A Novel Kinase Activity Associated with Nef Derived from Neurovirulent Simian Immunodeficiency Virus

VIROLOGY

251, 165±175 (1998) VY989408

ARTICLE NO.

A Novel Kinase Activity Associated with Nef Derived from Neurovirulent Simian Immunodeficiency Virus Sheila A. Barber,* Maureen T. Flaherty,* Scott M. Plafker,² and Janice E. Clements*,³,1 *Division of Comparative Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205; ²Department of Internal Medicine, Howard Hughes Medical Institute, University of Virginia, Charlottesville, Virginia 22908; and ³Departments of Molecular Biology and Genetics and Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 Received June 5, 1998; returned to author for revision July 23, 1998; accepted September 1, 1998 The Nef proteins of Simian immunodeficiency virus (SIV) and human immunodeficiency virus (HIV) have been shown to associate with several cellular kinases. Further, the ability of SIVmac239 Nef to associate with a p21-activated kinase (PAK)-related kinase has been correlated with pathogenic progression to AIDS in rhesus macaques. Because the ability of Nef to associate with the PAK-related kinase is viral isolate dependent, we reasoned that viral isolates derived from distinct physiological locations may encode Nef proteins that exhibit distinct kinase association profiles. In this study, we compared kinase activities associated with Nef proteins derived from the prototypic lymphocyte-tropic SIVmac239 and a macrophagetropic, neurovirulent clone, SIV/17E-Fr. Our findings not only support previous studies that have documented the association of SIVmac239 Nef with a PAK-related kinase and a Nef-associated kinase complex (NAKC) but describe a novel serine kinase activity detectable only in conjunction with the Nef protein derived from the neurovirulent clone, SIV/17E-Fr. The latter Nef protein does not associate with PAK, and unlike PAK or NAKC, this novel kinase activity is enhanced in association with nonmyristoylated forms of Nef and can utilize both ATP and GTP as phosphodonors. We also show that at least one substrate for the kinase is Nef itself and demonstrate that the SIV/17E-Fr Nef protein is phosphorylated in SIV-infected cells. These results suggest that the ability to associate with cellular kinases in general may be a conserved feature of Nef, but particular kinase/Nef associations may evolve with changes in the host environment concomitant with viral spread. © 1998 Academic Press

molecular determinants that contributed to the evolution of SIV/17E-Br, a panel of infectious molecular clones was subsequently prepared and characterized. SIV/17E-Cl, a clone containing a portion of SIV/17E-Br env in the genetic background of SIVmac239, was able to replicate in macrophages but failed to induce neurological pathology in macaques (Anderson et al., 1993; Flaherty et al., 1997; Mankowski et al., 1997). SIV/17E-Fr, a clone containing the env, nef, and 39 LTR of SIV/17E-Br in the genetic background of SIVmac239, replicated to higher titers in macrophages than SIV/17E-Cl and caused encephalitis in macaques (Flaherty et al., 1997; Mankowski et al., 1997). Further, four of four macaques that developed severe neurological disease after dual infection with SIV/17E-Fr and SIV/DeltaB670 (a swarm of both macrophage-tropic and lymphocyte-tropic viruses) had detectable SIV/17E-Fr DNA in the brain, and in three of these four macaques, SIV/17E-Fr DNA was the only detectable SIV genotype in the brain (Zink et al., 1997). Collectively, these studies suggest that although env sequences of macrophage-tropic, neurovirulent SIV/ 17E-Br are sufficient to confer macrophage tropism on lymphocyte-tropic SIVmac239, the additional presence of a full-length nef gene (not present in SIV/17E-Cl) is required for optimal viral replication in vitro and may contribute to the development of CNS disease in vivo. The nef genes of HIV and SIV encode a nonstructural

INTRODUCTION Encephalitits and dementia are among the common pathological complications associated with human immunodeficiency virus (HIV) infection (Power et al., 1993; Price et al., 1988). Virus isolated from the central nervous system (CNS) of AIDS dementia patients is ostensibly macrophage tropic; however, not all individuals infected with macrophage-tropic strains of HIV develop neurological disease (Cheng-Mayer et al., 1989; Price et al., 1988, 1990). Thus all macrophage-tropic viruses are not simultaneously neurovirulent, suggesting other viral factors (distinct from tropism determinants) contribute to the development of dementia. The Simian immunodeficiency virus (SIV) macaque model provides a powerful research system to study AIDS dementia because SIV-induced CNS disease mirrors key clinical and pathological features of HIV-induced encephalitis (Letvin and King, 1990; Ringler et al., 1998). Previously, we described the derivation of the macrophage-tropic, neurovirulent virus SIV/17E-Br from multiple in vivo passages of the prototypic lymphocyte-tropic virus SIVmac239 (Sharma et al., 1992). To identify the

1 To whom reprint requests should be addressed at 720 Rutland Avenue, Traylor G-60. Fax: (410) 955-9823.

165

0042-6822/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

166

BARBER ET AL.

protein that is functionally interchangeable between these two lentiviruses (Sinclair et al., 1997). Previous studies have determined that the nef gene is required for viral replication and pathogenesis in vivo (Daniel et al., 1992; Jamieson et al., 1994; Kestler et al., 1991) and that Nef contributes to the down-regulation of MHC I and CD4 (the primary receptor for HIV and SIV) from the surface of infected cells (Guy et al., 1987; Schwartz et al., 1996), enhanced virion infectivity (Chowers et al., 1994), and optimal replication in vitro (Flaherty et al., 1997). Most recently, it has been suggested that Nef also plays a role in the induction of c-kit in HIV- and SIV-infected astrocytes, perhaps promoting apoptosis in these cells (He et al., 1997). Despite such demonstrations of functional versatility, the mechanisms by which Nef mediates these and other potential functions are not well defined. Nef is expressed in two forms in the infected cell: a membrane-associated, myristoylated form and a nonmyristoylated form (Guy et al., 1987). The myristoylated form is required for virion incorporation of Nef, as well as Nef-mediated enhancement of viral infectivity and downregulation of CD4 (Flaherty et al., 1997 ; Harris and Neil, 1994 ; Pandori et al., 1996 ). Nef also contains multiple potential phosphorylation sites (Guy et al., 1987), is itself regulated by phosphorylation (Bandres et al., 1994; Luo et al., 1997) and can be phosphorylated in vitro by protein kinase C (PKC) and lck (Bodeus et al., 1995; Coates et al., 1997; Coates and Harries, 1995; Collette et al., 1995; Guy et al., 1987, 1990; Luo et al., 1997). Moreover, Nef has been shown to alter signaling pathways (De and Marsh, 1994; Graziani et al., 1996; Greenway et al., 1995; Iafrate et al., 1997; Luria et al., 1991; Niederman et al., 1992; Richard et al., 1996; Romero et al., 1998) and modulate the activities of a number of cellular kinases, including lck (Collette et al., 1995; Greenway et al., 1996), hck (Briggs et al., 1997; Moarefi et al., 1997), mitogen-activated protein kinase (MAPK) (Greenway et al., 1996), PKC-u (Smith et al., 1996), a p21-activated kinase (PAK)related kinase (Nunn and Marsh, 1996; Sawai et al., 1994), and an unidentified serine kinase present in the Nef-associated kinase complex (NAKC; Baur et al., 1997). Association with the NAKC and possibly the PAK-related kinase correlates with Nef-mediated enhancement of viral infectivity (Baur et al., 1997; Luo and Garcia, 1996). Further, mutations of R137 and R138 in SIVmac239 nef, which inhibit the association of Nef with the PAK-related kinase, revert after inoculation of SIVmac239 into rhesus macaques (Sawai et al., 1996), suggesting the importance of this kinase association for disease progression in vivo, at least with respect to virus isolated from the peripheral blood of infected animals. We have long been interested in studying the mechanisms or processes leading to the development of SIVinduced CNS disease. In doing so, we described the isolation of a neurovirulent macrophage-tropic clone, SIV/17E-Fr, and reported that the Nef protein of this clone

