Heterogeneous nuclear ribonuclear protein K interacts ... - CiteSeerX

Journal of General Virology (2008), 89, 2540–2549

DOI 10.1099/vir.0.2008/003673-0

Heterogeneous nuclear ribonuclear protein K interacts with the enterovirus 71 59 untranslated region and participates in virus replication Jing-Yi Lin,1,2,3 Mei-Ling Li,4 Peng-Nien Huang,1,2,3 Kun-Yi Chien,3 Jim-Tong Horng2,3 and Shin-Ru Shih1,2,3 Correspondence Shin-Ru Shih [email protected]

1

Department of Medical Biotechnology and Laboratory Science, Chang Gung University, Taoyuan, Taiwan, ROC

2

Research Center for Emerging Viral Infections, Chang Gung University, Taoyuan, Taiwan, ROC

3

Graduate Program in Biomedical Science, Chang Gung University, Taoyuan, Taiwan, ROC

4

Department of Molecular Genetics, Microbiology and Immunology, UMDNJ-Robert Wood Johnson Medical School, Piscataway, NJ, USA

Received 29 April 2008 Accepted 18 June 2008

Enterovirus 71 (EV71) is a picornavirus that can cause severe neurological complications in children. Like other picornaviruses, the genomic RNA of EV71 contains a long 59 untranslated region (UTR). Cellular proteins interact with the EV71 59 UTR, and these interactions are important for virus replication. Using an RNA pull-down assay and proteomics approaches, this study identified the heterogeneous nuclear ribonucleoprotein K (hnRNP K) as one of the EV71 59 UTR-associated proteins. The interaction between hnRNP K and the 59 UTR was further confirmed by mapping the interaction regions to stem–loops I–II and IV in the 59 UTR. During EV71 infection, hnRNP K was enriched in the cytoplasm where virus replication occurs, whereas hnRNP K was localized in the nucleus in mock-infected cells. Viral yields were found to be significantly lower in hnRNP K knockdown cells and viral RNA synthesis was delayed in hnRNP K knockdown cells in comparison with negative-control cells treated with small interfering RNA. These results suggest that hnRNP K interacts with the EV71 59 UTR and participates in virus replication.

INTRODUCTION Enterovirus 71 (EV71) infections usually cause hand-footand-mouth disease or herpangina. However, EV71 has been implicated as the aetiological agent in several largescale outbreaks of severe neurological disorders in children worldwide (Ho et al., 1999; McMinn, 2002; McMinn et al., 2001). In recent years, an increase in EV71 epidemic activity has been noted throughout the Asia–Pacific region. Severe neurological complications, including brainstem encephalitis, meningitis, poliomyelitis and even death have occurred in these epidemics (McMinn, 2002). In 1998, an EV71 epidemic occurred in Taiwan, with the virus infecting over 120 000 people and killing 78 children (Ho et al., 1999). Many EV71 epidemics on a smaller scale also occurred after 1998 on the island (Chen et al., 2007). EV71, a member of the family Picornaviridae, is a positivestranded RNA virus (Brown & Pallansch, 1995). The viral RNA has a small protein called VPg covalently attached to its 59 end and is polyadenylated at its 39 terminus (Flanegan et al., 1977; Sarnow, 1989). The genomic RNA is around 7500 nt. The 59 untranslated region (UTR) is 2540

745 nt and is highly structured, containing a cloverleaf-like structure important for viral RNA synthesis and an internal ribosomal entry site (IRES) that is critical for the direction of viral mRNA translation (Thompson & Sarnow, 2003). Many cellular proteins interact with the picornavirus 59 UTR and regulate virus replication; for instance, poly(rC)binding protein (PCBP), an RNA-binding protein that contains three heterogeneous nuclear ribonucleoprotein K (hnRNP K) homology (KH) motifs, has been demonstrated to interact with the 59 UTR of poliovirus and rhinovirus (Walter et al., 1999, 2002). Various IRESspecific trans-acting factors, such as polypyrimidine tractbinding protein (PTB), PCBP, autoantigen La and upstream N-ras protein (Unr), have been reported to be functionally important for picornaviruses (Boussadia et al., 2003; Costa-Mattioli et al., 2004; Toyoda et al., 1994; Walter et al., 1999). Knowledge of cellular proteins that associate with the 59 UTR of picornaviruses would facilitate an understanding of virus–host interactions that are crucial molecular targets for antiviral drug development. In this study, streptavidin was used to pull down biotin-labelled EV71 59 UTR and its 0002-008/ G 2008 SGM Printed in Great Britain

hnRNP K interacts with the EV71 59 UTR

associated cellular proteins. Fourteen bands representing 11 proteins were identified as potential EV71 59 UTRassociated proteins. hnRNP K was one such protein and was chosen for further study. hnRNP K has been reported to be involved in many viral infections (Bryant et al., 2000; Burnham et al., 2007; Chang et al., 2001; Hsieh et al., 1998; Shimada et al., 2004; Zhang et al., 2008), but its role in picornavirus infection remains unclear. The interaction domains in both the viral RNA and the hnRNP K protein were mapped. The impact of hnRNP K on EV71 virus replication was also addressed.

