FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.06-6433fje.
This Article
Right arrow Abstract Freely available
Right arrow Summary
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.06-6433fjev1
20/14/2603    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jiang, H.
Right arrow Articles by Fueyo, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jiang, H.
Right arrow Articles by Fueyo, J.
(The FASEB Journal. 2006;20:2603-2605.)
© 2006 FASEB

A novel CRM1-dependent nuclear export signal in adenoviral E1A protein regulated by phosphorylation

Hong Jiang*,1, Melissa V. Olson*, Diana R. Medrano*, Ok-Hee Lee*, Jing Xu*, Yuji Piao*, Marta M. Alonso*, Candelaria Gomez-Manzano*, Mien-Chie Hung{dagger}, W K Alfred Yung* and Juan Fueyo*

* Department of Neuro-Oncology, Brain Tumor Center and

{dagger} Department of Molecular and Cellular Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, USA

1Correspondence: Department of Neuro-Oncology, Unit 1002, Brain Tumor Center, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, USA. E-mail: hjiang{at}mdanderson.org

ABSTRACT

Adenoviral E1A is a versatile protein that can reprogram host cells for efficient viral replication. The nuclear import of E1A is mediated by a nuclear localization signal; however, whether E1A can be actively exported from the nucleus is unknown. We first reported a CRM1-dependent nuclear export signal (NES) in E1A that is conserved in the group C adenoviruses. We showed that CRM1 and E1A coimmunoprecipitated and that blockage of CRM1 function by leptomycin B or small interfering RNA resulted in the nuclear localization of E1A. Through mutational analyses, we identified an active canonical NES element within the E1A protein spanning amino acids 70–80. We further demonstrated that phosphorylation of adjacent serine (S)89 resulted in the cytoplasmic accumulation of E1A. Interestingly, coincident with the accumulation of cells in the S/G2/M phase and histone H1 phosphorylation, E1A was relocated to the cytoplasm at the late stage of the viral cycle, which was blocked by the CDC2/CDK2 inhibitor roscovitine. Importantly, titration of the progenies of the viruses in infected cells showed that the replication efficiency of the NES mutant adenovirus was up to 500-fold lower than that of the wild-type adenovirus. Collectively, our data demonstrate the existence of a NES in E1A that is modulated by the phosphorylation of the S89 residue and the NES plays a role for an efficient viral replication in the host cells.—Jiang, H., Olson, M. V., Medrano, D. R., Lee, O. H., Xu, J., Piao, Y., Alonso, M. M., Gomez-Manzano, C., Hung, M. C., Yung, W. K. A., Fueyo, J. A novel CRM1-dependent nuclear export signal in adenoviral E1A protein regulated by phosphorylation.


Key Words: leptomycin B • NES • NLS

ADENOVIRAL E1A PROTEINS are indispensable for the virus-host interaction (1) . The interaction between E1A and critical cellular proteins has been extensively studied (1 , 2) . To provide a suitable environment for viral replication, E1A proteins physically interact with multiple cellular proteins, including the pRb family of pocket proteins, p300/CBP, cyclin/cyclin-dependent kinase (Cdk), the carboxyl-terminal binding protein, transcriptional regulator YY1, RACK1, and SWI/SNF complex (1) . Because these proteins are involved in critical cellular processes, E1A binding inactivates or changes their function to facilitate viral replication. E1A proteins are extensively phosphorylated at multiple serines mapped to positions 89, 96, 132, 185, 188, and 219 in human Ad5 E1A (3) . Serines 132, 185, and 188 (S132, -185, -and 188) are the only phosphorylation sites known to affect biological function of Ad5 E1A. For example, phosphorylation of serine 132, which is adjacent to the LXCXE motif, enhances binding to Rb (4) , and may increase its ability to disrupt Rb–E2F complexes (5) . Similarly, phosphorylation of S185 and S188 specifically increases transcription of the E4 gene (6) . The functional relevance of the phosphorylation of S89, -96, and -219 is still unclear because they fall in the regions the function of which is not well characterized (3) .

Although E1A localizes in both the cytoplasmic and nuclear compartments as its cellular targets (1 , 2 , 7) , there remains a lack of knowledge of how E1A coordinately modulates these targets through its precise and timely subcellular localization. Thus, the nucleocytoplasmic transport of E1A is not thoroughly understood. The nuclear import of E1A has been reported to be mediated by importin-{alpha} 3 (8) , which binds to the C-terminal nuclear localization signal (NLS) composed of the sequence KRPRP (8 , 9) . Further, the interaction between E1A and importin-{alpha} 3 is regulated by acetylation at the lysine residue (7) . Additional noncanonical NLS sequences have been identified within residues 23–120 and in the conserved region 3 (10 , 11) . In many instances, nuclear import of the proteins mediated by the importins is complemented by nuclear export via CRM1/exportin1 (12 , 13) . Most commonly, CRM1 mediates nuclear export by interacting with protein cargos through the HIV-Rev-like leucine-rich nuclear export signal (NES), which can be blocked by a CRM- specific inhibitor leptomycin B (LMB; refs 14 15 16 ). Accumulating evidence has shown that viral proteins function through the CRM1-mediated nuclear export pathway (17) . For human subgroup C adenoviruses, E1B-55K and E4orf6 proteins are actively exported from the nucleus through a Rev-like signal sequence and are implicated in the selective export of late viral mRNAs (18 , 19) . Thus we hypothesized that E1A utilizes a similar mechanism. In this study, our aim was to identify a CRM1-dependent NES within E1A protein and to study the regulation of nuclear export activity and its role in the adenoviral life cycle.

