocking nuclear import in general. Regular traffic can flow unimpeded! ACKNOWLEDGMENTS We are supported by an Irvington Institute Fellowship of the Cancer Research Institute (M.D.D.) and the Howard Hughes Medical Institute (M.D.D., H.S.M.).
REFERENCES Basler, C.F., and Amarasinghe, G.K. (2009). J. Interferon Cytokine Res. 29, 511–520. Cook, A., Bono, F., Jinek, M., and Conti, E. (2007). Annu. Rev. Biochem. 76, 647–671. Daugherty, M.D., and Malik, H.S. (2012). Annu. Rev. Genet. 46, 677–700. Elde, N.C., and Malik, H.S. (2009). Nat. Rev. Microbiol. 7, 787–797. Najjar, I., and Fagard, R. (2010). Biochimie 92, 425–444. Reich, N.C., and Liu, L. (2006). Nat. Rev. Immunol. 6, 602–612. Reid, S.P., Leung, L.W., Hartman, A.L., Martinez, O., Shaw, M.L., Carbonnelle, C., Volchkov, V.E., Nichol, S.T., and Basler, C.F. (2006). J. Virol. 80, 5156–5167. Reid, S.P., Valmas, C., Martinez, O., Sanchez, F.M., and Basler, C.F. (2007). J. Virol. 81, 13469– 13477. Xu, W., Edwards, M.R., Borek, D.M., Feagins, A.R., Mittal, A., Alinger, J.B., Berry, K.N., Yen, B., Hamilton, J., Brett, T.J., et al. (2014). Cell Host Microbe in press. Yarbrough, M.L., Mata, M.A., Sakthivel, R., and Fontoura, B.M. (2014). Traffic 15, 127–140.
Previews absence of HP1, the level of compaction typical for heterochromatin cannot be achieved, and the H3K9me3 mark will be lost rapidly. Heterochromatin has been linked to a G1/S phase transition checkpoint in other organisms. It is conceivable that a similar checkpoint is in place in the blood-stage malaria parasite, as it is crucial to closely monitor the integrity of its heterochromatic domains to prevent revelation of its entire antigenic repertoire simultaneously. The 50% replication defect in the Hda2 knockdown probably reflects the proportion of the parasite population not fit for S phase entry due to insufficient heterochromatin integrity. In the HP1 knockdown, which exhibits a complete arrest in the cycle following HP1 depletion, none of the asexual parasites fulfil the checkpoint criteria, exemplified by the complete breakdown of var gene silencing and the resulting hyperexpression of PfEMP1 on the surface of the host cell. Similar to the 50% replication defect in the Hda2 knockdown, half of the HP1 knockdown population commits to game-
tocytogenesis in the cycle of HP1 depletion. This probably reflects the proportion of the population expressing AP2-G to levels sufficient to activate early gametocyte genes. These two papers report the induction of gametocytogenesis by gene knockdown. Until now, the only way of inducing gametocytogenesis in vitro has been applying environmental stress. The question remains: what induces gametocytogenesis in vivo? Is there an external signal? Is it merely stochastic? Also, it remains to be explored experimentally whether there is a direct link between the frequency of var gene switching and the rate of gametocyte conversion. ACKNOWLEDGMENTS We would like to thank Dr. Kathrin Witmer (Imperial College) for critical reading of the manuscript and acknowledge financial support from the Wellcome Trust (Ref: WT094752). REFERENCES Brancucci, N.M.B., Bertschi, N.L., Zhu, L., Niederwieser, I., Chin, W.H., Wampfler, R., Freymond, C.,
