- Email: [email protected]
Transportin-SR2 and Integrase
Current Biology Vol 18 No 16 R710
synchrony, which has become more popular in recent years [13,14], while not being mutually exclusive with the idea of binding by synchrony. When two neuronal populations are synchronized and provide convergent input to downstream neurons, this results in a larger depolarisation of those postsynaptic neurons, and hence enhanced activation of later processing stages [15]. In a similar vein, when two areas of the brain are synchronized, it is ensured that both neuronal membranes are at a mutually optimal excitabilty state to receive input or send output — for a detailed illustration of these ideas see [14], and see [16] for a related effect — and this should lead to more efficient transmission of neural activity. In that respect, the reported enhanced synchrony of high-frequency oscillations between auditory and visual cortex (to audiovisual looming stimuli) may explain the behavioural benefits of multimodally presented audio-visual looming stimuli against incongruent or purely visual or auditory looming signals. The question remains, how is this synchrony established? It is a well known property of oscillators that these easily adjust their phase even in the presence of only relatively weak coupling between them [17] and the STS is connected with the auditory belt region. So, is it all about facilitation of information transmission, or might the role of synchrony between the two processing streams be more generic — for integrating their sensory representations into a common percept? At this point, this remains pure speculation. Not much is currently known about how multisensory representations are formed from unimodal inputs. One computational model [18] assumes convergent projections of unisensory areas onto a multimodal map which will then combine its inputs by recursive activations between the multisensory area and the unisensory areas and can thereby reproduce important findings from psychophysical research [19]. This particular model explicitly does not make any assumptions about the relative timing of the respective inputs and outputs. Irrespective of the details of the model, however, the existence of such recursive modes of processing between multisensory and unisensory areas is quite likely given the findings in the literature [1] and it would be of interest to investigate whether
selective temporal coordination of the inputs — as observed by Maier et al. [9] — is correlated with the efficiency of how inputs are combined in the working brain and may therefore provide a solution to the ‘‘correspondence problem’’ [19]. Future experiments should investigate whether synchrony between two sensory processing streams covaries with behavioural measures of fusion between the sensory representations and, for example, whether this can be flexibly established depending on the task requirements or by using bistable stimuli that sometimes fuse and sometimes do not — under identical physical (stimulus) conditions. The remarkable finding from Maier et al. [9] is that they establish the existence of stimulus specific synchronization between auditory and visual brain areas and that synchrony seems to correspond with a behavioural effect of audio-visual integration [10]. References 1. Driver, J., and Noesselt, T. (2008). Multisensory interplay reveals crossmodal influences on ‘sensory-specific’ brain regions, neural responses, and judgments. Neuron 57, 11–23. 2. McGurk, H., and MacDonald, J. (1976). Hearing lips and seeing voices. Nature 264, 746–748. 3. Wallace, M.T., and Stein, B.E. (1996). Sensory organization of the superior colliculus in cat and monkey. Prog. Brain Res. 112, 301–311. 4. Calvert, G.A., Campbell, R., and Brammer, M.J. (2000). Evidence from functional magnetic resonance imaging of crossmodal binding in the human heteromodal cortex. Curr. Biol. 10, 649–657. 5. Macaluso, E., Frith, C.D., and Driver, J. (2002). Crossmodal spatial influences of touch on extrastriate visual areas take current gaze direction into account. Neuron 34, 647–658. 6. Driver, J., and Spence, C. (2000). Multisensory perception: Beyond modularity and convergence. Curr. Biol. 10, R731–R735.
7. Singer, W. (1999). Neuronal synchrony: a versatile code for the definition of relations? Neuron 24, 49–65, 111–125. 8. Riesenhuber, M., and Poggio, T. (1999). Hierarchical models of object recognition in cortex. Nat. Neurosci. 2, 1019–1025. 9. Maier, J.X., Chandrasekaran, C., and Ghazanfar, A. (2008). Integration of bimodal looming signals through neuronal coherence in the temporal lobe. Curr. Biol. 18, 963–968. 10. Maier, J.X., Neuhoff, J.G., Logothetis, N.K., and Ghazanfar, A.A. (2004). Multisensory integration of looming signals by rhesus monkeys. Neuron 43, 177–181. 11. Bruce, C., Desimone, R., and Gross, C.G. (1979). Visual properties of neurons in a polysensory area in superior temporal sulcus of the macaque. J. Neurophysiol. 46, 369–384. 12. Engel, A.K., and Singer, W. (2001). Temporal binding and the neural correlates of sensory awareness. Trends Cogn. Sci. 5, 16–25. 13. Salinas, E., and Sejnowski, T.J. (2001). Correlated neuronal activity and the flow of neural information. Nat. Rev. Neurosci. 2, 539–550. 14. Fries, P. (2005). A mechanism for cognitive dynamics: neuronal communication through neuronal coherence. Trends. Cogn. Sci. 9, 474–480. 15. Womelsdorf, T., Schoffelen, J.M., Oostenveld, R., Singer, W., Desimone, R., Engel, A.K., and Fries, P. (2007). Modulation of neuronal interactions through neuronal synchronization. Science 316, 1609–1612. 16. Lakatos, P., Chen, C.M., O’Connell, M.N., Mills, A., and Schroeder, C.E. (2007). Neuronal oscillations and multisensory interaction in primary auditory cortex. Neuron 53, 279–292. 17. Pikovsky, A., Rosenblum, M., and Kurths, J. (2001). Synchronization - A Universal Concept in Nonlinear Sciences (Cambridge: Cambridge University Press). 18. Deneve, S., and Pouget, A. (2004). Bayesian multisensory integration and cross-modal spatial links. J. Physiol. 98, 249–258. 19. Ernst, M.O., and Bu¨lthoff, H.H. (2004). Merging the senses into a robust percept. Trends Cogn. Sci. 8, 162–169.