METHODS In vitro transcription. The T7 promoter–EV71 59 UTR fragment

flanked by EcoRI sites was excised from the vector pCRII-TOPO. RNA transcripts were synthesized using a MEGAscript T7 kit (Ambion), following the protocol provided by the manufacturer. Biotinylated RNA was synthesized in a 20 ml MEGAscript transcription reaction by adding 1.25 ml 20 mM biotinylated UTP (Biotin-16UTP, which is a substrate for SP6, T3 and T7 RNA polymerases and can replace UTP in in vitro transcription for RNA labelling; Roche). Synthesized RNAs were purified using an RNeasy Protect Mini kit (Nobel) and analysed on 1 % agarose gels. Isolation of proteins associated with EV71 5§ UTR RNA sequences by affinity binding to biotinylated RNA. A reaction mixture containing 200 mg cell extract and 12.5 pmol biotinylated

EV71 59 UTR RNA probe was prepared. The mixture (with a final volume of 100 ml) contained 5 mM HEPES (pH 7.1), 40 mM KCl, 0.1 mM EDTA, 2 mM MgCl2, 2 mM dithiothreitol, 1 U RNasin and 0.25 mg heparin ml21 (RNA mobility shift buffer) and was incubated for 15 min at 30 uC and then added to 400 ml Streptavidin MagneSphere Paramagnetic Particles (Promega) for 10 min at room temperature to allow binding. The protein–RNA complexes were washed five times with the RNA mobility shift buffer without heparin. After the final wash, 30 ml 26 SDS-PAGE sample buffer was added to the beads and incubated for 10 min at room temperature to dissociate the proteins from the RNA. The sample containing the eluted proteins was then boiled and subjected to 8–16 % SDS-PAGE and visualized by silver staining or Western blotting. Proteins bands were excised and identified by in-gel trypsin digestion and analysed by Bruker Ultraflex matrix-assisted laser desorption/ionization time-offlight mass spectrometry (MALDI-TOF MS). Database-searching algorithm. After the masses derived from the

standards, trypsin, the matrix and keratins had been removed, the monoisotopic mass lists for each protonated peptide were used to search the database. Mass lists were exported to the Biotool 2.0 software package to perform peptide mass fingerprinting using Mascot (http:// www.matrixscience.com) algorithm scoring to identify the proteins. Western blot analysis. PVDF membranes were blocked and probed

with antibodies against hnRNP K (diluted 1 : 200; Santa Cruz Biotechnology), IGF-II mRNA-binding protein (IMP-1; diluted 1 : 500; Santa Cruz Biotechnology) and actin (diluted 1 : 5000; Chemicon). Co-immunoprecipitation. Cell extracts from EV71-infected RD cells

for use in co-immunoprecipitation assays were prepared at 6 h postinfection (p.i.). Lysates were pre-cleared by incubation on ice for 1 h with protein A–agarose (50 % in lysis buffer) bound to non-specific antibody. Non-specific complexes were pelleted by centrifugation at 10 000 g at 4 uC for 10 min. The supernatants were removed and used in the immunoprecipitation assay. Next, 100 ml pre-cleared lysate was http://vir.sgmjournals.org

diluted with 450 ml lysis buffer and then added to 15 ml hnRNP K antibody, followed by incubation on ice for 2 h. Pre-washed protein A– agarose (100 ml in PBS; 50 : 50) was added to each sample, which was then incubated on ice for 1 h. Immune complexes were pelleted by centrifugation at 1000 g at 4 uC for 5 min and washed three times with lysis buffer. Each pellet [or 100 ml pre-cleared lysate (total RNA)] was resuspended in 400 ml proteinase K buffer [100 mM Tris/HCl (pH 7.5), 12.5 mM EDTA, 150 mM NaCl, 1 % SDS] and incubated with 100 mg pre-digested proteinase K for 30 min at 37 uC. RNA was extracted with phenol/chloroform, precipitated in ethanol at 220 uC for 1 h, washed in 70 % ethanol, dried and resuspended in 20 ml DEPC H2O. Fluorescence microscopy analysis. RD cells grown on glass

coverslips were infected with EV71 for 1 h at an m.o.i. of 40. At 6 h p.i., the culture medium was removed and the cells were washed and fixed. The cells were then permeabilized in 5 % Triton X-100 at room temperature for 5 min. For hnRNP K and EV71 3A immunostaining, the samples were blocked in PBS containing 5 % BSA for 60 min at room temperature and then incubated with anti-hnRNP K antibody (diluted 1 : 200) and anti-EV71 3A antibody (diluted 1 : 200) for 1.5 h at room temperature. The samples were then reacted with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG or tetramethyl rhodamine isothiocyanate (TRITC)-conjugated goat antirabbit IgG (Jackson ImmunoResearch Laboratories) for 1 h at room temperature. After washing with PBS, the samples were treated with the nuclear stain Hoechst 33258 for 15 min and washed again three times with PBS. Images were captured using a confocal laser-scanning microscope (LSM 510 META; Zeiss). Virus growth and plaque assay. RD cells were transfected with hnRNP K and negative-control (NC) small interfering RNA (siRNA) for 48 h, trypsinzed and then counted. Cells (26105) were seeded into 12-well plates. After 24 h, the cells were challenged with EV71 (strain 4643/TW/1998) at an m.o.i. of 40 or 0.1 p.f.u. per cell. At various times p.i., the supernatants of cell culture medium and cell lysates were collected to determine viral titres by plaque assay on RD cells. The sequences of siRNA used were as follows: hnRNP K siRNA: 59-AAUUCCUCCUGCUAGACUCUGAUGA-39; NC siRNA: 59-AACUGGGUAAGCGGGCGCAAAUU-39; IMP-1 siRNA: 59UACUGUACCAUACUGAGCCAGCAGG-39. Evaluation of RNA replication by slot blotting. RD cells were

