* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download HIV-1 Infection of Nondividing Cells: C-Terminal
Survey
Document related concepts
Cellular differentiation wikipedia , lookup
Cell culture wikipedia , lookup
Cytokinesis wikipedia , lookup
Cell encapsulation wikipedia , lookup
Organ-on-a-chip wikipedia , lookup
Cell nucleus wikipedia , lookup
Tyrosine kinase wikipedia , lookup
Endomembrane system wikipedia , lookup
Signal transduction wikipedia , lookup
Phosphorylation wikipedia , lookup
Transcript
Cell, Vol. 80, 379-388, February 10, 1995, Copyright © 1995 by Cell Press HIV-1 Infection of Nondividing Cells: C-Terminal Tyrosine Phosphorylation of the Viral Matrix Protein Is a Key Regulator Philippe Gallay, Simon Swingler, Christopher Aiken, and Didier Trono Infectious Disease Laboratory The Salk Institute for Biological Studies 10010 North Torrey Pines Road La Jolla, California 92037 Summary The HIV-1 matrix (MA) protei n contains two subcellular localization signals with opposing effects. A myristoylated N-terminus governs particle assembly at the plasma membrane, and a nucleophilic motif facilitates import of the viral preintegration complex into the nucleus of nondividing cells. Here, we show that myristoylation acts as the MA dominant targeting signal in HIV-1 producer cells. During virus assembly, a subset of MA is phosphorylated on the C-terminal tyrosine by a virion-associated cellular protein kinase. Tyrosinephosphorylated MA is then preferentially transported to the nucleus of target cells. An MA tyrosine mutant virus grows normally in dividing cells, but is blocked for nuclear import in terminally differentiated macrophages. MA tyrosine phosphorylation thus reveals the karyophilic properties of this protein within the HIV-1 preintegration complex, thereby playing a critical role for infection of nondividing cells. Introduction Retroviruses are single-stranded RNAviruses whose replication depends on the integration of a double-stranded DNA intermediate, termed the provirus, into the host cell genome. For oncoretroviruses, this process depends on cell proliferation (Humphries and Temin, 1972, 1974), because the breakdown of the nuclear envelope at mitosis is necessary for bringing the viral preintegration complex in contact with the cell chromosomes (Roe et al., 1993; Lewis and Emerman, 1994). In contrast, human immunodeficiency virus type 1 (HIV-1) has the ability to infect nondividing cells (Lewis et al., 1992). This property is shared with other lentiviruses and reflects the existence of determinants that govern the active transport of the viral preintegration complex through the nucleopore (Bukrinsky et al., 1992). It likely plays a crucial role in AIDS pathogenesis because it allows the spread of HIV-1 in such critical targets as terminally differentiated macrophages (Weinberg et al., 1991). Two viral proteins, matrix (MA) and Vpr, have been shown to participate in this process in a partly redundant manner (Bukrinsky et al., 1993a; Heinzinger et al., 1994; von Schwedler et al., 1994). MA is the N-term inal cleavage product of the HIV-1 Gag precursor by the viral protease and contains a nuclear localization signal (NLS) (Myers et al., 1992) that closely resembles the prototypic SV40 large T antigen NLS (Kalderon et al., 1984). A peptide based on the MA NLS acts as a nuclear targeting sequence when coupled to a heterologous protein in vitro (Bukrinsky et al., 1993a). In cells acutely infected with H IV-l, MA was detected in fractions containing partially purified HIV-1 preintegration complexes (Bukrinsky et al., 1993b), as well as in the nucleus (Sharova and Bukrinskaya, 1991). In the absence of a functional vpr gene, the critical role of the MA NLS is further revealed: vpr-MA-NLS double mutant viruses show a defect in nuclear import and, as a result, have an impaired ability to infect nondividing cells such as macrophages (Bukrinsky et al., 1993a; Heinzinger et al., 1994; von Schwedler et al., 1994). In addition to fulfilling this critical function at an early step of the infection process, MA also plays an essential role in virus morphogenesis. Indeed, a myristate residue at the N-terminus of MA directs Gag to the plasma membrane. This targeting is essential for the proper assembly of viral particles and for their release into the extracellular space (Varmus and Swanstrom, 1984). In the process, MA also recruits the envelope glycoprotein at the surface of virions (Yu et al., 1992a, 1992b; Dorfman et al., 1994). Correspondingly, MA has been localized to the periphery of mature HIV-1 particles, bound to the inner leaflet of the virus lipid bilayer (Gelderblom et al., 1987). Since the myristoylation signal and the N LS have opposing influences on the subcellular localization of MA, their respective effects must be tightly regulated. The present work reveals the mechanism of this regulation. We find that myristoylation acts as a dominant targeting signal for MA in HIV-1 producer cells, preventing the NLS-mediated nuclear import of the protein. We further demonstrate that a small subset of MA molecules undergo C-terminal tyrosine phosphorylation at the time of particle assembly. Both membrane association and proteolytic cleavage of Gag are necessary for this modification, which can be observed in the absence of other viral proteins. Accordingly, tyrosine phosphorylation can be observed when recombinant MA is exposed to cellular extracts, as well as viral lysates. Although only 1% of all MA molecules internalized by newly infected cells migrate to the nucleus, the majority of those phosphorylated on tyrosine are rapidly translocated to this compartment. Mutating the C-terminal tyrosine of MA to phenylalanine does not alter Gag processing, nor replication of the resulting virus in dividing cells, such as activated T lymphocytes. However, in the absence of a functional vpr gene, the MA tyrosine mutant is profoundly defective for growth in terminally differentiated macrophages, owing to a block in the nuclear import of the viral preintegration complex. These results thus demonstrate that tyrosine phosphorylation reveals the karyophilic function of MA in the context of the HIV-1 preintegration complex, thereby playing a crucial role for infection of nonproliferating cells such as macrophages. Cell 380 Figure 1. Myristoylation Prevents the NLSMediatedNuclearImportof HIV-1 MA in Producer Cells MA: (A) Schematic representationof MA variants TTK~LK. analyzedin these experiments. MAG2A/KK27Tr: A ~ iii~ ~|~ (B) MA,H~sand MA~, variantsare nonmyristoyT~KLK* rated. [3H]myristate-labeledextractsfrom 293 cells transfected with proviral constructs exCA: pressing eitherwild-typeMA, or the two N-terCX WT MAG2A minus mutatedforms of the protein(see [A]), were analyzed by immunoprecipitationwith MA-specific antibody. (C) Subcellular localizationof MA in trans@ ~ ~ ~ tubulir fected cells. Cytoplasmic (CX) and nuclear Irealrnent: -(NX) fractions of 293 cells transfected with DNAs expressing the various forms of MA shown in (A) (minus: mock)were analyzedby 1 2 3 4 5 6 1 2 3 4 5 6 7 1 2 3 4 5 6 Western blotting,with antibodiesagainst MA (top), CA (middle),and tubulin (bottom).Wildtype MA (lane 