UCL Institute of Cognitive Neuroscience, UCL, 17 Queen Square, London WC1N 3AR, UK, and Wellcome Trust Centre for Neuroimaging, Institute of Neurology, UCL, 12 Queen Square, London WC1N 3BG, UK. E-mail: [email protected] DOI: 10.1016/j.cub.2008.06.051
HIV-1 Infection: Going Nuclear with TNPO3/Transportin-SR2 and Integrase Factors necessary for HIV-1 nuclear import have been sought for many years. Recent reports suggest that TNPO3/Transportin-SR2 binds to HIV-1 integrase and is required for HIV-1 infection of interphase cells. Jeremy Luban Everyone agrees that HIV-1 infects non-dividing cells [1], yet viral factors and host factors that promote HIV-1 nuclear import have been very difficult
to pin down [2]. Recent studies now show that TNPO3/Transportin-SR2 plays a role in HIV-1 replication [3,4] and, via an interaction with HIV-1 integrase, promotes the nuclear import of HIV-1 [4]. The discovery of a host
Dispatch R711
factor that regulates HIV-1 replication always stimulates a lot of activity in the research community. Predictably, behind the scenes at the most recent Cold Spring Harbor Retroviruses meeting, researchers were buzzing about TNPO3/Transportin-SR2. TNPO3/Transportin-SR2 was one of about 250 genes pulled out in an RNA interference (RNAi) screen of 21,121 genes for effects on HIV-1 replication [3]. The enormity of this screen, conducted by Elledge and colleagues, attracted great attention. For molecular biologists wishing to focus on individual host proteins that regulate HIV-1 replication, many attractive factors were revealed, amongst which TNPO3/Transportin-SR2 was one of the few that received the authors’ immediate attention [3]. HIV-1 infectivity was decreased in response to transfection with all eight small interfering RNAs (siRNAs) that targeted different regions of the TNPO3/Transportin-SR2 mRNA. Although this work showed that TNPO3/Transportin-SR2 promotes HIV-1 infection, it was not clear exactly what this host factor did that was so important for the virus. TNPO3/ Transportin-SR2 is an importin-b family member that recognizes serine–arginine-rich repeats within precursor-mRNA splicing factors and transports these factors into the nucleus [5,6], suggesting that the likely function of TNPO3/Transportin-SR2 is to transport the HIV-1 preintegration complex through the nuclear pore into the nucleus. The observation that TNPO3/Transportin-SR2 knockdown had no effect on murine leukemia virus (MLV) [3] is consistent with this possibility, given that, unlike HIV-1, MLV cannot traverse nuclear pore complexes and thus cannot infect non-dividing cells [7]. Furthermore, the inhibition of HIV-1 replication due to RNAi-mediated knockdown of TNPO3/ Transportin-SR2 occurred after reverse transcription, but before viral cDNA was covalently attached to host chromosomal DNA [3]. This finding suggests that TNPO3/Transportin-SR2 is either required for transport of viral cDNA into the nucleus or for optimal integration activity, although the authors could not distinguish between these two possibilities in the manuscript [3]. The authors of the RNAi screen gave no indication which viral component might be interacting with TNPO3/
Transportin-SR2 [3], and, among HIV-1 proteins, no target could be proposed on the basis of the presence of incriminating serine–arginine repeats. One might hunt for the viral target within the literature concerning HIV-1 nuclear import, but the search for HIV-1 components that promote nuclear import has a long and confusing history. HIV-1 matrix, integrase, Vpr, and cis-acting polypurine sequences have all been reported to contribute to HIV-1 nuclear import [2]. Yet, the Emerman lab disrupted all of these elements in HIV-1, or replaced them with MLV homologues, and found that such viruses infected non-dividing cells with almost the same relative efficiency as dividing cells [8]. Interpretation of these experiments was complicated by the fact that the absolute infectivity of the chimeric viruses was severely compromised. More recently, experiments from the Emerman lab suggest that the capsid protein is the critical determinant that distinguishes HIV-1 from MLV [9]. How the capsid protein might regulate HIV-1 nuclear import, and whether the capsidbinding protein cyclophilin A is relevant to this function [10], is currently the subject of investigation in several labs. In the absence of an experiment that generates a clear consensus otherwise, HIV-1 integrase remains a reasonable candidate for a functionally relevant target of nuclear import factors. Integrase is the only trans-acting HIV-1 component known to play an essential role in the nucleus during the early steps of infection. In a study published in this issue of Current Biology, Debyser and colleagues [4] report the results of a screen for proteins that interact with HIV-1 integrase and identified TNPO3/ Transportin-SR2 as an integrasebinding protein. This group also found that knockdown of TNPO3/ Transportin-SR2 compromised HIV-1 infectivity with no effect on MLV. Consistent with integrase determining HIV-1 dependence on TNPO3/ Transportin-SR2, recombinant TNPO3/ Transportin-SR2 interacted with HIV-1 integrase, but not with MLV integrase [4]. Even more convincing evidence in support of the importance of this interaction could have been provided by showing that mutant versions of HIV-1 integrase that fail to bind to TNPO3/Transportin-SR2 recapitulate the phenotype observed
CCA-3′ tRNA
Nucleases, starvation
TNPO3/Transportin-SR2
Nuclear pore
tRNA degradation/repair Current Biology
Figure 1. Nuclear import of non-functional tRNAs by TNPO3/Transportin-SR2. In response to nuclease/starvation-mediated removal of the 3’ CCA acceptor arm that is required for esterification to amino acids, tRNAs are transported back into the nucleus by TNPO3/Transportin-SR2 for repair or degradation.
following knockdown of TNPO3/ Transportin-SR2. Two experiments were carried out by Debyser and colleagues [4] to determine whether TNPO3/ Transportin-SR2 promotes nuclear import of HIV-1. The first experiment revealed that TNPO3/Transportin-SR2 knockdown led to a reduction in the formation of the circular DNAs that result when viral cDNA encounters nuclear DNA-repair enzymes [4]. The second experiment exploited an assay that was recently developed by Cereseto and colleagues [11] to directly visualize HIV-1 preintegration complexes in cells. In this assay, integrase-defective virions are complemented during virion production with wild-type integrase that has been fused to green fluorescent protein (GFP) and to the HIV-1 virion-targeting protein Vpr.