transfected with hnRNP K and NC siRNA for 48 h and then trypsinzed and counted. The cells (26105) were seeded into 12-well plates. After 24 h, the cells were challenged with EV71 (strain 4643/ TW/1998) at an m.o.i. of 40 and harvested at 2, 3, 4, 5, 6, 7, 8 and 9 h p.i. RNA were extracted and dissolved in 206 SSC containing formaldehyde for 30 min at 60 uC. The reaction was then loaded onto a nitrocellulose membrane in the slot-blot manifold. After washing twice, the membrane was removed, air dried and cross-linked in a Stratalinker (Stratagene) at 200 J for 9 min. The membrane was prehybridized at 68 uC for 30 min in DIG Easy Hyb (Roche). Digoxygenin (DIG)-labelled RNA probes, specific for the genome or anti-genome, were produced using a DIG Northern Starter kit (Roche). After addition of the probes at 100 ng ml21, the blots were incubated at 68 uC for 16 h. After the hybridization, the membrane was immediately submerged in a tray containing low-stringency buffer (26 SSC containing 0.1 % SDS) at room temperature for 5 min with shaking. The blot was then incubated twice (15 min each with shaking) in high-stringency buffer (0.16 SSC containing 0.1 % SDS) at 68 uC. The membrane was then incubated with Washing Buffer (Roche) for 2 min at room temperature with shaking. After the membrane had been blocked with Blocking Solution (Roche) for 30 min, it was incubated with alkaline phosphatase-conjugated antiDIG antibody solution for 30 min and then washed twice with Maleic Acid Buffer (0.1 M maleic acid, 0.15 M NaCl, 0.3 % Tween 20, pH 7.5; Roche). It was then equilibrated for 5 min in 20 ml Detection 2541

J.-Y. Lin and others Buffer [0.1 M Tris/HCl (pH 9.5), 0.1 M NaCl; Roche]. Finally, chemiluminescent substrate (CDP-Star; Roche) was added and the membrane was exposed to Kodak film.

RESULTS Interaction of hnRNP K with EV71 viral RNA in the 5§ UTR In order to detect cellular proteins interacting with EV71 viral RNA in the 59 UTR, which is important for virus replication, streptavidin beads were used to capture the fulllength, biotin-labelled EV71 59 UTR and associated cellular proteins. Fig. 1(a) outlines the design of the pull-down assay. In comparison with the control reactions of no RNA, biotin16-UTP only and unlabelled EV71 59 UTR (Fig. 1b, lanes 1– 3), 14 bands of distinct sizes (Fig. 1b, lane 4) were observed in the mixture of biotinylated EV71 59 UTR RNA and RD cell lysate, which may represent cellular proteins that specifically associate with the EV71 59 UTR. These protein bands were excised, digested with trypsin and subjected to MALDI-TOF MS analysis. The results are shown in Table 1 with their accession numbers obtained from the NCBI protein database. They included four proteins known to interact with the IRES of picornaviruses: Unr (band 2), PTB1 and PTB-2 (bands 8-1, 8-2, 9-1 and 9-2), PCBP2 (bands 11 and 12) and PCBP1 (band 13). Band 7 representing hnRNP K was chosen for further investigation in this study, as many studies have demonstrated that hnRNP K participates in viral infection (Bryant et al., 2000; Burnham et al., 2007; Chang et al., 2001; Hsieh et al., 1998; Shimada et al., 2004; Zhang et al., 2008), but the role of hnRNP K in picornavirus replication remains to be determined. The specific interaction between hnRNP K and the EV71 59 UTR was further verified by a competition assay and Western blotting. As shown in Fig. 1(c), this interaction was outcompeted by non-biotinylated EV71 59 UTR (Fig. 1c, lanes 3–5) but not by yeast tRNA (Fig. 1c, lanes 8–10) or EV71 VP1 RNA (Fig. 1c, lanes 13–15). The above reaction was performed using uninfected RD cell lysate. To confirm the interaction also occurred in infected cells, anti-hnRNP K antibody was used to pull down the EV71 59 UTR. The pull-down reaction was subjected to RNA extraction and analysis. The RT-PCR results obtained using EV71 59 UTR-specific primers are shown in Fig. 1(d, lanes 1–4). EV71 59 UTR RNA was pulled down from infected-cell lysate using anti-hnRNP K antibody. No specific band representing EV71 59 UTR was detected when the reaction proceeded without anti-hnRNP K antibody. As a negative control, primers specific to ribosomal protein S16 (RPS16) were used for RT-PCR (Fig. 1d, lanes 5–8). hnRNP K localizes in the cytoplasm where EV71 replication occurs hnRNP K is a nuclear protein, whereas EV71 replication occurs in the cytoplasm. EV71-infected cells were fixed at 2542