2) is found exclusivelyin the cytoplasm,whereaslarge amountsof the two nonmyristoylatedforms of MA (lanes3 and 5) migrate to the nucleus. Nucleartransportof nonmyristoylatedMA is greatly reducedby a mutationin the NLS (lanes4 and 6). CA and the tubulin control are restrictedto the cytoplasm. (D) Wild-type MA is tightly bound to the plasmamembrane.293 ceils expressingwild-typeand a nonmyristoylatedform of MA were separated into plasmamembrane,cytosol,and nucleusfractions,followedby immunoprecipitationwith MA-specificantibodyand Western blotting. (E) Nuclearimportof nonmyristoylatedMA is blockedby NLS peptideand wheatgerm agglutinin(WGA), 293 cellsexpressingMAGu,were treated overnightas indicatedabove each lane (minus: mock). Nuclearfractionswere then analyzedby immunoprecipitationwith MA-specificantibody, followedby Western blotting.The effect of WGA could be preventedby additionof N-acetylglucosamine,as described(Forbes, 1992). KK*~KLKH ..... ~ , ,,J-C'~~ ' !!!!!!!!iia~mm~I S, C Results Myristoylation Prevents the NLS-Mediated Transport of MA to the Nucleus of HIV-1 Producer Cells Altered forms of the HIV-1 MA protein were generated in the context of a full-length proviral construct (Figure 1A). First, two nonmyristoylated variants were created, one by replacing the MA N-terminal glycine by alanine, the second by introducing six histidines at the proximal end of Gag. Both mutations effectively prevented myristoylation (Figure 1B). N LS-defective versions of these proteins were also derived by replacing two lysines in the NLS by threonines, a change previously shown to abrogate the function of this motif (von Schwedler et al., 1994). Cytoplasmic and nuclear extracts of cells transfected with these constructs were then analyzed by Western blotting (Figure 1C). Wildtype MA was found exclusively in the cytoplasm (Figure 1C, lane 2), as was the viral capsid (CA); tubulin, a protein normally restricted to this compartment, sewed as a control. Further separation of the cytoplasm in membrane and cytosol fractions revealed that myristoylated MA was associated with the plasma membrane, as previously described (Varmus and Swanstrom, 1984; Spearman et al., 1994) (Figure 1D, lanes 1-3). In contrast, large amounts of both forms of nonmyristoylated MA were associated with the nucleus (Figure 1C, lanes 3 and 5; Figure 1D, lanes 4-6). This process was NLS-dependent, since mutations in this motif induced the partial redistribution of nonmyristoylated MA to the cytoplasm (Figure 1C, lanes 4 and 6). Furthermore, the nuclear import of nonmyristoylated MA was effectively blocked by peptide containing the prototypic SV40 large T antigen NLS (Figure 1E, lanes 1-4). This correlates the previous demonstration that such peptide can inhibit the MA-mediated nuclear import of the HIV-1 preintegration complex (Gulizia et al., 1994), as well as that of other karyophiles harboring a related signal (Michaud and Goldfarb, 1993). Finally, nuclear import of nonmyristoylated MA could be prevented by wheat germ agglutinin (Figure 1E, lanes 5-7), a lectin that specifically blocks transport through the nuclear pore (reviewed by Forbes, 1992). In these experiments, MA was expressed together with the other products of the gag and pol genes, raising the possibility that these additional polypeptides altered its localization. These studies were thus repeated with HIV-1derived constructs in which the sequence encoding wildtype or nonmyristoylated MA was immediately followed by a stop codon. The respective distribution of these proteins was similar to that observed in the context of full-length precursors, indicating that MA subcellular targeting is independent from other Gag and Pol proteins (data not shown). MA Molecules Imported to the Nucleus of Newly Infected Cells Are Myristoylated The above data indicate that MA has intrinsic nucleophilic properties conferred by the NLS, but in virus producer cells, these are masked by myristoylation. Yet, once HIV-1 enters target cells, a functional MA NLS prevails and plays a major role in facilitating the nuclear import of the viral preintegration complex (Bukrinsky et al., 1993a; von Schwedler et al., 1994). One possible explanation for this paradox was that some unmyristoylated MA was incorporated in vidons and was selectively imported to the nucleus of newly infected cells. Alternatively, proteolytic cleavage could occur on a subset of MA molecules, removing the myristoylated N-terminus. To test these hypotheses, growtharrested cells were acutely infected with viruses labeled with either [3sS]cysteine or [3H]myristic acid. Cytoplasmic MA TyrosinePhosphorylationand HIV NuclearImport 381 A 35S-cysteine-labelledvirions V cpm: CX NX treatment: (z-MA: _ o ~ x 14,422 134 (0.93%) c~-pTyr: B 3H-myristate-labelledvitions: V CX p17-IP' NX 1 2 3 MA-"~ B cpm: 19,516 192 (0.98%) Figure 2. MAMoleculeslmportedtotheNucleusofHIV-1TargetCells Are Myristoylated Cytoplasmic(CX)and nuclear(NX)fractionsof y-irradiatedP4-2cells acutely infectedwith virions purifiedfrom Molt IIIB cells labeledwith either pS]cysteine (A) or [3H]myristate(B) were immunoprecipitated with MA-specificantibody. V, virions. After autoradiography,bands were excised and counted; the number of cpm recoveredin each case is shown.Approximately1% of internalizedMA molecules,fully myristoylated, migrateto the nucleusof acutelyinfectedcells. @ c~-MA: pt7"1~' c~-pTyr: pl 7 " 1 ~ 1 KKKYKLKH ~ 2 3 4 Y The C-Terminal Tyrosine of MA Is Phosphorylated Taken together, these results demonstrate that myristoylation acts as a dominant targeting signal for MA during viral particle formation, yet for 1% of the molecules, it becomes recessive once HIV-1 enters target cells. This suggests a model in which some modification of MA, during or after assembly, is responsible for silencing the influence of myristoylation, thereby revealing the NLS effect. Interestingly, the membrane association of HIV-1 Gag depends not only on myristate, but also on nearby positively charged residues predicted to interact with negatively charged phospholipids in the plasma membrane (Zhou et al., 1994). Such an electrostatic mechanism implies that reversible membrane binding of MA might be achieved by introduction of appropriately placed negative charges into the protein, for instance through phosphorylation. By analogy, platelet-derived growth factor stimulation of 3T3 cells is accompanied by N-terminal phosphorylation of myristoylated pp60°src, which is then released from the cell membrane (Walker et al., 1993). Similarly, phosphorylation of membrane-bound myristoylated alanine-rich C kinase substrate (MARCKS) by protein kinase C releases this protein into the cytosol (Thelen et al., 1991). Based on these precedents, the phosphorylation status of HIV-1 MAwas examined. It was thus found that, in HIV-1 virions purified from acutely infected T lymphoid cells, MA 6 7 VSQNYPIVQN myr-G and nuclear extracts, prepared at 6 hr postinfection, were then analyzed for their respective MA contents. This revealed that approximately 99% of internalized MA molecules were found in a cytoplasmic fraction (Figure 2), mostly in association with the plasma membrane (data not shown). In contrast, 1% of MA molecules were present in a nuclear fraction. An identical nucieocytoplasmic ratio for MA was obtained when radioactive counts obtained from [3sS]cysteine- or from [3H]myristic acid-labeled protein were used to calculate this number. On that basis, it can be inferred that MA molecules associated with the nucleus of newly infected cells are fully myristoylated. 