Regulation of Translation by Ribosome Shunting thr
http://www.100md.com
病菌学杂志 2005年第9期
Department of Microbiology, New York University School of Medicine, New York, New York 10016
ABSTRACT
Adenovirus simultaneously inhibits cap-dependent host cell mRNA translation while promoting the translation of its late viral mRNAs during infection. Studies previously demonstrated that tyrosine kinase activity plays a central role in the control of late adenovirus protein synthesis. The tyrosine kinase inhibitor genistein decreases late viral mRNA translation and prevents viral inhibition of cellular protein synthesis. Adenovirus protein 100k blocks cellular mRNA translation by disrupting the cap-initiation complex and promotes viral mRNA translation through an alternate mechanism known as ribosome shunting. 100k protein interaction with initiation factor eIF4G and the viral 5' noncoding region on viral late mRNAs, known as the tripartite leader, are both essential for ribosome shunting. We show that adenovirus protein 100k promotes ribosome shunting in a tyrosine phosphorylation-dependent manner. The primary sites of phosphorylated tyrosine on protein 100k were mapped and mutated, and two key sites are shown to be essential for protein 100k to promote ribosome shunting. Mutation of the two tyrosine phosphorylation sites in 100k protein does not impair interaction with initiation factor 4G, but it severely reduces association of 100k with tripartite leader mRNAs. 100k protein therefore promotes ribosome shunting and selective translation of viral mRNAs by binding specifically to the adenovirus tripartite leader in a phosphotyrosine-dependent manner.
INTRODUCTION
Human adenovirus (Ad) selectively translates viral late mRNAs while simultaneously suppressing translation of cellular mRNAs during the late phase of infection. Late Ad mRNAs are capped but are translated despite inhibition of cellular protein synthesis, because they possess a common 5' noncoding region (5' NCR) known as the tripartite leader (6, 7, 12, 14, 23). The tripartite leader confers on mRNAs the ability to be selectively translated during late Ad infection and during cell stress by an alternate means of initiation known as ribosome shunting (26, 27). The general mechanism for ribosome shunting in Ad involves loading of 40S ribosome subunits to the 5' end of the capped mRNA, possibly with limited scanning followed by direct translocation of 40S subunits to the downstream initiation codon, directed by "shunting elements" in the tripartite leader. In uninfected cells, the tripartite leader directs translation by conventional 5' scanning of 40S ribosome subunits and by 40S ribosome shunting at roughly equal levels. However, in late Ad-infected cells, when the cap initiation complex is altered (4), the tripartite leader directs translation solely by ribosome shunting (26, 27).
The protein 100k is a late adenovirus polypeptide encoded by the L4 transcription unit in very large amounts (15). 100k is the first late viral protein translated with the onset of the late phase of infection (3). Recent studies showed that 100k protein is largely responsibly for inhibition of cellular protein synthesis (25), and it is also involved in promoting translation on tripartite leader mRNAs by ribosome shunting (25). 100k binds the carboxyl terminus of eIF4G at or near the site normally occupied by the kinase Mnk1, which is responsible for phosphorylating cap binding protein eIF4E. 100k therefore competitively displaces Mnk1 from cap initiation complexes, which prevents eIF4E phosphorylation (4). Dephosphorylation of eIF4E is often associated with inhibition of cellular cap-dependent mRNA translation, although the mechanism is not understood. 100k protein also binds mRNAs in the cytoplasm (1, 20) and demonstrates a preference for late Ad tripartite leader mRNAs (25). The C terminus of 100k (amino acid 726 to 805) contains a general RNA-binding domain, whereas its middle region (345 to 726) contains a tripartite leader-specific mRNA binding domain (25). During late viral infection, the modified 100k cap initiation complex associates with higher specificity to mRNAs that contain the tripartite leader 5' NCR. The 100k-tripartite leader complex enhances association with the initiation factor 4G (eIF4G) and poly(A) binding protein, which is associated with increased translation by ribosome shunting on late viral mRNAs (25).
Ad protein 100k is a tyrosine- and serine-phosphorylated protein (10). In addition, Ad stimulates tyrosine kinase activity during the late phase of infection, which has been linked to inhibition of cellular protein synthesis (9). To investigate whether the ability of Ad 100k protein to promote ribosome shunting and/or inhibit cellular protein synthesis is regulated by tyrosine phosphorylation, we performed biochemical and mutational studies of 100k protein. We show that tyrosine phosphorylation does not affect 100k protein association with eIF4G. Rather, tyrosine dephosphorylation of 100k protein at sites Y365 and Y682 is shown to impair its ability to promote ribosome shunting on Ad late mRNAs by decreasing 100k-RNA interaction. In addition, tyrosine phosphorylation was found to promote ribosome shunting by enhancing preferential binding of 100k protein to tripartite leader mRNAs. 100k protein therefore links the requirement for tyrosine kinase activity during late Ad infection to translational control.
MATERIALS AND METHODS
Antisera and cells. Rabbit polyclonal antisera were raised against the C-terminal protein fragment of human eIF4GI (amino acids [aa] 1045 to 1560). Mouse monoclonal anti-phosphotyrosine antibody was provided by E. Skolnik (New York University), and mouse monoclonal anti-rabbit eIF4A was provided by W. Merrick (Case Western Reserve, Cleveland, Ohio). Other antisera were from commercial sources and included horseradish peroxidase-conjugated donkey anti-rabbit or sheep anti-mouse secondary antibodies (Amersham), mouse mononclonal anti-FLAG antibody (Sigma), and the enhanced chemiluminescence system (ECL; Amersham). 293 cells comprise a human embryonic kidney cell line that expresses the E1 region of Ad5. 293 cells were cultured in Dulbecco's modified Eagle medium (DMEM) containing 10% calf serum and 50 μg/ml gentamicin.
Plasmids. Plasmid pCMV-Ad?-gal (8), pCMV-CR3?-gal, pTL-Flag100kN726, pTL-Flag100kN345, pTL-100kY365F, pCMV-LUC, and pCMV-AdB202LUC (25) were described previously. pFLAG-CM2 was from Sigma. pIRES-EGFP was from Clontech. pTL-Flag100kY365FY682F (renamed Flag100kYYF) and pTL-Flag100kN726Y365FY368F (renamed 100kN726YYF) were constructed by inserting a BstEII-AscI 100k DNA fragment containing an aa 365Y-to-F mutation into BstEII-AscI-digested pTL-Flag100kY682F and pTL-Flag100kN726Y682F vectors, respectively. Plasmids pTL-Flag100kY682F and pTL-Flag100N726Y682F were derived from pTL-Flag100k and pTL-Flag100kN726 using a site-directed mutagenesis kit (Stratagene). The primers for cloning were 100kY682F (5'-end oligo; 5'-CGAAAGGGACGGGGGGTTTTCCTGGACCCCCAGTCC-3') and 100kY682F (3'-end oligo; 5'-GCCGGACTGGGGGTCCAGGAAAACCCCCCGTCCCTTTCG-3'). All of the plasmids were confirmed by DNA sequencing.
Mammalian cell transfection and luciferase assay. 293 cells were transfected with plasmid constructs using Lipofectamine Plus (Invitrogen) according to manufacturer protocols. Cells were harvested 36 h posttransfection and subjected to detergent lysis (0.5% NP-40, 50 mM HEPES, pH 7.0, 250 mM NaCl, 2 mM EDTA, 2 mM sodium orthovanadate, 25 mM glycerophosphate, 1 tablet of protease inhibitor [Roche] per 10 ml) at 4°C for 20 min, nuclei were pelleted by microcentrifuge centrifugation, and supernatant lysates were collected. Luciferase assays were performed as per manufacturer protocol (Promega).
Immunoprecipitation and immunoblot analysis. Equal amounts of lysates were incubated with 2 μg mouse monoclonal antibodies to FLAG (Sigma) for 1 h at 4°C or 20 μl rabbit polyclonal serum against the C-terminal fragment of eIF4G (4) and RNase A (20 μg/ml) overnight at 4°C. Either protein G or A agarose was added and incubated for 1 h at 4°C. Precipitates were washed four times with lysis buffer, boiled in sodium dodecyl sulfate (SDS)-sample buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 25% glycerol, 0.01% bromophenol blue), and analyzed by SDS-polyacrylamide gel electrophoresis (PAGE). Gels were electroblotted onto Immunobilon-P membrane (Millipore) and blocked overnight in blocking buffer (150 mM NaCl, 10 mM Tris-HCl, pH 8.0, 5% bovine serum albumin at 4°C). Primary antibodies used were mouse monoclonal anti-phosphotyrosine antibody (1:500), mouse monoclonal anti-FLAG antibody (1 μg/ml), and rabbit polyclonal serum against the C terminus of eIF4G (1:250). Studies used horseradish peroxidase-conjugated donkey anti-rabbit or sheep anti-mouse secondary antibodies or horseradish peroxidase-conjugated protein A (1:5,000) as secondary antibody. An enhanced chemiluminescence detection method (Amersham) was used, and the membrane was exposed to film.