6 h p.i. and stained with antibody against hnRNP K to determine its location. The results obtained by confocal microscopy clearly demonstrated that hnRNP K was localized in the nucleus of mock-infected cells (Fig. 2, upper panels); however, hnRNP K was enriched in the cytoplasm in EV71-infected cells (Fig. 2, lower panels). Staining with antibody against viral 3A protein was used to identify EV71-infected cells. Identification of interacting regions between the EV71 5§ UTR and hnRNP K protein To confirm and elucidate in greater detail the interaction between hnRNP K and the EV71 59 UTR, the binding regions in the viral RNA and the hnRNP K protein were mapped. Fig. 3(a) shows the proposed secondary structure for the EV71 59 UTR as predicted by Mfold. Various truncated forms of viral RNA as indicated in Fig. 3(b) were synthesized by in vitro transcription and labelled using biotin. The pull-down assay was used as described above using streptavidin beads to capture the biotinylated viral RNA and associated cellular proteins from RD cells. Western blotting was then carried out to determine which regions of hnRNP K were interacting. Probes covering nt 1–167, 91–445, 91–561, 91–636, 91–745 and 242–445 (Fig. 3b, lanes 4, 8, 10, 12, 14 and 16, respectively) pulled down hnRNP K from RD cell lysate, whilst the other probes did not. These results suggest that hnRNP K may interact with the regions nt 1–167 (stem–loops I and II) and nt 242–445 (stem–loop IV) of the EV71 59 UTR. To determine which functional domains of hnRNP K interact with the EV71 59 UTR, plasmids expressing different truncated forms of hnRNP K fused to the Flag epitope (Fig. 4a) were constructed. These plasmids were transfected individually into RD cells and the cell lysates were used in pull-down assays. Expression of full-length hnRNP K and its truncated forms was visualized by Western blotting using anti-Flag antibody (Fig. 4b, lanes 1, 4, 7, 10 and 13). Streptavidin was shown to capture the biotinylated EV71 59 UTR and its associated full-length hnRNP K (Fig. 4b, lane 3), as well as the two truncated forms of nt 1–385 and 144–465 (Fig. 4b, lanes 12 and 15), but not the forms of nt 1–92 and 1–197 (Fig. 4b, lanes 6 and 9). These results suggest that hnRNP K may interact with the EV71 59 UTR through a region containing KH2 and the proline-rich domain plus one neighbouring KH domain. Lower virus replication rates in hnRNP K knockdown cells After the interaction between hnRNP K and EV71 59 UTR had been confirmed, we wanted to determine whether hnRNP K plays a role in EV71 replication. siRNA specific for hnRNP K was used to knock down its expression (Fig. 5a, lower panels). As described in Methods, siRNAtransfected cells were collected, counted and reseeded Journal of General Virology 89


into plates for further infection. RD cells that had been treated with NC siRNA or hnRNP K siRNA were then challenged with a high (m.o.i. of 40) or low (m.o.i. of 0.1) titre of EV71. Viral yields were measured at various times p.i. by plaque assay. Lower replication rates were observed in hnRNP K knockdown cells than in NC siRNA-treated cells (Fig. 5a), indicating that hnRNP K may play a positive role in EV71 replication. As a control, siRNA against IMP-1 (another cellular protein identified http://vir.sgmjournals.org

in Table 1) was used to treat the cells. The EV71 replication rate did not differ significantly in IMP-1 siRNAtransfected cells (Fig. 5b). Viral RNA synthesis is delayed when hnRNP K is knocked down hnRNP K was shown to interact with nt 1–167 of the 59 UTR, reported to be involved in viral RNA replication. 2543

J.-Y. Lin and others

Fig. 1. Interaction of hnRNP K with the EV71 59 UTR. (a) Outline of the pull-down assay. (b) Pull down of hnRNP K with the EV71 59 UTR in RD cell lysate. The EV71 59 UTR was labelled with biotin and incubated with 200 mg RD cell lysate for 15 min at 30 6C. Streptavidin was added to pull down the biotin-labelled RNA (EV71 59 UTR) and associated cellular proteins (lane 4). After they had been washed and dissociated from RNA, the eluted proteins were boiled, subjected to SDS-PAGE (8–16 % gradient gel) and silver stained. Negative controls: lane 1, no RNA; lane 2, biotin-16-UTP only; lane 3, unlabelled RNA. Comparison with the three control lanes revealed a number of protein bands that appeared only in lane 4. These bands were excised, digested in gel and analysed by MALDI-TOF MS. (c) A specific association between hnRNP K and the EV71 59 UTR was confirmed by Western blotting and a competition assay. Various amounts of unlabelled RNA were added to compete with the biotin-labelled EV71 59 UTR interacting with hnRNP K. A pull-down assay was then carried out as described in (a) and the eluted proteins were separated by 12 % SDS-PAGE. Antibody against hnRNP K was used in the Western blot analysis. Lanes 1, 6 and 11 contained cell lysate (200 mg) only. Increasing amounts of unlabelled RNA were added as follows: lanes 3–5, unlabelled EV71 59 UTR; lanes 8–10, unlabelled yeast tRNA; lanes 13–15, unlabelled EV71 VP1 RNA (891 nt). (d) EV71 59 UTR RNA was pulled down with hnRNP K from EV71-infected cell lysate. RD cells were infected with EV71 and the cell lysate was collected at 6 h p.i. Anti-hnRNP K antibody was used in the immunoprecipitation assay. Following washing and dissociation, the RNA was extracted and subjected to the RT-PCR using EV71 59 UTR-specific or RPS16-specific primers. Lanes: 1, cell lysate without immunoprecipitation as a positive RT-PCR control; 2, anti-hnRNP K antibody incubated with 200 mg infected-cell lysate; 3, negative control with no antibody; 4, negative control using water as template.