5 ~ C -Fill -F~I. Figure3. HIV-1 MA Is Phosphorylatedon C-TerminalTyrosine (A) Lysates of virions purified from HIV-l-infectedCEM cells were analyzedby Western blottingwith MA- and phosphotyrosine-specific antibodies,as indicated,with or withoutprior treatmentwith phosphatase (CIP), or phosphataseplus orthovanadate(CIP + OV). (B) The same analysis, on virions produced by transfection of 293 cells with constructs containing indicatedtyrosine-to-phenylalanine mutations in MA. The C-terminaltyrosineof MA (Y132) is the target of phosphorylation. (C) Two-dimensionalphosphoaminoacid analysisof s2P-labeledMA (wild type and Y132F) from virions isolatedfrom infectedCEM cells. Wild type (WT) containsphosphoserineand phosphotyrosine;MAy132F contains phosphoserineonly. reacted with an anti-phosphotyrosine antibody in Western blots (Figure 3A, lane 1). The signal could be abolished by prior treatment of viral extracts with calf intestine phosphatase (Figure 3A, lane 2), unless the phosphatase inhibitor orthovanadate was added (Figure 3A, lane 3). MA tyrosine phosphorylation was not restricted to T cell-produced virus, as it was also observed with virions purified from the supernatant of transfected fibroblasts (Figure 3B). HIV-1 MA contains three highly conserved tyrosines, at positions 29, 86, and 132, respectively (Myers et al., 1992). Each of these tyrosines was changed individually to phenylalanine, in the context of a full-length proviral construct. Virions obtained from cells transfected with the resulting plasmids were analyzed by anti-phosphotyrosine Western blotting (Figure 3B). This identified the C-terminal tyrosine of MA, at position 132, as the target of phosphorylation. Phosphoamino acid analysis of virions produced from 32p. labeled cells confirmed that MA is phosphorylated on Tyr- Cell 382 A A 'k'~"c "~-~'~"~" '~'Y'~ c~-MA: c~-CA: B ~,0 p55 -~' (~-pTyr: p55 ~-pTyr: 12345 4-'11 +q p18 p17 ,~" 4 1 2 3 ~J 4 ##~o o~-MA: %&"~ t~" t*~" 17 p24(ng): 70 7070 8 0 lane: 2 3 4 5 p68~ Fraction No. c~-pTyr: p68 "~' 1 2 3 CD4 Figure 4. Requirements of MA Tyrosine Phosphorylation (A) 293 cells stably expressing various forms of Gag (Figure 1; MAsTop encodes a truncated Gag precursor, owing to a stop codon right after the MA coding sequence), or a protease-defective provirus (APro), were analyzed by Western blotting with MA-, CA- and phosphotyrosinespecific antibodies as indicated. MA tyrosine phosphorylation depends on myristoylation and proteolytic cleavage of Gag. (B) The same analysis on 293 cells transiently transfected with vectors expressing a CD4/MA chimera represented below blot, with tyrosine (44MA) or phenylalanine (44MAy~32F)at the MA C-terminus. MA undergoes tyrosine phosphorylation when recruited to the membrane as part of an integral membrane protein. 132, as well as on serine (Figure 3C). Furthermore, radioactive measurements, d o n e by comparing the amounts of [~S]cysteine and 32p incorporated in virus-associated MA when producer cells w e r e labeled for 2 days, indicated that a p p r o x i m a t e l y lS/o of M A molecules present in virions are phosphorylated (data not shown). Requirements for MA Tyrosine Phosphorylation The requirements for MA tyrosine phosphorylation were then determined. It revealed that this modification was d e p e n d e n t on the plasma m e m b r a n e association of Gag, as nonmyristoylated forms of MA were not tyrosine phosphorylated (Figure 4A, lanes 1 and 2). If MA was produced in the absence of other Gag and Psi proteins, its e x p o s e d C-terminus underwent tyrosine phosphorylation (Figure 4A, lane 4). However, no phosphotyrosine was detected in the p55 Gag precursor, w h e t h e r it was produced in normal a m o u n t s by wild-type virus or accumulated by a proteasedefective mutant (Figure 4A, lane 5). The need for Gag proteolytic c l e a v a g e implies that MA tyrosi ne phosphorylation is coupled with virus maturation, as might be expected Figure 5. MA Protein Kinase Activity in Cells and in Virions (A) In vitro phosphorylation of recombinant wild-type and Y132F MA protein, using [7-32P]ATP and pelleted supernatant from human T lymphoid CEM cells (PCS). After reactions were performed as indicated, MA was purified on a nickel column, resolved by SDS-PAGE, and transferred to a PVDF membrane that was exposed to X-ray film. (B) Two-dimensional phosphoamino acid analysis of in vitro phosphorylated recombinant MA recovered from (A), showing incorporation of phosphate on serine, threonine and tyrosine in the case of wild-type, and on serine and threonine only in the case of MAY132F. (C, left panel) in vitro phosphorylation of recombinant MA, using [y;32P]ATPand fractions from a Sephacryl S-1000 gel filtration purification of HIV-1 virions from chronically infected Molt IIIB cells. Curve indicates p24 antigen content of each fraction. Phosphoamino acid analysis (inset) shows that serine, threonine, and tyrosine kinase activities are associated with HIV-1 particles. (C, right panel) Virions present in fractions 3 and 4 were further immunoprecipitated with a mixture of gp120 and gp41 antibodies, or with a control antiserum; the products were then used in an in vitro kinase assay with recombinant MA as substrate. Lane 1, crude viral pellet; lane 2, Sephacryl S-1000-purified virions; lane 3, as lane 2, followed by immunoprecipitation with envelope-specific antibodies; lane 4, as lane 2, followed by immunoprecipitation with control serum, lane 5, negative control (no lysate). The amount of p24 antigen present in each sample is indicated below each lane. for a regulatory event controlling, in a timely manner, the karyophilic properties of this protein. To explore further the factors important for MA tyrosine phosphorylation, a chimeric integral m e m b r a n e protein was constructed, with an extracellular and transmembrane region derived from CD4 and a cytoplasmic domain made by MA. A variant of this chimera, in which the C-terminal tyrosine of MA was changed to phenylalanine, served as control. These proteins, called 4 4 M A and 44MAyt32F, respectively, w e r e then analyzed by Western blotting with MA- and phosphotyrosine-specific antibodies (Figure 4B). The result showed that the tyrosine phosphorylation of the MA C-terminus occurred efficiently in the 4 4 M A protein. It confirmed that this process is independent of the recognition of a specific receptor by the myristoylated end of MA. Furthermore, it demonstrated that MA tyrosine phosphorylation does not require particle formation per se, nor the presence of any other viral protein. MA TyroeinePhosphorylationand HIV NuclearImport 383 A Co-culture: hours; 0 1 2 3 6 0 1 2 3 6 ~-pTyr B Cell free infection: hours: 0 2 4 8 0 2 4 8 o~-MA - ~ c~-pTyr Cytoplasm C Nucleus + NLS pepticle: hours: 0 2 4 0 2 4 8 ~-MA -~" c(-pTyr Cytoplasm The pattern of tryptic