MATERIALS AND METHODS

Cell culture
293 cells, which were obtained from Qbiogene, Inc. (Carlsbad, CA, USA), were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS (Hyclone Laboratories, Inc., Logan, UT, USA), 100 µg/ml penicillin, and 100 µg/ml streptomycin (Invitrogen, Carlsbad, CA, USA). U-2 OS, MRC-5, and A549 cells, which were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA), were cultured in DMEM/F-12 supplemented with 10% (MRC-5 and A549) or 20% (U-2 OS) FBS and antibiotics. The cells were maintained at 37°C in a humidified atmosphere of 5% CO2.

Reagents and antibodies
LMB and roscovitine were obtained from EMD Calbiochem (La Jolla, CA, USA). Monoclonal anti-E1A (M58) and anti-CRM1 were obtained from BD Bioscience PharMingen (San Diego, CA, USA); rat monoclonal anti-E1B55K from Calbiochem; polyclonal anti-E1A (13 S-5) and anti-lamin B (C-20) from Santa Cruz Biotechnology (Santa Cruz, CA, USA); polyclonal antiphospho-histone H1 from Upstate (Lake Placid, NY, USA); monoclonal anti-{alpha}-tubulin (B-5–1-2) from Sigma-Aldrich (St Louis, MO, USA); and Texas Red-conjugated secondary antibody (Ab) from Molecular Probes, Inc. (Eugene, OR, USA).

Plasmids and mutagenesis
The 13S Ad5 E1A cDNA was a kind gift from Dr. Ronald M. Evans (The Salk Institute for Biological Studies, San Diego, CA, USA). It was later subcloned into the pEGFP-C1 plasmid (BD Bioscience Clontech, Palo Alto, CA, USA). The p2xEGFP-C1 plasmid with two EGFPs in tandem was generated by inserting a copy of EGFP cDNA downstream of EGFP in the pEGFP-C1 plasmid. Mutations, deletions, or insertions in the plasmids were accomplished using the QuikChange II XL site-directed mutagenesis kits (Stratagene, La Jolla, CA, USA).

Nuclear and cytoplasmic protein extraction
The nuclear and cytoplasmic proteins were extracted from cells with the nuclear extract kit (Active Motif, Carlsbad, CA, USA) according to the manufacturer’s instructions.

Immunoprecipitation
Nuclear proteins were extracted as described previously, except that as a final step, the nuclear proteins were dissolved in TESP buffer (10 mM Tris-HCl, pH7.5; 1 mM EDTA; 0.5 M NaCl; and 1 mM DTT) plus a protease inhibitor cocktail (Sigma-Aldrich). Two volumes of TEMP buffer (10 mM Tris-Hcl, pH7.5; 1 mM EDTA; and 4 mM MgCl2) plus protease inhibitor cocktail were added to the nuclear extract. For each sample, 1.5 mg of proteins were incubated with 8 µg of anti-E1A monoclonal antibody (mAb) in the presence of 200 µM {gamma}GTP at 4°C for 4 h. The precipitated immunocomplex was then washed five times before the proteins were eluted with 1x sodium dodecyl sulfate (SDS) loading buffer. The samples were then analyzed by immunoblotting.

Immunoblotting
Equal amounts of proteins were loaded and separated in SDS-PAGE and probed with antibodies. Finally, the protein bands were visualized using the enhanced chemiluminescence (ECL) Western blotting detection system (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Protein expression was quantified by densitometric analysis with Scion Image Beta 4.02 Win computer software (available at http://www.scioncorp.com/frames/fr_scion_products.htm).

Immunofluorescence
Cells grown in chamber slides were washed twice with ice-cold PBS, and then fixed with cold 4% paraformaldehyde in PBS for 30 min at 4°C. The cells were then permeabilized by incubation in 0.2% Triton X-100 in PBS plus 1% normal goat serum for 5 min on ice and blocked with 5% normal goat serum in PBS for 0.5 h at room temperature. Later, the cells were stained for 1 h at room temperature with primary Ab followed by incubation with Texas Red-conjugated secondary Ab for 50 min. Finally, the slides were mounted with cover slips using ProLong Gold Antifade reagent (Molecular Probes) containing 4',6'-diam idino-2-phenylidole (DAPI) and viewed with a ZEISS Axioskop 40 fluorescence microscope (Carl Zeiss, Oberkochen, Germany).

Subcellular localization of EGFP-E1A fusion proteins
EGFP fusion protein-expressing plasmids (1 µg/well in a 24-well plate) were transiently transfected into subconfluent U-2 OS cells by FuGENE 6 (Roche Molecular Biochemicals, Indianapolis, IN). Twenty-four hours later, cells were treated with10 ng/ml LMB for 3 h or left untreated. EGFP fluorescence was analyzed in living cells with a ZEISS Axiovert 200 fluorescence microscope. Alternatively, the cells were cultured in chamber slides, fixed, permeabilized, and mounted as described in "Immunofluorescence studies."

RNA interference
Small-interfering RNA (siRNA) for the depletion of CRM1 (siCRM1) was prepared as described previously (20) . The siRNA was obtained from Dharmacon (Chicago, IL, USA). U-2 OS cells in 24-well plates or 4-well chamber slides were mock or siRNA transfected with Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol.