Here, viral RNA synthesis in hnRNP K knockdown cells was investigated further. Viral RNA synthesis in hnRNP K siRNA-treated cells was compared with that in NC siRNAtreated cells. The results in Fig. 6 (lower right panel) showed that hnRNP K was knocked down by its specific siRNA. EV71 (m.o.i. of 40) was used to infect cells and viral RNAs were extracted at different times p.i. A slot-blot assay using a specific RNA probe against either positive-sense or negative-sense EV71 viral RNA was employed to monitor viral RNA synthesis. The results in Fig. 6 demonstrated that the synthesis of both positive- and negative-strand EV71 viral RNA was delayed in hnRNP K knockdown cells. No significant difference in actin RNA amounts between NC and hnRNP K siRNA-treated cells was observed. These results suggest that hnRNP K may enhance viral RNA synthesis.

DISCUSSION Host-cell proteins are regarded as important to picornavirus replication. For instance, cellular PCBPs are required for poliovirus RNA synthesis. PCBP binds to the cloverleaflike secondary structure in the 59 UTR of the positivestranded poliovirus RNA, which is necessary for the binding of viral protein 3CD to the opposite side of the same cloverleaf (Murray et al., 2001; Parsley et al., 1997; Silvera et al., 1999). Another cellular protein critical for the synthesis of poliovirus RNA is poly(A)-binding protein (PABP). PABP interacts with PCBP, 3CD and the viral poly(A) tail to form a ribonucleoprotein complex that is required for negative-strand RNA synthesis (Herold & Andino, 2001; Wang et al., 1999). hnRNP C1 has also been found to interact with a secondary structure element that is

Table 1. MALDI-TOF MS results of cellular proteins associating with the EV71 59 UTR Band number 1 2 3 4 5 6 7 8-1 8-2 9-1 9-2 10 11 12 13 14

2544

Protein name TBP-associated factor TAFII 150 (TBP-AFII 150) N-ras upstream protein (Unr) Glycyl-tRNA synthetase (G tRNA synthetase) Glycyl-tRNA synthetase (G tRNA synthetase) Keratin IGF-II mRNA-binding protein 1 (IMP-1) Heterogeneous nuclear ribonucleoprotein K (hnRNP K) Polypyrimidine tract-binding protein 1 (PTB-1) Polypyrimidine tract-binding protein 2 (PTB-2) Polypyrimidine tract-binding protein 1 (PTB-1) Polypyrimidine tract-binding protein 2 (PTB-2) ErbB3-binding protein (EBP1) Poly(rC)-binding protein 2 (PCBP2) Poly(rC)-binding protein 2 (PCBP2) Poly(rC)-binding protein 1 (PCBP1) Heterogeneous nuclear ribonucleoprotein A1 (hnRNP A1)

NCBI accession no.

Sequence coverage (%)

Mass (Da)

AAC13540 O75534 P41250 P41250

12 20 31 33

138 380 89 684 83 828 83 828

AAD09826 P61978 Q9BUQ0 S23016 Q9BUQ0 S23016 Q9UQ80 Q6IPF4 Q6IPF4 S58529 Q6IPF2

25 37 37 38 45 45 34 30 22 31 43

63 783 51 281 59 767 59 712 59 767 59 712 43 970 39 026 39 026 37 987 34 273

Journal of General Virology 89


Fig. 2. hnRNP K is located in the cytoplasm of EV71-infected cells. RD cells were mock-infected (upper panels) or infected with EV71 at an m.o.i. of 40 (lower panels). At 6 h p.i., cells were fixed with formaldehyde, washed and immunostained with antibody against hnRNP K or EV71 3A protein. FITC-conjugated goat anti-mouse IgG or TRITC-conjugated goat anti-rabbit IgG was used as secondary antibody. Hoechst 33258 was used to stain the nucleus. Images were captured by confocal laserscanning microscopy. A phase-contrast image is also shown.

predicted to form at the 39 end of the poliovirus negativestrand RNA and is functionally important for viral RNA synthesis (Brunner et al., 2005). In the present study of EV71 replication, another cellular protein, hnRNP K, was identified as interacting with the first 167 nt of the positive-stranded RNA genome. EV71 viral RNA synthesis was decreased when expression of the host cellular hnRNP K was shut down by siRNA. Whether hnRNP K interacts with other cloverleaf-associated proteins requires investigation to determine in detail the mechanism by which hnRNP K influences EV71 viral RNA synthesis.

In addition to the first 167 nt, hnRNP K also interacted with nt 242–445 of the EV71 59 UTR. Based on the RNA secondary structure predicted by Mfold, nt 242–445 may form stem–loop IV within the IRES, which is crucial for viral protein synthesis. hnRNP K participates in c-myc IRES activity (Evans et al., 2003), but has not been thought to be essential for poliovirus and rhinovirus viral protein translation (Choi et al., 2004). Our work did not clarify why hnRNP K interacts with the region nt 242–445. It may stabilize EV71 viral RNA through this interaction and thereby promote viral RNA synthesis.