Rottmann, M., Felger, I., Bozdech, Z., and Voss, T.S. (2014). Cell Host Microbe 16, this issue, 165–176. Canzio, D., Chang, E.Y., Shankar, S., Kuchenbecker, K.M., Simon, M.D., Madhani, H.D., Narlikar, G.J., and Al-Sady, B. (2011). Mol. Cell 41, 67–81. Coleman, B.I., Skillman, K.M., Jiang, R.H.Y., Childs, L.M., Altenhofen, L.M., Ganter, M., Leung, Y., Goldowitz, I., Kafsack, B.F.C., Marti, M., et al. (2014). Cell Host Microbe 16, this issue, 177–186. Flueck, C., Bartfai, R., Volz, J., Niederwieser, I., Salcedo-Amaya, A.M., Alako, B.T., Ehlgen, F., Ralph, S.A., Cowman, A.F., Bozdech, Z., et al. (2009). PLoS Pathog. 5, e1000569. Guizetti, J., and Scherf, A. (2013). Cell. Microbiol. 15, 718–726. Kafsack, B.F., Rovira-Graells, N., Clark, T.G., Bancells, C., Crowley, V.M., Campino, S.G., Williams, A.E., Drought, L.G., Kwiatkowski, D.P., Baker, D.A., et al. (2014). Nature 507, 248–252. Lopez-Rubio, J.J., Mancio-Silva, L., and Scherf, A. (2009). Cell Host Microbe 5, 179–190. Sinha, A., Hughes, K.R., Modrzynska, K.K., Otto, T.D., Pfander, C., Dickens, N.J., Religa, A.A., Bushell, E., Graham, A.L., Cameron, R., et al. (2014). Nature 507, 253–257. Tonkin, C.J., Carret, C.K., Duraisingh, M.T., Voss, T.S., Ralph, S.A., Hommel, M., Duffy, M.F., Silva, L.M., Scherf, A., Ivens, A., et al. (2009). PLoS Biol. 7, e84.
How a Virus Blocks a Cellular Emergency Access Lane to the Nucleus, STAT! Matthew D. Daugherty1,2,* and Harmit S. Malik1,2 1Division
of Basic Sciences Hughes Medical Institute Fred Hutchinson Cancer Research Center, Seattle, WA 98109, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.chom.2014.07.013 2Howard
Early in viral infection, the STAT1 transcription factor is rapidly transported into the nucleus using a nonconventional import mechanism to establish an antiviral state. In this issue, Xu et al. (2014) show how the Ebola virus VP24 protein precisely blocks specialized STAT1 import while leaving other cellular import processes intact. As anyone stuck in traffic watching an ambulance race past can appreciate, it is important to make sure that emergency responders have special access to roadways to ensure their rapid response. A similar emergency response is mounted in cells that sense viral infection via the
interferon (IFN) signaling cytokines, allowing the host to rapidly establish an antiviral state by upregulating the expression of hundreds of genes, known collectively as interferon-stimulated genes (ISGs). Key to this signaling cascade is the phosphorylation and nuclear import of STAT1
150 Cell Host & Microbe 16, August 13, 2014 ª2014 Elsevier Inc.
and STAT2 (signal transducer and activator of transcription) proteins, which are powerful transcription activators that switch on ISG expression. As part of the rapid immune response, STAT1 nuclear import is mediated in a noncanonical way, akin to an ‘‘emergency access
Cell Host & Microbe
Previews A
B
C
Figure 1. Ebola VP24 Specifically Blocks the Noncanonical Nuclear Import of STAT1 (A) Most cellular proteins rely on a classical nuclear localization signal (cNLS) for their import into the nucleus, via cNLS binding to armadillo (ARM) repeats 2–8 (green) of karyopherin-a (KPNA) proteins, including the NPI-1 subfamily of KPNAs. (B) Upon IFN binding to the IFN receptor on a cell surface, Janus kinases (JAK1 and TYK2) become activated and phosphorylate STAT1. Phosphorylated STAT1 dimerizes and then presents a nonclassical NLS (ncNLS), which is recognized by ARMs 8–10 of NPI-1 KPNAs (blue) for nuclear import. STAT1 binding to KPNAs does not interfere with cNLS binding, suggesting that both cargos may be imported simultaneously. Once nuclear, dimeric STAT1 activates expression of several hundred interferon stimulated genes (ISGs), which establish an antiviral state in the cell. (C) Upon Ebola infection, STAT1 is phosphorylated normally but is unable to bind KPNA due to competition with Ebola VP24 for specific binding to ARMs 8–10. This competitive mechanism of VP24 still allows cNLS-containing cargoes to bind KPNAs, thus specifically blocking only STAT1 import.