differs in sequence from the parental lymphocyte-tropic clone SIVmac239 Nef (Flaherty et al., 1997). In light of recent publications describing Nef-associated kinase activities and the fact that the expression of certain kinases and hence, certain signaling pathways, is cell type specific (Hardie and Hanks, 1995), we considered the possibility that Nef proteins derived from neurovirulent viruses and classically lymphocyte-tropic viruses may selectively associate with different cellular kinases, with the idea that kinase associations may evolve during in vivo passage, but those important for viral growth in certain tissues (i.e., the CNS or periphery) may be conserved. In this report, we characterize a novel serine kinase activity associated with SIV/17E-Fr Nef that phosphorylates Nef and is distinct from both the PAK-related kinase and NAKC. We also demonstrate that SIV/17E-Fr Nef is phosphorylated in SIV-infected cells. RESULTS SIV/17E-Fr and SIVmac239 Nef proteins exhibit distinct in vitro kinase assay profiles in SIV-infected primary rhesus macaque macrophages To compare the kinase activities associated with distinct Nef proteins, we performed in vitro kinase assays on Nef derived from SIV/17E-Fr and another clone, SIV/ Fr-2. SIV/Fr-2 is identical to SIV/17E-Fr except that the SIV/17E-Fr nef has been replaced with SIVmac239 nef open, resulting in five amino acid coding differences (Flaherty et al., 1997). Like SIV/17E-Fr, SIV/Fr-2 is macrophage tropic and thus grows well in primary rhesus macaque macrophages. Nef immunoprecipitates (Nef IPs), prepared from macrophages infected with SIV/ 17E-Fr or SIV/Fr-2 for 5 days, were compared in in vitro kinase assays. SIVmac239 Nef (present in SIV/Fr-2) associated with phosphoprotein (pp)62 and pp97 (Fig. 1A, top arrowheads), as shown in previous reports (Nunn and Marsh, 1996; Sawai et al., 1994) that subsequently identified pp62 as a PAK-related kinase. A third phosphoprotein, pp35, also associated with the SIV/Fr-2 Nef but invariably to a lesser extent (light band, Fig. 1A, lowest arrowhead). This kinase activity likely represents a serine kinase activity of NAKC, recently shown to phosphorylate SIVmac239 Nef (MW ;35 kDa) in a complex with PAK (Baur et al., 1997). In contrast to SIV/Fr-2 Nef, SIV/17E-Fr Nef associated only with pp35 and not with pp62 or pp97. To ensure that the observed differences in Nef-associated kinase activities did not simply reflect differences in levels of Nef due to variability in viral replication or sequence-specific differences in IP efficiency, we transfected CEM3174 cells with infectious SIV/17E-Fr or SIV/Fr-2 DNA and prepared whole-cell lysates when both cultures had reached high RT levels and thus were most likely to express the same amounts of Nef protein. Western blot analysis (Fig. 1B, bottom) confirmed that comparable amounts of Nef were present

KINASE ACTIVITY OF Nef FROM NEUROVIRULENT SIV

FIG. 1. (A) In vitro kinase assay of Nef immunoprecipitates prepared from primary rhesus macaque macrophages infected for 5 days with SIV/17E-Fr or SIV/Fr-2, as described in Materials and Methods. RT levels (cpm/ml) at the time of cells lysis were 16953 (SIV/17E-Fr) and 8276 (SIV/Fr-2). (B, Top) In vitro kinase assay of Nef immunoprecipitates prepared from CEM3174 cells transfected with SIV/17E-Fr or SIV/Fr-2 DNA, as described in Materials and Methods. RT levels (in cpm/ml) at the time of cells lysis were 219247 (SIV/17E-Fr) and 254144 (SIV/Fr-2). (Bottom) Western blot analysis of Nef immunoprecipitates prepared in parallel from the same lysates used in the in vitro kinase assays shown in the top. Results are representative of at least three independent experiments performed on Nef immunoprecipitates prepared from SIVinfected rhesus macaque macrophages and transfected CEM3174 or U937 cells. The computer-generated images were produced using Adobe Photoshop 4.0 and Adobe Illustrator 6.0.

167

Nef and thus potentially just below the level of detection in assays of SIV/Fr-2 (2myr) Nef. In marked contrast, increased levels of pp35 associated with SIV/17E-Fr (2myr) Nef, demonstrating for the first time that a nonmyristoylated form of Nef favors a particular kinase association in infected cells. Comparable levels of Nef were immunoprecipitated from [35S]methionine-labeled lysates prepared in parallel from the same cultures of the myristoylation mutants (not shown). We consistently observed this disparate effect of Nef myristoylation on pp35 levels in in vitro kinase assays prepared from SIV-infected U937 cells, primary rhesus macaque macrophages, and CEM3174 cells (not shown), which suggested that the SIV/17E-Fr Nef/pp35 kinase and SIV/Fr-2 Nef/pp35 kinase (of NAKC) may be different. The specificity of the kinase assays for Nef was confirmed by the lack of specific phosphoproteins in kinase assays of SIV/Dnef IPs; all bands except pp97, pp65, and pp35 appeared in SIV/Dnef IPs after longer film exposure (not shown). SIV/17E-Fr Nef-associated kinase can utilize GTP and is inhibited by staurosporine Although ATP is the preferred phosphodonor for most kinases, some kinases can utilize both ATP and GTP. To determine whether the p35-kinase(s), associating with

in Nef IPs prepared from the same lysates where parallel Nef IPs exhibited distinct in vitro kinase assay patterns (Fig. 1B, top). Myristoylation mutants of SIV/17E-Fr Nef exhibit increased association with pp35 Both SIV and HIV Nef proteins exist in two forms: a membrane-associated, myristoylated form and a nonmyristoylated (2myr) form (Guy et al., 1987). To determine whether the association of SIV/17E-Fr Nef with pp35 is dependent on the myristoylation of Nef, infectious clones were prepared that contained mutations of the amino-terminal glycine residue required for Nef myristoylation (Bhatnagar and Gordon, 1997). In vitro kinase assays were performed with Nef IPs prepared (at high RT levels) from U937 cells transfected with the infectious DNA of SIV/Fr-2, SIV/17E-Fr, myristoylation mutants SIV/ 17E-Fr (2myr) and SIV/Fr-2 (2myr), and SIV/Dnef, a clone containing a 182-bp deletion plus a mutation at the initiating methionine of SIV/17E-Fr nef (Flaherty et al., 1997). Consistent with a previous report (Sawai et al., 1995), the ability of SIVmac239 Nef, present in clone Fr-2, to associate with pp62 was largely dependent on Nef-myristoylation (Fig. 2), as decreased levels of pp62 were detected in cells transfected with DNA from SIV/Fr-2 (2myr). The ability of SIV/Fr-2 Nef to associate with pp97 and pp35 was also dependent on myristoylated Nef but possibly to a lesser extent for pp35 because this band was consistently lighter than pp62 and pp97 in assays of SIV/Fr-2

FIG. 2. In vitro kinase assays of Nef immunoprecipitates prepared from U937 cells transfected with DNA from each of the indicated SIV recombinant clones, as described in Materials and Methods. Nef myristoylation mutants are designated (2myr). The RT levels (cpm/ml) at the time of lysis were: 17E-Fr, 190380; 17E-Fr (2myr), 75798; Fr-2, 153280; Fr-2 (2myr), 96051; Dnef, 85535. The results are representative of at least three independent experiments performed on Nef immunoprecipitates from transfected CEM3174 cells or U937 cells. The computer-generated image of the scanned autoradiograph was produced using Adobe Photoshop 4.0 and Adobe Illustrator 6.0.