Fig. 3. Determination of the regions of the EV71 59 UTR that interact with hnRNP K. (a) Mfold software was used to predict the RNA secondary structure of the EV71 59 UTR. The first and the last nucleotides in each stem– loop are numbered. (b) Analysis of the regions of interaction in the EV71 59 UTR. Various truncated forms of the RNA, as indicated, were transcribed in vitro and biotinylated. RD cell lysates were incubated with these biotinlabelled RNAs and non-biotinylated RNA probes were used as controls. After being pulled down by streptavidin, the protein complex was separated by SDS-PAGE as described in Methods and Western blotting was carried out to detect hnRNP K in the pulled-down complex. http://vir.sgmjournals.org

2545


Fig. 4. Determination of domains in hnRNP K that interact with the EV71 59 UTR. (a) hnRNP K domains. The hnRNP K protein comprises three KH domains and a proline-rich domain. Four truncated forms, hnRNP K1–92, hnRNP K1–197, hnRNP K1–385 and hnRNP K144–465, were generated and fused with the Flag epitope at their N terminus. (b) Expression of the truncated forms of hnRNP K in RD cells and mapping of the regions interacting with the EV71 59 UTR. Plasmids carrying wild type (lane 1) or the truncated forms of hnRNP K (lanes 4, 7, 10 and 13) were transfected into RD cells. Western blotting using anti-Flag antibody was used to examine protein expression levels. Cell extracts from transfected cells were collected at 48 h post-transfection and incubated with biotinylated RNA (EV71 59 UTR). Unlabelled RNA was used as a negative control. Streptavidin was used in the pull-down assay and the complex was analysed by SDSPAGE.

hnRNP K is known to be a pre-mRNA-binding protein and shuttles intermediates between the nucleus and cytoplasm. It is involved in diverse molecular and cellular functions such as transcription, translation and nuclear–cytoplasmic shuttling (Bomsztyk et al., 2004). Our work revealed that hnRNP K localized mainly in the nucleus in mock-infected cells, whilst the protein was found mainly in the cytoplasm of EV71-infected cells at 6 h p.i. According to the results of a [35S]Met-labelling experiment (data not shown), viral protein synthesis dominated around this time point. Accordingly, hnRNP K becomes enriched in the cytoplasm when virus replication is proceeding. hnRNP K also reportedly changes its localization following Sindbis virus infection (Burnham et al., 2007). It is not clear how hnRNP K moves from the nucleus to the cytoplasm when the host cells are challenged by a virus. Identification of viral or host factors that participate in the mechanism of hnRNP K transportation would be of interest. Recently, hnRNP K was also found to be important in regulating viral infection. For example, the overexpression of hnRNP K augmented hepatitis B virus (HBV) replication, whereas gene silencing of endogenous hnRNP K resulted in a significant reduction in the HBV viral load. Cytidine deaminase APOBEC3B interacts with hnRNP K 2546

and suppresses HBV expression (Zhang et al., 2008). hnRNP K has also been demonstrated to interact with Sindbis virus non-structural protein and viral subgenomic mRNA, and a reduction in hnRNP K expression in HeLa cells by siRNA treatment weakens expression of green fluorescent protein driven by a viral subgenomic promoter (Burnham et al., 2007). In addition to these examples, hnRNP K also interacts with several viral proteins, including human herpesvirus 6 immediate-early protein 2 (Shimada et al., 2004), dengue virus core protein (Chang et al., 2001), herpes simples virus type 1 IE63 protein (Bryant et al., 2000) and hepatitis C virus core protein (Hsieh et al., 1998). The hnRNP K protein has three KH domains. A prolinerich domain is flanked by the KH2 and KH3 domains. KH stands for K homology, with reference to the initial identification of the motif in hnRNP K (Adinolfi et al., 1999; Messias & Sattler, 2004; Siomi et al., 1993; Tomonaga & Levens, 1995). The KH domain is a 45–55 aa motif that has been shown to have nucleic acid-binding activity. Numerous RNA-binding proteins contain a KH domain. However, they exhibit a diverse architecture (Dejgaard & Leffers, 1996; Gibson et al., 1993; Leopoldino et al., 2007; Siomi et al., 1993). Moreover, various KH domains bind to Journal of General Virology 89


Fig. 5. EV71 exhibits a lower growth rate in hnRNP K knockdown cells but not in IMP-1 knockdown cells. (a) RD cells were treated with siRNA against hnRNP K for 48 h and then challenged with EV71 at an m.o.i. of 40 or 0.1 as indicated. At various times p.i., virus was harvested and the titre was measured by plaque assay. Experiments were carried out in triplicate and the results are shown as means±SD. The lower panels demonstrated that hnRNP K was knocked down following siRNA treatment. Cell extracts were collected at 6 h p.i. for Western blot analysis. (*P,0.05 and **P,0.01, Student’s two-tailed unpaired t-test). (b) Experiments were performed as in (a) using IMP-1 knockdown cells.

different nucleic acids. Here, we found that the interaction domain of hnRNP K with the EV71 59 UTR is located in the region containing KH2 and the proline-rich domain plus one neighbouring KH domain. The KH domain is responsible for RNA binding (Siomi et al., 1993), whilst the proline-rich domain is considered to be critical for the interaction of hnRNP K with other proteins. hnRNP K may bind to the EV71 59 UTR directly or through other viral and/or cellular proteins. hnRNP K is not only important for virus replication; it has also frequently been reported to play a crucial role in cancer biology. A loss-of-function screening assay by randomized intracellular antibodies has demonstrated that hnRNP K is a potential target for metastasis (Inoue et al., 2007). Enhanced interaction between hnRNP K and nucleolin regulates gastrin mRNA turnover, which is related to gastrointestinal tract malignancies (Lee et al., 2007). hnRNP K drives translational activation of http://vir.sgmjournals.org

specifically silenced mRNAs, such as the L2 mRNA of human papillomavirus type 16 in squamous epithelial cells (Collier et al., 1998). These cited studies highlight the importance of hnRNP K in disease formation. We believe that further investigation of the details of the mechanism by which the hnRNP K switches among its multiple roles in various cells under different circumstances would be helpful in the development of therapeutic strategies against cancers and infectious diseases.