phosphopeptides of phosphorylated MA molecules obtained in these experiments was examined and found to be comparable to that observed with virus-associated MA labeled in vivo (data not shown). Significant levels of MA serine, threonine, and tyrosine kinase activities were also found in HIV-1 virions, purified from a chronically infected human T lymphoid cell line by size fractionation (Figure 5C, left). Moreover, the MA protein kinase activity remained associated with the viral particles when these were purified further by immunoprecipitation with antibodies against the viral envelope (Figure 5C, right). Nucleus Figure 6. Tyrosine-PhosphorylatedMA Is RapidlyTransportedto the Nucleus of Acutely InfectedCells Cytoplasmicand nuclear fractions of ,'/-irradiatedP4-2 cells acutely infected either by coculture(at a ratio of 50:1 with Molt IIIB cells) (A) or with cell-freevirions (5 Ilg of p24) (B), harvestedat the indicated times postinfection,were analyzedby immunoprecipitationwith antiMA antibody,followedby Western blottingwith MA- or phoephotyrosine-specific antibody. Whereas the majority of MA remains in the cytoplasm (associatedwith the plasma membrane,.data not shown), most tyrosine-phosphorylatedMA rapidly migratesto the nucleus. ((3) Cell-freeinfection repeatedin the presenceof NLS peptide(500 IIM). Nuclear import of MA is blocked, but tyroeine-phosphorylated species are still detected,demonstratingthat they are not generated in the nucleus. MA Tyrosine Phosphorylation Is Accomplished by a Cellular Protein Kinase Present in Virions The last point implied that a cellular kinase is responsible for MA tyrosine phosphorylation. This was tested by in vitro kinase assays employing recombinant MA protein as substrate (Figure 5). When incubated in the presence of reaction buffer and ATP, MA did not autophosphorylate. In contrast, phosphate moieties were transferred from [y-32P]ATP to MA, when either crude extracts of CEM or 293 cells or pelleted CEM supernatant were added (Figure 5A). Greater levels of 3~p incorporation were seen when pelleted cell supernatant, as opposed to cellular extract, was used as the source of protein kinase activity. However, this was due to higher levels of endogenous ATP in the cells, which interfered with the incorporation of 32p into the MA substrate: when reactions were normalized by adding nonradioactive ATP to 20 raM, cell extracts were found to be more effective at phosphorylating MA (data not shown). In these in vitro assays, incorporation of phosphate was less efficient on tyrosine than on serine, and some phosphothreonine was also observed, whereas none was found with in vivo labeled MA. However, the C-terminal tyrosine of MA was specifically modified, as no phosphotyrosine was recovered when reactions were performed on a recombinant MAy~32Fvariant (Figure 5B). Tyrosine-Phosphorylated MA Is Rapidly Imported to the Nucleus of Newly Infected Cells To ask whether tyrosine phosphorylation influences the karyophilic properties of MA, growth-arrested cells were exposed to HIV-1 and fractionated in cytoplasmic and nuclear fractions, at various times postinfection. These fractions were then analyzed by Western blotting, with MAand phosphotyrosine-specific antibodies (Figures 6A and 6B). As seen previously (Figure 3), the vast majority of MA molecules remained associated with the cytoplasm. In contrast, most of those phosphorylated on tyrosine were rapidly transported to the nucleus. If cells were treated with NLS peptide, total levels of phosphotyrosine-containing MA were not significantly changed, but this protein was retained in the cytoplasm (Figure 6C). This indicates that the nuclear accumulation of tyrosine-phosphorylated MA does not reflect de novo phosphorylation in this compartment, but instead the rapid and preferential translocation of this protein from the cytoplasm to the nucleus of acutely infected cells. An HIV-1 Strain Defective for MA Tyrosine Phosphorylation Is Blocked for Growth in Terminally Differentiated Macrophages Owing to a Defect in Nuclear Import Our findings were compatible with a role for MA tyrosine phosphorylation in HIV-1 nuclear import and, hence, for infection of nondividing cells. To probe this issue, the replication of viruses containing wild-type and phosphotyrosine-defective versions of MA were compared under conditions of cell proliferation and growth arrest. A virus with a mutation in the MA NLS served as additional control, as it was previously shown to be defective specifically in nondividing cells (von Schwedler et al., 1994). Since Vpr can partly substitute for MA in facilitating the nuclear migration of HIV-1 in nonmitotic targets (Heinzinger et al., 1994), these experiments were done with viral strains carrying a defective vprgene. Substituting a phenylalaninefor the C-terminal tyrosine of MA did not alter Gag processing (Figure 3) nor replication of the resulting HIV-1 mutant in activated peripheral blood lymphocytes (Figure 7, top). However, like the MA NLS mutant, this variant was unable to grow in terminally differentiated macrophages (Figure 7, bottom). To examine further this parallel and to determine the level at which the replication of the tyrosine mutant was Cell 384 3o I tLme: 28 Target: PHA-activeted PBL 26 24 • 2021 -- O-" --O"18" "-•'" p,\ R7Bal ; MAKK27.n-.R7Bal ,,./ 24 h 32 h 48 h a F " ~ ¢ l a b clla b c ' • MAyI32F.RTBal R7 *" 16 ' '" *. ,;' 12" 14" 10" 8" 6" 3h 8h ar'-~-]c ~ y ~ O LL -Ib /, /4 4' ¢ 2 4 6 ~ 30" 28 " Target:Macrophages 8 10 12 Ci 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 24 22 2O 18 16 14 12 10 8 62 42 22 5 10 15 20 25 30 days post-infection Figure 7. MA Phosphotyrosine Mutant HIV-1 Is Unable to Grow in Figure 8. NuclearImportof MA PhosphotyrosineMutant HIV-1 Is Defective in Macrophages Terminally differentiated macrophages infected with wild-type MA (a: R7BaL), MA phosphotyrosine-defective(b: MA¥132~R7BaL),or NLS-defective (c: MAKK2TI-FR7BaL)viruses (all of which encode a nonfunctional Vpr protein) were analyzed by PCR at the indicated timepoints, with primersspecificfor variousproductsof reversetranscription. EL, early linear products (elongated minus-standDNA); LL, late linear products (generated after the second template switch);Ci, two-LTRcircles(formed in the nucleus).Both MA mutantsinducewildtype levelsof linear DNA but dramaticallyreducedamountsof circles. Macrophages Growth of NLS-defective (MAKK2rcrR7Bak) and MA phosphotyrosine- defectiveviruses(MA¥132FR7BaL)were comparedwith that of the wildtype parent (R7Bak) in PHA-activated PBL (top) and terminally differentiated macrophages(bottom).The T cell tropic strain R7, which cannot infect macrophages, served as control. All viruses encode a truncated Vpr that cannnotsubstitutefor MA NLS functionin macrophages(yon Schwedler et al., 1994). blocked, viral DNA synthesis in freshly infected macrophages was studied. The accepted model for reverse transcription includes a series of steps, beginning with the synthesis of DNA complementary to the U5 and R regions of the viral long terminal repeat (minus-strand strong-stop DNA), continuing with the elongation of this minus-strand after a first template switch, and culminating in the