Current Biology Vol 18 No 16 R712
Plasma membrane
Reverse transcription
TNPO3/Transportin-SR2
Nuclear pore Establishment of provirus Current Biology
Figure 2. Nuclear import of HIV-1 is dependent on TNPO3/Transportin-SR2 and nonfunctional tRNAs. Upon fusion with the plasma membrane, the HIV-1 RNA genome (two thin, vertical, black lines) is reverse transcribed to produce the viral cDNA (two thick, vertical, red lines). Recent publications indicate a role for TNPO3/ Transportin-SR2 in the nuclear import of the viral preintegration complex that may involve interactions among TNPO3/Transportin-SR2, tRNA, and HIV-1 integrase (green rectangle). Unknown characteristics of the HIV-1 capsid protein (blue lines shown encasing the viral nucleic acid) — perhaps its interaction with cyclophilin A — also appear to be important for nuclear import.
Visualization of HIV-1 virions with Vpr– GFP fusions has been exploited most extensively by Hope and colleagues [12]. By targeting cells with the labeled virions, visualizing the nuclear lamina, and deconvolving confocal microscopy images to reveal the position of individual subvirion particles relative to the nuclear envelope, it was shown that TNPO3/ Transportin-SR2 knockdown reduces the percentage of subvirion particles detected within the nuclear envelope following acute infection [4]. Although questions remain concerning the functionality of these fluorescent subvirion particles, development of this assay to visualize HIV-1 preintegration complexes in the nucleus is a significant advance that nicely complements other advanced microscopy techniques in the literature [13,14].
Oddly enough, the Saccharomyces cerevisiae orthologue of TNPO3/ Transportin-SR2, MTR10, made an appearance in the retroviral literature several years ago. Parent and colleagues [15] had demonstrated that the Rous sarcoma virus Gag polyprotein is imported into the nucleus during virion assembly and that one of the nuclear localization signals required MTR10 for nuclear import in yeast. Although there is a report suggesting that HIV-1 Gag may also enter the nucleus during the virion assembly process [16], Debyser and colleagues [4] were unable to detect any effect of TNPO3/Transportin-SR2 on HIV-1 virion assembly. Hopper and colleagues [17] made the surprising observation that MTR10 shuttles tRNAs from the cytoplasm back into the nucleus (Figure 1), especially under conditions of nutrient deprivation. These observations are especially intriguing when one considers that Fassati and colleagues [18] screened cytoplasmic fractions for the ability to promote nuclear import of purified HIV-1 preintegration complexes in permeabilized HeLa cells and found tRNA in the active fraction. tRNAs lacking the 3’ terminal CCA trinucleotide that constitutes the amino-acid acceptor site were associated with HIV-1 particles and not with an HIV-1 chimera in which gag sequences were replaced with those from MLV. Synthetic tRNAs with 3’ truncations functioned in the permeabilized cell transport assay to transport HIV-1 preintegration complexes into the nucleus. Interestingly, 3’ defective tRNAs were imported into the nucleus in the absence of HIV-1, indicating that this tRNA retrograde transport system probably functions in tRNA quality control as it does in S. cerevisiae. This last result was the first demonstration of retrograde tRNA transport in mammalian cells and demonstrates once again the enormous potential of HIV-1 studies to reveal previously unsuspected information about basic cellular function. Taken together, these results raise the very interesting possibility that, by associating with defective tRNAs, HIV-1 hijacks a TNPO3/Transportin-SR2-dependent pathway for nuclear import (Figure 2). How TNPO3/Transportin-SR2, integrase, and tRNA might be functionally connected is unclear. One possibility is that integrase and tRNA
each independently interact with TNPO3/Transportin-SR2 and via this transport factor promote HIV-1 nuclear import. Finally, the identification of a new HIV-1 host factor reveals hitherto unimagined dependencies of the virus, and thereby offers the promise of new therapeutic approaches that limit infection and associated pathology. In the past year, the first inhibitors of HIV-1 integrase were approved for clinical use but, as with any drug that targets an HIV-1 protein, mutations associated with drug resistance are a concern. Host factors essential for viral replication offer potential drug targets that preclude problems with drug resistance: HIV-1 clones bearing mutations that confer resistance to integrase inhibitors were inhibited as effectively as wild-type virus by knockdown of TNPO3/TransportinSR2 [4]. This experiment foretells the future development of HIV-1 inhibitors based on the disruption of the interaction between HIV-1 integrase and TNPO3/Transportin-SR2. References 1. Weinberg, J.B., Matthews, T.J., Cullen, B.R., and Malim, M.H. (1991). Productive human immunodeficiency virus type 1 (HIV-1) infection of nonproliferating human monocytes. J. Exp. Med. 174, 1477–1482. 2. Suzuki, Y., and Craigie, R. (2007). The road to chromatin - nuclear entry of retroviruses. Nat. Rev. Microbiol. 5, 187–196. 3. Brass, A.L., Dykxhoorn, D.M., Benita, Y., Yan, N., Engelman, A., Xavier, R.J., Lieberman, J., and Elledge, S.J. (2008). Identification of host proteins required for HIV infection through a functional genomic screen. Science 319, 921–926. 4. Christ, F., Thys, W., De Rijck, J., Gijsbers, R., Albanese, A., Arosio, D., Emiliani, S., Rain, J.