ACKNOWLEDGEMENTS The authors would like to thank the National Science Council of the Republic of China, Taiwan, for financially supporting this research under contract no. NSC-94-IDP002-1. The Proteomic Center of Chang Gung University, Taiwan, is appreciated for its technical support, especially Dr Jau-Song Yu. Ted Knoy and Dr Guang-Wu Chen are thanked for their editorial assistance. 2547


Fig. 6. Viral EV71 RNA synthesis is decreased in hnRNP K knockdown cells. RD cells were transfected with hnRNP K and NC siRNA for 48 h and the cells reseeded into plates. After 24 h, the cells were infected with EV71 (strain 4643/TW/1998) at an m.o.i. of 40 and RNA was extracted at different times p.i. The RNA was loaded onto a nitrocellulose membrane in a slot-blot manifold. After washing, the membrane was air dried, cross-linked and incubated with pre-hybridization solution. DIG-labelled RNA probes specific for either the genome or anti-genome were added and incubated. Following washing and blocking, the blots were incubated with chemiluminescent substrate for visualization. Y, Yeast tRNA as a negative control. The bar charts show the quantitative results of the slot blot.

REFERENCES

Chang, C. J., Luh, H. W., Wang, S. H., Lin, H. J., Lee, S. C. & Hu, S. T. (2001). The heterogeneous nuclear ribonucleoprotein K (hnRNP K)

Adinolfi, S., Bagni, C., Castiglione Morelli, M. A., Fraternali, F., Musco, G. & Pastore, A. (1999). Novel RNA-binding motif: the KH

interacts with dengue virus core protein. DNA Cell Biol 20, 569–577.

module. Biopolymers 51, 153–164. Bomsztyk, K., Denisenko, O. & Ostrowski, J. (2004). hnRNP K: one

protein multiple processes. Bioessays 26, 629–638. Boussadia, O., Niepmann, M., Creancier, L., Prats, A. C., Dautry, F. & Jacquemin-Sablon, H. (2003). Unr is required in vivo for efficient

initiation of translation from the internal ribosome entry sites of both rhinovirus and poliovirus. J Virol 77, 3353–3359. Brown, B. A. & Pallansch, M. A. (1995). Complete nucleotide sequence

of enterovirus 71 is distinct from poliovirus. Virus Res 39, 195–205. Brunner, J. E., Nguyen, J. H., Roehl, H. H., Ho, T. V., Swiderek, K. M. & Semler, B. L. (2005). Functional interaction of heterogeneous nuclear

ribonucleoprotein C with poliovirus RNA synthesis initiation complexes. J Virol 79, 3254–3266. Bryant, H. E., Matthews, D. A., Wadd, S., Scott, J. E., Kean, J., Graham, S., Russell, W. C. & Clements, J. B. (2000). Interaction

between herpes simplex virus type 1 IE63 protein and cellular protein p32. J Virol 74, 11322–11328.

Chen, K. T., Chang, H. L., Wang, S. T., Cheng, Y. T. & Yang, J. Y. (2007). Epidemiologic features of hand-foot-mouth disease and

herpangina caused by enterovirus 71 in Taiwan, 1998–2005. Pediatrics 120, e244–e252. Choi, K., Kim, J. H., Li, X., Paek, K. Y., Ha, S. H., Ryu, S. H., Wimmer, E. & Jang, S. K. (2004). Identification of cellular proteins enhancing

activities of internal ribosomal entry sites by competition with oligodeoxynucleotides. Nucleic Acids Res 32, 1308–1317. Collier, B., Goobar-Larsson, L., Sokolowski, M. & Schwartz, S. (1998). Translational inhibition in vitro of human papillomavirus

type 16 L2 mRNA mediated through interaction with heterogenous ribonucleoprotein K and poly(rC)-binding proteins 1 and 2. J Biol Chem 273, 22648–22656. Costa-Mattioli, M., Svitkin, Y. & Sonenberg, N. (2004). La autoanti-

gen is necessary for optimal function of the poliovirus and hepatitis C virus internal ribosome entry site in vivo and in vitro. Mol Cell Biol 24, 6861–6870. Dejgaard, K. & Leffers, H. (1996). Characterisation of the nucleic-

Burnham, A. J., Gong, L. & Hardy, R. W. (2007). Heterogeneous nuclear

acid-binding activity of KH domains. Different properties of different domains. Eur J Biochem 241, 425–431.

ribonuclear protein K interacts with Sindbis virus nonstructural proteins and viral subgenomic mRNA. Virology 367, 212–221.

Evans, J. R., Mitchell, S. A., Spriggs, K. A., Ostrowski, J., Bomsztyk, K., Ostarek, D. & Willis, A. E. (2003). Members of the poly (rC) binding

2548

Journal of General Virology 89

hnRNP K interacts with the EV71 59 UTR protein family stimulate the activity of the c-myc internal ribosome entry segment in vitro and in vivo. Oncogene 22, 8012–8020.

Murray, K. E., Roberts, A. W. & Barton, D. J. (2001). Poly(rC) binding

Flanegan, J. B., Petterson, R. F., Ambros, V., Hewlett, N. J. & Baltimore, D. (1977). Covalent linkage of a protein to a defined

Parsley, T. B., Towner, J. S., Blyn, L. B., Ehrenfeld, E. & Semler, B. L. (1997). Poly(rC) binding protein 2 forms a ternary complex with the

nucleotide sequence at the 59-terminus of virion and replicative intermediate RNAs of poliovirus. Proc Natl Acad Sci U S A 74, 961–965.