Adenoviruses
The V74A mutation in the E1A gene was introduced into pXC1 plasmid (Microbix Biosystems Inc., Ontario, Canada) through site-directed mutagenesis. The resulting pXC1-V74A plasmid was cotransfected with pBHG10 (Microbix Biosystems Inc.) into 293 cells to generate the AdV74A adenovirus. The mutation in E1A was confirmed by sequencing the polymerase chain reaction (PCR) product of E1A region including the mutated site. The WT adenovirus (Adwt) and AdV74A were prepared as described previously (21) . Briefly, the virus was amplified in A549 cells, purified by three successive bandings on cesium, and stored at –80°C. The viral titer was assayed through tissue culture infection dose 50 (TCID50) and determined as plaque-forming units (pfu)/ ml according to the validated method developed by Quantum Biotechnology (Carlsbad, CA).

Viral replication assays
The cells were plated in 6-well plates at 1 x 105 cell per well. Twenty-four hours later, the cells were infected with virus at 10 multiplicity of infection for 30 min. The medium was then removed and replaced with fresh medium. Forty-eight hours after infection, the cells were harvested and assayed for viral titer through TCID50 as described previously (22) . Final viral titers were determined as pfu/ml.

Cell-cycle analysis
Cells were trypsinized and fixed in 70% ice-cold ethanol and incubated with propidium iodide (25 µg/ml) and RNase A (15 µg/ml) for 30 min at 37°C. The DNA content of the cells was analyzed by FacsCalibur (Becton-Dickinson, San Jose, CA).

Statistical analyses
A two-tailed Student’s t test was performed to determine the statistical significance of the percentage of nuclear exclusion cells. Data are mean ± SD.

RESULTS

E1A protein displays CRM1-dependent nuclear export activity
To test our hypothesis that the nuclear export of E1A depends on CRM1, we performed an immunoprecipitation assay with nuclear extract from 293 cells that constitutively express adenoviral E1 proteins, which showed that E1A coprecipitated with CRM1 (Fig. 1 A). Our finding was confirmed by the further observation that E1A accumulated in the nucleus of 293 cells treated with LMB (16 , 23 ; Fig. 1B ). To facilitate our study of the nucleocytoplasmic transport of E1A, we added an EGFP tag at the N terminus of E1A. Because of the finding in a previous study showing that a K -> A mutation in the C-terminal NLS changed the localization of the E1A protein from a nuclear to a pancellular localization (7) and to further override the strong nuclear import activity of E1A, we generated and tested the effect of an EGFP-E1AK fusion protein with the same mutation (K285A in 13s E1A; Fig. 1C ). LMB treatment resulted in the relocation of EGFP-E1AK into the nucleus (Fig. 1C ), strongly indicating the existence of an NES and other functional NLSs in E1A (10 , 11) . Consistent with these observations, siRNA targeting CRM1 induced the dramatic nuclear localization of EGFP-E1AK (Fig. 1D, E ). Taken together, these data show that E1A depends at least in part on CRM1 for nuclear export.


Figure 1
View larger version (38K):
[in this window]
[in a new window]

  		Figure 1. CRM1-dependent nuclear export activity of E1A. A) Coimmunoprecipitation of E1A and CRM1 from nuclear extract of 293 cells. mIgG, mouse IgG; Input, 2% of nuclear extract proteins used for immunoprecipitation. B) Nuclear accumulation of E1A induced by LMB in 293 cells. Cells were treated with 10 ng/ml LMB for 16 h. Then nuclear and cytoplasmic proteins were extracted and analyzed by immunoblotting (20 µg proteins per lane, left panel). The response of E1B to LMB was used as a positive control for LMB treatment. Lamin B and {alpha}-tubulin were used as loading controls for nuclear and cytoplasmic proteins, respectively. Nuclear/cytoplasmic (N/C) ratio of proteins is shown in the right panel. N/C ratios of LMB-untreated cells were regarded as 1. C) Subcellular localization of EGFP, EGFP-E1A and EGFP-E1AK fusion proteins in U-2 OS cells and response to LMB. 4',6'-diam idino-2-phenylidole stain was used to indicate the position of nucleus. D) CRM1-dependent cytoplasmic localization of EGFP-E1AK. U-2 OS cells were first transfected with 100 nM siRNAs for 24 h and then transfected with the EGFP-E1AK vector for another 24 h. Cells were stained with an anti-CRM1 antibody (Ab) and Texas Red-conjugated secondary Ab. CsiRNA, control siRNA; siCRM1, siRNA for CRM1. E) Down-regulation of CRM1 by siRNA. U-2 OS cells were transfected with 100 nM siRNA or mock transfected. Proteins in cell lysates were analyzed with immunoblotting 48 h after transfection. {alpha}-tubulin was used as loading control.