168

BARBER ET AL.

kinase inhibitor staurosporine, the tyrosine kinase inhibitor genistein, and the vehicle control DMSO. The kinase activity associated with SIV/17E-Fr Nef was inhibited only by staurosporine, as indicated by the loss of pp35 (Fig. 3B), suggesting that this kinase is a serine/threonine kinase. SIV/Fr-2-associated pp35 was also inhibited by staurosporine, whereas the intensities of nonspecific bands were essentially unaffected by the presence of staurosporine (not shown). Inhibition by staurosporine is a strong indicator that the novel kinase associated with SIV/17E-Fr Nef is a serine/threonine kinase. To identify the amino acid(s) phosphorylated by the SIV/17E-Fr Nef-associated kinase, phosphoamino acid analysis was performed on pp35 excised from a rehydrated in vitro kinase gel. We analyzed pp35 associated with SIV/17E-Fr Nef immunoprecipitated from primary rhesus macaque macrophages infected with SIV/17E-Fr and from CEM3174 and U937 cells transfected with infectious SIV/17E-Fr DNA. The results were identical in all cases; pp35 was phosphorylated exclusively on serine(s) (Fig. 3C). SIV/17E-Fr Nef-associated kinase phoshorylates Nef

FIG. 3. In vitro kinase assays of Nef immunoprecipitates prepared from CEM3174 cells transfected with SIV/17E-Fr or SIV/Fr-2 DNA, as described in Materials and Methods, in the presence of (A) 200 mM [g-32P] ATP (3000 Ci/mol) or 200 mM [g-32P] GTP (6000 Ci/mol) or (B) 10 mM staurosporine (STAUR.), 100 mM genistein (GEN.), or DMSO, used at the highest concentration present in either inhibitor (1%). RTs (cpm/ ml) at the time of lysis were 108167 for SIV/17E-Fr and 112866 for SIV/Fr-2 (A) and 80754 for SIV/17E-Fr and 80383 for SIV/Fr-2 (B). Results are representative of at least three similar independent experiments. (C) Phosphoamino acid analysis of pp35 extracted from in vitro kinase gels of Nef immunoprecipitates prepared from CEM3174 cells transfected with SIV/17E-Fr DNA, as described in Materials and Methods. The circles corresponding to Ser (serine), Thr (threonine), and Tyr (tyrosine) represent the positions of the phosphoamino acid standards. Results are representative of three independent analyses of pp35 derived from Nef in vitro kinase assays of infected primary rhesus macaque macrophages and transfected CEM3174 or U937 cells. The computer-generated images of the scanned autoradiographs were produced using Adobe Photoshop 4.0 and Adobe Illustrator 6.0.

the SIV/17E-Fr and SIV/Fr-2 Nef proteins, could utilize GTP and ATP, we compared in vitro kinase assays in the presence of [g-32P]ATP or [g-32P]GTP (Fig. 3A). Although pp35 was readily detectable in ATP-based assays of both Nef proteins, pp35 was only detectable in GTP-based assays of the SIV/17E-Fr Nef protein and not the SIV/Fr-2 Nef protein, confirming our suspicions that the SIV/ 17E-Fr and SIV/Fr-2 Nef-associated pp35 kinases were distinct. Even after prolonged exposure, pp35 was never observed in association with SIV/Fr-2 Nef in the GTPbased assays. To further characterize the SIV/17E-Fr Nef-associated kinase activity, in vitro kinase assays were performed in the presence of the serine/threonine

Because SIV/17E-Fr-associated pp35 comigrates with [35S]methionine-labeled Nef in SDS±PAGE (not shown) and with the pp35 kinase assay product associated with SIVmac239 Nef (present in SIV/Fr-2; Figs. 1 and 2), which likely represents NAKC-phosphorylated Nef (Baur et al., 1997), we examined the possibility that the SIV/17E-Frassociated pp35 also represents phosphorylated Nef. To this end, we utilized a combination of two-dimensional protein separation techniques and Western blot analysis. First, Nef IPs prepared from [35S]methionine-labeled CEM3174 cells transfected with the DNA of SIV/Fr-2 or SIV/17E-Fr (2myr) were separated in one dimension using standard SDS±PAGE (Fig. 4A). The only Nef-specific protein band (not present in the Mock-transfected cells), of ;35 kDa was excised from the polyacrylamide gel and separated in the second dimension by isoelectric focusing (Fig. 4C). The [35S]-label from both SIV/Fr-2 and SIV/17E-Fr p35 bands concentrated essentially into single spots (see Materials and Methods) that migrated to pIs between 5.1 and 6.0. The different pIs likely reflect the amino acid differences between the two Nef proteins and are consistent with the pIs of both SIV/Fr-2 Nef (5.35) and SIV/17E-Fr Nef (6) as predicted by the computer analysis program, MacVector, in the absence of posttranslational modifications. These data indicate that the predominant [35S]methionine-labeled 35-kDa protein present in Nef IPs is Nef itself. Supporting this conclusion is our observation that [32P]-labeled SIV/17E-Fr Nef (cleaved and purified from a GST±Nef fusion protein) migrates to the same region in IEF gels (not shown). We next performed two-dimensional analysis of pp35 derived from in vitro kinase assays of Nef IPs prepared from

KINASE ACTIVITY OF Nef FROM NEUROVIRULENT SIV

169

parallel but unlabeled cultures of transfected CEM3174 cells. Separation of pp35 in the first dimension showed the typical difference in pp35 intensity seen when comparing SIV/Fr-2 and SIV/17E-Fr (2myr) Nef kinase assays (Fig. 4B). Separation of pp35 in the second dimension (Fig. 4D) again resulted in the concentration of the radioactive label, [32P], into spots that migrated to the same relative positions in the IEF gels (i.e., between 5.1 and 6.0) and whose difference in [32P] intensity reflected the difference observed in pp35 after separation in the first dimension (Fig. 4B). These results strongly suggested that pp35 was phosphorylated Nef. To confirm this result using a more direct approach, the proteins in the IEF gel were transferred to Immobilon and subjected to Western blot analysis (using Nef antiserum) as described in Materials and Methods. After ECL detection, in which the blots were exposed to film for 1±3 s (Fig. 4E), two spots of equal intensity appeared that were superimposable over the spots observed in the IEF gel (Fig. 4D). This result is consistent with our findings that equal levels of Nef are present in transfected CEM3174 cells at peak virus production and demonstrates unequivocally that pp35 is phosphorylated Nef. SIV/17E-Fr Nef is phosphorylated in SIV-infected cells

FIG. 4. (A) SDS±PAGE separation of Nef immunoprecipitates prepared from [35S]methionine-labeled CEM3174 cells transfected with SIV/17E-Fr (2myr) DNA or SIV/Fr-2 DNA, as described in Materials and Methods. (B) In vitro kinase assays of Nef immunoprecipitates prepared from parallel, but unlabeledm cultures of CEM'174 cells transfected with SIV/Fr-2 DNA or SIV/17E-Fr (2myr) DNA, as described in Materials and Methods. (C) Isoelectric focusing of p35 band excised from the gel shown above in A. (D) Isoelectric focusing of pp35 excised from the gel shown above in B. (E) Western blot analysis of the gel shown in D. Results are representative of three independent twodimensional analyses of pp35. The computer-generated images of the scanned autoradiographs were produced using Adobe Photoshop 4.0 and Adobe Illustrator 6.0.

Thus far, our studies have revealed a novel serine kinase activity present in IPs of SIV/17E-Fr Nef that both phosphorylates Nef and is distinct from the two previously described serine kinase activities associated with SIVmac239 Nef in in vitro kinase assays. As the first step in assessing the potential relevance of this in vitro observation to events occurring during viral replication, we examined the phosphorylation state of Nef in infected CEM3174 cells. Nef IPs prepared from 32P-orthophosphate-labeled CEM3174 cells transfected with the DNA of SIV/Fr-2 or SIV/17E-Fr were separated using standard SDS±PAGE and analyzed by autoradiography (Fig. 5). The ;35-kDa phosphoprotein was detected only in Nef IPs from SIV/17E-Fr lysates, supporting the possibility that our in vitro observations may reflect the activity(ies) of Nef in SIV-infected cells. There were no other specific phosphoproteins detected after prolonged exposure, whereas in vitro kinase assays of SIV/17E-Fr and SIV/ Fr-2, Nef IPs prepared from parallel unlabeled lysates exhibited the same characteristic phosphoprotein patterns described above in Figs. 1 and 2 (not shown). These results indicate that the major in vitro kinase products associated with SIVmac239 Nef, namely, pp62, pp97, and pp35, may not represent Nef-associated phosphoproteins during SIV infection, that their level of phosphorylation during infection may be below the level of detection, or that in vivo kinase substrates may differ from those in in vitro kinase assays.