59-terminal sequences of poliovirus RNA and the viral 3CD proteinase. RNA 3, 1124–1134.

Gibson, T. J., Thompson, J. D. & Heringa, J. (1993). The KH domain

Sarnow, P. (1989). Role of 39-end sequences in infectivity of

occurs in a diverse set of RNA-binding proteins that include the antiterminator NusA and is probably involved in binding to nucleic acid. FEBS Lett 324, 361–366.

poliovirus transcripts made in vitro. J Virol 63, 467–470.

proteins mediate poliovirus mRNA stability. RNA 7, 1126–1141.

Shimada, K., Kondo, K. & Yamanishi, K. (2004). Human herpesvirus

Herold, J. & Andino, R. (2001). Poliovirus RNA replication requires

6 immediate-early 2 protein interacts with heterogeneous ribonucleoprotein K and casein kinase 2. Microbiol Immunol 48, 205–210.

genome circularization through a protein–protein bridge. Mol Cell 7, 581–591.

Silvera, D., Gamarnik, A. V. & Andino, R. (1999). The N-terminal K

Ho, M., Chen, E. R., Hsu, K. H., Twu, S. J., Chen, K. T., Tsai, S. F., Wang, J. R. & Shih, S. R. (1999). An epidemic of enterovirus 71

infection in Taiwan. Taiwan Enterovirus Epidemic Working Group. N Engl J Med 341, 929–935. Hsieh, T. Y., Matsumoto, M., Chou, H. C., Schneider, R., Hwang, S. B., Lee, A. S. & Lai, M. M. (1998). Hepatitis C virus core protein

interacts with heterogeneous nuclear ribonucleoprotein K. J Biol Chem 273, 17651–17659. Inoue, A., Sawata, S. Y., Taira, K. & Wadhwa, R. (2007). Loss-of-

function screening by randomized intracellular antibodies: identification of hnRNP-K as a potential target for metastasis. Proc Natl Acad Sci U S A 104, 8983–8988. Lee, P. T., Liao, P. C., Chang, W. C. & Tseng, J. T. (2007). Epidermal

homology domain of the poly(rC)-binding protein is a major determinant for binding to the poliovirus 59-untranslated region and acts as an inhibitor of viral translation. J Biol Chem 274, 38163– 38170. Siomi, H., Matunis, M. J., Michael, W. M. & Dreyfuss, G. (1993). The

pre-mRNA binding K protein contains a novel evolutionarily conserved motif. Nucleic Acids Res 21, 1193–1198. Thompson, S. R. & Sarnow, P. (2003). Enterovirus 71 contains a type

I IRES element that functions when eukaryotic initiation factor eIF4G is cleaved. Virology 315, 259–266. Tomonaga, T. & Levens, D. (1995). Heterogeneous nuclear ribonu-

cleoprotein K is a DNA-binding transactivator. J Biol Chem 270, 4875–4881. Toyoda, H., Koide, N., Kamiyama, M., Tobita, K., Mizumoto, K. & Imura, N. (1994). Host factors required for internal initiation of

growth factor increases the interaction between nucleolin and heterogeneous nuclear ribonucleoprotein K/poly(C) binding protein 1 complex to regulate the gastrin mRNA turnover. Mol Biol Cell 18, 5004–5013.

translation on poliovirus RNA. Arch Virol 138, 1–15.

Leopoldino, A. M., Carregaro, F., Silva, C. H., Feitosa, O., Mancini, U. M., Freitas, J. M. & Tajara, E. H. (2007). Sequence and transcriptional

Differential utilization of poly(rC) binding protein 2 in translation directed by picornavirus IRES elements. RNA 5, 1570–1585.

study of HNRPK pseudogenes, and expression and molecular modeling analysis of hnRNP K isoforms. Genome 50, 451–462.

Walter, B. L., Parsley, T. B., Ehrenfeld, E. & Semler, B. L. (2002).

McMinn, P. C. (2002). An overview of the evolution of enterovirus 71

and its clinical and public health significance. FEMS Microbiol Rev 26, 91–107.

Walter, B. L., Nguyen, J. H., Ehrenfeld, E. & Semler, B. L. (1999).

Distinct poly(rC) binding protein KH domain determinants for poliovirus translation initiation and viral RNA replication. J Virol 76, 12008–12022. Wang, Z., Day, N., Trifillis, P. & Kiledjian, M. (1999). An mRNA

McMinn, P., Lindsay, K., Perera, D., Chan, H. M., Chan, K. P. & Cardosa, M. J. (2001). Phylogenetic analysis of enterovirus 71 strains

stability complex functions with poly(A)-binding protein to stabilize mRNA in vitro. Mol Cell Biol 19, 4552–4560.

isolated during linked epidemics in Malaysia, Singapore, and Western Australia. J Virol 75, 7732–7738.

Zhang, W., Zhang, X., Tian, C., Wang, T., Sarkis, P. T., Fang, Y., Zheng, S., Yu, X. F. & Xu, R. (2008). Cytidine deaminase APOBEC3B interacts

Messias, A. C. & Sattler, M. (2004). Structural basis of single-stranded

with heterogeneous nuclear ribonucleoprotein K and suppresses hepatitis B virus expression. Cell Microbiol 10, 112–121.

RNA recognition. Acc Chem Res 37, 279–287.

http://vir.sgmjournals.org

2549