E1A protein contains a functional NES element spanning amino acids 70–80 within E1A protein
To identify the sequence within E1A responsible for nuclear export activity, we performed systematic mutational analyses followed by examination of the subcellular localization of EGFP-E1AK mutants with deletions of amino acids 1–51 ({Delta}aa1–51) or 52–91 ({Delta}aa52–91; Fig. 2 A). The {Delta}aa1–51 mutant assumed a predominantly cytoplasmic or pancellular pattern and accumulated in the nucleus after LMB treatment (Fig. 2A ). However, the {Delta}aa52–91 mutant resulted in the drastic relocation of the protein to the nucleus (Fig. 2A ), clearly indicating the existence of a putative NES within this region. Careful examination of the sequence of E1A within this region revealed a canonical NES sequence (VMLAVQEGIDL) spanning amino acids 70 to 80 that matched the putative consensus of NES (16 , 24 ; Fig. 2B ). Thus, since the {Delta}aa70–80 mutant with deletion of this candidate NES relocated to the nucleus (Fig. 2A ), it is highly likely that this E1A sequence is responsible for nuclear export activity. To confirm the NES activity of the identified sequence, we fused amino acids 62–81 to two EGFPs in tandem and examined the subcellular localization of the fusion protein. Consistent with the findings from previous experiments, the fusion protein showed complete nuclear exclusion that was blocked by LMB (Fig. 2A ).


Figure 2
View larger version (39K):
[in this window]
[in a new window]

  		Figure 2. Identification of a functional nuclear export element in E1A. A) Subcellular localization of different EGFP fusion proteins and the response to LMB are shown in the left panel. The right panel shows a schematic illustration of E1A deletions and EGFP tandem fused with the putative E1A region including NES sequence. B) Comparison of nuclear export signals identified in E1A, E1B, and HIV-1 Rev proteins with the putative NES consensus. {alpha} -Helix and coil indicate secondary structure in E1A NES. {Phi} indicates a large hydrophobic residue, such as leucine, isoleucine, valine, phenylalanine or methionine, and X indicates any amino acid (19) .

To determine the contribution of the hydrophobic residues within the NES to the nuclear export activity, we substituted alanine for these residues in the EGFP-E1AK fusion protein (Fig. 3 A). The L80A mutant, as well as V70A and L72A mutants (data not shown), either distributed diffusely or were restricted to the cytoplasm of the cells and were just as sensitive to LMB as was EGFP-E1AK (Fig. 3B ). In contrast, the V74A and I78A mutants were mainly restricted to the nucleus, indicating the abrogation of NES activity in these two mutants (Fig. 3B ). Thus, our results demonstrate that the E1A protein encompasses a functional NES element spanning amino acids 70–80 within which V74 and I78 are indispensable for the novel NES activity.


Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		Figure 3. Mutational analysis of E1A NES. A) Illustration of the alanine replacements in E1A NES. Numbers represent position of amino acids in E1A. *Indispensable hydrophobic amino acids for E1A NES activity. B) Subcellular localization of EGFP-E1AK with indicated mutations and response to LMB in U-2 OS cells. Shown are mutants of critical hydrophobic amino acids (V74 and I78) as well as a representative example (L80) of other hydrophobic amino acids in the NES.

E1A NES activity is regulated by phosphorylation at S89
It has been shown that phosphorylation in the vicinity of an NLS or NES may play a role in the intracellular distribution of proteins (25) . In E1A, the phosphorylation sites S89 and S96 are close to the NES (26 , 27 ; Fig. 4 A). Therefore, we were interested in ascertaining whether this novel NES was modulated by phosphorylation of these serine residues. To this end, we replaced S89 or S96 with alanine in the EGFP-E1AK fusion protein. Using nuclear exclusion of the EGFP fusion proteins as an indicator of NES activity, we showed that the S89A mutation resulted in a 3-fold reduction of the nuclear export activity of EGFP-E1AK (Fig. 4B ). The effect of this activity was specific to S89; S96 inactivation did not result in any noticeable change in the subcellular localization of EGFP-E1AK (Fig. 4B ; P=0.1). We next substituted glutamic acid for these two serine residues to partially mimic the negative charge of phosphorylation. The S89E mutation had the opposite effect of the S89A mutation on the subcellular localization of EGFP-E1AK (Fig. 4B ), further verifying the role of S89 in regulating E1A NES activity. The S96E mutation, however, did not change the subcellular distribution of EGFP-E1AK (Fig. 4B ).


Figure 4
View larger version (22K):
[in this window]
[in a new window]

  		Figure 4. Regulation of E1A NES activity by phosphorylation at S89. A) Sequence of E1A showing the NES sequence and vicinity region. Numbers represent positions of amino acids in E1A. S89 and S96 are 2 verified phosphorylation sites in E1A. B) Analysis of regulation of E1A NES activity by phosphorylation through mutation and roscovitine treatment. Mutations at S89 or S96 were introduced in EGFP-E1AK. Representative subcellular localization of EGFP fusion proteins shows that S89A mutation and roscovitine dramatically decreased nuclear exclusion of EGFP-E1AK (upper panel). For each fusion protein, percentage of cells showing nuclear exclusion of the EGFP fusion protein was calculated in four fields. Total number of cells counted for each experiment was ~250 (lower panel). Data are mean ± SD of cell numbers in the 4 fields.

Because E1A was phosphorylated at S89 by CDC2 and the phosphorylation of the protein was greatest in mitotic cells with maximal levels of CDC2 kinase activity (5 , 28) , we next studied whether the inhibition of CDC2 modified the subcellular localization of EGFP-E1AK. We observed that the CDC2/CDK2-specific inhibitor roscovitine indeed induced a 5-fold reduction in the percentage of cells showing the nuclear exclusion of EGFP-E1AK (Fig. 4B ), an effect similar to that observed with S89A mutant. Furthermore, while roscovitine did not modify the subcellular localization of the S89A mutant (P=0.5), the S89E mutant showed resistance to roscovitine compared with EGFP-E1AK (P<0.05; Fig. 4B ). We concluded that phosphorylation at S89 by CDC2 enhances NES activity.

