Polypyrimidine tract binding protein functions as a negative regulator of feline calicivirus translation - PubMed (original) (raw)
Polypyrimidine tract binding protein functions as a negative regulator of feline calicivirus translation
Ioannis Karakasiliotis et al. PLoS One. 2010.
Abstract
Background: Positive strand RNA viruses rely heavily on host cell RNA binding proteins for various aspects of their life cycle. Such proteins interact with sequences usually present at the 5' or 3' extremities of the viral RNA genome, to regulate viral translation and/or replication. We have previously reported that the well characterized host RNA binding protein polypyrimidine tract binding protein (PTB) interacts with the 5'end of the feline calicivirus (FCV) genomic and subgenomic RNAs, playing a role in the FCV life cycle.
Principal findings: We have demonstrated that PTB interacts with at least two binding sites within the 5'end of the FCV genome. In vitro translation indicated that PTB may function as a negative regulator of FCV translation and this was subsequently confirmed as the translation of the viral subgenomic RNA in PTB siRNA treated cells was stimulated under conditions in which RNA replication could not occur. We also observed that PTB redistributes from the nucleus to the cytoplasm during FCV infection, partially localizing to viral replication complexes, suggesting that PTB binding may be involved in the switch from translation to replication. Reverse genetics studies demonstrated that synonymous mutations in the PTB binding sites result in a cell-type specific defect in FCV replication.
Conclusions: Our data indicates that PTB may function to negatively regulate FCV translation initiation. To reconcile this with efficient virus replication in cells, we propose a putative model for the function of PTB in the FCV life cycle. It is possible that during the early stages of infection, viral RNA is translated in the absence of PTB, however, as the levels of viral proteins increase, the nuclear-cytoplasmic shuttling of PTB is altered, increasing the cytoplasmic levels of PTB, inhibiting viral translation. Whether PTB acts directly to repress translation initiation or via the recruitment of other factors remains to be determined but this may contribute to the stimulation of viral RNA replication via clearance of ribosomes from viral RNA.
Conflict of interest statement
Competing Interests: The authors have declared that no competing interests exist.
Figures
Figure 1. Schematic representation of the feline calicivirus genome.
Diagrammatic illustration of the three open reading frames present within the feline calicivirus genome. The mature components of ORF1 are highlighted along with the recently proposed NS1-7 nomenclature used for murine norovirus . The positions of the leader of the capsid protein (LC), as well as the major and minor capsid proteins (VP1 and VP2) are also highlighted.
Figure 2. Sequence and structure of the 5′ end of the feline calicivirus genome.
(A) RNA sequence of nucleotides 1-245 of the feline calicivirus genomic RNA, highlighting the four potential PTB binding sites (BS1-4). Note that the pyrimidine rich sequences around the core PTB binding sequence (UCUU) are highlighted. The ORF1 AUG initiation codon at position 20 is shown in bold italics. The positions of single strand specific RNases T1,T2 and A, as well as the double-strand specific RNase V1, obtained from RNase sensitivity mapping (shown in panel B and C) are highlighted as detailed in the symbol legend. (B) In vitro transcribed RNA representing nucleotides 1-245 of FCV genomic RNA was subjected to limited RNase digestion and primer extension analysis using either a primer binding between nucleotides 223 to 204 (B) or nucleotides 245–228 (C). A sequencing ladder was generated using same primers and run on the gel along with primer extension products to identify the cleavage sites of RNases. The positions of RNase cleavage sites which show disparity with the predicted RNA structure are highlights with a black filled circle. Note that only cleavage sites which were apparent in multiple experiments are depicted.
Figure 3. Identification of PTB binding sites using truncation analysis.
EMSA analysis of GST-PTB binding to probes encompassing various regions of the FCV 5′ end. Radiolabeled RNA probes were generated by in vitro transcription of PCR products encompassing the various regions of the FCV genome, followed by gel purification. EMSA reactions were set up as described in the text and the reactions analyzed by native PAGE. Gels were then dried, exposed to phosphorimager screen and the amount of probe in a complex with PTB determined using ImageQuant software. (A) Diagrammatic representation of the truncations under study highlighting the relative affinity of PTB for the transcript. # highlights that the affinity is expressed relative to WT 1–245 probe (set at 1). nd denotes that the relative affinity could not be determined due to insufficient complex formation. EMSA gels using the 1–245 transcript along with the 3′ end truncations (B) or 5′ end truncations (C). EMSAs were performed a minimum of three times and the data obtained from one representative experiment shown. Asterisks are used to highlight the position of alternative conformations of the RNA probe which make up a minor component of the refolded RNA transcript.
Figure 4. Identification of PTB binding sites using synonymous mutations.