generation of a full-length double-stranded DNA molecule following a second template switch. Once the viral DNA is transported to the nucleus, full-length linear molecules can circularize either by ligation of their two ends or through recombinatorial events (Varmus and Swanstrom, 1984). The appearance of these various intermediates and products of reverse transcription was thus monitored by polymerase chain reaction (PCR), in terminally differentiated macrophages acutely infected with the HIV-l-derived viruses (Figure 8). This revealed that the phosphotyrosinedefective virus directed the synthesis of wild-type amounts of viral linear DNA, indicating that its entry and reverse transcription proceeded normally. In contrast, and like the NLS mutant, this variant produced much lower levels of viral DNA circles compared with wild type. These results indicated that the absence of tyrosine at the C-terminal end of MA specifically impairs the nuclear import of HIV-1 DNA. These data thus allow us to conclude that tyrosine phosphorylation reveals the karyophilic properties of MA in the context of the HIV-1 preintegration complex, thereby playing an esential role for infection of terminally differentiated macrophages. Discussion The HIV-1 MA protein fulfills two crucial functions at different stages of the virus life cycle, as it is required for both particle assembly and nuclear import of the viral preintegration complex. The myristoylation signal and the NLS, which exert contradictory influences on the subcellular localization of MA, are each essential for one of these two functions. Here, we present a model for the regulatory mechanism that allows MA to play its dual role. We first show that myristoylation acts as a dominant targeting signal for MA in virus producer cells, leading to particle formation at the plasma membrane and release of new virions into the extracellular space. We then demonstrate that a small subset of MA molecules undergo tyrosine phosphorylation. The dependence of this process on both membrane association and proteolytic cleavage of Gag suggests that it occurs concomitantly with virus maturation. A protein kinase activity that transfers phosphate to the C-terminal tyrosine of recombinant MA can be detected in cells as well as in purified virions. A crucial link between MA tyrosine phosphorylation and the subsequent import of the HIV-1 preintegration complex is revealed by two lines of evidence. First, there is rapid translocation of tyrosine-phosphorylated MA to the nucleus of newly infected cells, whereas the bulk of MA, not phosphorylated on tyrosine, remains at the plasma membrane. Second, a mutant virus in which the critical MA tyrosine is changed to phenylalanine is selectively impaired for growth in terminally differentiated macrophages, owing to a block in nuclear import. MA Tyrosine Phosphorylation and HIV-1 Nuclear Import: Potential Mechanisms The nucleophilic properties of several other proteins are derepressed following phosphorylation, For instance, treat- MA Tyrosine Phosphorylationand HIV Nuclear Import 385 ment of cells with interferon-a (IFN-~) induces tyrosine phosphorylation and nuclear translocation of the IFN-ccstimulated gene factor 3 (ISGF3), a transcriptional regulator involved in mediating cellular responses to the cytokine (Schindler et al., 1992). In that case, an interaction between the phosphotyrosine and an SH2 domain mediates the dimerization of the protein, perhaps exposing a yet unidentified nucleophilic motif (Shuai et al., 1994). Also, the nuclear import of SV40 large T antigen is significantly reduced when sequences flanking the NLS and containing casein kinase II (CKII) phosphorylation sites are removed (Rihs and Peters, 1989). This might reflect an increased affinity of phosphorylated T antigen for a cytoplasmic receptor responsible for its routing to the nuclear envelope. It is likely that tyrosine phosphorylation enhances the nuclear transport of HIV-1 MA through a different mechanism. Indeed, nonmyristoylated MA is effectively imported to the nucleus even though it is not tyrosine phosphorylated (Figures 1 and 4). Based on this, it is probable that phosphorylation triggers the redistribution of MA from the membrane to the inner region of the virus, thereby allowing the NLS to play its role in the context of the viral preintegration complex. This could be accomplished by at least two nonmutually exclusive mechanisms: by simply disrupting interactions between MA and the viral membrane, or by stimulating the binding of MA to another component of the viral core nucleoprotein complex. Membrane release of MA could result from a major conformational change in the protein following phosphorylation. However, a more subtle mechanism could achieve the same result. Indeed, the proximal region of HIV-1 Gag contains a bipartite membrane targeting signal, comprising the myristoylated N-terminal 14 residues, plus an adjacent sequence of 17 amino acids, a number of which are positively charged (Spearman et al., 1994; Zhou et al., 1994). This more distal region, which has been shown to bind to acidic phospholipids in vitro (Zhou et al., 1994), is predicted to establish electrostatic interactions with negatively charged phospholipids located on the inner leaflet of the membrane bilayer. Thus, by introducing negative charges in MA, phosphorylation could disrupt these interactions, thereby releasing the protein from the plasma membrane. A similar mechanism has been shown to account for the release of myristoylated pp60~~'°and MARCKS from the plasma membrane into the cytosol following phosphorylation by protein kinases A and C, respectively (Walker et al., 1993; Thelen et al., 1991). With MA, detachment from the membrane would unleash the karyophilic potential conferred by the NLS. However, two arguments can be advanced against this simple model. First, the tyrosine that becomes phosphorylated is at a significant distance from the region of MA involved in membrane binding, not only considering a linear map of the protein but also its recently proposed structure (Matthews et al., 1994). Second, tyrosine-phosphorylated MA is detected in virus producer or MA-expressing cells (Figure 1; data not shown), yet in these cells it resides in the cytoplasm but not the nucleus. This suggests that an additional factor potentiates the effect of MA tyrosine phosphorylation once HIV-1 leaves producer cells. On the one hand, there could be an additive effect of MA serine phosphorylation, and in agreement with such a model, our preliminary studies indicate that this modification too is coupled to virus maturation (S. S. and D. T., unpublished data). Alternatively, it could be that MA is preferentially released from the membrane of virions rather than from the membrane of virus producer cells. In that respect, the fluidity and lipidic composition of the HIV-1 membrane have been shown to significantly differ from those of the host cell plasma membrane (Aloia et al., 1993). Finally, tyrosine phosphorylation could trigger an association between MA and one of the proteins that are directed to the central region of the core during virus maturation and are ultimately involved in forming