-C., Benarous, R., Cereseto, A., et al. (2008). Transportin-SR2 imports HIV into the nucleus. Curr. Biol. 18, 1192–1202. 5. Kataoka, N., Bachorik, J.L., and Dreyfuss, G. (1999). Transportin-SR, a nuclear import receptor for SR proteins. J. Cell Biol. 145, 1145–1152. 6. Lai, M.C., Lin, R.I., Huang, S.Y., Tsai, C.W., and Tarn, W.Y. (2000). A human importin-beta family protein, transportin-SR2, interacts with the phosphorylated RS domain of SR proteins. J. Biol. Chem. 275, 7950–7957. 7. Roe, T., Reynolds, T.C., Yu, G., and Brown, P.O. (1993). Integration of murine leukemia virus DNA depends on mitosis. EMBO J. 12, 2099–2108. 8. Yamashita, M., and Emerman, M. (2005). The cell cycle independence of HIV infections is not determined by known karyophilic viral elements. PLoS Pathog. 1, e18. 9. Yamashita, M., Perez, O., Hope, T.J., and Emerman, M. (2007). Evidence for direct involvement of the capsid protein in HIV infection of nondividing cells. PLoS Pathog. 3, 1502–1510. 10. Luban, J. (2007). Cyclophilin A, TRIM5, and resistance to human immunodeficiency virus type 1 infection. J. Virol. 81, 1054– 1061. 11. Albanese, A., Arosio, D., Terreni, M., and Cereseto, A. (2008). HIV-1 pre-integration
Dispatch R713
complexes selectively target decondensed chromatin in the nuclear periphery. PLoS ONE 3, e2413. 12. McDonald, D., Vodicka, M.A., Lucero, G., Svitkina, T.M., Borisy, G.G., Emerman, M., and Hope, T.J. (2002). Visualization of the intracellular behavior of HIV in living cells. J. Cell Biol. 159, 441–452. 13. Arhel, N., Genovesio, A., Kim, K.A., Miko, S., Perret, E., Olivo-Marin, J.C., Shorte, S., and Charneau, P. (2006). Quantitative fourdimensional tracking of cytoplasmic and nuclear HIV-1 complexes. Nat. Methods 3, 817–824. 14. Bell, P., Montaner, L.J., and Maul, G.G. (2001). Accumulation and intranuclear distribution of unintegrated human
immunodeficiency virus type 1 DNA. J. Virol. 75, 7683–7691. 15. Butterfield-Gerson, K.L., Scheifele, L.Z., Ryan, E.P., Hopper, A.K., and Parent, L.J. (2006). Importin-beta family members mediate alpharetrovirus gag nuclear entry via interactions with matrix and nucleocapsid. J. Virol. 80, 1798–1806. 16. Dupont, S., Sharova, N., DeHoratius, C., Virbasius, C.M., Zhu, X., Bukrinskaya, A.G., Stevenson, M., and Green, M.R. (1999). A novel nuclear export activity in HIV-1 matrix protein required for viral replication. Nature 402, 681–685. 17. Shaheen, H.H., and Hopper, A.K. (2005). Retrograde movement of tRNAs from the cytoplasm to the nucleus in Saccharomyces
Vision: Attention Makes the Cup Flow Over Scalp potentials are surprisingly informative about visual attention: a recent study that used them to record neural responses to up to four superimposed visual patterns simultaneously has now revealed the flow of attentional signals back to visual cortex. Jochen Braun and Mircea Ariel Schoenfeld When we fix our gaze on a complex visual scene, we can alter our phenomenal experience by focussing mentally on different parts or aspects of the scene. Neural correlates of this ‘selective visual attention’ have been observed in an anatomically distributed, but functionally integrated, network of brain sites, including the lateral geniculate and pulvinar nuclei of the thalamus, visual areas in occipital and temporal cortex, and higher order areas in frontal and parietal cortex [1,2]. It is thought that attention signals originate in frontal and parietal cortex and are then transmitted by feedback and recurrent projections backwards to earlier stages of the visual pathways. These efferent signals seem to selectively enhance the amplitude, and perhaps also the temporal synchronicity, of neural responses to the ‘attended’ parts or aspects of a visual scene, at the expense of the neural responses to all other parts or aspects of the scene. Except in the most simplistic displays, however, the attentional enhancement of neural responses is not limited to the desired information, but extends also to some other stimulus features that may be present in the display but that are irrelevant to the task at hand. This ‘spill-over’ to
some irrelevant features (but not to others) is of considerable interest, as it presumably reflects the organization of the projection patterns that communicate attentional signals back to visually responsive neurons. One pattern of spill-over goes by the name of ‘object attention’. Typically, object attention is encountered when two visual patterns are superimposed transparently, that is, such that each pattern remains recognizable individually. To take an idealized example, an array of red items moving coherently in one direction might be superimposed over an array of blue items moving coherently in another direction (Figure 1A). Because of the shared colour and motion, each array is phenomenally experienced as a distinct visual object. In viewing such a display, observers can choose which array they attend and, thus, which array they experience more fully. Attentional spill-over becomes apparent when observers are asked about one particular attribute of one array, the shape of the red array items. In this case, the attentional enhancement — as measured either behaviourally or neurophysiologically — applies not only to the relevant attribute (shape) but also to the irrelevant attributes (motion, colour) of the target array. All attributes of the other array are suppressed, however. Thus, in this
cerevisiae. Proc. Natl. Acad. Sci. USA 102, 11290–11295. 18. Zaitseva, L., Myers, R., and Fassati, A. (2006). tRNAs promote nuclear import of HIV-1 intracellular reverse transcription complexes. PLoS Biol. 4, e332.