Northern blot analysis. 293 cells were lysed directly into Trizol extraction reagent (Invitrogen). RNA samples were electrophoresed on a denaturing formaldehyde gel and transferred to a nylon membrane (Perkin Elmer Life Science, Inc.). Membranes were probed overnight at 68°C with [-32P]dCTP-labeled probes directed to the ?-galactosidase and GAPDH coding regions. Following washes, membranes were exposed to film at –70°C.
In vivo RNA-binding assay. 293 cells were lysed directly into detergent lysis buffer (0.5% NP-40, 50 mM HEPES, pH 7.0, 250 mM NaCl, 2 mM EDTA, 2 mM sodium orthovanadate, 25 mM glycerophosphate, 1 tablet of protease inhibitor [Roche] per 10 ml, and 40 U/ml Rnasin [Promega]) at 4°C for 20 min, nuclei were pelleted by centrifugation, and supernatant lysates were collected. Wild-type or mutant 100k protein was recovered from equal amounts of lysate by immunoprecipitation with anti-FLAG antibody using lysis buffer. mRNAs were extracted from the immunoprecipitated protein using Trizol extraction reagent. Semiquantitative reverse transcription-PCR (RT-PCR) amplification of immunoprecipitated ?-galactosidase (?-gal) mRNA was carried out as described previously (22).
RESULTS
The ability of Ad 100k protein to promote ribosome shunting is regulated by tyrosine phosphorylation. Previous studies showed that during Ad infection, genistein, a broad inhibitor of tyrosine kinases, can decrease late viral mRNA translation and prevent viral inhibition of host cell protein synthesis without significantly affecting short-term viral mRNA abundance or replication (9). These results suggested that tyrosine kinase activity may play a role in promoting late viral protein synthesis. Studies were carried out to determine whether 100k is a target of tyrosine kinases. 293 cells were cotransfected with vector alone or a vector expressing Flag-tagged wild-type 100k protein, and plasmids expressing a luciferase reporter mRNA containing a mutant Ad tripartite leader 5' NCR (B202) which has a strong hairpin structure inserted in the 3' end of the leader (Fig. 1A). The B202 5' NCR can only translate by ribosome shunting, as the hairpin blocks initiation by ribosome scanning (25, 27). In uninfected cells, translation by ribosome shunting accounts for roughly half the translation directed by the tripartite leader (26, 27). The B202 tripartite leader therefore translates at about half the level of the wild-type leader. The plasmid pIRES-EGFP was included as an internal control and expresses green fluorescent protein (GFP) directed by the encephalomyocarditis virus (EMCV) internal ribosome entry site (IRES). Cells were treated with genistein (300 μM), an inhibitor of tyrosine kinase activity (2), or the more specific Src kinase inhibitor PP2 (30 μM) for 4 h overnight (21), with mock-treated cells serving as a control. At 36 h posttransfection, lysates and total RNAs were prepared. Flag-100k protein was immunoprecipitated, resolved by SDS-PAGE, and examined by immunoblot analysis using antibody against Flag or phosphotyrosine (Fig. 1B). 100k immune complexes exhibited strong tyrosine phosphorylation, which was decreased significantly by treatment with genistein (approximately eightfold) and less so with PP2 (approximately two- to threefold) (Fig. 1B). These data are consistent with involvement of Src and other tyrosine kinases in 100k phosphorylation. Equal amounts of 100k were recovered in the immunoprecipitations, as shown.
100k was previously shown to bind tightly to the C-terminal end of eIF4G (4), which is important for inhibition of cellular protein synthesis (5) and promotion of ribosome shunting (25). To explore whether genistein inhibition of 100k protein tyrosine phosphorylation affects its interaction with eIF4G, cells were transfected with a plasmid expressing Flag-100k protein and treated with genistein, and then endogenous eIF4G was immunoprecipitated from equal amounts of cell lysates and proteins were detected by immunoblot analysis. Equal amounts of eIF4G were immunoprecipitated as shown. Genistein treatment had little, if any, ability to decrease association of 100k protein with eIF4G, while PP2 clearly had no effect (Fig. 1B). These data indicate that genistein-sensitive tyrosine phosphorylation of 100k protein has little effect on its interaction with eIF4G. However, analysis of luciferase synthesized by ribosome shunting from the B202 mRNA revealed an increase in ribosome shunting promoted by 100k protein on tripartite leader mRNAs (Fig. 1C) as shown previously (25). The ability of 100k to promote ribosome shunting was conversely decreased threefold by treatment with genistein (300 μm). In contrast, treatment of cells with PP2 (30 μM) had little, if any, effect (Fig. 1C).
These data suggest that tyrosine phosphorylation of 100k might play a role in its ability to promote ribosome shunting. These results also suggest that the key regulatory phosphotyrosines in 100k protein are predominantly phosphorylated by kinases other than Src, since PP2 specifically inhibits Src kinase activity (21) and had little effect on 100k ribosome shunting activity.
100k protein phosphotyrosine sites are located between aa 345 and aa 726. To determine whether 100k tyrosine phosphorylation directly regulates its activity in translational control, we first identified the major sites of tyrosine phosphorylation. 100k protein has 21 tyrosines, several of which are within strong consensus sites for phosphorylation. The regions in 100k that are most heavily tyrosine phosphorylated were preliminarily mapped by utilizing 100k protein truncation mutants 100kN726 and 100kN345 (Fig. 2A). Plasmids expressing wild-type or mutant Flag-100k proteins were transfected into 293 cells. At 32 h posttransfection, cells were treated with 150 μM genistein or with vehicle control for 4 h before lysis. Wild-type or mutant 100k protein was recovered from equal amounts of cell lysates by immunoprecipitation with anti-Flag antibody. 100k proteins were analyzed by SDS-PAGE and immunoblot analysis using anti-Flag and anti-phosphotyrosine (P-Tyr) antibodies (Fig. 2B). Mutant 100kN726 protein was found to be tyrosine phosphorylated to the same extent as wild-type 100k protein (Fig. 2B, lanes 2 and 3), whereas no tyrosine phosphorylation signal was detected in the 100kN345 protein (Fig. 2B, lanes 4 and 8). Genistein treatment significantly decreased the tyrosine phosphorylation level of wild-type and N726 100k proteins (approximately five- to ninefold) (Fig. 2B, lanes 6 and 7). These results suggest that the tyrosine phosphorylation sites in protein 100k are located between amino acids 345 and 726.
Tyrosine mutations at aa 365 and aa 682 impair the ability of 100k protein to promote ribosome shunting. Based on the tyrosine phosphorylation sites predicted by two different programs, Prosite (www.expasy.org/prosite/) and NetPhos2.0 (genome.cbs.dtu.dk/services/NetPhos/), we mutated tyrosine (Y) sites at positions 365 and 682 to phenylalanine (F), as described in Materials and Methods. To investigate the role of tyrosine phosphorylation of 100k in promoting ribosome shunting, 293 cells were cotransfected with vector alone or a vector expressing Flag-tagged wild-type or mutant 100k expression vector and a plasmid expressing luciferase mRNA containing the B202 mutant tripartite leader 5' NCR. Lysates and total RNA were prepared. Analysis of wild-type or mutant 100k immune complexes demonstrated that tyrosine phosphorylation of mutant 100k proteins 100kY365F and 100kY682F was strongly decreased but not abolished (Fig. 3A). These results suggest that aa 365 and aa 682 are indeed two significant tyrosine phosphorylation sites. Analysis of luciferase reporter activity revealed that the two 100k mutants were each partially decreased (50%) in their ability to promote ribosome shunting compared to wild-type 100k protein (Fig. 3C). Taken together, these data indicate that 100k protein undergoes tyrosine phosphorylation at both aa 365 and aa 682, and mutation of either site independently and moderately impairs 100k protein ability to promote ribosome shunting.
To determine whether a double point mutation at both aa 365 and 682 sites more severely affects 100k's ability to promote ribosome shunting, a Flag-100k protein containing both the Y365F and Y682F sites were developed. 293 cells were transfected as described above, and Flag-100k protein abundance and tyrosine phosphorylation were examined. Flag-100k was immunoprecipitated and probed by immunoblot analysis. The 100k double mutation 100kYYF almost fully abolished its tyrosine phosphorylation compared to wild-type 100k protein (Fig. 3B) or either single tyrosine phosphorylation mutant. Control studies (data not shown) showed that similar amounts of all 100k proteins were expressed and immunoprecipitated. To examine the ability of the double tyrosine mutant to promote ribosome shunting, luciferase activity was measured from equal amounts of the same cell lysates and normalized to similar mRNA levels. There was a >3-fold reduction in ribosome shunting carried out by the double mutant (Fig. 3C). These findings suggest that although there are several tyrosines that are phosphorylated in 100k protein, tyrosine phosphorylation at aa 365 and aa 682 together are essential for 100k protein to strongly promote ribosome shunting.