170

BARBER ET AL.

FIG. 5. SDS±PAGE separation of Nef immunoprecipitates prepared from [32P]orthophosphate-labeled CEM3174 cells transfected with SIV/ 17E-Fr DNA or SIV/Fr-2 DNA, as described in Materials and Methods. RTs (in cpm/ml) at the time of labeling were 154446 for SIV/17E-Fr and 186956 for SIV/Fr-2. Results are representative of at least two similar independent experiments. The computer-generated images of the scanned autoradiographs were produced using Adobe Photoshop 4.0 and Adobe Illustrator 6.0

DISCUSSION Although nef is required for viral replication and pathogenesis in vivo (Daniel et al., 1992; Jamieson et al., 1994; Kestler et al., 1991), the various mechanisms of Nef functions during SIV or HIV in vivo infections are not known and still under investigation. Nevertheless, many versatile functions of Nef have been described. For example, Nef plays a role in the down-regulation of CD4 and MHC I molecules, in part through accelerated endocytosis and the promotion of clathrin-coated pit formation (Aiken et al., 1994; Foti et al., 1997; Le Gall et al., 1998; Rhee and Marsh, 1994; Schwartz et al., 1995, 1996). It is thought that down-regulation of CD4, the primary receptor for SIV and HIV, may serve to modulate CD4/TCRmediated signaling and discourage viral superinfection (Aiken et al., 1994; Benson et al., 1993; Rhee and Marsh, 1994). Dependent on the viral entry pathway, Nef also enhances virion infectivity (Aiken, 1997; Luo et al., 1998), which correlates with its ability to stimulate proviral DNA synthesis (Aiken and Trono, 1995; Chowers et al., 1995) and to associate with NAKC and possibly with a PAKrelated kinase (Baur et al., 1997; Chowers et al., 1994; Flaherty, Barber, and Clements, 1997; Luo and Garcia, 1996; Miller et al., 1994). Nef association with the activated PAK-related kinase is viral-isolate dependent, as shown here and by Luo and Garcia (1996). In the present study, we compared in vitro kinase assays of Nef IPs prepared from rhesus macaque macrophages infected with the neurovirulent clone SIV/17E-Fr or SIV/Fr-2, a clone molecularly identical to SIV/17E-Fr except that it contains SIVmac239 nef open (Flaherty et al., 1997). Our results indicate that SIV17E-Fr Nef associates with a novel kinase activity that results in the phosphorylation of Nef itself. Nef phosphorylation can be achieved in the presence of ATP or GTP (Fig. 3A), is inhibited by staurosporine (Fig. 3B), and occurs exclusively on serine(s) (Fig. 3C). We also found SIV/17E-Fr Nef to be phosphorylated in SIV-infected cells and are currently mapping the relevant phosphorylation sites in Nef derived from 32P-

labeled SIV-infected cells and in vitro kinase assays. Of particular interest was the enhanced kinase association or activation by nonmyristoylated forms of SIV/17E-Fr Nef (Fig. 2) because most of the previously described functions of Nef are dependent on the membrane-associated, myristoylated form of Nef (i.e., CD4 down-regulation, enhanced viral infectivity, virion incorporation, and association with the PAK-related kinase; Chowers et al., 1994; Flaherty, Barber, and Clements, 1997; Harris and Neil, 1994; Pandori et al., 1996; Sawai et al., 1995; Bukovsky et al., 1997). This characteristic may prove helpful in determining the identity of the kinase and may focus future attention on examining the effect(s) of nonmyristoylated forms of Nef present in the cytosol and nucleus (Franchini et al., 1986; Kaminchik et al., 1991; Kienzle et al., 1992; Ranki et al., 1994; Yu and Felsted, 1992). Previous studies have shown that SIVmac239 Nef (present in clone SIV/Fr-2) associates with pp62, pp97, and pp35 in in vitro kinase assays (Baur et al., 1997; Nunn and Marsh, 1996; Sawai et al., 1994). Our results both confirm and extend these observations to include SIV-infected rhesus macaque macrophages. It is thought that pp62 is a PAK-related kinase (Nunn and Marsh, 1996; Sawai et al., 1996), that pp97 represents a PAKkinase substrate that coprecipitates with the Nef/kinase complex (Nunn and Marsh, 1996), and that pp35 is Nef phosphorylated by an unidentified serine kinase of NAKC (Baur et al., 1997), henceforth referred to as SK-NAKC. Of note, SIV/17E-Fr Nef did not associate with the activated PAK-related kinase, nor did we observe any active srcrelated kinases in association with either Nef protein; src-related kinases autophosphorylate and thus result in tyrosine phosphoproteins in the 55- to 62-kDa range (Eiseman and Bolen, 1990). Our data also corroborate the findings of Baur et al. (1997) and demonstrate for the first time in the context of natural infection that in vitro SIVmac239 Nef (present in SIV/Fr-2) is phosphorylated by at least one of the Nef-associated kinase(s) (Fig. 4). Two experimental results appear to distinguish the SK-NAKC activity associated with SIV/Fr-2 Nef from the kinase activity associated with SIV/17E-Fr Nef. First, SIV/ 17E-Fr (2myr) Nef, but not SIV/Fr-2 (2myr) Nef, exhibited enhanced kinase association/activity (Fig. 2). Second, the kinase activity associated with SIV/17E-Fr Nef, but not SK-NAKC, was detectable in GTP-based in vitro kinase assays (Fig. 3). We have also confirmed that enhanced kinase activity occurs with SIV/17E-Fr (2myr) Nef in the GTP-based assays (not shown). These findings do not, however, rule out the possibility that SIV/17E-Fr Nef associates with both the novel kinase and SK-NAKC. We are currently designing experiments to address this possibility. Although we have identified pp35 as phosphorylated Nef, the identity of the kinase remains to be determined. We are presently considering the following possibilities: (1) both SIVmac239 Nef and SIV/17E-Fr Nef autophos-