Department of Microbiology and Molecular Medicine, University of Geneva, 1 Rue Michel Servet, CH-1211 Geneva 4, Switzerland. E-mail: [email protected]
DOI: 10.1016/j.cub.2008.07.037
simplified example, object attention enhances all responses to the attended array and suppresses all responses to the unattended array. Object attention has been documented most extensively with purely behavioural measures [3–6], although a few studies have encountered its characteristic pattern of spill-over enhancement also in single-unit activity of visual cortex [7] and in visual evoked potentials [8,9]. Note that electrophysiological studies of attentional spill-over face an enormous hurdle: they must distinguish the neural responses not just to two superimposed patterns, but to relevant and irrelevant attributes of these patterns. Over the last decade, the measurement of visual evoked potentials on the scalp has been refined to the point that it can now overcome this hurdle. A key to the singular informativeness of this method is the oscillatory response evoked by a flickering pattern that is known as a ‘steady-state visual evoked potential’ or SSVEP [10]. As the frequency of the oscillatory response matches that of the driving flickering pattern, two patterns flickering at different frequencies elicit distinguishable oscillatory contributions to the visual evoked potential. When observers are required to discriminate one pattern, the neural response to the attended pattern (as measured by the SSVEP) increases relative to the response to the unattended pattern [8,9,11]. As they reported recently in Current Biology, Andersen et al. [12] have been able to distinguish neural responses to four superimposed arrays, setting a new standard for evoked potential methods and affording an even more penetrating insight into attentional
synchrony, which has become more popular in recent years [13,14], while not being mutually exclusive with the idea of binding by synchrony. When two neuronal populations are synchronized and provide convergent input to downstream neurons, this results in a larger depolarisation of those postsynaptic neurons, and hence enhanced activation of later processing stages [15]. In a similar vein, when two areas of the brain are synchronized, it is ensured that both neuronal membranes are at a mutually optimal excitabilty state to receive input or send output — for a detailed illustration of these ideas see [14], and see [16] for a related effect — and this should lead to more efficient transmission of neural activity. In that respect, the reported enhanced synchrony of high-frequency oscillations between auditory and visual cortex (to audiovisual looming stimuli) may explain the behavioural benefits of multimodally presented audio-visual looming stimuli against incongruent or purely visual or auditory looming signals. The question remains, how is this synchrony established? It is a well known property of oscillators that these easily adjust their phase even in the presence of only relatively weak coupling between them [17] and the STS is connected with the auditory belt region. So, is it all about facilitation of information transmission, or might the role of synchrony between the two processing streams be more generic — for integrating their sensory representations into a common percept? At this point, this remains pure speculation. Not much is currently known about how multisensory representations are formed from unimodal inputs. One computational model [18] assumes convergent projections of unisensory areas onto a multimodal map which will then combine its inputs by recursive activations between the multisensory area and the unisensory areas and can thereby reproduce important findings from psychophysical research [19]. This particular model explicitly does not make any assumptions about the relative timing of the respective inputs and outputs. Irrespective of the details of the model, however, the existence of such recursive modes of processing between multisensory and unisensory areas is quite likely given the findings in the literature [1] and it would be of interest to investigate whether
selective temporal coordination of the inputs — as observed by Maier et al. [9] — is correlated with the efficiency of how inputs are combined in the working brain and may therefore provide a solution to the ‘‘correspondence problem’’ [19]. Future experiments should investigate whether synchrony between two sensory processing streams covaries with behavioural measures of fusion between the sensory representations and, for example, whether this can be flexibly established depending on the task requirements or by using bistable stimuli that sometimes fuse and sometimes do not — under identical physical (stimulus) conditions. The remarkable finding from Maier et al. [9] is that they establish the existence of stimulus specific synchronization between auditory and visual brain areas and that synchrony seems to correspond with a behavioural effect of audio-visual integration [10]. References 1. Driver, J., and Noesselt, T. (2008). Multisensory interplay reveals crossmodal influences on ‘sensory-specific’ brain regions, neural responses, and judgments. Neuron 57, 11–23. 2. McGurk, H., and MacDonald, J. (1976). Hearing lips and seeing voices. Nature 264, 746–748. 3. Wallace, M.T., and Stein, B.E. (1996). Sensory organization of the superior colliculus in cat and monkey. Prog. Brain Res. 112, 301–311. 4. Calvert, G.A., Campbell, R., and Brammer, M.J. (2000). Evidence from functional magnetic resonance imaging of crossmodal binding in the human heteromodal cortex. Curr. Biol. 10, 649–657. 5. Macaluso, E., Frith, C.D., and Driver, J. (2002). Crossmodal spatial influences of touch on extrastriate visual areas take current gaze direction into account. Neuron 34, 647–658. 6. Driver, J., and Spence, C. (2000). Multisensory perception: Beyond modularity and convergence. Curr. Biol. 10, R731–R735.
7. Singer, W. (1999). Neuronal synchrony: a versatile code for the definition of relations? Neuron 24, 49–65, 111–125. 8. Riesenhuber, M., and Poggio, T. (1999). Hierarchical models of object recognition in cortex. Nat. Neurosci. 2, 1019–1025. 9. Maier, J.X., Chandrasekaran, C., and Ghazanfar, A. (2008). Integration of bimodal looming signals through neuronal coherence in the temporal lobe. Curr. Biol. 18, 963–968. 10. Maier, J.X., Neuhoff, J.G., Logothetis, N.K., and Ghazanfar, A.A. (2004). Multisensory integration of looming signals by rhesus monkeys. Neuron 43, 177–181. 11. Bruce, C., Desimone, R., and Gross, C.G. (1979). Visual properties of neurons in a polysensory area in superior temporal sulcus of the macaque. J. Neurophysiol. 46, 369–384. 12. Engel, A.K., and Singer, W. (2001). Temporal binding and the neural correlates of sensory awareness. Trends Cogn. Sci. 5, 16–25. 13. Salinas, E., and Sejnowski, T.J. (2001). Correlated neuronal activity and the flow of neural information. Nat. Rev. Neurosci. 2, 539–550. 14. Fries, P. (2005). A mechanism for cognitive dynamics: neuronal communication through neuronal coherence. Trends. Cogn. Sci. 9, 474–480. 15. Womelsdorf, T., Schoffelen, J.M., Oostenveld, R., Singer, W., Desimone, R., Engel, A.K., and Fries, P. (2007). Modulation of neuronal interactions through neuronal synchronization. Science 316, 1609–1612. 16. Lakatos, P., Chen, C.M., O’Connell, M.N., Mills, A., and Schroeder, C.E. (2007). Neuronal oscillations and multisensory interaction in primary auditory cortex. Neuron 53, 279–292. 17. Pikovsky, A., Rosenblum, M., and Kurths, J. (2001). Synchronization - A Universal Concept in Nonlinear Sciences (Cambridge: Cambridge University Press). 18. Deneve, S., and Pouget, A. (2004). Bayesian multisensory integration and cross-modal spatial links. J. Physiol. 98, 249–258. 19. Ernst, M.O., and Bu¨lthoff, H.H. (2004). Merging the senses into a robust percept. Trends Cogn. Sci. 8, 162–169.