100k protein tyrosine mutants associate with eIF4G to the same level as wild-type 100k protein. To investigate the role of tyrosine phosphorylation in 100k protein association with eIF4G, wild-type or mutant 100k proteins were expressed in 293 cells from transfected plasmids. Lysates were prepared 36 h posttransfection, and endogenous eIF4G was recovered from equal amounts of RNase A-treated lysates by immunoprecipitation with anti-eIF4G antibody. The eIF4G complex was analyzed by immunoblotting using anti-Flag and anti-eIF4G antibodies. RNase A treatment assured that 100k protein association with eIF4G occurred independently of RNA binding, and it must occur by direct protein-protein interaction. The tyrosine mutant 100k proteins associated with eIF4G to an extent similar to that of the wild-type 100k protein (Fig. 4). Transfection of smaller amounts of plasmid DNA, resulting in lower expression of proteins, did not alter the interaction results (data not shown). These data indicate that tyrosine phosphorylation of 100k protein does not affect 100k protein association with eIF4G, and results were not obscured by protein overexpression.
Tyrosine phosphorylation of 100k protein promotes its preferential late viral interaction with tripartite leader mRNAs. We recently showed that 100k has two distinct RNA-binding motifs (25). A nonspecific RNA-binding motif which confers general RNA interaction was mapped to the C terminus of 100k protein (mutant 100kN726). A specific tripartite leader RNA-binding motif was mapped to the middle of the protein. The tripartite leader-specific RNA-binding motif allows the 100k-eIF4F complex to bind more strongly to Ad viral mRNAs than to a nontripartite leader mRNAs in vivo (25). However, specific tripartite leader RNA binding of wild-type or mutant 100kN726 100k protein could not be detected using an in vitro RNA filter binding assay and recombinant 100k proteins (data not shown). These results suggested that posttranslational modifications, possibly 100k protein tyrosine phosphorylation, might be required to confer selective binding to tripartite leader RNA, which is essential to promote ribosome shunting. To investigate whether tyrosine phosphorylation of 100k protein affects the ability of 100k to bind RNA, the RNA-binding ability of both the wild-type and 100kN726 mutant proteins were examined following genistein treatment. Studies were also carried out to characterize the RNA binding ability in vivo of mutant 100kN726YYF. 293 cells were cotransfected with plasmids expressing wild-type or mutant Flag-100k proteins and ?-gal reporter mRNAs controlled by the tripartite leader or a 5' UTR that is strongly dependent on eIF4F, known as CR3 (8, 25). Cells were mock treated or treated with genistein for 4 h, and then wild-type and mutant 100k proteins were recovered from equal amounts of lysates by immunoprecipitation with anti-Flag antibody. Tyrosine phosphorylation levels of both 100k protein and 100kN726 protein decreased upon treatment with genistein (Fig. 5A, compare lanes 2 to 3 and 4 to 5). However, mutant 100kN726YYF protein tyrosine phosphorylation levels were unchanged by genistein treatment (Fig. 5A, compare lanes 6 and 7). This finding further suggests that tyrosines at aa 365 and aa 682 are major sites for phosphorylation in 100k protein. The mRNA that associates with wild-type or mutant 100k protein was isolated from the corresponding 100k immune complexes by treatment with Trizol and identified by quantitative RT-PCR with specific primers for the coding region of ?-gal (Fig. 5B). Total mRNA levels of CR3 or tripartite leader reporter mRNAs in whole-cell lysates were determined by Northern blot analysis (Fig. 5C). Consistent with previous studies (25), 100k associated with tripartite leader mRNAs to a three- to fivefold higher level than to mRNAs containing the CR3 5' NCR (Fig. 6B, lanes 3 and 5). The 100kN726 protein did not detectably associate with CR3 mRNA (Fig. 5B, lane 7). Importantly, 100k association with tripartite leader mRNA strongly decreased with genistein treatment (Fig. 5B, lanes 5 and 6). The ability of mutant 100kN726 to associate with tripartite leader mRNA was also inhibited by genistein treatment (Fig. 5B, lanes 9 and 10). In contrast, genistein treatment did not change 100k association with CR3 mRNA (Fig. 5B, lanes 3 and 4), suggesting that tyrosine phosphorylation of 100k is important for its interaction with tripartite leader mRNAs. Moreover, the mutant 100kN726YYF failed to associate with the tripartite leader mRNA in the presence or absence of genistein (Fig. 5B, lanes 11 to 14). These results imply that tyrosine phosphorylated 100k protein promotes ribosome shunting by enhancing the specific association between 100k protein and late viral tripartite leader mRNA (3LDR).
Mutation of tyrosine 365 and 682 are sufficient to abolish 100k protein preferential binding to tripartite leader RNA. Studies presented above showed that 100kYYF protein failed to promote ribosome shunting compared to wild-type 100k protein. Studies therefore determined whether tyrosine phosphorylation of 100k plays an essential role in its ability to promote ribosome shunting by affecting its RNA-binding ability. The RNA-binding ability of mutant 100kYYF protein was investigated using an in vivo RNA-binding assay as described above. As shown earlier, 100kN726 protein preferentially bound only to tripartite leader mRNA (Fig. 6B, lane 5). The preferential RNA-binding ability of this mutant was abolished with genistein treatment (Fig. 6B, lane 6), which decreased its tyrosine phosphorylation level (Fig. 6A, lanes 3 to 6). Genistein treatment did not change the tyrosine phosphorylation level of the 100kYYF protein (Fig. 6A, lanes 7 to 10). 100kYYF-eIF4F complexes associated with CR3 and tripartite leader mRNA with similar but very low affinities, which was not affected by genistein treatment (Fig. 6B, lanes 7 to 10). The general RNA-binding motif in the C terminus of 100k, which binds to all mRNAs, remains intact in the 100kYYF mutant, and its RNA-binding ability is obviously not affected by tyrosine dephosphorylation. These data therefore confirm that tyrosine phosphorylation of 100k protein promotes ribosome shunting by enhancing preferential binding of 100k protein to tripartite leader containing mRNA.
DISCUSSION
Ad protein 100k has been shown to selectively promote late Ad mRNA translation by specifically binding to the tripartite leader 5' NCR through a preferential viral mRNA-binding domain and by specifically binding to initiation factor eIF4G (25). 100k-eIF4F complexes are recruited to Ad late mRNAs, which can apparently utilize the modified complex to promote tripartite leader-directed translation by ribosome shunting during late Ad infection. Studies reported here show that tyrosine phosphorylation of several specific sites in Ad 100k protein confer the ability to promote ribosome shunting on tripartite leader mRNAs. Tyrosine phosphorylation of 100k protein does not affect its ability to associate with eIF4G, rather it enhances 100k protein binding to tripartite leader mRNAs.
Earlier work has shown that genistein, a broad-spectrum tyrosine kinase inhibitor, decreases late Ad protein synthesis during viral infection and prevents inhibition of host cell mRNA translation. These data suggested that tyrosine kinase activity plays a role in the regulation of late Ad protein synthesis (9). Cytoplasmic 100k protein is tyrosine phosphorylated, and its tyrosine phosphorylation level strongly decreases after genistein treatment (Fig. 1B). Since PP2, a Src kinase-specific inhibitor, did not prevent 100k protein from promoting ribosome shunting whereas genistein did (Fig. 1A), these data indicate that the function of late Ad 100k protein is to promote ribosome shunting in a manner regulated by non-Src tyrosine phosphorylation. It is possible that genistein affects factors other than Ad 100k protein, such as translation initiation factors that are involved in ribosome shunting. However, we found that single tyrosine mutation at aa 365 or aa 682 caused moderate defects in promoting ribosome shunting (Fig. 3C) and that mutation of both tyrosines nearly abolished the ability of 100k to promote ribosome shunting (Fig. 3B and C). Furthermore, the extent of inhibition of 100k mediated ribosome shunting activity correlated with decreased tyrosine phosphorylation in the corresponding mutants (Fig. 3). It is important to note that these data do not fully exclude the possibility that tyrosine phosphorylation of non-Ad proteins might be involved in ribosome shunting or that mutation of the phosphorylated tyrosines might alter 100k structure, thereby impairing its activity. However, several lines of evidence all support the most straightforward interpretation that Ad 100k protein tyrosine phosphorylation is important for promoting ribosome shunting.
Tyrosine kinase signaling is stimulated shortly after Ad enters the late phase of infection (9), and cytoplasmic 100k protein is tyrosine phosphorylated (10) (as shown here). The late phase of Ad infection is marked by the onset of viral DNA replication, from 8 to 14 h after infection (24), when most structural proteins involved in production of viral particles are synthesized. The functional significance of regulating ribosome shunting by tyrosine phosphorylation of 100k protein in the cytoplasm might be to link enhanced synthesis of viral structural proteins to viral-induced cell stress responses, which would enable rapid production of viral particles before the death of Ad-infected cells. It should also be noted that although E1A and E1B protein activities on translation and mRNA transport might influence tripartite leader-100k interaction, we suspect that there is no critical requirement for early proteins in 100k translation activity. 100k was found to stimulate tripartite leader-directed translation in HeLa cells which do not express E1A and E1B (data not shown). It is also not known which protein tyrosine kinase phosphorylates 100k protein at positions Y365 and Y682. The tyrosine phosphorylation site characterizes a fairly general consensus site and could in fact be targeted by multiple tyrosine kinases.