KINASE ACTIVITY OF Nef FROM NEUROVIRULENT SIV

phorylate, but to different extents; (2) SIV/17E-Fr Nef associates with a kinase that does not autophosphorylate and is in such low abundance that it remains undetectable in 35S-based IP analyses; and (3) both Nefs associate with a kinase of low abundance that does not autophosphorylate, but only SIV/17E-Fr Nef can serve as a substrate for that kinase. Indeed, among the amino acid differences between SIV/Fr-2 Nef and SIV/17E-Fr Nef (P12-S, E75-K, E93-Q, M119-I, and E150-K, respectively, for Fr-2 - 17E-Fr) is the presence of a serine in SIV/17E-Fr Nef that does not exist in SIV/Fr-2 Nef. ATPase and GTPase functions have been described for SIV and HIV Nef (Guy et al., 1987; Nebrada et al., 1991); however, the ability of Nef to autophosphorylate has been controversial in the literature (Kaminchik et al., 1990; Sawai et al., 1994) and not actively pursued in the field. Thus, one of the two latter possibilities is more likely. Two previously reported Nef-associated serine/ threonine kinases are MAPK (Greenway et al., 1996) and PKCu (Smith et al., 1996). It is doubtful that MAPK is the SIV/17E-Fr Nef-associated kinase because MAPK activity is inhibited by HIV Nef (Greenway et al., 1996) and MAPK phosphorylates myelin basic protein (mbp), whereas the novel kinase does not (data not shown). We are currently examining PKCu, which is expressed in hematopoetic cells, although primarily lymphocytes (Smith et al., 1996), but of yet have no evidence to implicate the involvement of this kinase. In conclusion, our study characterized a novel kinase activity associated with the Nef protein of the macrophage-tropic, neurovirulent clone, SIV/17E-Fr. The existence of at least two identified active Nef-associated serine/threonine kinases (PAK, PKCu) suggests that the ability of Nef to associate with kinases may be intrinsically important for its complete function. Uniquely, the novel kinase preferentially associates with nonmyristoylated Nef, suggesting a novel function for this Nef form, perhaps in keeping with its different subcellular distribution. The identification of this kinase should help to define this function of Nef. In addition, the distinct Nefassociated kinase assay patterns described above for SIV/Fr-2 and SIV/17E-Fr indicate that at least one of the five amino acid differences between these two Nef proteins is crucial for detectable association with activated PAK, SK-NAKC, and/or the novel kinase. The relevant amino acids (described above) are distinct from the R-R and PXXP motifs previously demonstrated to be important for Nef/PAK interactions (Lang et al., 1997; Sawai et al., 1995) but may overlap the N-terminal region required for Nef/SK-NAKC interaction (Baur et al., 1997). Further, because SIV/17E-Fr nef is derived from SIV/17E-Br, a neurovirulent virus obtained by in vivo passage of SIVmac239, our results suggest that although there may be selective pressure toward Nef proteins that associate with active PAK in cells isolated from the peripheral blood of SIV-infected macaques (Sawai et al., 1996),

171

there is no absolute pressure to maintain this association during infection of the central nervous system (CNS). This conclusion questions the requirement of PAK-associated Nefs for all SIV-induced pathology and thus supports the recent findings of Lang et al. (1997), which demonstrate that macaques inoculated with SIV, harboring PXXP mutations in Nef that abrogate Nef/PAK association, develop AIDS despite a low frequency of nef revertants. Intriguingly, we have shown that Nef-associated kinase activity is also packaged into SIV/17E-Fr virions despite the absence of Nef-associated PAK and SKNAKC in SIV/Fr-2 virions, supporting further a unique role for this SIV/17E-Fr Nef (Flaherty et al., 1997). In the same study, we also determined that the infectivity of SIV/ 17E-Fr virions is only slightly less than SIV/Fr-2 virions (12% vs 17%, respectively, compared with 2% and 4% for the nonmyristoylated Nefs, respectively, and 2% for SIV/ 17E-FrDnef), suggesting that the novel kinase may also modulate viral infectivity. We are currently addressing this possibility and evaluating the replicative ability of SIV/17E-Fr mutants unable to associate with this kinase. To assess the relevance of this kinase to CNS disease, it is of considerable interest to examine Nef-associated kinase activity in other neurotropic, as well as nonneurotropic, strains of SIV, and in infected cells of the CNS, primarily brain macrophages or microglia (Gabuzda et al., 1986; Gartner et al., 1986; Price et al., 1988), astrocytes (He et al., 1997; Ranki et al., 1995; Saito et al., 1994; Takahashi et al., 1996; Tornatore et al., 1994), and endothelial cells (Mankowski et al., 1994; Moses et al., 1993). Such results may help to define the critical virus±cell interactions required to support productive infection in the CNS leading to SIV- or HIV-induced neurological disease. MATERIALS AND METHODS Cell culture and metabolic labeling Macrophages derived from rhesus macaque PBMCs were prepared as described previously (Flaherty et al., 1997). CEM3174 cells were grown in RPMI media (Paragon Biotech Inc., Baltimore, MD) supplemented with 10% fetal bovine serum (Atlanta Biologicals, Norcross, GA), 50 mg/ml gentamicin (Life Technologies, Gaithersburg, MD), 2 mM pyruvate (Sigma, St. Louis, MO), and 2 mM glutamine (Life Technologies). For 35S-labeling, CEM3174 cells were incubated in methionine-free RPMI (Life Technologies) containing 2% fetal bovine serum, 2 mM pyruvate, and 2 mM glutamine for 1 h and then supplemented for 16 h with Trans35S-Label (ICN pharmaceuticals, Irvine, CA). For 32P-labeling, CEM3174 cells were incubated in phosphate-free RPMI (Life Technologies) containing 10% dialyzed fetal bovine serum (Life Technologies) for 1 h and then supplemented for 16 h

172

BARBER ET AL.

with 32P-orthophosphate (NEN Life Science Products, Boston, MA). Infection of rhesus macaque macrophages Monolayers of rhesus macaque macrophages were incubated in six-well plates with 100,000 cpm of reverse transcriptase (RT) activity of the SIV molecular clones for 8±16 h, after which the inoculum was removed and the macrophages were washed twice with media. Virus growth was monitored by analyzing supernatants for RT activity as described previously (Clabough et al., 1991). Transfection of SIV recombinant clones The construction of the molecular clones SIV/17E-Fr, SIV/Fr-2, SIV/17E-Fr(2myr), SIV/Fr-2(2myr), and SIV 17EFr(Dnef) have been described previously (Flaherty et al., 1997). CEM3174 or U937 cells were transfected with 12 mg of infectious DNA by electroporation, and the cell-free supernatants from the transfected CEM3174 cells were aliquoted and stored at 280°C for virus stocks. Immunoprecipitation, Western blotting, and in vitro kinase assays Transfected cells were pelleted by low-speed centrifugation, washed in ice-cold PBS containing 200 mM sodium orthovanadate (SOV, Aldrich Chemical, Milwaukee, WI), and lysed in ice-cold lysis buffer containing 50 mM Tris±HCl (pH 8.0), 0.5% Nonidet P-40 (ICN Biomedicals, Aurora, OH), 2 mM EDTA, 137 mM NaCl, 10% glycerol, 2 mg/ml aprotinin (Sigma), 100 mM leupeptin (Sigma), and 2 mM SOV. Whole-cell lysates were prepared from infected rhesus macaque macrophages in the same manner, except that the adherent macrophages were scraped into ice-cold PBS containing 200 mM SOV before low-speed centrifugation. Insoluble material was pelleted by high-speed centrifugation (;10,000g), and protein content was determined using the BioRad protein assay system (BioRad Laboratories, Hercules, CA). Equal amounts of lysate proteins were precleared by incubation with normal rabbit serum for 1 h at 4°C followed by incubation with Protein A±Sepharose 4 Fast Flow (Pharmacia, Piscataway, NY) for 1 h at 4°C. Polyclonal antiserum against SIVmac239 Nef was prepared in rabbits immunized with purified Nef protein (a gift from Casey Morrow, University of Alabama at Birmingham). The pre-cleared lysates were incubated first with polyclonal Nef antiserum overnight at 4°C and then with Protein A±Sepharose for 1 h at 4°C. Immunoprecipitates (IPs) prepared from radiolabeled cells or IPs to be used in Western blot analyses were washed four times with lysis buffer, solubilized in SDS±PAGE loading buffer (Laemmli, 1970), and heated at .90°C for 5 min before electrophoretic separation. Separated proteins were transfered to Immobilon (Millipore Corporation, Bedford, MA) using the BioRad semidry transfer apparatus (Bio-