UCL Institute of Cognitive Neuroscience, UCL, 17 Queen Square, London WC1N 3AR, UK, and Wellcome Trust Centre for Neuroimaging, Institute of Neurology, UCL, 12 Queen Square, London WC1N 3BG, UK. E-mail: [email protected] DOI: 10.1016/j.cub.2008.06.051
HIV-1 Infection: Going Nuclear with TNPO3/Transportin-SR2 and Integrase Factors necessary for HIV-1 nuclear import have been sought for many years. Recent reports suggest that TNPO3/Transportin-SR2 binds to HIV-1 integrase and is required for HIV-1 infection of interphase cells. Jeremy Luban Everyone agrees that HIV-1 infects non-dividing cells [1], yet viral factors and host factors that promote HIV-1 nuclear import have been very difficult
to pin down [2]. Recent studies now show that TNPO3/Transportin-SR2 plays a role in HIV-1 replication [3,4] and, via an interaction with HIV-1 integrase, promotes the nuclear import of HIV-1 [4]. The discovery of a host
Dispatch R711
factor that regulates HIV-1 replication always stimulates a lot of activity in the research community. Predictably, behind the scenes at the most recent Cold Spring Harbor Retroviruses meeting, researchers were buzzing about TNPO3/Transportin-SR2. TNPO3/Transportin-SR2 was one of about 250 genes pulled out in an RNA interference (RNAi) screen of 21,121 genes for effects on HIV-1 replication [3]. The enormity of this screen, conducted by Elledge and colleagues, attracted great attention. For molecular biologists wishing to focus on individual host proteins that regulate HIV-1 replication, many attractive factors were revealed, amongst which TNPO3/Transportin-SR2 was one of the few that received the authors’ immediate attention [3]. HIV-1 infectivity was decreased in response to transfection with all eight small interfering RNAs (siRNAs) that targeted different regions of the TNPO3/Transportin-SR2 mRNA. Although this work showed that TNPO3/Transportin-SR2 promotes HIV-1 infection, it was not clear exactly what this host factor did that was so important for the virus. TNPO3/ Transportin-SR2 is an importin-b family member that recognizes serine–arginine-rich repeats within precursor-mRNA splicing factors and transports these factors into the nucleus [5,6], suggesting that the likely function of TNPO3/Transportin-SR2 is to transport the HIV-1 preintegration complex through the nuclear pore into the nucleus. The observation that TNPO3/Transportin-SR2 knockdown had no effect on murine leukemia virus (MLV) [3] is consistent with this possibility, given that, unlike HIV-1, MLV cannot traverse nuclear pore complexes and thus cannot infect non-dividing cells [7]. Furthermore, the inhibition of HIV-1 replication due to RNAi-mediated knockdown of TNPO3/ Transportin-SR2 occurred after reverse transcription, but before viral cDNA was covalently attached to host chromosomal DNA [3]. This finding suggests that TNPO3/Transportin-SR2 is either required for transport of viral cDNA into the nucleus or for optimal integration activity, although the authors could not distinguish between these two possibilities in the manuscript [3]. The authors of the RNAi screen gave no indication which viral component might be interacting with TNPO3/
Transportin-SR2 [3], and, among HIV-1 proteins, no target could be proposed on the basis of the presence of incriminating serine–arginine repeats. One might hunt for the viral target within the literature concerning HIV-1 nuclear import, but the search for HIV-1 components that promote nuclear import has a long and confusing history. HIV-1 matrix, integrase, Vpr, and cis-acting polypurine sequences have all been reported to contribute to HIV-1 nuclear import [2]. Yet, the Emerman lab disrupted all of these elements in HIV-1, or replaced them with MLV homologues, and found that such viruses infected non-dividing cells with almost the same relative efficiency as dividing cells [8]. Interpretation of these experiments was complicated by the fact that the absolute infectivity of the chimeric viruses was severely compromised. More recently, experiments from the Emerman lab suggest that the capsid protein is the critical determinant that distinguishes HIV-1 from MLV [9]. How the capsid protein might regulate HIV-1 nuclear import, and whether the capsidbinding protein cyclophilin A is relevant to this function [10], is currently the subject of investigation in several labs. In the absence of an experiment that generates a clear consensus otherwise, HIV-1 integrase remains a reasonable candidate for a functionally relevant target of nuclear import factors. Integrase is the only trans-acting HIV-1 component known to play an essential role in the nucleus during the early steps of infection. In a study published in this issue of Current Biology, Debyser and colleagues [4] report the results of a screen for proteins that interact with HIV-1 integrase and identified TNPO3/ Transportin-SR2 as an integrasebinding protein. This group also found that knockdown of TNPO3/ Transportin-SR2 compromised HIV-1 infectivity with no effect on MLV. Consistent with integrase determining HIV-1 dependence on TNPO3/ Transportin-SR2, recombinant TNPO3/ Transportin-SR2 interacted with HIV-1 integrase, but not with MLV integrase [4]. Even more convincing evidence in support of the importance of this interaction could have been provided by showing that mutant versions of HIV-1 integrase that fail to bind to TNPO3/Transportin-SR2 recapitulate the phenotype observed
CCA-3′ tRNA
Nucleases, starvation
TNPO3/Transportin-SR2
Nuclear pore
tRNA degradation/repair Current Biology
Figure 1. Nuclear import of non-functional tRNAs by TNPO3/Transportin-SR2. In response to nuclease/starvation-mediated removal of the 3’ CCA acceptor arm that is required for esterification to amino acids, tRNAs are transported back into the nucleus by TNPO3/Transportin-SR2 for repair or degradation.
following knockdown of TNPO3/ Transportin-SR2. Two experiments were carried out by Debyser and colleagues [4] to determine whether TNPO3/ Transportin-SR2 promotes nuclear import of HIV-1. The first experiment revealed that TNPO3/Transportin-SR2 knockdown led to a reduction in the formation of the circular DNAs that result when viral cDNA encounters nuclear DNA-repair enzymes [4]. The second experiment exploited an assay that was recently developed by Cereseto and colleagues [11] to directly visualize HIV-1 preintegration complexes in cells. In this assay, integrase-defective virions are complemented during virion production with wild-type integrase that has been fused to green fluorescent protein (GFP) and to the HIV-1 virion-targeting protein Vpr.