the virus preintegration complex, such as the integrase and the reverse transcriptase. The nuclear magnetic resonance analysis of MA predicts that the C-terminal 20 amino acids of the protein do not adopt a rigid conformation and, therefore, might be available for such an interaction (Matthews et al., 1994). An implication of all the above models is that tyrosinephosphorylated MA molecules might be found inside the viral core. Previous immunoelectron microscopic studies have instead localized MA at the periphery of virions, tightly bound to the inner leaflet of the virus membrane (Gelderblom et al., 1987). However, of the 1500 MA molecules estimated to be present in each virion (Gelderblom, 1991), on the basis of our data, we would not expect more than 1 ° , or approximately 15, to be associated with the core. Indeed, this represents the fraction subsequently imported to the nucleus of newly infected cells (Figure 3). It is reasonable to assume that such a small number of MA molecules would have escaped detection by electron microscopy. Another prediction derived from our experiments is that, following fusion of the virus and target cell membranes, tyrosine-phosphorylated MA molecules are the ones whose NLS is recognized by a cytoplasmic chaperone responsible for leading the preintegration complex to the nucleopore (Adam and Gerace, 1991; Forbes, 1992). Our preliminary analyses show that MA molecules contained in the preintegration complex are indeed phosphorylated on tyrosine, supporting this hypothesis (P. G. and D. T., unpublished data). A Virus-Associated Cellular Protein Kinase Responsible for MA Tyrosine Phosphorylation The data presented here provide strong evidence indicating that MA tyrosine phosphorylation is accomplished by a cellular protein kinase. The MA tyrosine kinase activity is not restricted to T lymphocytes but is also found in fibreblasts. It fails to phosphorylate nonmyristoylated MA, and so it is likely associated with the cell membrane. This activity can also be detected in the cell supernatant, most probably due to cellular breakdown, rather than to secretion. Remarkably, the MA tyrosine kinase is present in HIV-1 virions purified by size fractionation followed by immuneprecipitation with anti-envelope antibodies. Whether this reflects the stable binding of the enzyme to its MA substrate or the passive incorporation of a membrane-bound Cell 386 e n z y m e in b u d d i n g v i r i o n s is still to be d e f i n e d . T h e t a r g e t o f t h e t y r o s i n e k i n a s e is t h e last a m i n o acid o f MA, a prev i o u s l y u n s e e n p l a c e m e n t for a p h o s p h o t y r o s i n e . O n l y the f r e e e n d o f M A is r e c o g n i z e d , e i t h e r a f t e r it is l i b e r a t e d by p r o t e o l y t i c c l e a v a g e o f the G a g p r e c u r s o r or w h e n it is n a t u r a l l y e x p o s e d , for i n s t a n c e in the MAsToP a n d 4 4 M A m o l e c u l e s ( F i g u r e 4). T h i s a p p e a r s to p e r m i t a c o u p l i n g between virus maturation and MA tyrosine phosphorylation. T h e M A t y r o s i n e k i n a s e r e m a i n s to be f o r m a l l y identified. H o w e v e r , t h e p r e s e n t w o r k e s t a b l i s h e s that it p l a y s a crucial role in facilitating HIV-1 r e p l i c a t i o n in t e r m i n a l l y d i f f e r e n t i a t e d m a c r o p h a g e s , cells t h o u g h t to be critical for both the p e r s i s t e n c e a n d t h e d i s s e m i n a t i o n o f the infection ( G a r t n e r a n d P o p o v i c , 1990). Efforts a i m e d at i d e n t i f y i n g c o m p o u n d s t h a t b l o c k M A t y r o s i n e p h o s p h o r y l a t i o n are t h u s w a r r a n t e d , as t h e y m a y lead to n o v e l t h e r a p i e s for s l o w i n g the s p r e a d o f the v i r u s in H I V - i n f e c t e d individuals. Experimental Procedures DNA Constructions PCR-mediated mutagenesis was used to introduce site-specific mutations in the MA- or protease-encoding regions of plasmid R7 (Kim et al., 1989; yon Schwedler et al., 1993), which contains the HIV-1HXB2D provirat DNA (Shaw et al., 1984) with a full-length nef reading frame, as well as in R7BaL, which encodes a macrophage-tropic virus (von Schwedler et al., 1994). Both R7 and R7BaL produce a truncated Vpr protein, 78 amino acids in length, previously shown to be unable to substitute for the NLS function of MA in macrophages (von Schwedler et al., 1994). The MAG~ variant has the N-terminal glycine of Gag changed to alanine. The proximal sequence of the MA,H,, protein is Met/Gly/(His)JSer, in which Ser is normally the sixth residue of Gag. APro has a 45 nt deletion in the pol gene (bases 2384-2428, as numbered by Ratner et al., 1985). The various mutations were also introduced in a modified proviral DNA construct, 3E-neo, which expresses an env-defective version of R7 and contains the sequence encoding for neomycin phosph0transferase instead of nef (Trono, 1992). The 44MA and 44MA~132Fchimeras were constructed by PCR, introducing a unique cloning site at the junction between the transmembrane region of CD4 and the N-terminal end of MA, in the context of a CMVbased expression vector (Aiken et al., 1994). Recombinant MA carrying an N-terminal histidine tag was produced in Escherichia coil using the bacterial expression vector pET-15b (Novagen) and purified by affinity chromatography on a nickel-Sepharose column according to the instructions of the manufacturer. Cells All cells were maintained as previously described (Aiken et al., 1994; von Schwedler et al., 1994). HIV-t-infected Molt IIIB human T lymphoid cell lines were obtained from C. Farnet through the NIH AIDS Research and Reference Program. HeLa-derived P4-2 cells (Charneau et al., 1990) were a gift from F. Clavel. Peripheral blood lymphocytes (PBLs) and monocyte-derived macrophages were purified as previously described (von Schwedler et al., 1994). Macrophages were kept in culture for 2 weeks before infection. For stimulation, PBLs were incubated with phytohemagglutinin (Bethesda Research Laboratories) at 3 ~g/ ml for 48 hr and then maintained in RPMI 1640 containing 10% fetal calf serum and recombinant interleukin-2 (Sigma) at 10 U/ml. To induce growth arrest, P4-2 cells were exposed for 20 min to a 61Co source calibrated at 200 rad/min. After irradiation, cells were diluted for infection to 1 x 106/ml and plated. The arrested state of the cells at the time of infection was verified by propidium iodide staining of their DNA and flow cytometry to ascertain that all cells were in the G2 phase of the cycle. Stable 293 cell lines expressing various MA variants, as well as APro, were created by transfection with LIE-neo-derived plasraids and selection in 0.5 mg/ml G418. TransfecUons and Infections Tranfections and preparations of viral stocks were as previously described (von Schwedler et al., 1993; Aiken et al., 1994). Infections of PBLs (2 x 106 cells) and macropbages (0.5 x 10e cells) were performed with 20 ng of p24 of 293 cell-produced viruses as previously described (von Schwedler et al., 1994). For virus growth curves, cells were washed extensively after virus adsorption and p24 antigen production