Rad Laboratories) and incubated with anti-Nef antiserum followed by goat anti-rabbit horseradish peroxidase (HRP)-conjugated Ig (DAKO, Carpinteria, CA). Detection of immune complexes was achieved using enhanced chemiluminescence (Amersham, Arlington Heights, IL). The IPs to be used for in vitro kinase assays were washed three times with lysis buffer and once with kinase buffer (50 mM Tris±HCl, pH 8.0, 10 mM NaCl, 1% Triton X-100, 5 mM MgCl2, 5 mM MnCl2, with or without 10 mM staurosporine; Calbiochem, La Jolla, CA), 100 mM genistein (Life Technologies), or 1% DMSO, and resuspended for 15 min in kinase buffer containing 2 mM SOV and 200 mM [g-32P]ATP (3000 mCi/mmol, NEN Research Products). Immune complexes were washed with icecold lysis buffer and separated by 9±12% SDS±PAGE. Gels were fixed in 10% acetic acid±30% methanol for 15±30 min, dried, and exposed to film (NEN Life Science Products). Phosphoamino acid analysis After autoradiography, pp35 was excised from the rehydrated kinase gel, digested with trypsin (Worthington, Freehold, NJ) overnight, and subjected to acid hydrolysis for 1 h at 110°C as described previously (Van Der Geer et al., 1991). Samples were lyophilized and resuspended in electrophoresis buffer before separation (along with phosphoamino acid standards) on cellulose TLC plates (EM Science, Gibbstown, NJ) as described previously (Jelinek and Weber, 1993; Van Der Geer et al., 1991). Phosphoamino acid standards were localized with ninhydrin spray, and the phosphorylated amino acid(s) from pp35 were detected by autoradiography. Two-dimensional electrophoresis IPs were processed as described above for in vitro kinase analysis or prepared from 35S-labeled transfected CEM3174 cells. Proteins present in the IPs were separated in the first dimension by SDS±PAGE (10% polyacrylamide gel). Gels containing the Nef kinase samples were processed as described above, and the gels containing the 35S-labeled Nef IPs were fixed in 20% isopropanol± 10% acetic acid for 1 h, washed 3±4 times in deionized water, and emersed in 1 M sodium salicylate for 30 min before drying. After autoradiography, the bands corresponding to 35S-labeled p35 or 32P-labeled p35 were excised from the dried gels and washed in 20% ethanol for 10 min, 23 isoelectric focusing (IEF) solublization buffer (Novex, San Diego, CA) containing 20% ethanol for 10 min, and IEF cathode buffer (Novex) for 10 min. The gel slices were then loaded vertically into the lanes of IEF gels (Novex) along with IEF standards (BioRad), and the proteins were separated according to the manufacturer's specifications. Note that separation of bands that appear to be horizontal in the first dimension will appear as spots when loaded vertically and separated in the

KINASE ACTIVITY OF Nef FROM NEUROVIRULENT SIV

second dimension. After separation, the IEF gels containing 32P-labeled p35 were processed as described above for the kinase assay gels, and the IEF gels containing the 35S-labeled p35 were emersed in 1M sodium salicylate for 30 min before drying and autoradiography. After autoradiography, the IEF gels to be used in Western anaylses were rehydrated for 30 min in SDS±PAGE running buffer (0.25 M Tris, 1.9 M glycine, 0.35 M SDS) and then soaked for 15 min in transfer buffer (0.025 M Tris, 0.19 M glycine, 20% methanol) before transfer and probing with Nef antiserum as described above. ACKNOWLEDGMENTS The authors would like to thank Kevin Maughan, Debbie Hauer, Maryann Brooks, Rebecca Bernbaum, Alma Matas, and Dr. Barry J. Margulies for their contributions to this work and the rest of the retrovirus laboratory for helpful discussions. This work was supported by NIH Grants NS35751, NS32208, and NS07392. Scott M. Plafker's contribution to this work was performed in the laboratory in the laboratory of Dr. Wade Gibson (Department of Pharmacology, Johns Hopkins University School of Medicine, Baltimore, MD) in the Pharmacology and Molecular Sciences training program and was supported by USPHS Grant GM07626.

REFERENCES Aiken, C. (1997). Pseudotyping human immunodeficiency virus type 1 (HIV-1) by the glycoprotein of vesicular stomatitis virus targets HIV-1 entry to an endocytic pathway and suppresses both the requirement for Nef and the sensitivity to cyclosporin A. J. Virol. 71, 5871±5877. 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. Aiken, C., and Trono, D. (1995). Nef stimulates human immunodeficiency virus type 1 proviral DNA synthesis. J. Virol. 69, 5048±5056. Anderson, M. G., Hauer, D., Sharma, D. P., Joag, S. V., Narayan, O., Zink, M. C., and Clements, J. E. (1993). Analysis of envelope changes acquired by SIVmac239 during neuroadaption in rhesus macaques. Virology 195, 616±626. Bandres, J. C., Luria, S., and Ratner, L. (1994). Regulation of human immunodeficiency virus Nef protein by phosphorylation. Virology 201, 157±161. Baur, A. S., Sass, G., Laffert, B., Willbold, D., Cheng-Meyer, C., and Peterlin, B. M. (1997). The N-terminus of Nef from HIV-1/SIV associates with a protein complex containing Lck and a serine kinase. Immunity 6, 283±291. Benson, R. E., Sanfridson, A., Ottinger, J. S., Doyle, C., and Cullen, B. R. (1993). Downregulation of cell-surface CD4 expression by simian immunodeficiency virus Nef prevents viral super infection. J. Exp. Med. 177, 1561±1566. Bhatnagar, R. S., and Gordon, J. I. (1997). Understanding covalent modifications of proteins by lipids: where cell biology and biophysics mingle. Trends Cell Biol. 7, 14±18. Bodeus, M., Marie-Cardine, A., Bougeret, B., Ramos-Morales, F., and Benarous, R. (1995). In vitro binding and phosphorylation of human immunodeficiency virus type 1 Nef protein by serine/threonine protein kinase. J. Gen. Virol. 76, 1337±1344. 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. In ªMethods in Enzymologyº (T. Hunter and B. M. Sefton, Eds.), Vol. 201, pp. 110±152. Academic Press, New York.

173

Briggs, S. D., Sharkey, M., Stevenson, M., and Smithgall, T. E. (1997). SH3-mediated Hck tyrosine kinase activation and fibroblast transformation by the Nef protein of HIV-1. J. Biol. Chem. 272, 17899±17902. Bukovsky, A. A., Dorfman, T., Weimann, A., and Gottlinger, H. G. (1997). Nef association with human immunodeficiency virus type 1 virions and cleavage by the viral protease. J. Virol. 71, 1013±1018. Cheng-Mayer, C., Weiss, C., Seto, D., and Levy, J. A. (1989). Isolates of HIV-1 from brain may constitute a special group of AIDS viruses. Proc. Natl. Acad. Sci. USA 86, 8575±8579. Chowers, M. Y., Pandori, M. W., Spina, C. A., Richman, D. D., and Guatelli, J. C. (1995). The growth advantage conferred by HIV-1 nef is determined at the level of viral DNA formation and is independent of CD4 downregulation. Virology 212, 451±457. Chowers, M. Y., Spina, C. A., Kwoh, T. J., Fitch, N. J. S., Richman, D. D., and Guatelli, J. C. (1994). Optimal infectivity in vitro of human immunodeficiency virus type 1 requires an intact nef gene. J. Virol. 68, 2906±2914. Clabough, D. L., Gebhard, D., Flaherty, M. T., Whetter, L. E., Perry, S. T., Coggins, L., and Fuller, F. J. (1991). Immune-mediated thrombocytopenia in horses infected with equine infectious anemia virus. J. Virol. 65, 6242±6251. Coates, K., Cooke, S. J., Mann, D. A., and Harris, M. P. G. (1997). Protein kinase C-mediated phosphorylation of HIV-1 Nef in human cell lines. J. Biol. Chem. 272, 12289±12294. Coates, K., and Harries, M. (1995). The human immunodeficiency virus type 1 Nef protein functions as a protein kinase C substrate in vitro. J. Gen. Virol. 76, 837±844. Collette, Y., Dutartre, H., Benziane, A., Ramos-Morales, F., Benarous, R., Harris, M., and Olive, D. (1995). Physical and functional interaction of Nef with Lck. J. Biol. Chem. 271, 6333±6341. Daniel, M. D., Kirchhoff, F., Czajak, C., Sehgal, P. K., and Desrosiers, R. C. (1992). Protective effects of a live attenuated SIV vaccine with a deletion in the nef gene. Science 258, 1938±1941. De, S. K., and Marsh, J. W. (1994). HIV-1 Nef inhibits a common pathway in NIH-3T3 cells. J. Biol. Chem. 269, 6656±6660. Eiseman, E., and Bolen, J. B. (1990). src-related tyrosine protein kinases as signaling components in hematopoietic cells. Cancer Cells 2, 303±310. Flaherty, M. T., Barber, S. A., and Clements, J. E. (1997). Neurovirulent simian immunodeficiency virus incorporates a Nef-associated kinase into virions. AIDS Res. Hum. Retrovirus 14, 163±170. Flaherty, M. T., Hauer, D. A., Mankowski, J. L., Zink, M. C., and Clements, J. E. (1997). Molecular and biological characterization of a neurovirulent molecular clone of SIV. J. Virol. 71, 5790±5798. Foti, M., Mangasarian, A., Piguet, V., Lew, D. P., Krause, K.-H., Trono, D., and Carpentier, J.-L. (1997). Nef-mediated clathrin-coated pit formation. J. Cell Biol. 139, 37±47. Franchini, G., Robert-Guroff, M., Ghrayeb, J., Chang, N. T., and WongStaal, F. (1986). Cytoplasmic localization of the HTLV-III 39 orf protein in cutlured T cells. Virology 155, 593±599. Gabuzda, D. H., Ho, D. D., de la Monte, S. M., Hirsch, M. S., Rota, T. R., and Sobel, R. A. (1986). Immunohistochemical identification of HTLVIII antigen in brains of patients with AIDS. Ann. Neurol. 20, 289±295. Gartner, S., Markovitz, P., Markovitz, D. M., Kaplan, M. H., Gallo, R. C., and Popovic, M. (1986). The role of mononuclear phagocytes in HTLV-III/LAV infection. Science 233, 215±219. Graziani, A., Galimi, F., Medico, E., Cottone, E., Gramaglia, D., Boccaccio, C., and Comoglio, P. M. (1996). The HIV-1 Nef protein interferes with phosphatidylinositol 3-kinase activation. J. Biol. Chem. 271, 6590±6593. Greenway, A., Azad, A., Mills, J., and McPhee, D. (1996). Human immunodeficiency virus type 1 Nef binds directly to Lck and Mitogenactivated protein kinase, inhibiting kinase activity. J. Virol. 70, 6701± 6708. Greenway, A. L., Azad, A., and McPhee, D. A. (1995). Human immunodeficiency virus type 1 Nef protein inhibits activation pathways in