lane’’ to the nucleus. In this issue of Cell Host & Microbe, Xu et al. (Xu et al., 2014) use insights from structural biology and biochemistry to uncover how Ebola virus blocks this specialized access mechanism to disable the innate immune system with exquisite specificity. The import of most proteins into the nucleus depends on a combination of karyopherin-a proteins (KPNA, also known as importin-a) that directly bind cargo proteins and karyopherin-b proteins that carry the karyopherin-a-cargo complex into the nucleus via the nuclear pore. Typically, KPNA proteins recognize and bind a classical nuclear localization signal (cNLS) primarily composed of one or two short regions of highly basic residues on import cargoes (Figure 1A) (Cook et al., 2007). STAT1, however, lacks such a cNLS and instead relies on a
noclassical NLS (ncNLS) for import by only a subset of KPNAs, known as NPI-1 karyopherins (KPNA1, KPNA5, and KPNA6 or importin-a 5, 6 and 7). The STAT1 ncNLS is exclusively presented in the STAT1 dimer, allowing discrimination between the preactivation STAT1 monomer and the postactivation STAT1 dimer. Interestingly, cNLS cargos and the STAT1 ncNLS bind physically distinct armadillo (ARM) repeat regions of KPNAs such that binding of STAT1 does not compete with cNLS binding, suggesting that both cargos can be carried at the same time by the same KPNA (Reich and Liu, 2006) (Figure 1B). Though this specialized access to the nucleus by STAT1 has been known for years, the details of what comprises an ncNLS and how NPI-1 KPNAs recognize such an import signal have been elusive.
Resolution of this conundrum has now emerged from an unexpected source: Ebola virus. Ebola virus and the related Marburg virus are members of the filoviridae, enveloped negative-sense RNA viruses that can cause viral hemorrhagic fevers that can reach fatality rates near 90%. One characteristic of Ebola virus infections is the lack of IFN response in infected cells, essentially nullifying the innate immune response to viral infection. Such dysregulation of the innate immune system is likely responsible for the unchecked replication of the virus and rapid dissemination throughout the body (Basler and Amarasinghe, 2009). Previous work by the Basler group and colleagues showed that the VP24 protein from Ebola, a minor matrix protein present in incoming viral particles, was singlehandedly able to disrupt signaling by
Cell Host & Microbe 16, August 13, 2014 ª2014 Elsevier Inc. 151
Cell Host & Microbe
Previews IFN-a/IFN-b and IFN-g by preventing phosphorylated STAT1 from entering the nucleus (Reid et al., 2006). Additional biochemical mapping using truncations and individual mutations revealed that VP24 required the same regions on KPNA1, KPNA5, and KPNA6 to bind to these karyopherins as phosphorylated STAT1 required for binding, suggesting that VP24, like STAT1, contains a ncNLS (Reid et al., 2007). In the current study, Xu et al. (Xu et al., 2014) reveal mechanistic details of Ebola virus inhibition of STAT signaling while simultaneously providing molecular insight into STAT1 nuclear import. By solving the crystal structure of VP24 bound to KPNA5, the authors showed that VP24 tightly binds KPNA5 using an extended surface to interact with the three C-terminal ARMs of KPNA5. Using these structural data, the authors showed that the same residues required for KPNA50 s interaction with VP24 were also required for interaction with STAT1, thus providing the first structural and biochemical glimpse of how KPNA can recognize an ncNLS. These data also elegantly explain the specificity of VP24 and STAT1 binding by only a subset of KPNAs, as many of the critical binding residues in KPNA5 are well conserved among NPI-1 KPNAs, but not well conserved among non-STAT1-binding KPNAs. Expanding from KPNA5 binding of an ncNLS to the in vivo function of VP24, the authors showed that residues in VP24 that make direct contacts to KPNA5 are required to block nuclear import of STAT1 and its transcriptional functions. Thus, consistent with a previously proposed model (Reid et al., 2007), VP24 binding to KPNA5 serves as a competitive inhibitor of phosphorylatedSTAT1 binding. Importantly, although VP24 directly interfered with nuclear import of STAT1 via its ncNLS, it did not block binding of a cNLS-containing cargo protein by KPNA5. Thus, Ebola virus does not inhibit all nuclear import via NPI-1 KPNA proteins; this would presumably be highly deleterious for the virus. Instead, Ebola VP24 protein specif-