was monitored by ELISA (DuPont). Experiments were repeated many times with cells from different donors, with similar resu Its. In rare instances, however, both NLS and tyrosine mutants showed a less pronounced phenotype, perhaps linked to the state of arrest or differentiation of the macrophages. To follow MA nuclear import, infections of P4-2 irradiated cells were performed either by coculture with Molt IIIB cells at a ratio of 50 to 1, or by the addition of cell-free virus purified from the same cells (5 p.g of p24 antigen per 3 x 107 P4-2 cells). In the experiments of Figure 2, infections were performed using cell-free virus produced by Molt IIIB cells radiolabeled for 20 hr, either with 200 p.Ci/ml of Tran~S-label (ICN) or with 500 i~Ci/ml of [3H]myristic acid (Amersham). Virions used for in vitro kinase assays were prepared by gel fractionation on a Sephacryl S-100O column as previously described (Trono, 1992). Western Blots and Immunoprecipitstions Rabbit anti-HIV-1 MA serum was obtained by immunization with the above described recombinant histidine-tagged HIV-1 MA protein. Monoclonal antibodies against p24/p55 (ST-3) and tubulin were purchased from Bethesda Research Laboratories and Boehringer Mannhelm, respectively. Rabbit anti-phosphotyrosine serum was a gift from B. Seffon, Salk institute. Of note, several commercially available antiphosphotyrosine monoclonal antibodies failed to recognize the tyrosine-phosphorylated form of MA. Western blot and immunoprecipitation studies were as previously described (Aiken et al., 1994). Subcellular FracUonation Cells were lysed in cold hypotonic buffer (20 mM potassium HEPES [pH 7.8], 5 mM potassium acetate, 0.5 mM MgCI2, 0.5 mM dithiothreitol, 1 mM sodium orthovanadate, 1 p.g/ml leupeptin, 1 p~g/ml aprotinin, 1 ~g/ml pepstatin A, 1 mM phenylmethylsulfonyl fluoride [PMSF]) using several strokes of a Dounce homogenizer as described by Sugasawa et al. (1990). Nuclear preservation during cell lysis was followed by phase-contrast microscopy. Cytoplasmic and nuclear extracts were separated by centrifugation at 750 x g for 5 rain, and nuclei were extracted with a hypertonic buffer (20 mM potassium HEPES [pH 7.8], 0.5 mM MgCI2, 500 mM potassium acetate, 0.5 mM dithiothreitol, 1 mM sodium orthovanadate, 1 p.g/ml leupeptin, 1 ~g/ml aprotinin, 1 p.g/ml pepstatin A, 1 mM PMSF). Membrane and cytosol fractions were isolated from cytoplasmic extracts as described (Lin et al., 1991). High pressure liquid chromatography (H PLC)-purified peptides corresponding to the prototypic SV40 T antigen NLS in the sense (PKKKRKVEDPYC) and reverse (PDEVKRKKKPYC) orientations were used as described (Gulizia et al., 1994). Wheat germ agglutinin (WGA) was used at 10 I~M, with or without addition of N-acetylglucosamine (450 I~M), as described in Finlay et al. (1987). Phosphorylation Studies HIV-1 DNA (40 I~g) was transfected into 5 × 106 CEM ceils by electroporation. At the peak of p24 viral antigen production, ceils were radiolabeled with 35 mCi of ~P H3PO4 (DuPont) for 48 hr at 37°C. Labeled virions were first filtered and pelleted, and then incubated for 10 rain in viral lysis buffer (12 mM Tris-HCI [pH 8.0], 0.5% Triton X-100, 300 mM NaCI, 1 mM MgCI2, 0.5 mM dithiothreitol, and 100 rim sodium orthovanadate). M A protein was the n immunoprecipitated with a monoclonal anti-HIV-1 IIIB MA antibody (Advanced Biotechnologies Inc.)on protein A-Sepharose beads. After SDS-PAGE and electrobiotting to polyvinylidene difluoride (PVDF) membrane, phosphorylated MA was visualized by autoradicgraphy and analyzed further for phosphoamino acid content, essentially as described by Boyle et al. (1991). PCR Analysis of Acutely Infected Cells PCR analysis was performed as previously described (yon $chwedler et al., 1993). The sequences of HIV-specific primers are as follows (positions of nucleotides in the HIV-1H×B~Dsequence, according to MA Tyrosine Phosphorylation and HIV Nuclear tmport 387 Ratner etal. [1985], are indicated in parentheses). Vif6: GGGAAAGCTAGGGGATGGI-]TTAT (5136-5159); Vif?: CAGGGTCTACTTGTGTGCTATTC (5340-5317); LTR5: GGCTAACTAGGGAACCCACTGCTT (496-516); 5NC2: CCGAGTCCTGCGTCGAGAGAGC (698-677); LTRS, TCCCAGGCTCAGATCTGGTCTAAC (488-465 and 9572-9549); L TRg, GCCTCAATAAAGCTTGCCTTG (522-542 and 9606-9626). Vif6 plus Vif7 amplify elongated minus-strand DNA, LTR5 plus 5NC2 amplify double-stranded molecules generated after the second template switch, and LTR8 plus LTR9 amplify two-LTR circles. Acknowledgments Correspondence should be addressed to D. T. We thank Tony Hunter for helpful suggestions during the course 0I t'nese studies, as well as critical reading of this manuscript. We are indebted to Bart Sefton for the gift of the anti-phosphoryrosine antibody, to Inder Verma, Tom Hope, and Rick Bushman for insightful cormnents, to Jinping Song for technical assistance, and to Verna Stitt for the artwork. This study was supported by a grant from the Berger Foundation to D. T. and by postdoctoral fellowships from the Swiss National Science Foundation to P. G. and from the National Institutes of Health to C. A. As a Pew Scholar in the Biomedical Sciences, D. T. receives support from the Pew Charitable Trust. Received November 9, 1994; revised December 22, 1994. References Adam, S. A., and Gerace, L. (1991). Cytosolic proteins that specifically bind nuclear location signals are receptors for nuclear import. Cell 66, 837-847. Aiken, C., Konner, J., Landau, N. R., Lenburg, M. E., and Trono, D. (1994). Nef induces CD4 endocytosis: requirement for a critical dileucine motif in the membrane-proximal CD4 cytoplasmic domain. Cell 76, 853-864. Aloia, R. C., Tian, H., and Jensen, F. C. (1993). Lipid composition and fluidity of the human immunodeficiency virus envelope and host cell plasma membranes. Prec. Natl. Acad. Sci. USA 90, 5181-5185. Boyle, W. J., Van Der Geer, P., and Hunter, T. (1991). Phosphopeptide mapping and phosphoamino acid analysis by two-dimensional separation on thin-layer cellulose plates. Meth. Enzymol. 201, 110-256. Bukrinsky, M I., Sharova, N., Dempsey, M. P., Stanwick, T. L., Bukrinskaya, A. G., Haggerty, S., and Stevenson, M. (1992). Active nuclear import of human immunodeficiency virus type 1 preintegration complexes. Proc. Natl. Acad. Sci. USA 89, 6580-6584. Bukrinsky, M. I., Haggerty, S., Dempsey, M. P., Sharova, N., Adzhubei,,, A., Spitz, L., Lewis, P., Goldfarb, D., Emerman, M., and Stevenson,,, M. (1993a). A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells. Nature 365, 666-669. Bukrinsky, M. I., Sharova, N., McDonald, T. L., Pushkarskaya,, T., Tarpley, W. G., and Stevenson, M. (1993b). Association of integTase, matrix, and reverse transcriptase antigens of human immu~odeficiency virus type 1 with viral nucleic acids following acute infection. Proc. Natl. Acad. Sci. USA 90, 6125-6129. Charneau, P., Alizon, M., and Clavel, F. (1990). A second origin of DNA plus-strand synthesis is required for optimal human immunodeficiency virus replication. J. Virol. 66, 2814-2820. Dorfman, T., Mammano, F., Haseltine, W. A., and GSttlinger, H. G. (1994). Role of the matrix protein in the virion association of the human immunodeficiency virus type 1 envelope glycopro,tein. J. Virol. 