Current Biology Vol 18 No 16 R712
Plasma membrane
Reverse transcription
TNPO3/Transportin-SR2
Nuclear pore Establishment of provirus Current Biology
Figure 2. Nuclear import of HIV-1 is dependent on TNPO3/Transportin-SR2 and nonfunctional tRNAs. Upon fusion with the plasma membrane, the HIV-1 RNA genome (two thin, vertical, black lines) is reverse transcribed to produce the viral cDNA (two thick, vertical, red lines). Recent publications indicate a role for TNPO3/ Transportin-SR2 in the nuclear import of the viral preintegration complex that may involve interactions among TNPO3/Transportin-SR2, tRNA, and HIV-1 integrase (green rectangle). Unknown characteristics of the HIV-1 capsid protein (blue lines shown encasing the viral nucleic acid) — perhaps its interaction with cyclophilin A — also appear to be important for nuclear import.
Visualization of HIV-1 virions with Vpr– GFP fusions has been exploited most extensively by Hope and colleagues [12]. By targeting cells with the labeled virions, visualizing the nuclear lamina, and deconvolving confocal microscopy images to reveal the position of individual subvirion particles relative to the nuclear envelope, it was shown that TNPO3/ Transportin-SR2 knockdown reduces the percentage of subvirion particles detected within the nuclear envelope following acute infection [4]. Although questions remain concerning the functionality of these fluorescent subvirion particles, development of this assay to visualize HIV-1 preintegration complexes in the nucleus is a significant advance that nicely complements other advanced microscopy techniques in the literature [13,14].
Oddly enough, the Saccharomyces cerevisiae orthologue of TNPO3/ Transportin-SR2, MTR10, made an appearance in the retroviral literature several years ago. Parent and colleagues [15] had demonstrated that the Rous sarcoma virus Gag polyprotein is imported into the nucleus during virion assembly and that one of the nuclear localization signals required MTR10 for nuclear import in yeast. Although there is a report suggesting that HIV-1 Gag may also enter the nucleus during the virion assembly process [16], Debyser and colleagues [4] were unable to detect any effect of TNPO3/Transportin-SR2 on HIV-1 virion assembly. Hopper and colleagues [17] made the surprising observation that MTR10 shuttles tRNAs from the cytoplasm back into the nucleus (Figure 1), especially under conditions of nutrient deprivation. These observations are especially intriguing when one considers that Fassati and colleagues [18] screened cytoplasmic fractions for the ability to promote nuclear import of purified HIV-1 preintegration complexes in permeabilized HeLa cells and found tRNA in the active fraction. tRNAs lacking the 3’ terminal CCA trinucleotide that constitutes the amino-acid acceptor site were associated with HIV-1 particles and not with an HIV-1 chimera in which gag sequences were replaced with those from MLV. Synthetic tRNAs with 3’ truncations functioned in the permeabilized cell transport assay to transport HIV-1 preintegration complexes into the nucleus. Interestingly, 3’ defective tRNAs were imported into the nucleus in the absence of HIV-1, indicating that this tRNA retrograde transport system probably functions in tRNA quality control as it does in S. cerevisiae. This last result was the first demonstration of retrograde tRNA transport in mammalian cells and demonstrates once again the enormous potential of HIV-1 studies to reveal previously unsuspected information about basic cellular function. Taken together, these results raise the very interesting possibility that, by associating with defective tRNAs, HIV-1 hijacks a TNPO3/Transportin-SR2-dependent pathway for nuclear import (Figure 2). How TNPO3/Transportin-SR2, integrase, and tRNA might be functionally connected is unclear. One possibility is that integrase and tRNA
each independently interact with TNPO3/Transportin-SR2 and via this transport factor promote HIV-1 nuclear import. Finally, the identification of a new HIV-1 host factor reveals hitherto unimagined dependencies of the virus, and thereby offers the promise of new therapeutic approaches that limit infection and associated pathology. In the past year, the first inhibitors of HIV-1 integrase were approved for clinical use but, as with any drug that targets an HIV-1 protein, mutations associated with drug resistance are a concern. Host factors essential for viral replication offer potential drug targets that preclude problems with drug resistance: HIV-1 clones bearing mutations that confer resistance to integrase inhibitors were inhibited as effectively as wild-type virus by knockdown of TNPO3/TransportinSR2 [4]. This experiment foretells the future development of HIV-1 inhibitors based on the disruption of the interaction between HIV-1 integrase and TNPO3/Transportin-SR2. References 1. Weinberg, J.B., Matthews, T.J., Cullen, B.R., and Malim, M.H. (1991). Productive human immunodeficiency virus type 1 (HIV-1) infection of nonproliferating human monocytes. J. Exp. Med. 174, 1477–1482. 2. Suzuki, Y., and Craigie, R. (2007). The road to chromatin - nuclear entry of retroviruses. Nat. Rev. Microbiol. 5, 187–196. 3. Brass, A.L., Dykxhoorn, D.M., Benita, Y., Yan, N., Engelman, A., Xavier, R.J., Lieberman, J., and Elledge, S.J. (2008). Identification of host proteins required for HIV infection through a functional genomic screen. Science 319, 921–926. 4. Christ, F., Thys, W., De Rijck, J., Gijsbers, R., Albanese, A., Arosio, D., Emiliani, S., Rain, J.-C., Benarous, R., Cereseto, A., et al. (2008). Transportin-SR2 imports HIV into the nucleus. Curr. Biol. 18, 1192–1202. 5. Kataoka, N., Bachorik, J.L., and Dreyfuss, G. (1999). Transportin-SR, a nuclear import receptor for SR proteins. J. Cell Biol. 145, 1145–1152. 6. Lai, M.C., Lin, R.I., Huang, S.Y., Tsai, C.W., and Tarn, W.Y. (2000). A human importin-beta family protein, transportin-SR2, interacts with the phosphorylated RS domain of SR proteins. J. Biol. Chem. 275, 7950–7957. 7. Roe, T., Reynolds, T.C., Yu, G., and Brown, P.O. (1993). Integration of murine leukemia virus DNA depends on mitosis. EMBO J. 12, 2099–2108. 8. Yamashita, M., and Emerman, M. (2005). The cell cycle independence of HIV infections is not determined by known karyophilic viral elements. PLoS Pathog. 1, e18. 9. Yamashita, M., Perez, O., Hope, T.J., and Emerman, M. (2007). Evidence for direct involvement of the capsid protein in HIV infection of nondividing cells. PLoS Pathog. 3, 1502–1510. 10. Luban, J. (2007). Cyclophilin A, TRIM5, and resistance to human immunodeficiency virus type 1 infection. J. Virol. 81, 1054– 1061. 11. Albanese, A., Arosio, D., Terreni, M., and Cereseto, A. (2008). HIV-1 pre-integration