There are a growing number of proteins whose ability to function in RNA binding is regulated by tyrosine phosphorylation (11, 13, 17, 28). While tyrosine phosphorylation of 100k protein increases its affinity to viral RNA, some RNA-binding proteins demonstrate a decrease in RNA-binding ability upon tyrosine phosphorylation (13, 17, 28). The pp68 protein SYNCRIP/NSAP1 (synaptotagmin-binding cytoplasmic RNA-interacting protein) may involved in regulation of mRNA translation or stability by insulin, and it is regulated by tyrosine phosphorylation (11). The MnSOD-BP protein (manganese superoxide dismutase TNA-binding protein) regulates MnSOD activity and is also regulated by its tyrosine phosphorylation states (13). Moreover, the mechanism by which hnRNPK acts as a translational silencer of human papilloma virus type 16 (HPV-15) L2 capsid protein and reticulocyte 15-lipooxygenase (LOX) occurs by blocking 60S ribosome subunits from joining 40S subunits, and it is regulated by c-Src (16). c-Src mediates tyrosine phosphorylation of hnRNPK and inhibits its RNA-binding activity, leading to activation of translation (17). QK1, a selective RNA-binding protein which belongs to the protein family known as signal transduction activator of RNA (STAR), is essential for myelination in the central nervous system. Tyrosine phosphorylation of QK1 was shown to be involved in regulating mRNA stability (28). The mechanism by which tyrosine phosphorylation regulates RNA-binding is not known in any of these examples or for Ad 100k protein. However, since two mutants which converted the key tyrosine sites of 100k to aspartic acid (100kY365D and 100kY682D) failed to promote ribosome shunting (data not shown), it is likely that a negative charge created by tyrosine phosphorylation is not in itself important for increasing 100k protein RNA-binding ability. 100k protein can dimerize (unpublished data), and it is possible that dimeric 100k protein is the form that binds to RNA, as found for several other RNA binding proteins (18, 19). However, all of the tyrosine mutants (100kY365F, 100kY682F, 100kYYF, 100kY365D, 100kY682D, and 100kYYD) dimerize to the same extent as wild-type 100k protein (data not shown), despite the fact that for those mutants tyrosine phosphorylation levels decreased compared to wild type 100k protein. Thus, it is therefore more likely that tyrosine phosphorylation of 100k protein is essential for maintaining certain conformational changes or interactions that confer on the protein the ability to bind to RNA.
REFERENCES
Adam, S. A., and G. Dreyfuss. 1987. Adenovirus proteins associated with mRNA and hnRNA in infected HeLa cells. J. Virol. 61:3276-3283.
Akiyama, T., J. Ishida, S. Nakagawa, H. Ogawara, S. Watanabe, N. Itoh, M. Shibuya, and Y. Fukami. 1987. Genistein, a specific inhibitor of tyrosine-specific protein kinases. J. Biol. Chem. 262:5592-5595.
Bablanian, R., and W. C. Russell. 1974. Adenovirus polypeptide synthesis in the presence of non-replicating poliovirus. J. Gen. Virol. 24:261-279.
Cuesta, R., Q. Xi, and R. J. Schneider. 2000. Adenovirus-specific translation by displacement of kinase Mnk1 from cap-initiation complex eIF4F. EMBO J. 19:3465-3474.
Cuesta, R., Q. Xi, and R. J. Schneider. 2004. Structural basis for competitive inhibition of eIF4G-Mnk1 interaction by the adenovirus 100-kilodalton protein. J. Virol. 78:7707-7716.
Dolph, P. J., J. T. Huang, and R. J. Schneider. 1990. Translation by the adenovirus tripartite leader: elements which determine independence from cap-binding protein complex. J. Virol. 64:2669-2677.
Dolph, P. J., V. Racaniello, A. Villamarin, F. Palladino, and R. J. Schneider. 1988. The adenovirus tripartite leader may eliminate the requirement for cap-binding protein complex during translation initiation. J. Virol. 62:2059-2066.
Feigenblum, D., and R. J. Schneider. 1996. Cap-binding protein (eukaryotic initiation factor 4E) and 4E-inactivating protein BP-1 independently regulate cap-dependent translation. Mol. Cell. Biol. 16:5450-5457.
Feigenblum, D., R. Walker, and R. J. Schneider. 1998. Adenovirus induction of an interferon-regulatory factor during entry into the late phase of infection. J. Virol. 72:9257-9266.
Gambke, C., and W. Deppert. 1981. Late nonstructural 100,000- and 33,000-dalton proteins of adenovirus type 2. II. Immunological and protein chemical analysis. J. Virol. 40:594-598.
Hresko, R. C., and M. Mueckler. 2002. Identification of pp68 as the tyrosine-phosphorylated form of SYNCRIP/NSAP1. A cytoplasmic RNA-binding protein. J. Biol. Chem. 277:25233-25238.
Huang, W., and S. J. Flint. 1998. The tripartite leader sequence of subgroup C adenovirus major late mRNAs can increase the efficiency of mRNA export. J. Virol. 72:225-235.
Knirsch, L., and L. B. Clerch. 2001. Tyrosine phosphorylation regulates manganese superoxide dismutase (MnSOD) RNA-binding protein activity and MnSOD protein expression. Biochemistry 40:7890-7895.
Logan, J., and T. Shenk. 1984. Adenovirus tripartite leader sequence enhances translation of mRNAs late after infection. Proc. Natl. Acad. Sci. USA 81:3655-3659.
Oosterom-Dragon, E. A., and H. S. Ginsberg. 1980. Purification and preliminary immunological characterization of the type 5 adenovirus, nonstructural 100,000-dalton protein. J. Virol. 33:1203-1207.
Ostareck, D. H., A. Ostareck-Lederer, I. N. Shatsky, and M. W. Hentze. 2001. Lipoxygenase mRNA silencing in erythroid differentiation: the 3'UTR regulatory complex controls 60S ribosomal subunit joining. Cell 104:281-290.
Ostareck-Lederer, A., D. H. Ostareck, C. Cans, G. Neubauer, K. Bomsztyk, G. Superti-Furga, and M. W. Hentze. 2002. c-Src-mediated phosphorylation of hnRNP K drives translational activation of specifically silenced mRNAs. Mol. Cell. Biol. 22:4535-4543.
Piron, M., T. Delaunay, J. Grosclaude, and D. Poncet. 1999. Identification of the RNA-binding, dimerization, and eIF4GI-binding domains of rotavirus nonstructural protein NSP3. J. Virol. 73:5411-5421.
Ramos, A., D. Hollingworth, S. A. Major, S. Adinolfi, G. Kelly, F. W. Muskett, and A. Pastore. 2002. Role of dimerization in KH/RNA complexes: the example of Nova KH3. Biochemistry 41:4193-4201.
Riley, D., and S. J. Flint. 1993. RNA-binding properties of a translational activator, the adenovirus L4 100-kilodalton protein. J. Virol. 67:3586-3595.
Sanna, P. P., F. Berton, M. Cammalleri, M. K. Tallent, G. R. Siggins, F. E. Bloom, and W. Francesconi. 2000. A role for Src kinase in spontaneous epileptiform activity in the CA3 region of the hippocampus. Proc. Natl. Acad. Sci. USA 97:8653-8657.
Sarkar, B., Q. Xi, H. Cheng, and R. J. Schneider. 2003. Selective degradation of AU-rich mRNAs promoted by the p37 AUF1 protein isoform. Mol. Cell. Biol. 23:6685-6693.
Schneider, R. J. 2000. Adenovirus inhibition of cellular protein synthesis and preferential translation of viral mRNAs, p. 901-914. In J. W. B. Hershey, M. B. Mathews, and N. Sonenberg (ed.), Translational regulation. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Schneider, R. J., and Y. Zhang. 1993. Translational regulation in adenovirus infected cells, vol. 2. Plenum Press, New York, N.Y.
Xi, Q., R. Cuesta, and R. J. Schneider. 2004. Tethering of eIF4G to adenoviral mRNAs by 100k protein drives ribosome shunting. Genes Dev. 18:1997-2009.
Yueh, A., and R. J. Schneider. 1996. Selective translation initiation by ribosome jumping in adenovirus-infected and heat-shocked cells. Genes Dev. 10:1557-1567.
Yueh, A., and R. J. Schneider. 2000. Translation by ribosome shunting on adenovirus and hsp70 mRNAs facilitated by complementarity to 18S rRNA. Genes Dev. 14:414-421.