174

BARBER ET AL.

peripheral blood mononuclear cells and T-cell lines. J. Virol. 1842-1850. Guy, B., Kieny, M. P., Riviere, Y., LePeuck, C., Dott, K., Girard, M., Montagnier, L., and Lecocq, J. P. (1987). HIV F/39 orf encodes a phosphorylated GTP-binding protein resemblying an oncogene product. Nature 330, 266±269. Guy, B., Riviere, Y., Dott, K., Regnault, A., and Kieny, M. P. (1990). Mutational analysis of the HIV nef protein. Virology 176, 413±425. Hardie, G., and Hanks, S. (1995). The protein kinase. In ªFacts Book: Protein-Serine Kinasesº (G. Hardie and S. Hanks, Eds.), Academic Press, San Diego, California. Harris, M. P. G., and Neil, J. C. (1994). Myristoylation-dependent binding of HIV-1 nef to CD4. J. Mol. Biol. 241, 136±142. He, J., DECastro, C. M., Vandenbark, G. R., Busciglio, J., and Gabuzda, D. (1997). Astrocyte apoptosis induced by HIV-1 transactivation of the c-kit protooncogene. Proc. Natl. Acad. Sci. USA 94, 3954±3959. Iafrate, A. J., Bronson, S., and Skowronski, J. (1997). Separable functions of Nef disrupt two aspects of T cell receptor machinery: CD4 expression and CD3 signaling. EMBO 16, 673±684. Jamieson, B. D., Aldrovandi, G. M., Planelles, V., Jowett, J. B. M., Gao, L., Bloch, L. M., Chen, I. S. Y., and Zack, J. A. (1994). Requirement of human immunodeficiency virus type 1 nef for in vivo replication and pathogenicity. J. Virol. 68, 3478±3485. Jelinek, T., and Weber, M. J. (1993). Optimization of the resolution of phosphoamino acids by one-dimensional thin-layer electrophoresis. BioTechniques 15, 628±629. Kaminchik, J., Bashan, N., Itach, A., Sarver, N., Gorecki, M., and Panet, A. (1991). Genetic characterization of human immunodeficiency virus type 1 nef gene products translated in vitro and expressed in mammalian cells. J. Virol. 65, 583±588. Kaminchik, J., Bashin, N., Pinchaasi, D., Amit, B., Sarver, N., Johnston, M. I., Fischer, M., Yavin, Z., Gorecki, M., and Panet, A. (1990). Expression and biochemical characterization of human immunodeficiency virus type 1 nef gene product. J. Virol. 64, 3447±3454. Kestler, H. W., Ringler, D. J., Mori, K., Panicali, D. L., Sehgal, P. D., Daniel, M. D., and Desrosiers, R. C. (1991). Importance of the nef gene for maintenance of high virus loads and for development of AIDS. Cell 65, 651±662. Kienzle, N., Bachmann, M., Muller, W. E. G., and Muller-Lanztch, N. (1992). Expression and cellular localization of the Nef protein from human immunodeficiency virus-type 1 in stably transfected B-cells. Arch. Virol. 124, 123±132. Laemmli, U. K. (1970). Cleavage of structural proteins duirng the assembly of the head of bacteriophage T4. Nature 227, 680±685. Lang, S. M., Iafrate, A. J., Stahl-Hennig, C., Kuhn, E. M., Nisslein, T., Kayp, F.-J., Haupt, M., Hunsmann, G., Skowronski, J., and Kirchoff, F. (1997). Association of simian immunodeficiency virus Nef with cellular serine/threonine kinases is dispensable for the development of AIDS in rhesus macaques. Nat. Med. 3, 860±865. Le Gall, S., Erdtmann, L., Benichou, S., Berlioz-Torrent, C., Liu, L., Benarous, R., Heard, J.-M., and Schwartz, O. (1998). Nef interacts with the u subunit of clathrin adaptor complexes and reveals a cryptic sorting signal in MHC I molecules. Immunity 8, 483±495. Letvin, N. L., and King, N. W. (1990). Immunologic and pathologic manifestations of the infection of rhesus monkeys with simian immunodeficiency virus of macaques. J. Acquir. Immune Defic. Syndr. 3, 1023±1040. Luo, T., Douglas, J. L., Livingston, R. L., and Garcia, J. V. (1998). Infectivity enhancement by HIV-1 Nef is dependent on the pathway of virus entry: Implications for HIV-based gene transfer systems. Virology 241, 224±233. Luo, T., Downing, J. R., and Garcia, J. V. (1997). Induction of phosphorylation of human immunodeficiency virus type 1 Nef and enhancement of CD4 downregulation by phorbol myristate acetate. J. Virol. 71, 2535±2539. Luo, T., and Garcia, J. V. (1996). The association of Nef with a cellular