EMSA analysis of GST-PTB binding to wild type (WT) or mutant derivatives of nucleotides 1–202 of the FCV 5′ end. Radiolabeled RNA probes were generated by in vitro transcription of PCR products, followed by gel purification. EMSA reactions were set up as described in the text and the reactions analyzed by native PAGE. Gels were then dried, exposed to phosphorimager screen and the amount of probe in a complex with PTB determined using ImageQuant software. (A) Diagrammatic representation of the region under study highlighting the binding sites (BS1-3), the mutations introduced (m) are underlined and the amino acid sequence coded by that region of the genome is also shown. The relative affinity of PTB for the transcript is shown based on quantification of PTB binding. # highlights that the affinity is expressed relative to WT 1–202 probe. (B) Quantification of PTB binding to WT or PTB binding site mutants (mBS1-3). The amount of radiolabeled probe in a complex with PTB was determined using phosphoimager and expressed as a percentage of the total. Note that for clarity only data for the single binding site mutant BS3 (mBS3) is shown but mBS1 and mBS2 displayed identical binding curves (data not shown). (C) Representative EMSA gel displaying the binding of GST-PTB to WT RNA probe or a probe containing synonymous changes in binding sites 1, 2 and 3 (mBS1+2+3). EMSAs were performed a minimum of three times and the data obtained from one representative experiment shown.
Figure 5. Feline calicivirus infection results in the redistribution of PTB.
CRFK cells were infected with FCV for 16 h at 32°C and at an m.o.i. of 0.5 (A) Infected and control cells were fixed and stained for NS6-7 (green) and PTB (DH17 monoclonal, red). DAPI was used for nuclei (blue) staining. The cells were observed using confocal microscopy. Quantification of nuclear (B) and cytoplasmic (C) PTB density (arbitrary intensity units/arbitrary volume units) against the density of NS6-7 in the cytoplasm of the same cell. Individual NS6-7 expressing cells were selected and a region of interest selected in the cytoplasm and nucleus. The levels of NS6-7 staining (green) and PTB staining (red) was quantified in arbitrary units and plotted accordingly (see materials and methods for further details).
Figure 6. Feline calicivirus infection causes both PTB specific and general effects on nuclear localization of proteins.
(A). GFP-PTB expressing CRFK cells were infected with recombinant FCV expressing a LC-dsRED fusion protein (m.o.i. 2) and examined during the course of the infection. (B) Cells transfected with cDNA constructs expressing PTB-GFP and dsRED fused to 3 copies of the nuclear localization sequence of the SV40 large T-antigen (DsRED2-Nuc) were infected with FCV at a m.o.i. of 2. Cells were then examined over the course of the infection. See supplementary files Figure S1 and Figure S2 for AVI data relating to A and B respectively.
Figure 7. PTB colocalizes with mature FCV replication complexes in FCV infected CRFK cells.
CRFK cells were infected with FCV for 16 h at 32°C and at an m.o.i. of 0.5. (A) Infected cells were fixed and stained for NS6-7 (green) and PTB (DH17 monoclonal, red). DAPI was used for nuclei (blue) staining. Note that only a fraction of the cytoplasm of an infected cell was observed using confocal microscopy. (B) Plot of pixel intensity for NS6-7 and PTB along a straight line in the above image for the visualization of the colocalization of PTB with the replication complexes (NS6-7). A line of quantification was arbitrarily drawn across the image (see insect picture for the position) and the staining intensity (arbitrary units) of NS6-7 and PTB plotted against the pixel position (from left to right). Two replication complexes which contain both NS6-7 and PTB are highlighted in A and B. (C) Viral RNA immunoprecipitation from FCV infected CRFK cells at 37°C with antibodies against PTB (DH17 and DH7) and the viral NS6-7. Antibody for GAPDH protein was used as negative control and the NS6-7 IP was PCR amplified without reverse transcription to confirm the authenticity of product (-RT). The presence of the viral RNA in immunoprecipitates was confirmed by RT-PCR amplification of nucleotide 1-245 of the FCV genome.
Figure 8. PTB inhibits FCV translation in vitro.
(A) Effect of his-PTB on FCV in vitro translation that demonstrates the reduced expression of FCV proteins produced by the translation of ORF1 (from the genomic RNA) and ORF2 (from the subgenomic RNA). The 35S methionine labeled FCV proteins were analyzed by SDS-PAGE. Addition of increasing amounts of in vitro transcribed RNA encompassing nucleotides 1–245 of the FCV genome (5′G1-245) led to the recovery of the FCV translation. The bar chart illustrates the efficiency of FCV translation as percentage of the control (0 ng PTB) as measured by phosphorimager quantification. The poliovirus protease-polymerase 3CD was used to control for any non-specific effects of adding recombinant protein to the reaction. Error bars represent the standard deviation from three different experiments. (B) Increasing amounts of his-PTB and 5′G 1-245 in the in vitro translation of a dicistronic RNA construct that expressed the CAT enzyme under the cap structure and firefly luciferase under the FMDV IRES. The 35S methionine labeled CAT and firefly luciferase were analyzed by SDS-PAGE. The bar chart illustrates the efficiency of cap- and FMDV IRES-dependent translation as a percentage of the control (0ng PTB) as measured by phosphorimager quantification. Error bars represent the standard deviation from three different experiments.