Dispatch R713
complexes selectively target decondensed chromatin in the nuclear periphery. PLoS ONE 3, e2413. 12. McDonald, D., Vodicka, M.A., Lucero, G., Svitkina, T.M., Borisy, G.G., Emerman, M., and Hope, T.J. (2002). Visualization of the intracellular behavior of HIV in living cells. J. Cell Biol. 159, 441–452. 13. Arhel, N., Genovesio, A., Kim, K.A., Miko, S., Perret, E., Olivo-Marin, J.C., Shorte, S., and Charneau, P. (2006). Quantitative fourdimensional tracking of cytoplasmic and nuclear HIV-1 complexes. Nat. Methods 3, 817–824. 14. Bell, P., Montaner, L.J., and Maul, G.G. (2001). Accumulation and intranuclear distribution of unintegrated human
immunodeficiency virus type 1 DNA. J. Virol. 75, 7683–7691. 15. Butterfield-Gerson, K.L., Scheifele, L.Z., Ryan, E.P., Hopper, A.K., and Parent, L.J. (2006). Importin-beta family members mediate alpharetrovirus gag nuclear entry via interactions with matrix and nucleocapsid. J. Virol. 80, 1798–1806. 16. Dupont, S., Sharova, N., DeHoratius, C., Virbasius, C.M., Zhu, X., Bukrinskaya, A.G., Stevenson, M., and Green, M.R. (1999). A novel nuclear export activity in HIV-1 matrix protein required for viral replication. Nature 402, 681–685. 17. Shaheen, H.H., and Hopper, A.K. (2005). Retrograde movement of tRNAs from the cytoplasm to the nucleus in Saccharomyces
Vision: Attention Makes the Cup Flow Over Scalp potentials are surprisingly informative about visual attention: a recent study that used them to record neural responses to up to four superimposed visual patterns simultaneously has now revealed the flow of attentional signals back to visual cortex. Jochen Braun and Mircea Ariel Schoenfeld When we fix our gaze on a complex visual scene, we can alter our phenomenal experience by focussing mentally on different parts or aspects of the scene. Neural correlates of this ‘selective visual attention’ have been observed in an anatomically distributed, but functionally integrated, network of brain sites, including the lateral geniculate and pulvinar nuclei of the thalamus, visual areas in occipital and temporal cortex, and higher order areas in frontal and parietal cortex [1,2]. It is thought that attention signals originate in frontal and parietal cortex and are then transmitted by feedback and recurrent projections backwards to earlier stages of the visual pathways. These efferent signals seem to selectively enhance the amplitude, and perhaps also the temporal synchronicity, of neural responses to the ‘attended’ parts or aspects of a visual scene, at the expense of the neural responses to all other parts or aspects of the scene. Except in the most simplistic displays, however, the attentional enhancement of neural responses is not limited to the desired information, but extends also to some other stimulus features that may be present in the display but that are irrelevant to the task at hand. This ‘spill-over’ to
some irrelevant features (but not to others) is of considerable interest, as it presumably reflects the organization of the projection patterns that communicate attentional signals back to visually responsive neurons. One pattern of spill-over goes by the name of ‘object attention’. Typically, object attention is encountered when two visual patterns are superimposed transparently, that is, such that each pattern remains recognizable individually. To take an idealized example, an array of red items moving coherently in one direction might be superimposed over an array of blue items moving coherently in another direction (Figure 1A). Because of the shared colour and motion, each array is phenomenally experienced as a distinct visual object. In viewing such a display, observers can choose which array they attend and, thus, which array they experience more fully. Attentional spill-over becomes apparent when observers are asked about one particular attribute of one array, the shape of the red array items. In this case, the attentional enhancement — as measured either behaviourally or neurophysiologically — applies not only to the relevant attribute (shape) but also to the irrelevant attributes (motion, colour) of the target array. All attributes of the other array are suppressed, however. Thus, in this
cerevisiae. Proc. Natl. Acad. Sci. USA 102, 11290–11295. 18. Zaitseva, L., Myers, R., and Fassati, A. (2006). tRNAs promote nuclear import of HIV-1 intracellular reverse transcription complexes. PLoS Biol. 4, e332.
Department of Microbiology and Molecular Medicine, University of Geneva, 1 Rue Michel Servet, CH-1211 Geneva 4, Switzerland. E-mail: [email protected]
DOI: 10.1016/j.cub.2008.07.037
simplified example, object attention enhances all responses to the attended array and suppresses all responses to the unattended array. Object attention has been documented most extensively with purely behavioural measures [3–6], although a few studies have encountered its characteristic pattern of spill-over enhancement also in single-unit activity of visual cortex [7] and in visual evoked potentials [8,9]. Note that electrophysiological studies of attentional spill-over face an enormous hurdle: they must distinguish the neural responses not just to two superimposed patterns, but to relevant and irrelevant attributes of these patterns. Over the last decade, the measurement of visual evoked potentials on the scalp has been refined to the point that it can now overcome this hurdle. A key to the singular informativeness of this method is the oscillatory response evoked by a flickering pattern that is known as a ‘steady-state visual evoked potential’ or SSVEP [10]. As the frequency of the oscillatory response matches that of the driving flickering pattern, two patterns flickering at different frequencies elicit distinguishable oscillatory contributions to the visual evoked potential. When observers are required to discriminate one pattern, the neural response to the attended pattern (as measured by the SSVEP) increases relative to the response to the unattended pattern [8,9,11]. As they reported recently in Current Biology, Andersen et al. [12] have been able to distinguish neural responses to four superimposed arrays, setting a new standard for evoked potential methods and affording an even more penetrating insight into attentional