E1A NES activity is relevant for the viral life cycle
To determine whether E1A NES is active during the adenovirus replication cycle, we monitored the localization of E1A for a period of 24 h after viral infection in U-2 OS cells. We observed that E1A was predominantly localized to the nucleus at an early stage of viral infection (8 h postinfection) but started to relocate to the cytoplasm at a late stage of infection (24 h postinfection) in the majority of the infected cells (Fig. 5 A). Incubation of infected cells with roscovitine and LMB inhibited the relocation of E1A to the cytoplasm (Fig. 5A ), indicating that the increased nuclear export activity caused by phosphorylation mediated the relocation of E1A to the cytoplasm at this time. Parallel experiments involving the examination of the cell-cycle profile of adenovirus-infected cells revealed that 24 h after infection, the cells accumulated in the S and G2/M phases (Fig. 5B ) characterized by high CDC2 and CDK2 kinase activity that can phosphorylate E1A proteins (5 , 28 29 30) . Consistently, immunoblotting analysis revealed that phosphorylated histone H1, an indicator of DNA replication and CDC2 and CDK2 activity (37 38 39) , was only detectable at 24 h after adenoviral infection (Fig. 5C ). Collectively, these data suggest that E1A is probably transferred to the cytoplasm as a result of NES activity enhanced by phosphorylation during a late stage of viral infection.


Figure 5
View larger version (24K):
[in this window]
[in a new window]

  		Figure 5. E1A nuclear export during viral infection. A) Subcellular localization of E1A in U-2 OS cells during adenoviral infection. Cells were infected with adenoviruses at 50 MOI. Sixteen hours postinfection (pi), cells were treated with 20 µM roscovitine or 20 ng/mL LMB for 8 h. The cells were then stained with monoclonal E1A Ab and Texas Red-conjugated secondary Ab. B) Cell cycle profile of U-2 OS cells during adenoviral infection. Cells that were mock infected or infected with viruses for 8 or 24 h were collected and the DNA content were analyzed through flow cytometry (upper panel). Percentage of cells in G1 phase and other phases of cell cycle is shown in lower panel. C) Phosphorylation of histone H1 during adenoviral infection. The phosphorylated histone H1 (p-Histone H1) in the U-2 OS cells at indicated time point after viral infection was analyzed by immunoblotting. {alpha}-tubulin was used as loading control.

Next, we were interested in examining whether NES plays a role in an efficient viral replication. Thus we constructed the adenovirus with the V74A mutation (AdV74A) to abrogate the NES activity. In serum-starved normal human lung fibroblasts MRC-5 cells, titration of the progenies of the viruses demonstrated that, 48 h after infection, the replication efficiency of AdV74A was up to 500-fold lower than Adwt (Fig. 6 A). In a parallel experiment, the difference between replication efficiency of the two viruses is much smaller in 293 cells where the WT E1A should compensate the functional alteration due to V74A mutation in AdV74A (Fig. 6A ), indicating the attenuated replication of AdV74A in the fibroblasts was mainly caused by the abrogation of NES activity of E1A instead of the initial viral infection dose difference. Coincident with the change in replication efficiency, immunostaining of E1A proteins in the infected fibroblasts showed that 48 h after infection, in majority of the infected cells, WT E1A relocated to the cytoplasm resulting in equal staining of E1A (60% of the cells) while the mutant E1A still displayed a predominant nuclear localization (75% of the cells; Fig. 6B, C ). The subcellular localization of E1A proteins was correlated with the stages of viral infection rather than the expression levels of the proteins (Fig. 6C, D ). Thus, we concluded that NES activity is required for an efficient adenoviral replication.


Figure 6
View larger version (25K):
[in this window]
[in a new window]

  		Figure 6. Requirement of NES activity for efficient adenoviral replication. A) Viral replication profile of Adwt and AdV74A in 293 cells and serum-starved normal human lung fibroblasts (MRC-5 cells). MRC-5 cells were starved in culture medium with 0.5% FBS since the time when cells were plated. Viral titer is expressed as logarithmic scale of pfu/ml. Shown are results from 2 independent experiments. B) Subcellular localization of E1A proteins in adenovirus-infected MRC-5 cells. Cells were plated in culture medium with 0.5% FBS. Twenty-four hours later, cells were infected with the indicated viruses at 100 multiplicity of infection for 48 h. Cells were then stained with monoclonal E1A Ab and Texas Red-conjugated secondary Ab. C) For each adenovirus, percentage of cells showing pancellular E1A was calculated in 5 fields at 24, 32, 40, and 48 h postinfection (pi). Total number of cells counted for each experiment was ~250. Data are mean ± SD of cell numbers in the 5 fields. D) Kinetics of E1A expressed by Adwt and AdV74A at time points indicated in C) was analyzed by immunoblotting (upper panel). {alpha}-tubulin was used as loading control. Relative expression level of E1As was normalized to {alpha}-tubulin. Expression level of Adwt at 24 h was regarded as 1 (lower panel).