Figure 9. RNAi mediated knockdown of PTB inhibit the translation from the FCV subgenomic RNA in cells.
(A) Genome schematic of the FCV genome highlighting the position of the three FCV open reading frames and the NS6-7 specific primer which bound to nucleotides between 3215–3244 used for RNase H-directed inactivation of the FCV genomic RNA. (B) In vitro translation of FCV RNA after RNase H digestion using an FCV genomic RNA specific DNA oligonucleotide (FP) and a control DNA oligonucleotide (CP). Immunoprecipitation (VP1-IP) with α-major capsid (VP1) from the in vitro translation of FCV RNA after RNase H digestion using an FCV genomic RNA specific DNA oligonucleotide and a control DNA oligonucleotide. The proteins were labeled with 35S methionine and the SDS-PAGE analysis was exposed to a phosphorimager screen. (C) Western blot for the major capsid protein (VP1) in infected CRFK cells, purified virus, non-transfected CRFK cells and CRFK cells treated with PTB or GFP siRNAs that have been transfected with FCV RNA digested with RNase H using an FCV specific DNA oligonucleotide. Control blots for PTB and GAPDH are shown to demonstrate PTB knockdown and equal loading of samples. Results displayed were obtained from cells incubated at 37°C but identical results were also observed when the experiments were performed at 32°C (data not shown) An asterisk is used to highlight a non-specific protein with reactivity to anti-VP1 antisera.
Figure 10. Mutations in PTB binding sites 2 and 3 results in a cell type specific defect in FCV replication.
(A) Single and multi-cycle (B) growth curve analysis of wild type FCV Urbana strain derived from cDNA (WT) or a derivative carrying synonymous nucleotide changes in PTB binding sites 2 and 3 at the 5′ end of the genomic RNA. Cells were infected with a m.o.i. of 3 for single-cycle and 0.01 for multi-cycle (based on the titre of viral stocks in CRFK cells) and samples harvested at various times post inoculation. Viral titre was subsequently determined by TCID50 on CRFK cells and expressed as TCID50 per ml. Infections were performed in triplicate and the error bars represent the standard deviation. *** P<0.001 *P<0.01 *P<0.05 by two-way ANOVA.
Figure 11. Characterisation of PTB distribution and expression levels in various cell types.
(A) Western blot and (B) confocal image analysis of the expression and cellular distribution of PTB in uninfected AK-D, CRFK and FEA cells. Approximately 1×105 cells were lysed and analysed by western blot using both monoclonal (DH17) and polyclonal (N20) antibodies to PTB (A). GAPDH was used to ensure equal loading. (B) Confocal analysis of AK-D, CRFK and FEA cells stained for DNA with DAPI (blue) and PTB (DH17 monoclonal, green). L, M and H highlight AK-D cells expressing low, medium and high levels of PTB respectively. Size bar represents 100 µm.
Figure 12. Proposed model for the function of PTB in the FCV life cycle.
Schematic representation of the potential regulatory function of PTB during the various stages of the FCV infectious cycle. (A) The viral RNA is translated via VPg-dependent translation to produce the various viral proteins. (B) As the viral proteins accumulate, they modulate (directly or indirectly) the nuclear-cytoplasmic shuttling of PTB. (C) Cytoplasmic PTB interacts with the viral RNA, possibly in combination with other cellular or viral proteins to subsequently inhibit the recruitment of the ribosomes to the 5′ end of the positive-strand viral RNA (D). Displacement of the ribosomes allows the synthesis of negative strand RNA (E) by the RNA-dependent RNA polymerase.
References
- Auweter SD, Oberstrass FC, Allain FH. Solving the structure of PTB in complex with pyrimidine tracts: an NMR study of protein-RNA complexes of weak affinities. J Mol Biol. 2007;367:174–186. - PubMed
- Mitchell SA, Spriggs KA, Coldwell MJ, Jackson RJ, Willis AE. The Apaf-1 internal ribosome entry segment attains the correct structural conformation for function via interactions with PTB and unr. Mol Cell. 2003;11:757–771. - PubMed
Publication types
MeSH terms
Substances
Grants and funding
- 081624/WT_/Wellcome Trust/United Kingdom
- BB/D007607/1/BB_/Biotechnology and Biological Sciences Research Council/United Kingdom
- BB/D007755/1/BB_/Biotechnology and Biological Sciences Research Council/United Kingdom
- ImNIH/Intramural NIH HHS/United States