Zhang, Y., Z. Lu, L. Ku, Y. Chen, H. Wang, and Y. Feng. 2003. Tyrosine phosphorylation of QKI mediates developmental signals to regulate mRNA metabolism. EMBO J. 22:1801-1810.(Qiaoran Xi, Rafael Cuesta)
ABSTRACT
Adenovirus simultaneously inhibits cap-dependent host cell mRNA translation while promoting the translation of its late viral mRNAs during infection. Studies previously demonstrated that tyrosine kinase activity plays a central role in the control of late adenovirus protein synthesis. The tyrosine kinase inhibitor genistein decreases late viral mRNA translation and prevents viral inhibition of cellular protein synthesis. Adenovirus protein 100k blocks cellular mRNA translation by disrupting the cap-initiation complex and promotes viral mRNA translation through an alternate mechanism known as ribosome shunting. 100k protein interaction with initiation factor eIF4G and the viral 5' noncoding region on viral late mRNAs, known as the tripartite leader, are both essential for ribosome shunting. We show that adenovirus protein 100k promotes ribosome shunting in a tyrosine phosphorylation-dependent manner. The primary sites of phosphorylated tyrosine on protein 100k were mapped and mutated, and two key sites are shown to be essential for protein 100k to promote ribosome shunting. Mutation of the two tyrosine phosphorylation sites in 100k protein does not impair interaction with initiation factor 4G, but it severely reduces association of 100k with tripartite leader mRNAs. 100k protein therefore promotes ribosome shunting and selective translation of viral mRNAs by binding specifically to the adenovirus tripartite leader in a phosphotyrosine-dependent manner.
INTRODUCTION
Human adenovirus (Ad) selectively translates viral late mRNAs while simultaneously suppressing translation of cellular mRNAs during the late phase of infection. Late Ad mRNAs are capped but are translated despite inhibition of cellular protein synthesis, because they possess a common 5' noncoding region (5' NCR) known as the tripartite leader (6, 7, 12, 14, 23). The tripartite leader confers on mRNAs the ability to be selectively translated during late Ad infection and during cell stress by an alternate means of initiation known as ribosome shunting (26, 27). The general mechanism for ribosome shunting in Ad involves loading of 40S ribosome subunits to the 5' end of the capped mRNA, possibly with limited scanning followed by direct translocation of 40S subunits to the downstream initiation codon, directed by "shunting elements" in the tripartite leader. In uninfected cells, the tripartite leader directs translation by conventional 5' scanning of 40S ribosome subunits and by 40S ribosome shunting at roughly equal levels. However, in late Ad-infected cells, when the cap initiation complex is altered (4), the tripartite leader directs translation solely by ribosome shunting (26, 27).
The protein 100k is a late adenovirus polypeptide encoded by the L4 transcription unit in very large amounts (15). 100k is the first late viral protein translated with the onset of the late phase of infection (3). Recent studies showed that 100k protein is largely responsibly for inhibition of cellular protein synthesis (25), and it is also involved in promoting translation on tripartite leader mRNAs by ribosome shunting (25). 100k binds the carboxyl terminus of eIF4G at or near the site normally occupied by the kinase Mnk1, which is responsible for phosphorylating cap binding protein eIF4E. 100k therefore competitively displaces Mnk1 from cap initiation complexes, which prevents eIF4E phosphorylation (4). Dephosphorylation of eIF4E is often associated with inhibition of cellular cap-dependent mRNA translation, although the mechanism is not understood. 100k protein also binds mRNAs in the cytoplasm (1, 20) and demonstrates a preference for late Ad tripartite leader mRNAs (25). The C terminus of 100k (amino acid 726 to 805) contains a general RNA-binding domain, whereas its middle region (345 to 726) contains a tripartite leader-specific mRNA binding domain (25). During late viral infection, the modified 100k cap initiation complex associates with higher specificity to mRNAs that contain the tripartite leader 5' NCR. The 100k-tripartite leader complex enhances association with the initiation factor 4G (eIF4G) and poly(A) binding protein, which is associated with increased translation by ribosome shunting on late viral mRNAs (25).
Ad protein 100k is a tyrosine- and serine-phosphorylated protein (10). In addition, Ad stimulates tyrosine kinase activity during the late phase of infection, which has been linked to inhibition of cellular protein synthesis (9). To investigate whether the ability of Ad 100k protein to promote ribosome shunting and/or inhibit cellular protein synthesis is regulated by tyrosine phosphorylation, we performed biochemical and mutational studies of 100k protein. We show that tyrosine phosphorylation does not affect 100k protein association with eIF4G. Rather, tyrosine dephosphorylation of 100k protein at sites Y365 and Y682 is shown to impair its ability to promote ribosome shunting on Ad late mRNAs by decreasing 100k-RNA interaction. In addition, tyrosine phosphorylation was found to promote ribosome shunting by enhancing preferential binding of 100k protein to tripartite leader mRNAs. 100k protein therefore links the requirement for tyrosine kinase activity during late Ad infection to translational control.
MATERIALS AND METHODS
Antisera and cells. Rabbit polyclonal antisera were raised against the C-terminal protein fragment of human eIF4GI (amino acids [aa] 1045 to 1560). Mouse monoclonal anti-phosphotyrosine antibody was provided by E. Skolnik (New York University), and mouse monoclonal anti-rabbit eIF4A was provided by W. Merrick (Case Western Reserve, Cleveland, Ohio). Other antisera were from commercial sources and included horseradish peroxidase-conjugated donkey anti-rabbit or sheep anti-mouse secondary antibodies (Amersham), mouse mononclonal anti-FLAG antibody (Sigma), and the enhanced chemiluminescence system (ECL; Amersham). 293 cells comprise a human embryonic kidney cell line that expresses the E1 region of Ad5. 293 cells were cultured in Dulbecco's modified Eagle medium (DMEM) containing 10% calf serum and 50 μg/ml gentamicin.
Plasmids. Plasmid pCMV-Ad?-gal (8), pCMV-CR3?-gal, pTL-Flag100kN726, pTL-Flag100kN345, pTL-100kY365F, pCMV-LUC, and pCMV-AdB202LUC (25) were described previously. pFLAG-CM2 was from Sigma. pIRES-EGFP was from Clontech. pTL-Flag100kY365FY682F (renamed Flag100kYYF) and pTL-Flag100kN726Y365FY368F (renamed 100kN726YYF) were constructed by inserting a BstEII-AscI 100k DNA fragment containing an aa 365Y-to-F mutation into BstEII-AscI-digested pTL-Flag100kY682F and pTL-Flag100kN726Y682F vectors, respectively. Plasmids pTL-Flag100kY682F and pTL-Flag100N726Y682F were derived from pTL-Flag100k and pTL-Flag100kN726 using a site-directed mutagenesis kit (Stratagene). The primers for cloning were 100kY682F (5'-end oligo; 5'-CGAAAGGGACGGGGGGTTTTCCTGGACCCCCAGTCC-3') and 100kY682F (3'-end oligo; 5'-GCCGGACTGGGGGTCCAGGAAAACCCCCCGTCCCTTTCG-3'). All of the plasmids were confirmed by DNA sequencing.
Mammalian cell transfection and luciferase assay. 293 cells were transfected with plasmid constructs using Lipofectamine Plus (Invitrogen) according to manufacturer protocols. Cells were harvested 36 h posttransfection and subjected to detergent lysis (0.5% NP-40, 50 mM HEPES, pH 7.0, 250 mM NaCl, 2 mM EDTA, 2 mM sodium orthovanadate, 25 mM glycerophosphate, 1 tablet of protease inhibitor [Roche] per 10 ml) at 4°C for 20 min, nuclei were pelleted by microcentrifuge centrifugation, and supernatant lysates were collected. Luciferase assays were performed as per manufacturer protocol (Promega).
Immunoprecipitation and immunoblot analysis. Equal amounts of lysates were incubated with 2 μg mouse monoclonal antibodies to FLAG (Sigma) for 1 h at 4°C or 20 μl rabbit polyclonal serum against the C-terminal fragment of eIF4G (4) and RNase A (20 μg/ml) overnight at 4°C. Either protein G or A agarose was added and incubated for 1 h at 4°C. Precipitates were washed four times with lysis buffer, boiled in sodium dodecyl sulfate (SDS)-sample buffer (62.5 mM Tris-HCl, pH 6.8, 2% SDS, 25% glycerol, 0.01% bromophenol blue), and analyzed by SDS-polyacrylamide gel electrophoresis (PAGE). Gels were electroblotted onto Immunobilon-P membrane (Millipore) and blocked overnight in blocking buffer (150 mM NaCl, 10 mM Tris-HCl, pH 8.0, 5% bovine serum albumin at 4°C). Primary antibodies used were mouse monoclonal anti-phosphotyrosine antibody (1:500), mouse monoclonal anti-FLAG antibody (1 μg/ml), and rabbit polyclonal serum against the C terminus of eIF4G (1:250). Studies used horseradish peroxidase-conjugated donkey anti-rabbit or sheep anti-mouse secondary antibodies or horseradish peroxidase-conjugated protein A (1:5,000) as secondary antibody. An enhanced chemiluminescence detection method (Amersham) was used, and the membrane was exposed to film.