68, 1689-1696. Finlay, D. R., Newmeyer, D. D., Price, T. M., and: Forbes, D. J. (1987). Inhibition of in vitro nuclear transport by a lectin that binds to nuclear pores. J. Cell Biol. 104, 189-200. Forbes, D. J. (1992). Structure and function of the nuclear pore complex. Annu. Rev. Cell Biol. 8, 495-527. Gartner, S., and Popovic, M. (1990). Macrophage. trapism of HIV-I. AIDS Res. Hum. Retroviruses 8, 1017-1021. Gelderblom, H. R. (1991). Assembly and morphology of HIV: potential effect of structure on viral function. AIDS 5, 617-638. Gelderblom, H. R., Hausmann, E. H. S., C)zel, M., Pauli, G., and Koch, M. A. (1987). Fine structure of human immunodeficiency virus (HIV) and immunolocalization of structural proteins. Virology 156, 171-176. Gulizia, J., Dempsey, M. P., Sharova, N., Bukrinsky, M. I., Spitz, L., Goldfarb, D., and Stevenson, M. (1994). Reduced nuclear import of human immunodeficiency virus type 1 preintegration complexes in the presence of a prototypic nuclear targeting signal. J. Virol. 68, 20212025. Heinzinger, N. K., Bukrinsky, M. I., Haggerty, S. A., Ragland, A. M., Kewalramani, V., Lee, M.-A., Gendelman, H. E., Ratner, L., Stevenson, M., and Emerman, M. (1994). The Vpr protein of human immunodeficiency virus type 1 influences nuclear localization of viral nucleic acids in nondividing host cells. Proc. Natl. Acad. Sci. USA 91,7311-7315. Humphries, E. H., and Temin, H. M. (1972). Cell cycle dependent activation of Rous sarcoma virus-infected stationary chicken cells: avian leukosis virus group-specific antigens and ribonucleic acid. J. Virol. 10, 82-87. Humphries, E. H., and Temin, H. M. (1974). Requirement for cell division for initiation of transcription of Rous sarcoma virus RNA, J. Virol. 14, 531-546. Kalderon, D., Roberts, B. L., Richardson, W. D., and Smith, A. E. (1984). A short amino acid sequence able to specify nuclear location. Cell 39, 499-509. Kim, S., Byrn, R., Groopman, J., and Baltimore, D. (1989). Temporal aspects of DNA and RNA synthesis during human immunodeficiency virus infection: evidence for differential gone expression. J. Virol. 63, 3708-3713. Lewis, P. F, and Emerman, M. (1994). Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J. Virol. 68, 510-516. Lewis, P, Hensel, M., and Emerman, M. (1992). Human immunodeficiency virus infection of cell arrested in the cell cycle. EMBO J. 11, 3053-3058. Lin, P. H., Selinfreund, R., and Wharton, W. (1991). Isolation of cell membrane for epidermal growth factor receptor studies. Meth. Enzymol. 198, 251-259 Matthews, S., Badow, P., Boyd, J., Barton, G., Russell, R., Mills, H., Cunningham, M., Meyers, N., Burns, N., Clark, N., Kingsman, S., Kingsman, A., and Campbell, I. (1994). Structural similarity between the p17 matrix protein of HIV-1 and interferon-~/. Nature 370, 666668. Michaud, N., and Goldfarb, D. S. (1993). Most nuclear proteins are imported by a single pathway. Exp. Cell Res. 208, 128-136. Myers, G., Berzofsky, J. A., Pavlakis, G. N., Korber, B., and Smith, R.F., eds. (1992). Human Retroviruses and AIDS 1992: A Compilation and Analysis of Nucleic Acid and Amino Acid Sequences (Los Alamos, New Mexico: Los Alamos National Laboratory). Ratner, L., Haseltine, W. A., Patarca, R., Livak, K. J., Starcich, B., Josephs, S. F., Doran, E. Z., Rafalski, J. A., Whitehorn, E. A., Baumeister, K., Ivanoff, L., Petteway, S. R., Pearson, M. L., Lautenberger, J. A., Papas, T. S., Ghrayeb, J., Chang, N., Gallo, R. C., and Wong-Staal, F. (1965). Complete nucleotide sequence of the AIDS virus, HTLV-III. Nature 313, 277-283. Rihs, H.-P., and Peters, R. (1989). Nuclear transport kinetics depend on phosphorylation-site-containing sequences flanking the karyophilic signal of the Simian virus 40 T-antigen. EMBO J. 7, 1479-1484. Roe, T., Reynolds, T. C., Yu, G., and Brown, P. O. (I993). Integration of murine leukemia virus DNA depends on mitosis'..EMBO J. 12, 20992108. Schindler, C., Shuai, K., Prezioso, V. R., and Darnell, J. E., Jr. (19921, Interferon-dependent tyrosine phosphoryla~ion of a latent cytoplasmic transcription factor. Science 257, 809-813. Sharova, N., and Bukrinskaya, A. (1991)~ p17 and p17-contai~ihg gag precursors of input human immunodeficiency virus are tra~,ported into the nuclei of i~fected cells. AIDS Res. Hum. Retroviruse~ 7, 303306. Shaw, G. M., Hahn, B. H., Arya, S. K., Groopman, J. E--~.Gallo, R. C., Cell 388 and Wong-Staal, F. (1984). Molecular characterization of human T-cell leukemia (lymphotropic) virus type III in the acquired immune deficiency syndrome. Science 226, 1165-1171. Shuai, K., Horvath, C. M., Tsai Huang, L. H., Qureshi, S. A., Cowburn, D., and DarneU, J. E., Jr. (1994). Interferon activation of the transcription factor Stat91 involves dimerization through SH2-phosphotyrosyl peptide interactions. Cell 76, 821-826. Spearman, P., Wang, J.-J., Vander Heyden, N., and Rather; L. (1994). Identification of human immunodeficienvy virus type 1 Gag protein domains essential to membrane binding and particle assembly. J. Virol. 68, 3232-3242. Sugasawa, K., Murakami, Y., Miyamoto, N., Hanaoka, F., and Ui, M. (1990). Assembly of nascent DNA into nucleosome structures in simian virus 40 chromosomes by HeLa cell extract. J. Virol. 64, 4820-4829. Thelen, M., Rosen, A., Nairn, A. C., and Aderem, A. (1991). Regulation by phosphorylation reversible association of a myristoylated protein kinase C substrate with the plasma membrane. Nature 351,320-322. Trono, D. (1992). Partial reverse transcripts in virions from human immunodeficiency and routine leukemia viruses. J. Virol. 66, 48934900. Varmus, H. E., and Swanstrom, R. (1984). Replication of Retroviruses. In RNA Tumor Viruses, Second Edition, R. A. Weiss, N. Teich, H. E. Varmus, and J. Coffin, eds. (Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press), pp. 369-512. von Schwedler, U., Song, J., Aiken, C., and Trono, D. (1993). vif is crucial for human immunodeficiency virus type 1 proviral DNA synthesis in infected cells. J. Virol. 67, 4945-4955. von Schwedler, U., Kornbluth, R. S., and Trono, D. (1994). The nuclear localization signal of the matrix protein of human immunodeficiency virus type 1 allows the establishment of infection in macrophages and quiescent T lymphocytes. Proc. Natl. Acad. Sci. USA 91, 6992-6996. Walker, F., deBlaquiere, J., and Burgess, A. W. (1993). Translocation of pp60~'`rcfrom the plasma membrane to the cytosol after stimulation by platelet-derived growth factor. J. Biol. Chem. 268, 19552-19558. 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. Yu, X., Yu, Q,-C., Lee, T.-H., and Essex, M. (1992a)~ The C terminus of human immunodeficiency virus type 1 matrix protein is involved in early steps of the virus life cycle. J. Virol. 66, 5667-5670. Yu, X., Yuan, X., Matsuda, Z., Lee, T.-H., and Essex, M. (1992b). The matrix protein of human immunodeficiency virus type 1 is required for incorporation of viral envelope protein into mature virions. J. Virol. 66, 5966-5971. Zhou, W., Parent, L. J., Wills, J. W., and Resh, M. D. (1994). Identification of a membrane-binding domain within the amino-terminal region of human immunodeficiency virus type 1 Gag protein which interacts with acidic phospholipids. J. Virol. 68, 2556-2569.