DISCUSSION

In this study, we first reported a functional CRM1-dependent NES in the E1A protein of human adenovirus type 5 that is modulated by phosphorylation at an adjacent S89. We further demonstrated that this NES activity is up-regulated during the late stage of the virus life cycle and is important for potent viral replication. Our observations of E1A NES activity in combination with previous findings regarding the NLS of E1A suggest a model for the nucleocytoplasmic transport of E1A (Fig. 7 ). In this model, the E1A protein binds to importin-{alpha} 3 through NLS in the C-terminus to localize to the nucleus (7 , 8) , a process that is tightly regulated by acetylation at the lysine residue (K285 in 13s E1A) within the NLS (7) . We show here that E1A is exported to the cytoplasm through a CRM1-dependent pathway that is regulated by phosphorylation and is important for efficient viral replication (Fig. 6A ) and probably for E1A degradation (Fig. 6D ).


Figure 7
View larger version (14K):
[in this window]
[in a new window]

  		Figure 7. Proposed model of E1A nucleocytoplasmic transportation. E1A is synthesized in the cytoplasm and transported to the nucleus through acetylation-regulated binding to importin-{alpha}3. Conversely, nuclear E1A is relocated to cytoplasm by phosphorylation-regulated nuclear export through CRM1.

Within the sequence of E1A protein, there are other regions that match with the putative motif of NES, for instance, amino acids 43–51 (LHELYDLDV) and amino acids 249–257 (LCPIKPVAV). We deleted the regions and found that they were not involved in the nuclear export of E1A protein (data not shown). Analysis of E1A protein secondary structure using PSIPRED predicted an {alpha} -helix in the N-terminal part of the identified NES (VMLAVQE) and a random coil in the C-terminal part (GIDL; Fig. 2B ) (31) . The structure is consistent with the previous analysis of the leucine-rich NESs showing a clear tendency for the signals to be {alpha}-helical in the N-terminal part but bend off differently in the C-terminal end containing the last hydrophobic residue (24) . The other two candidates, however, fell in the middle of random coils, which could be the reason why they are not functional.

Although it has been repeatedly reported that S89 and S96 are phosphorylated in vivo (27 , 28) , the two serines map to a region of E1A the role of which is unclear (3) . The function regulated by the phosphorylation of these sites is also not clear (27) . The identification of a functional NES close to these sites clarified the functional relevance of the phosphorylation (Fig. 4) . Our data show that at the late phase of viral infection, E1A began to relocate to the cytoplasm, which correlates with the accumulation of the infected cells at the S and G2/M phases and histone H1 phosphorylation (Fig. 5) . This phenomenon can be explained by results of a previous report in which CDC2 phosphorylated E1A at S89 and in which E1A was phosphorylated at the highest level in vivo in mitotic cells, which display maximal levels of CDC2 kinase activity (28) .

Overall, E1A subcellular localization results from the dynamic movement of the protein, such as nuclear import, nuclear export, and diffusion. Moreover, the interaction of E1A with the proteins that have their own nucleocytoplasmic activity should also influence the localization of E1A. In addition, other posttranslational modifications affecting its shuttling could also contribute to the result. To this end, acetylation of the lysine in the C-terminal NLS locates the protein in cytoplasm (7) . Modulation of other NLS sites should also change the localization of E1A and the modulation could be related to the cell cycle. Considering all these factors, it is understandable that there is a variance in the subcellular localization of E1A in the cells (Figs. 4 , 6) . Nevertheless, NES activity can be demonstrated by statistically calculating the percentage of certain E1A distribution patterns in the cells (Figs. 4 , 6) .

Serum-starved cells represent an unfavorable environment for viral replication. The viral cycle was longer in serum-starved MRC-5 cells than in U-2 OS cells (Figs. 5 , 6) . The viruses need to activate and reprogram the host cells to achieve an efficient viral replication. Our data show that the active nuclear export of E1A is required for efficient viral replication in serum-starved normal human lung fibroblast (Fig. 6A ). Because E1A proteins physically interact with multiple cellular proteins involved in the regulation of gene transcription and the cell cycle (1 , 2) , it is possible that the virus uses the nuclear export activity of E1A to move some of the E1A-binding proteins from the nucleus to the cytoplasm. Because the majority of the identified E1A-binding proteins execute their functions in the nucleus (1 , 2) , moving them to the cytoplasm should further attenuate their control on transcription and the cell cycle, and thus creates a favorable environment for viral replication. On the other hand, the active nuclear export of E1A indicates that E1A may also exert its pleiotropic effects on the regulation of viral replication in part by affecting cytoplasmic processes through interacting with the cytoplasmic targets, such as Rack1 (32) , Yak-related kinases (33 , 34) , and the RII subunit of protein kinase A (34) . In addition, late in adenovirus infection, in the cytoplasm, the cytoskeleton is destroyed (35) and thus translocation of E1A to the cytoplasm could be required for efficient cell lysis.

Finally, the putative NES contains some sequence from CR1 and is also composed of non conserved sequence. The critical hydrophobic backbone of the NES is only preserved in human subgroup C adenoviruses (hAd2, 5, and 6) but not in other subgroups such as Ad12, in which the oncogenic function of E1A is more aggressively manifested (3 , 36) . Increased levels of phosphorylated signal transducer and activator of transcription-1 protein and expression of IFN regulatory factor-7 were found in Ad5- vs. Ad12-transformed BRK cells (36) , which could be the critical alteration caused by the nuclear export activity of E1A that leads to the plethora of oncogenic differences. Thus, the functional NES within Ad5 E1A could be one of the negative modulators for tumorigenesis.

ACKNOWLEDGMENTS

We thank Vickie J. Williams (Department of Scientific Publications, The University of Texas M.D. Anderson Cancer Center) for editorial assistance.