serine/threonine kinase and its enhancement of infectivity are viral isolate dependent. J. Virol. 70, 6493±6496. Luria, S., Chambers, I., and Berg, P. (1991). Expression of the type 1 human immunodeficiency virus Nef protein prevents antigen receptor-mediated induction of interleukin 2 mRNA. Proc. Natl. Acad. Sci. USA 88, 5326±5330. Mankowski, J. L., Flaherty, M. T., Spelman, J. P., Hauer, D. A., Didier, P. J., Martin Amedee, A., Murphey-Corb, M., Kirstein, L. M., Munoz, A., Clements, J. E., and Zink, M. C. (1997). Pathogenesis of simian immunodeficiency virus encephalitis: Viral determinants of neurovirulence. J. Virol. 71, 6055±6060. Mankowski, J. L., Spelman, J. P., Ressetar, H. G., Strandberg, J. D., Laterra, J., Clements, J. E., and Zink, M. C. (1994). Neurovirulent simian immunodeficiency virus replicates productively in endothelial cells of the central nervous system in vivo and in vitro. J. Virol. 68, 8202±8208. Miller, M. D., Warmerdam, M. T., Gaston, I., Greene, W. C., and Feinberg, M. B. (1994). The human immunodeficiency virus-1 nef gene product: a positive factor for viral infection and replication in primary lymphocytes and macrophages. J. Exp. Med. 179, 101±113. Moarefi, I., LaFevre-Bernt, M., Sicheri, F., Huse, M., Lee, C.-H., Kuriyan, J., and Miller, W. T. (1997). Activation of the Src-family tyrosine kinase Hck by SH3 domain displacement. Nature 385, 650±653. Moses, A. V., Bloom, F. E., Pauza, C. D., and Nelson, J. A. (1993). Human immunodeficiency virus infection of human brain capillary endothelial cells occurs via a CD4/galacytosylceramide-independent mechanism. Proc. Natl. Acad. Sci. USA 90, 10474±10478. Nebrada, A. R., Bryan, T., Segade, F., Wingfield, P., Vankatesan, S., and Santos, E. (1991). Biochemical and biological comparison of HIV-1 nef and ras gene products. Virology 183, 151±159. Niederman, T. M. J., J.Victor, G., Hastings, W. R., Luria, S., and Ratner, L. (1992). Human immunodeficiency virus type 1 Nef protein inhibits NF-kB induction in human T cells. J. Virol. 66, 6213±6219. Nunn, M. F., and Marsh, J. W. (1996). Human immunodeficiency virus type 1 Nef associates with a member of the p21-activated kinase family. J. Virol. 70, 6157±6161. Pandori, M. W., Fitch, N. J. S., Criag, H. M., Richman, D. D., Spina, C. A., and Guatelli, J. C. (1996). Producer-cell modification of human immunodeficiency virus type 1: Nef is a virion protein. J. Virol. 70, 4283± 4290. Power, C., Kong, P. A., Crawford, T. O., Wessilingh, S., Glass, J. D., McArthur, J. C., and Trapp, B. D. (1993). Cerebral white matter changes in acquired immunodeficiency syndrome dementia: alterations of the blood-brain barrier. Ann. Neurol. 34, 339±350. Price, R. W., Brew, B., Sidtis, J. J., Rosenblum, M., Scheck, A. C., and Cleary, P. (1988). The brain in AIDS: central nervous system HIV-1 infection and AIDS dementia complex. Science 239, 586±592. Price, R. W., Brew, B. J., and Rosenblum, M. (1990). The AIDS dementia complex and HIV-1 brain infection: a pathogenic model of virusimmune interaction. Immunol. Mechanism Neurol. Psychiatr. Dis. 269±289. Ranki, A., Lagerstedt, A., Ovod, V., Aavik, E., and Krohn, K. J. E. (1994). Expression kinetics and subcellular localization of HIV-1 regulatory proteins Nef, Tat and Rev in acutely and chronically infected lymphoid cell lines. Arch. Virol. 139, 365±378. Ranki, A., Nyberg, M., Ovod, V., Haltia, M., Elovaara, I., Raininko, R., Haapasalo, H., and Krohn, K. (1995). Abundant expression of HIV Nef and Rev proteins in brain astrocytes in vivo is associated with dementia. AIDS 9, 1001±1008. Rhee, S. S., and Marsh, J. W. (1994). Human immunodeficiency virus type 1 Nef-induced down modulation of CD4 is due to rapid internalization and degradation of surface CD4. J. Virol. 68, 5156±5163. Richard, A., Robichaud, G., Lapointe, R., Bourgoin, S., Darveau, A., and Poulin, L. (1996). Interference of HIV-1 Nef in the sphingomyelin transduction pathway activated by tumour necrosis factor-alpha in human glial cells. AIDS 11, F1±F7. Ringler, D. J., Hunt, R. D., Desrosiers, R. C., Daniel, M. D., Chalifoux, L. V.,

KINASE ACTIVITY OF Nef FROM NEUROVIRULENT SIV and King, N. W. (1988). Simian immunodeficiency virus-induced meningoencephalitis: natural history and retrospective study. Ann. Neurol. 23 (Suppl), S101±S107. Romero, I. A., Teixeira, A., Strosberg, A. D., Cazaubon, S., and Couraud, P.-O. (1998). The HIV-1 Nef protein inhibits extracellular signal-regulated kinase-dependent DNA synthesis in a human astrocytic cell line. J. Neurochem. 70, 778±785. Saito, Y., Sharer, L. R., Epstein, L. G., Michaels, J., Mintz, M., Louder, M., Golding, K., Cvetkovich, T. A., and Blumberg, B. M. (1994). Overexpressionof nef as a marker for restricted HIV-1 infection of astrocytes in postmortem pediatric central nervous system. Neurology 44, 474± 481. Sawai, E. T., Baur, A., Struble, H., Peterlin, B. M., Levy, J. A., and Cheng-Mayer, C. (1994). Human immunodeficiency virus type 1 Nef associates with a cellular serine kinase in T lymphocytes. Proc. Natl. Acad. Sci. USA 91, 1539±1543. Sawai, E. T., Baur, A. S., Peterlin, B. M., Levy, J. A., and Cheng-Mayer, C. (1995). A conserved domain and membrane targeting of Nef from HIV and SIV are required for association with a cellular serine kinase activity. J. Biol. Chem. 270, 15307±15314. Sawai, E. T., Khan, I. H., Montbriand, P. M., Cheng-Mayer, C., and Luciw, P. A. (1996). Activation of PAK by HIV and SIV Nef: Importance for AIDS in rhesus macaques. Curr. Biol. 6, 1519±1527. Schwartz, O., Dautry-Varsat, A., Goud, B., Marechal, V., Subtil, A., Heard, J.-M., and Danos, O. (1995). Human immunodeficiency virus type 1 Nef induces accumulation of CD4 in early endosomes. J. Virol. 69, 528±533. Schwartz, O., Marechal, V., Le Gall, S., Lemonnier, F., and Heard, J. M.

175

(1996). Endocytosis of major histocompatibility complex I molecules is induced by HIV-1 Nef protein. Nat. Med. 2, 338±342. Sharma, D. P., Zink, M. C., Anderson, M. G., Adams, R., Clements, J. E., Joag, S. V., and Narayan, O. (1992). Derivation of neurotropic SIV from exclusively lymphocyte-tropic parental virus: pathogenesis of infection in macaques. J. Virol. 66, 3550±3556. Sinclair, E., Barbosa, P., and Feinberg, M. B. (1997). The nef gene products of both simian and human immunodeficiency viruses enhance infectivity and are functionally interchangeable. J. Virol. 71, 3641±3651. Smith, B. L., Krushelnycky, B. W., Mochly-Rosen, D., and Berg, P. (1996). The HIV Nef protein associates with protein kinase C theta. J. Biol. Chem. 271, 16753±16757. Takahashi, K., Wesselingh, S. L., Griffin, D. E., McArthur, J. C., Johnson, R. T., and Glass, J. D. (1996). Localization of HIV-1 human brain using polymerase chain reaction/in situ hybridization and immunocytochemistry. Ann. Neurol. 39, 705±711. Tornatore, C., Chandra, R., Berger, J. R., and Major, E. O. (1994). HIV-1 infection of subcortical astrocytes in the pediatric central nervous system. Neurology 44, 481±487. Yu, G., and Felsted, R. L. (1992). Effect of myristoylation on p27nef subcellular distribution and suppression of HIV-LTR transcription. Virology 187, 46±55. Zink, M. C., Martin, A. A., Mankowski, J. L., Craig, L., Munoz, A., Didier, P., Carter, D. L., Murphy-Corb, M., and Clements, J. E. (1997). Pathogenesis of SIV encephalitis: selection and replication of neurovirulent SIV. Am. J. Pathol. 151, 793±803.