ically blocks STAT1’s ‘‘emergency access lane’’ to the nucleus with surgical precision (Figure 1C). Why would Ebola virus choose this strategy to antagonize the host immune system? The finding that Ebola virus disrupts STAT1 function is not unexpected, given that IFN signaling is such an important axis of the innate immune system. Indeed, numerous viruses antagonize STAT signaling by mechanisms such as STAT1 and STAT2 protein degradation or dephosphorylation of activated STAT1 (Najjar and Fagard, 2010), and these would appear to be at least as effective a strategy. However, encoding a competitive inhibitor of STAT1-KPNA interaction might present some evolutionary advantages to Ebola. This would especially be true if the constraint to maintain STAT1KPNA interactions imposes greater difficulty for hosts to evolve away from antagonism via ‘‘mimicry’’ by Ebola virus VP24 (Elde and Malik, 2009). In contrast, changes in STAT proteins at viral protein binding interfaces might be sufficient to escape antagonism by other mechanisms, and therefore easier to evolve in principle (Daugherty and Malik, 2012). Interestingly, Ebola virus may not be unique in employing this strategy; measles virus and rotavirus may also block STAT1 nuclear import, although it remains unclear whether they also employ a competitive inhibitor (Yarbrough et al., 2014). By establishing a clear mechanism for Ebola VP24 blockage of IFN signaling in infected cells and structurally defining the interaction between KPNAs and ncNLSs, this work continues the long tradition of using viruses to molecularly dissect cellular pathways that are often highly pleiotropic, especially in the field of nucleocytoplasmic transport (Yarbrough et al., 2014). For instance, one of the important implications of this work is a deeper understanding of the evolution and functions of the ‘‘emergency access lane’’ for nuclear import. Although we do not know whether an ncNLS-mediated STAT1 import is a superior strategy compared to import using a cNLS, such
152 Cell Host & Microbe 16, August 13, 2014 ª2014 Elsevier Inc.
a specialized system may have evolved so that STAT1 does not disrupt, or is not disrupted by, the import of normal cellular cargos upon a cellular emergency such as IFN induction (Reich and Liu, 2006). By using Ebola virus VP24 as a tool, and specific mutations of the newly defined ncNLS-binding residues on KPNAs, future work will be able to dissect which additional cargoes use this specialized system for nuclear import and with what consequences. One thing is for certain: with the ‘‘emergency access lane’’ in full service, it allows pathogens like Ebola virus to specifically block it without paying any of the pleiotropic costs of blocking nuclear import in general. Regular traffic can flow unimpeded! ACKNOWLEDGMENTS We are supported by an Irvington Institute Fellowship of the Cancer Research Institute (M.D.D.) and the Howard Hughes Medical Institute (M.D.D., H.S.M.).
REFERENCES Basler, C.F., and Amarasinghe, G.K. (2009). J. Interferon Cytokine Res. 29, 511–520. Cook, A., Bono, F., Jinek, M., and Conti, E. (2007). Annu. Rev. Biochem. 76, 647–671. Daugherty, M.D., and Malik, H.S. (2012). Annu. Rev. Genet. 46, 677–700. Elde, N.C., and Malik, H.S. (2009). Nat. Rev. Microbiol. 7, 787–797. Najjar, I., and Fagard, R. (2010). Biochimie 92, 425–444. Reich, N.C., and Liu, L. (2006). Nat. Rev. Immunol. 6, 602–612. Reid, S.P., Leung, L.W., Hartman, A.L., Martinez, O., Shaw, M.L., Carbonnelle, C., Volchkov, V.E., Nichol, S.T., and Basler, C.F. (2006). J. Virol. 80, 5156–5167. Reid, S.P., Valmas, C., Martinez, O., Sanchez, F.M., and Basler, C.F. (2007). J. Virol. 81, 13469– 13477. Xu, W., Edwards, M.R., Borek, D.M., Feagins, A.R., Mittal, A., Alinger, J.B., Berry, K.N., Yen, B., Hamilton, J., Brett, T.J., et al. (2014). Cell Host Microbe in press. Yarbrough, M.L., Mata, M.A., Sakthivel, R., and Fontoura, B.M. (2014). Traffic 15, 127–140.