Received for publication June 14, 2006. Accepted for publication August 22, 2006.

REFERENCES

  1. Sang, N., Caro, J., Giordano, A. (2002) Adenoviral E1A: everlasting tool, versatile applications, continuous contributions and new hypotheses. Front. Biosci. 7,d407-d413[Medline]
  2. Frisch, S. M., Mymryk, J. S. (2002) Adenovirus-5 E1A: paradox and paradigm. Nat. Rev. Mol. Cell. Biol. 3,441-452[CrossRef][Medline]
  3. Avvakumov, N., Kajon, A. E., Hoeben, R. C., Mymryk, J. S. (2004) Comprehensive sequence analysis of the E1A proteins of human and simian adenoviruses. Virology 329,477-492[Medline]
  4. Whalen, S. G., Marcellus, R. C., Barbeau, D., Branton, P. E. (1996) Importance of the Ser-132 phosphorylation site in cell transformation and apoptosis induced by the adenovirus type 5 E1A protein. J. Virol. 70,5373-5383[Abstract/Free Full Text]
  5. Mal, A., Piotrkowski, A., Harter, M. L. (1996) Cyclin-dependent kinases phosphorylate the adenovirus E1A protein, enhancing its ability to bind pRb and disrupt pRb-E2F complexes. J. Virol. 70,2911-2921[Abstract]
  6. Whalen, S. G., Marcellus, R. C., Whalen, A., Ahn, N. G., Ricciardi, R. P., Branton, P. E. (1997) Phosphorylation within the transactivation domain of adenovirus E1A protein by mitogen-activated protein kinase regulates expression of early region 4. J. Virol. 71,3545-3553[Abstract]
  7. Madison, D. L., Yaciuk, P., Kwok, R. P., Lundblad, J. R. (2002) Acetylation of the adenovirus-transforming protein E1A determines nuclear localization by disrupting association with importin-alpha. J. Biol. Chem. 277,38755-38763[Abstract/Free Full Text]
  8. Kohler, M., Gorlich, D., Hartmann, E., Franke, J. (2001) Adenoviral E1A protein nuclear import is preferentially mediated by importin alpha3 in vitro. Virology 289,186-191[CrossRef][Medline]
  9. Lyons, R. H., Ferguson, B. Q., Rosenberg, M. (1987) Pentapeptide nuclear localization signal in adenovirus E1a. Mol. Cell. Biol. 7,2451-2456[Abstract/Free Full Text]
  10. Richter, J. D., Young, P., Jones, N. C., Krippl, B., Rosenberg, M., Ferguson, B. (1985) A first exon-encoded domain of E1A sufficient for posttranslational modification, nuclear-localization, and induction of adenovirus E3 promoter expression in Xenopus oocytes. Proc. Natl. Acad. Sci. U. S. A. 82,8434-8438[Abstract/Free Full Text]
  11. Standiford, D. M., Richter, J. D. (1992) Analysis of a developmentally regulated nuclear localization signal in Xenopus. J. Cell Biol. 118,991-1002[Abstract/Free Full Text]
  12. Fried, H., Kutay, U. (2003) Nucleocytoplasmic transport: taking an inventory. Cell. Mol. Life Sci. 60,1659-1688[CrossRef][Medline]
  13. Lo, H. W., Hung, M. C. (2006) Nuclear EGFR signalling network in cancers: linking EGFR pathway to cell cycle progression, nitric oxide pathway and patient survival. Br. J. Cancer 94,184-188[CrossRef][Medline]
  14. Wen, W., Meinkoth, J. L., Tsien, R. Y., Taylor, S. S. (1995) Identification of a signal for rapid export of proteins from the nucleus. Cell 82,463-473[CrossRef][Medline]
  15. Fischer, U., Huber, J., Boelens, W. C., Mattaj, I. W., Luhrmann, R. (1995) The HIV-1 Rev activation domain is a nuclear export signal that accesses an export pathway used by specific cellular RNAs. Cell 82,475-483[CrossRef][Medline]
  16. Fornerod, M., Ohno, M., Yoshida, M., Mattaj, I. W. (1997) CRM1 is an export receptor for leucine-rich nuclear export signals. Cell 90,1051-1060[CrossRef][Medline]
  17. Fontoura, B. M., Faria, P. A., Nussenzveig, D. R. (2005) Viral interactions with the nuclear transport machinery: discovering and disrupting pathways. IUBMB Life 57,65-72[Medline]
  18. Dobbelstein, M., Roth, J., Kimberly, W. T., Levine, A. J., Shenk, T. (1997) Nuclear export of the E1B 55-kDa and E4 34-kDa adenoviral oncoproteins mediated by a rev-like signal sequence. EMBO J. 16,4276-4284[CrossRef][Medline]
  19. Kratzer, F., Rosorius, O., Heger, P., Hirschmann, N., Dobner, T., Hauber, J., Stauber, R. H. (2000) The adenovirus type 5 E1B–55K oncoprotein is a highly active shuttle protein and shuttling is independent of E4orf6, p53 and Mdm2. Oncogene 19,850-857[CrossRef][Medline]
  20. Lund, E., Guttinger, S., Calado, A., Dahlberg, J. E., Kutay, U. (2004) Nuclear export of microRNA precursors. Science 303,95-98[Abstract/Free Full Text]
  21. Jones, N., Shenk, T. (1978) Isolation of deletion and substitution mutants of adenovirus type 5. Cell 13,181-188[CrossRef][Medline]