Northern blot analysis. 293 cells were lysed directly into Trizol extraction reagent (Invitrogen). RNA samples were electrophoresed on a denaturing formaldehyde gel and transferred to a nylon membrane (Perkin Elmer Life Science, Inc.). Membranes were probed overnight at 68°C with [-32P]dCTP-labeled probes directed to the ?-galactosidase and GAPDH coding regions. Following washes, membranes were exposed to film at –70°C.
In vivo RNA-binding assay. 293 cells were lysed directly into detergent lysis buffer (0.5% NP-40, 50 mM HEPES, pH 7.0, 250 mM NaCl, 2 mM EDTA, 2 mM sodium orthovanadate, 25 mM glycerophosphate, 1 tablet of protease inhibitor [Roche] per 10 ml, and 40 U/ml Rnasin [Promega]) at 4°C for 20 min, nuclei were pelleted by centrifugation, and supernatant lysates were collected. Wild-type or mutant 100k protein was recovered from equal amounts of lysate by immunoprecipitation with anti-FLAG antibody using lysis buffer. mRNAs were extracted from the immunoprecipitated protein using Trizol extraction reagent. Semiquantitative reverse transcription-PCR (RT-PCR) amplification of immunoprecipitated ?-galactosidase (?-gal) mRNA was carried out as described previously (22).
RESULTS
The ability of Ad 100k protein to promote ribosome shunting is regulated by tyrosine phosphorylation. Previous studies showed that during Ad infection, genistein, a broad inhibitor of tyrosine kinases, can decrease late viral mRNA translation and prevent viral inhibition of host cell protein synthesis without significantly affecting short-term viral mRNA abundance or replication (9). These results suggested that tyrosine kinase activity may play a role in promoting late viral protein synthesis. Studies were carried out to determine whether 100k is a target of tyrosine kinases. 293 cells were cotransfected with vector alone or a vector expressing Flag-tagged wild-type 100k protein, and plasmids expressing a luciferase reporter mRNA containing a mutant Ad tripartite leader 5' NCR (B202) which has a strong hairpin structure inserted in the 3' end of the leader (Fig. 1A). The B202 5' NCR can only translate by ribosome shunting, as the hairpin blocks initiation by ribosome scanning (25, 27). In uninfected cells, translation by ribosome shunting accounts for roughly half the translation directed by the tripartite leader (26, 27). The B202 tripartite leader therefore translates at about half the level of the wild-type leader. The plasmid pIRES-EGFP was included as an internal control and expresses green fluorescent protein (GFP) directed by the encephalomyocarditis virus (EMCV) internal ribosome entry site (IRES). Cells were treated with genistein (300 μM), an inhibitor of tyrosine kinase activity (2), or the more specific Src kinase inhibitor PP2 (30 μM) for 4 h overnight (21), with mock-treated cells serving as a control. At 36 h posttransfection, lysates and total RNAs were prepared. Flag-100k protein was immunoprecipitated, resolved by SDS-PAGE, and examined by immunoblot analysis using antibody against Flag or phosphotyrosine (Fig. 1B). 100k immune complexes exhibited strong tyrosine phosphorylation, which was decreased significantly by treatment with genistein (approximately eightfold) and less so with PP2 (approximately two- to threefold) (Fig. 1B). These data are consistent with involvement of Src and other tyrosine kinases in 100k phosphorylation. Equal amounts of 100k were recovered in the immunoprecipitations, as shown.
100k was previously shown to bind tightly to the C-terminal end of eIF4G (4), which is important for inhibition of cellular protein synthesis (5) and promotion of ribosome shunting (25). To explore whether genistein inhibition of 100k protein tyrosine phosphorylation affects its interaction with eIF4G, cells were transfected with a plasmid expressing Flag-100k protein and treated with genistein, and then endogenous eIF4G was immunoprecipitated from equal amounts of cell lysates and proteins were detected by immunoblot analysis. Equal amounts of eIF4G were immunoprecipitated as shown. Genistein treatment had little, if any, ability to decrease association of 100k protein with eIF4G, while PP2 clearly had no effect (Fig. 1B). These data indicate that genistein-sensitive tyrosine phosphorylation of 100k protein has little effect on its interaction with eIF4G. However, analysis of luciferase synthesized by ribosome shunting from the B202 mRNA revealed an increase in ribosome shunting promoted by 100k protein on tripartite leader mRNAs (Fig. 1C) as shown previously (25). The ability of 100k to promote ribosome shunting was conversely decreased threefold by treatment with genistein (300 μm). In contrast, treatment of cells with PP2 (30 μM) had little, if any, effect (Fig. 1C).
These data suggest that tyrosine phosphorylation of 100k might play a role in its ability to promote ribosome shunting. These results also suggest that the key regulatory phosphotyrosines in 100k protein are predominantly phosphorylated by kinases other than Src, since PP2 specifically inhibits Src kinase activity (21) and had little effect on 100k ribosome shunting activity.
100k protein phosphotyrosine sites are located between aa 345 and aa 726. To determine whether 100k tyrosine phosphorylation directly regulates its activity in translational control, we first identified the major sites of tyrosine phosphorylation. 100k protein has 21 tyrosines, several of which are within strong consensus sites for phosphorylation. The regions in 100k that are most heavily tyrosine phosphorylated were preliminarily mapped by utilizing 100k protein truncation mutants 100kN726 and 100kN345 (Fig. 2A). Plasmids expressing wild-type or mutant Flag-100k proteins were transfected into 293 cells. At 32 h posttransfection, cells were treated with 150 μM genistein or with vehicle control for 4 h before lysis. Wild-type or mutant 100k protein was recovered from equal amounts of cell lysates by immunoprecipitation with anti-Flag antibody. 100k proteins were analyzed by SDS-PAGE and immunoblot analysis using anti-Flag and anti-phosphotyrosine (P-Tyr) antibodies (Fig. 2B). Mutant 100kN726 protein was found to be tyrosine phosphorylated to the same extent as wild-type 100k protein (Fig. 2B, lanes 2 and 3), whereas no tyrosine phosphorylation signal was detected in the 100kN345 protein (Fig. 2B, lanes 4 and 8). Genistein treatment significantly decreased the tyrosine phosphorylation level of wild-type and N726 100k proteins (approximately five- to ninefold) (Fig. 2B, lanes 6 and 7). These results suggest that the tyrosine phosphorylation sites in protein 100k are located between amino acids 345 and 726.
Tyrosine mutations at aa 365 and aa 682 impair the ability of 100k protein to promote ribosome shunting. Based on the tyrosine phosphorylation sites predicted by two different programs, Prosite (www.expasy.org/prosite/) and NetPhos2.0 (genome.cbs.dtu.dk/services/NetPhos/), we mutated tyrosine (Y) sites at positions 365 and 682 to phenylalanine (F), as described in Materials and Methods. To investigate the role of tyrosine phosphorylation of 100k in promoting ribosome shunting, 293 cells were cotransfected with vector alone or a vector expressing Flag-tagged wild-type or mutant 100k expression vector and a plasmid expressing luciferase mRNA containing the B202 mutant tripartite leader 5' NCR. Lysates and total RNA were prepared. Analysis of wild-type or mutant 100k immune complexes demonstrated that tyrosine phosphorylation of mutant 100k proteins 100kY365F and 100kY682F was strongly decreased but not abolished (Fig. 3A). These results suggest that aa 365 and aa 682 are indeed two significant tyrosine phosphorylation sites. Analysis of luciferase reporter activity revealed that the two 100k mutants were each partially decreased (50%) in their ability to promote ribosome shunting compared to wild-type 100k protein (Fig. 3C). Taken together, these data indicate that 100k protein undergoes tyrosine phosphorylation at both aa 365 and aa 682, and mutation of either site independently and moderately impairs 100k protein ability to promote ribosome shunting.
To determine whether a double point mutation at both aa 365 and 682 sites more severely affects 100k's ability to promote ribosome shunting, a Flag-100k protein containing both the Y365F and Y682F sites were developed. 293 cells were transfected as described above, and Flag-100k protein abundance and tyrosine phosphorylation were examined. Flag-100k was immunoprecipitated and probed by immunoblot analysis. The 100k double mutation 100kYYF almost fully abolished its tyrosine phosphorylation compared to wild-type 100k protein (Fig. 3B) or either single tyrosine phosphorylation mutant. Control studies (data not shown) showed that similar amounts of all 100k proteins were expressed and immunoprecipitated. To examine the ability of the double tyrosine mutant to promote ribosome shunting, luciferase activity was measured from equal amounts of the same cell lysates and normalized to similar mRNA levels. There was a >3-fold reduction in ribosome shunting carried out by the double mutant (Fig. 3C). These findings suggest that although there are several tyrosines that are phosphorylated in 100k protein, tyrosine phosphorylation at aa 365 and aa 682 together are essential for 100k protein to strongly promote ribosome shunting.
100k protein tyrosine mutants associate with eIF4G to the same level as wild-type 100k protein. To investigate the role of tyrosine phosphorylation in 100k protein association with eIF4G, wild-type or mutant 100k proteins were expressed in 293 cells from transfected plasmids. Lysates were prepared 36 h posttransfection, and endogenous eIF4G was recovered from equal amounts of RNase A-treated lysates by immunoprecipitation with anti-eIF4G antibody. The eIF4G complex was analyzed by immunoblotting using anti-Flag and anti-eIF4G antibodies. RNase A treatment assured that 100k protein association with eIF4G occurred independently of RNA binding, and it must occur by direct protein-protein interaction. The tyrosine mutant 100k proteins associated with eIF4G to an extent similar to that of the wild-type 100k protein (Fig. 4). Transfection of smaller amounts of plasmid DNA, resulting in lower expression of proteins, did not alter the interaction results (data not shown). These data indicate that tyrosine phosphorylation of 100k protein does not affect 100k protein association with eIF4G, and results were not obscured by protein overexpression.
Tyrosine phosphorylation of 100k protein promotes its preferential late viral interaction with tripartite leader mRNAs. We recently showed that 100k has two distinct RNA-binding motifs (25). A nonspecific RNA-binding motif which confers general RNA interaction was mapped to the C terminus of 100k protein (mutant 100kN726). A specific tripartite leader RNA-binding motif was mapped to the middle of the protein. The tripartite leader-specific RNA-binding motif allows the 100k-eIF4F complex to bind more strongly to Ad viral mRNAs than to a nontripartite leader mRNAs in vivo (25). However, specific tripartite leader RNA binding of wild-type or mutant 100kN726 100k protein could not be detected using an in vitro RNA filter binding assay and recombinant 100k proteins (data not shown). These results suggested that posttranslational modifications, possibly 100k protein tyrosine phosphorylation, might be required to confer selective binding to tripartite leader RNA, which is essential to promote ribosome shunting. To investigate whether tyrosine phosphorylation of 100k protein affects the ability of 100k to bind RNA, the RNA-binding ability of both the wild-type and 100kN726 mutant proteins were examined following genistein treatment. Studies were also carried out to characterize the RNA binding ability in vivo of mutant 100kN726YYF. 293 cells were cotransfected with plasmids expressing wild-type or mutant Flag-100k proteins and ?-gal reporter mRNAs controlled by the tripartite leader or a 5' UTR that is strongly dependent on eIF4F, known as CR3 (8, 25). Cells were mock treated or treated with genistein for 4 h, and then wild-type and mutant 100k proteins were recovered from equal amounts of lysates by immunoprecipitation with anti-Flag antibody. Tyrosine phosphorylation levels of both 100k protein and 100kN726 protein decreased upon treatment with genistein (Fig. 5A, compare lanes 2 to 3 and 4 to 5). However, mutant 100kN726YYF protein tyrosine phosphorylation levels were unchanged by genistein treatment (Fig. 5A, compare lanes 6 and 7). This finding further suggests that tyrosines at aa 365 and aa 682 are major sites for phosphorylation in 100k protein. The mRNA that associates with wild-type or mutant 100k protein was isolated from the corresponding 100k immune complexes by treatment with Trizol and identified by quantitative RT-PCR with specific primers for the coding region of ?-gal (Fig. 5B). Total mRNA levels of CR3 or tripartite leader reporter mRNAs in whole-cell lysates were determined by Northern blot analysis (Fig. 5C). Consistent with previous studies (25), 100k associated with tripartite leader mRNAs to a three- to fivefold higher level than to mRNAs containing the CR3 5' NCR (Fig. 6B, lanes 3 and 5). The 100kN726 protein did not detectably associate with CR3 mRNA (Fig. 5B, lane 7). Importantly, 100k association with tripartite leader mRNA strongly decreased with genistein treatment (Fig. 5B, lanes 5 and 6). The ability of mutant 100kN726 to associate with tripartite leader mRNA was also inhibited by genistein treatment (Fig. 5B, lanes 9 and 10). In contrast, genistein treatment did not change 100k association with CR3 mRNA (Fig. 5B, lanes 3 and 4), suggesting that tyrosine phosphorylation of 100k is important for its interaction with tripartite leader mRNAs. Moreover, the mutant 100kN726YYF failed to associate with the tripartite leader mRNA in the presence or absence of genistein (Fig. 5B, lanes 11 to 14). These results imply that tyrosine phosphorylated 100k protein promotes ribosome shunting by enhancing the specific association between 100k protein and late viral tripartite leader mRNA (3LDR).
Mutation of tyrosine 365 and 682 are sufficient to abolish 100k protein preferential binding to tripartite leader RNA. Studies presented above showed that 100kYYF protein failed to promote ribosome shunting compared to wild-type 100k protein. Studies therefore determined whether tyrosine phosphorylation of 100k plays an essential role in its ability to promote ribosome shunting by affecting its RNA-binding ability. The RNA-binding ability of mutant 100kYYF protein was investigated using an in vivo RNA-binding assay as described above. As shown earlier, 100kN726 protein preferentially bound only to tripartite leader mRNA (Fig. 6B, lane 5). The preferential RNA-binding ability of this mutant was abolished with genistein treatment (Fig. 6B, lane 6), which decreased its tyrosine phosphorylation level (Fig. 6A, lanes 3 to 6). Genistein treatment did not change the tyrosine phosphorylation level of the 100kYYF protein (Fig. 6A, lanes 7 to 10). 100kYYF-eIF4F complexes associated with CR3 and tripartite leader mRNA with similar but very low affinities, which was not affected by genistein treatment (Fig. 6B, lanes 7 to 10). The general RNA-binding motif in the C terminus of 100k, which binds to all mRNAs, remains intact in the 100kYYF mutant, and its RNA-binding ability is obviously not affected by tyrosine dephosphorylation. These data therefore confirm that tyrosine phosphorylation of 100k protein promotes ribosome shunting by enhancing preferential binding of 100k protein to tripartite leader containing mRNA.
DISCUSSION
Ad protein 100k has been shown to selectively promote late Ad mRNA translation by specifically binding to the tripartite leader 5' NCR through a preferential viral mRNA-binding domain and by specifically binding to initiation factor eIF4G (25). 100k-eIF4F complexes are recruited to Ad late mRNAs, which can apparently utilize the modified complex to promote tripartite leader-directed translation by ribosome shunting during late Ad infection. Studies reported here show that tyrosine phosphorylation of several specific sites in Ad 100k protein confer the ability to promote ribosome shunting on tripartite leader mRNAs. Tyrosine phosphorylation of 100k protein does not affect its ability to associate with eIF4G, rather it enhances 100k protein binding to tripartite leader mRNAs.
Earlier work has shown that genistein, a broad-spectrum tyrosine kinase inhibitor, decreases late Ad protein synthesis during viral infection and prevents inhibition of host cell mRNA translation. These data suggested that tyrosine kinase activity plays a role in the regulation of late Ad protein synthesis (9). Cytoplasmic 100k protein is tyrosine phosphorylated, and its tyrosine phosphorylation level strongly decreases after genistein treatment (Fig. 1B). Since PP2, a Src kinase-specific inhibitor, did not prevent 100k protein from promoting ribosome shunting whereas genistein did (Fig. 1A), these data indicate that the function of late Ad 100k protein is to promote ribosome shunting in a manner regulated by non-Src tyrosine phosphorylation. It is possible that genistein affects factors other than Ad 100k protein, such as translation initiation factors that are involved in ribosome shunting. However, we found that single tyrosine mutation at aa 365 or aa 682 caused moderate defects in promoting ribosome shunting (Fig. 3C) and that mutation of both tyrosines nearly abolished the ability of 100k to promote ribosome shunting (Fig. 3B and C). Furthermore, the extent of inhibition of 100k mediated ribosome shunting activity correlated with decreased tyrosine phosphorylation in the corresponding mutants (Fig. 3). It is important to note that these data do not fully exclude the possibility that tyrosine phosphorylation of non-Ad proteins might be involved in ribosome shunting or that mutation of the phosphorylated tyrosines might alter 100k structure, thereby impairing its activity. However, several lines of evidence all support the most straightforward interpretation that Ad 100k protein tyrosine phosphorylation is important for promoting ribosome shunting.
Tyrosine kinase signaling is stimulated shortly after Ad enters the late phase of infection (9), and cytoplasmic 100k protein is tyrosine phosphorylated (10) (as shown here). The late phase of Ad infection is marked by the onset of viral DNA replication, from 8 to 14 h after infection (24), when most structural proteins involved in production of viral particles are synthesized. The functional significance of regulating ribosome shunting by tyrosine phosphorylation of 100k protein in the cytoplasm might be to link enhanced synthesis of viral structural proteins to viral-induced cell stress responses, which would enable rapid production of viral particles before the death of Ad-infected cells. It should also be noted that although E1A and E1B protein activities on translation and mRNA transport might influence tripartite leader-100k interaction, we suspect that there is no critical requirement for early proteins in 100k translation activity. 100k was found to stimulate tripartite leader-directed translation in HeLa cells which do not express E1A and E1B (data not shown). It is also not known which protein tyrosine kinase phosphorylates 100k protein at positions Y365 and Y682. The tyrosine phosphorylation site characterizes a fairly general consensus site and could in fact be targeted by multiple tyrosine kinases.
There are a growing number of proteins whose ability to function in RNA binding is regulated by tyrosine phosphorylation (11, 13, 17, 28). While tyrosine phosphorylation of 100k protein increases its affinity to viral RNA, some RNA-binding proteins demonstrate a decrease in RNA-binding ability upon tyrosine phosphorylation (13, 17, 28). The pp68 protein SYNCRIP/NSAP1 (synaptotagmin-binding cytoplasmic RNA-interacting protein) may involved in regulation of mRNA translation or stability by insulin, and it is regulated by tyrosine phosphorylation (11). The MnSOD-BP protein (manganese superoxide dismutase TNA-binding protein) regulates MnSOD activity and is also regulated by its tyrosine phosphorylation states (13). Moreover, the mechanism by which hnRNPK acts as a translational silencer of human papilloma virus type 16 (HPV-15) L2 capsid protein and reticulocyte 15-lipooxygenase (LOX) occurs by blocking 60S ribosome subunits from joining 40S subunits, and it is regulated by c-Src (16). c-Src mediates tyrosine phosphorylation of hnRNPK and inhibits its RNA-binding activity, leading to activation of translation (17). QK1, a selective RNA-binding protein which belongs to the protein family known as signal transduction activator of RNA (STAR), is essential for myelination in the central nervous system. Tyrosine phosphorylation of QK1 was shown to be involved in regulating mRNA stability (28). The mechanism by which tyrosine phosphorylation regulates RNA-binding is not known in any of these examples or for Ad 100k protein. However, since two mutants which converted the key tyrosine sites of 100k to aspartic acid (100kY365D and 100kY682D) failed to promote ribosome shunting (data not shown), it is likely that a negative charge created by tyrosine phosphorylation is not in itself important for increasing 100k protein RNA-binding ability. 100k protein can dimerize (unpublished data), and it is possible that dimeric 100k protein is the form that binds to RNA, as found for several other RNA binding proteins (18, 19). However, all of the tyrosine mutants (100kY365F, 100kY682F, 100kYYF, 100kY365D, 100kY682D, and 100kYYD) dimerize to the same extent as wild-type 100k protein (data not shown), despite the fact that for those mutants tyrosine phosphorylation levels decreased compared to wild type 100k protein. Thus, it is therefore more likely that tyrosine phosphorylation of 100k protein is essential for maintaining certain conformational changes or interactions that confer on the protein the ability to bind to RNA.
REFERENCES
Adam, S. A., and G. Dreyfuss. 1987. Adenovirus proteins associated with mRNA and hnRNA in infected HeLa cells. J. Virol. 61:3276-3283.
Akiyama, T., J. Ishida, S. Nakagawa, H. Ogawara, S. Watanabe, N. Itoh, M. Shibuya, and Y. Fukami. 1987. Genistein, a specific inhibitor of tyrosine-specific protein kinases. J. Biol. Chem. 262:5592-5595.
Bablanian, R., and W. C. Russell. 1974. Adenovirus polypeptide synthesis in the presence of non-replicating poliovirus. J. Gen. Virol. 24:261-279.
Cuesta, R., Q. Xi, and R. J. Schneider. 2000. Adenovirus-specific translation by displacement of kinase Mnk1 from cap-initiation complex eIF4F. EMBO J. 19:3465-3474.
Cuesta, R., Q. Xi, and R. J. Schneider. 2004. Structural basis for competitive inhibition of eIF4G-Mnk1 interaction by the adenovirus 100-kilodalton protein. J. Virol. 78:7707-7716.
Dolph, P. J., J. T. Huang, and R. J. Schneider. 1990. Translation by the adenovirus tripartite leader: elements which determine independence from cap-binding protein complex. J. Virol. 64:2669-2677.
Dolph, P. J., V. Racaniello, A. Villamarin, F. Palladino, and R. J. Schneider. 1988. The adenovirus tripartite leader may eliminate the requirement for cap-binding protein complex during translation initiation. J. Virol. 62:2059-2066.
Feigenblum, D., and R. J. Schneider. 1996. Cap-binding protein (eukaryotic initiation factor 4E) and 4E-inactivating protein BP-1 independently regulate cap-dependent translation. Mol. Cell. Biol. 16:5450-5457.
Feigenblum, D., R. Walker, and R. J. Schneider. 1998. Adenovirus induction of an interferon-regulatory factor during entry into the late phase of infection. J. Virol. 72:9257-9266.
Gambke, C., and W. Deppert. 1981. Late nonstructural 100,000- and 33,000-dalton proteins of adenovirus type 2. II. Immunological and protein chemical analysis. J. Virol. 40:594-598.
Hresko, R. C., and M. Mueckler. 2002. Identification of pp68 as the tyrosine-phosphorylated form of SYNCRIP/NSAP1. A cytoplasmic RNA-binding protein. J. Biol. Chem. 277:25233-25238.
Huang, W., and S. J. Flint. 1998. The tripartite leader sequence of subgroup C adenovirus major late mRNAs can increase the efficiency of mRNA export. J. Virol. 72:225-235.
Knirsch, L., and L. B. Clerch. 2001. Tyrosine phosphorylation regulates manganese superoxide dismutase (MnSOD) RNA-binding protein activity and MnSOD protein expression. Biochemistry 40:7890-7895.
Logan, J., and T. Shenk. 1984. Adenovirus tripartite leader sequence enhances translation of mRNAs late after infection. Proc. Natl. Acad. Sci. USA 81:3655-3659.
Oosterom-Dragon, E. A., and H. S. Ginsberg. 1980. Purification and preliminary immunological characterization of the type 5 adenovirus, nonstructural 100,000-dalton protein. J. Virol. 33:1203-1207.
Ostareck, D. H., A. Ostareck-Lederer, I. N. Shatsky, and M. W. Hentze. 2001. Lipoxygenase mRNA silencing in erythroid differentiation: the 3'UTR regulatory complex controls 60S ribosomal subunit joining. Cell 104:281-290.
Ostareck-Lederer, A., D. H. Ostareck, C. Cans, G. Neubauer, K. Bomsztyk, G. Superti-Furga, and M. W. Hentze. 2002. c-Src-mediated phosphorylation of hnRNP K drives translational activation of specifically silenced mRNAs. Mol. Cell. Biol. 22:4535-4543.
Piron, M., T. Delaunay, J. Grosclaude, and D. Poncet. 1999. Identification of the RNA-binding, dimerization, and eIF4GI-binding domains of rotavirus nonstructural protein NSP3. J. Virol. 73:5411-5421.
Ramos, A., D. Hollingworth, S. A. Major, S. Adinolfi, G. Kelly, F. W. Muskett, and A. Pastore. 2002. Role of dimerization in KH/RNA complexes: the example of Nova KH3. Biochemistry 41:4193-4201.
Riley, D., and S. J. Flint. 1993. RNA-binding properties of a translational activator, the adenovirus L4 100-kilodalton protein. J. Virol. 67:3586-3595.
Sanna, P. P., F. Berton, M. Cammalleri, M. K. Tallent, G. R. Siggins, F. E. Bloom, and W. Francesconi. 2000. A role for Src kinase in spontaneous epileptiform activity in the CA3 region of the hippocampus. Proc. Natl. Acad. Sci. USA 97:8653-8657.
Sarkar, B., Q. Xi, H. Cheng, and R. J. Schneider. 2003. Selective degradation of AU-rich mRNAs promoted by the p37 AUF1 protein isoform. Mol. Cell. Biol. 23:6685-6693.
Schneider, R. J. 2000. Adenovirus inhibition of cellular protein synthesis and preferential translation of viral mRNAs, p. 901-914. In J. W. B. Hershey, M. B. Mathews, and N. Sonenberg (ed.), Translational regulation. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Schneider, R. J., and Y. Zhang. 1993. Translational regulation in adenovirus infected cells, vol. 2. Plenum Press, New York, N.Y.
Xi, Q., R. Cuesta, and R. J. Schneider. 2004. Tethering of eIF4G to adenoviral mRNAs by 100k protein drives ribosome shunting. Genes Dev. 18:1997-2009.
Yueh, A., and R. J. Schneider. 1996. Selective translation initiation by ribosome jumping in adenovirus-infected and heat-shocked cells. Genes Dev. 10:1557-1567.
Yueh, A., and R. J. Schneider. 2000. Translation by ribosome shunting on adenovirus and hsp70 mRNAs facilitated by complementarity to 18S rRNA. Genes Dev. 14:414-421.
Zhang, Y., Z. Lu, L. Ku, Y. Chen, H. Wang, and Y. Feng. 2003. Tyrosine phosphorylation of QKI mediates developmental signals to regulate mRNA metabolism. EMBO J. 22:1801-1810.(Qiaoran Xi, Rafael Cuesta)