  22. Jiang, H., Alemany, R., Gomez-Manzano, C., Medrano, D. R., Lemoine, M. G., Olson, M. V., Alonso, M. M., Lee, O. H., Conrad, C. C., Yung, W. K., Fueyo, J. (2005) Downmodulation of E1A protein expression as a novel strategy to design cancer-selective adenoviruses. Neoplasia 7,723-729[CrossRef][Medline]
  23. Wolff, B., Sanglier, J. J., Wang, Y. (1997) Leptomycin B is an inhibitor of nuclear export: inhibition of nucleo-cytoplasmic translocation of the human immunodeficiency virus type 1 (HIV-1) Rev protein and Rev-dependent mRNA. Chem. Biol. 4,139-147[CrossRef][Medline]
  24. La Cour, T., Kiemer, L., Molgaard, A., Gupta, R., Skriver, K., Brunak, S. (2004) Analysis and prediction of leucine-rich nuclear export signals. Protein Eng. Des. Sel. 17,527-536[Abstract/Free Full Text]
  25. Jans, D. A., Hubner, S. (1996) Regulation of protein transport to the nucleus: central role of phosphorylation. Physiol. Rev. 76,651-685[Abstract/Free Full Text]
  26. Tremblay, M. L., Dumont, D. J., Branton, P. E. (1989) Analysis of phosphorylation sites in the exon 1 region of E1A proteins of human adenovirus type 5. Virology 169,397-407[CrossRef][Medline]
  27. Dumont, D. J., Marcellus, R. C., Bayley, S. T., Branton, P. E. (1993) Role of phosphorylation near the amino terminus of adenovirus type 5 early region 1A proteins. J. Gen. Virol. 74,583-595[Abstract/Free Full Text]
  28. Dumont, D. J., Branton, P. E. (1992) Phosphorylation of adenovirus E1A proteins by the p34cdc2 protein kinase. Virology 189,111-120[CrossRef][Medline]
  29. Doree, M., Hunt, T. (2002) From Cdc2 to Cdk1: when did the cell cycle kinase join its cyclin partner?. J. Cell Sci. 115,2461-2464[Abstract/Free Full Text]
  30. Padmanabhan, R., Tanimoto, A., Sasaguri, Y. (2003) Transactivation of human cdc2 promoter by adenovirus E1A. Curr. Top. Microbiol. Immunol. 272,365-397[Medline]
  31. Jones, D. T. (1999) Protein secondary structure prediction based on position-specific scoring matrices. J. Mol. Biol. 292,195-202[CrossRef][Medline]
  32. Sang, N., Severino, A., Russo, P., Baldi, A., Giordano, A., Mileo, A. M., Paggi, M. G., De Luca, A. (2001) RACK1 interacts with E1A and rescues E1A-induced yeast growth inhibition and mammalian cell apoptosis. J. Biol. Chem. 276,27026-27033[Abstract/Free Full Text]
  33. Zhang, Z., Smith, M. M., Mymryk, J. S. (2001) Interaction of the E1A oncoprotein with Yak1p, a novel regulator of yeast pseudohyphal differentiation, and related mammalian kinases. Mol. Biol. Cell. 12,699-710[Abstract/Free Full Text]
  34. Fax, P., Carlson, C. R., Collas, P., Tasken, K., Esche, H., Brockmann, D. (2001) Binding of PKA-RIIalpha to the Adenovirus E1A12S oncoprotein correlates with its nuclear translocation and an increase in PKA-dependent promoter activity. Virology 285,30-41[CrossRef][Medline]
  35. Mangel, W. F., Baniecki, M. L., McGrath, W. J. (2003) Specific interactions of the adenovirus proteinase with the viral DNA, an 11-amino-acid viral peptide, and the cellular protein actin. Cell. Mol. Life Sci. 60,2347-2355[CrossRef][Medline]
  36. Vertegaal, A. C., Kuiperij, H. B., Houweling, A., Verlaan, M., van der Eb, A. J., Zantema, A. (2003) Differential expression of tapasin and immunoproteasome subunits in adenovirus type 5- versus type 12-transformed cells. J. Biol. Chem. 278,139-146[Abstract/Free Full Text]
  37. Halmer, L., Gruss, C. (1996) Effects of cell cycle dependent histone H1 phosphorylation on chromatin structure and chromatin replication. Nucleic Acids Res. 24,1420-1427[Abstract/Free Full Text]
  38. Shimada, A., Ohsumi, K., Kishimoto, T. (1998) An indirect role for cyclin B-Cdc2 in inducing chromosome condensation in Xenopus egg extracts. Biol. Cell. 90,519-530[CrossRef][Medline]
  39. Alexandrow, M. G., Hamlin, J. L. (2005) Chromatin decondensation in S-phase involves recruitment of Cdk2 by Cdc45 and histone H1 phosphorylation. J. Cell Biol. 168,875-886[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Mol. Cell. Biol.Home page
A. Kulisz and H.-G. Simon
An Evolutionarily Conserved Nuclear Export Signal Facilitates Cytoplasmic Localization of the Tbx5 Transcription Factor
Mol. Cell. Biol., March 1, 2008; 28(5): 1553 - 1564.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Summary
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.06-6433fjev1
20/14/2603    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jiang, H.
Right arrow Articles by Fueyo, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jiang, H.
Right arrow Articles by Fueyo, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS