| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biochemistry, University of Cambridge, 80 Tennis Court Road, Cambridge CB2 1GA, UK
Correspondence
Richard J. Jackson
rjj{at}mole.bio.cam.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Present address: Department of Biological Sciences, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, UK. 
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Translation from a picornavirus IRES requires the eukaryotic canonical initiation factors, apart from the cap-binding protein eIF4E (Pestova et al., 1996; Ohlmann et al., 1996), although hepatitis A virus shows some dependence on eIF4E (Ali et al., 2001). No viral proteins are required, as evidenced by the ability of picornavirus IRESs to direct translation of heterologous reporter cistrons in vitro (Jang et al., 1988; Pelletier & Sonenberg, 1988). However, host cell trans-acting factors are required for efficient translation, although the requirement for these factors depends on the picornavirus species. Within the picornavirus family, there are two major groups with respect to 5'-UTR structure: the entero-/rhinoviruses, and the cardio-/aphthoviruses; hepatitis A virus is classified in its own group. There is considerable sequence homology and a high degree of secondary structure homology in the 5'-UTR between members of each of these groups, but very little homology between the different groups (Jackson & Kaminski, 1995). This is reflected in the different trans-acting factor requirements of the different classes of picornavirus RNA. Initiation of translation from the cardio-/aphthovirus IRESs can take place in a wide range of cell-free extracts, including rabbit reticulocyte lysate (RRL) (Svitkin & Agol, 1978; Kaminski et al., 1990), but translation initiation from entero- and rhinovirus IRESs is neither efficient nor accurate in RRL without supplementation with HeLa cell factors (Brown & Ehrenfeld, 1979; Dorner et al., 1984; Borman & Jackson, 1992; Bailly et al., 1996).
We have reported the purification of two separable activities from HeLa cells that are able to stimulate translation from the HRV-2 IRES in vitro. The A-type activity was identified as polypyrimidine tract-binding protein (PTB), a regulatory protein involved in alternative splicing as well as translation (Hunt & Jackson, 1999), and the B-type activity was identified as Unr (Hunt et al., 1999). Unr is an essential cytoplasmic RNA-binding protein that contains five cold-shock domains (Boussadia et al., 1997; Jacquemin-Sablon et al., 1994; Doniger et al., 1992). It has been shown to stimulate translation from the HRV-2 IRES in a dicistronic mRNA assay on its own, and synergistically with PTB in vitro (Hunt et al., 1999). It has also been shown to bind to the HRV-2 5'-UTR directly as it can be cross-linked to the HRV-2 5'-UTR by UV radiation (Hunt et al., 1999). The requirement for Unr for translation from the HRV-2 IRES has also been elegantly demonstrated in vivo. Knock out of both copies of the unr gene in murine embryonic stem cells resulted in poor translation from both HRV-2 and poliovirus IRESs, which could be restored by expression of Unr in trans (Boussadia et al., 2003).
The HRV-2 5'-UTR is approximately 600 nt long and has a complex secondary structure consisting of five stem–loop subdomains, an unstructured polypyrimidine tract and a sixth stem–loop that encompasses the initiation codon (Fig. 1; Belsham & Jackson, 2000). The IRES boundaries were defined to include subdomains 2 through the polypyrimidine tract (Borman & Jackson, 1992), with subdomain 1 being shown to be required for viral RNA replication, at least in polioviruses (Andino et al., 1990). To investigate the role of Unr in IRES-dependent translation, the binding sites of Unr on the 5'-UTR were mapped, initially to individual subdomains, and subsequently by identification of the specific nucleotides involved in Unr binding.
|
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
) encodes a monocistronic mRNA consisting of the HRV-2 5'-UTR followed by nucleotides 47–889 of influenza A virus segment 8 (GenBank accession number CY003692), which encodes a slightly truncated form of the influenza virus NS1, designated NS', and a 3'-UTR. It was linearized with BamHI prior to transcription to generate transcripts consisting of the HRV-2 5'-UTR (from nt 10 to the NS' initiation codon terminating in AUGGAUC). pJHRV-d1, -d2, -d3, -d4, -d5, -d5+P, -d5+6 encode monocistronic mRNAs consisting of individual subdomains of the HRV-2 5'-UTR upstream of NS'. These were generated by PCR of nt 1–102 (d1), 102–176 (d2), 179–224 (d3), 232–431 (d4), 439–547 (d5), 439–578 (d5+P) and 439–611 (d5+6) of the HRV-2 5'-UTR, respectively, and subcloning into a pGEM-based vector upstream of the NS' coding region. Linearization of these constructs with XbaI (pJHRV-d1) or BamHI (pJHRV-d2 through d5+6) prior to transcription was carried out to generate transcripts consisting of the individual subdomains for use in gel-shift and UV cross-linking assays.
pXLJHRV10-611, encoding a dicistronic mRNA consisting of the Xenopus laevis cyclin B2 open reading frame, nt 10–611 of the HRV-2 5'-UTR and NS' open reading frame and 3'-UTR, has also been described previously (Borman & Jackson, 1992). It was linearized with EcoRI prior to transcription to generate dicistronic mRNAs for in vitro translation assays.
pJHRV-CAC and pXLJHRV-CAC (subdomain 2 mutants), pJHRV-UAA and pXLJHRV-UAA (subdomain 5 mutants), and pJHRV-CAC-UAA and pXLJHRV-CAC-UAA (subdomain 2+5 mutants) were generated by site-directed mutagenesis (QuikChange, Stratagene) of pJHRV-2 and pXLJ10-611.
In vitro transcription and translation.
Uncapped mRNAs for use in translation assays were generated by in vitro transcription of the linearized plasmid template with T7 RNA polymerase, as described previously (Dasso & Jackson, 1989). Translation assays were carried out in RRL that had been treated with micrococcal nuclease according to the protocol of Jackson & Hunt (1983), and were set up as previously described (Hunt & Jackson, 1999).
UV cross-linking assays.
High specific activity 32P-labelled RNA probes were synthesized as previously described (Hunt & Jackson, 1999) and the yield of RNA quantified as described by Dasso & Jackson (1989). All RNA probes were adjusted to a concentration of 10 nM. UV cross-linking reactions were also carried out according to Hunt & Jackson (1999), in which the final concentration of RNA probe was 1 nM.
Gel-shift assays.
32P-labelled RNA probes were synthesized as described above for UV cross-linking, except that after quantification, the probes were adjusted to a concentration of 2 nM. Gel-shift reactions (10 µl) were set up containing 10 mM HEPES-KOH pH 7.5, 3 mM MgCl2, 5 % glycerol, 1 mM DTT, 90 mM KCl, 1 µg E. coli 23S rRNA, 16 U RNAguard (Pharmacia) and the protein(s) to be assayed; the final concentration of RNA probe was 0.2 nM. The reactions were incubated at room temperature for 15 min, and then analysed by native PAGE in TBE at 4 °C, followed by autoradiography of the dried gel.
Modification interference analysis.
In vitro transcribed RNA (2 µg) was dephosphorylated with 20 U alkaline phosphatase (Roche) at 37 °C for 30 min under the conditions recommended by the supplier. The phosphatase was inactivated by incubation at 85 °C for 10 min. This RNA was then 5'-end-labelled using 50 µCi (1.85 MBq) of [
-32P]ATP and 10 U polynucleotide kinase (Roche) under the supplier's recommended conditions. The reaction was incubated at 37 °C for 30 min and then phenol extracted, stored in ethanol and resuspended in 20 µl H2O just prior to modification. Each chemical modification was carried out on approximately 25 % of the 5'-end-labelled RNA generated in the reaction described above.
a) Modification of adenine bases by diethyl pyrocarbonate (DEPC).
Buffer I (200 µl; 200 mM HEPES-KOH, 10 mM MgCl2, 50 mM KCl pH 7.5) was added to the RNA, followed by 10 µl of DEPC (Sigma) and incubated at 37 °C for 20 min.
b) Modification of uracil and cytosine bases by hydrazine hydrate.
Hydrazine monohydrate (16 µl; Fluka) was added to the RNA and the reaction incubated on ice for 10 min.
c) Modification of guanine and adenine bases by dimethyl sulphate (DMS).
Buffer I (200 µl) was added to the RNA, followed by 3 µl DMS (Sigma). The reaction was incubated at 37 °C for 5 min and then placed on ice.
All modification reactions were stopped by ethanol/sodium acetate precipitation.
The modified RNAs were used in gel-shift reactions with recombinant Unr, as described above. Following native PAGE for 1 h, autoradiography was carried out on the wet gel, to determine the positions of the free RNA and the RNA–protein complexes. These were cut out of the gel and the RNA eluted in 200 µl 0.5 M sodium acetate, 10 mM Tris/HCl pH 7.8, 5 mM EDTA, 1 % SDS, 100 µg ml–1 proteinase K at room temperature for 1 h. Each sample was extracted with phenol and ethanol-precipitated with the help of carrier RNA (4 µg calf liver tRNA). The RNA was resuspended in 20 µl 1 M aniline solution (10 % v/v aniline, 10 % v/v glacial acetic acid in H2O), incubated at 60 °C for 10 min in the dark and the reaction was stopped by ethanol precipitation. In order to cleave exclusively at guanine bases following modification by DMS, the modified RNA was reduced prior to aniline cleavage. Following ethanol precipitation, the RNA was resuspended in 10 µl 1 M Tris/HCl pH 8. Freshly prepared NaBH4 solution (10 µl; 8 mg ml–1) in ethanol was added, and the reaction incubated on ice for 10 min, followed by aniline cleavage as described above. The RNA cleavage products were analysed by denaturing urea-PAGE, ensuring an equal number of c.p.m. were loaded in each lane, and were detected by autoradiography.
Overexpression of recombinant proteins.
His-tagged Unr (pET21dUnr-5), the isoform which lacks exon 5 sequences (Boussadia et al., 1993), and His-tagged PTB (pET28aPTB) were overexpressed in E. coli BL21(DE3) cells and purified using Ni2+-NTA agarose (Qiagen) as described previously (Boussadia et al., 1993; Jacquemin-Sablon et al., 1994). His-tagged Unr was also overexpressed in insect (Sf9) cells by infection with His-tagged Unr-expressing recombinant baculovirus. This baculovirus was generated according to the supplier's recommendations, using linearized baculovirus DNA (Invitrogen) and a pBlueBac4 vector modified to contain the open reading frame of Unr and a C-terminal histidine tag. The baculovirus-expressed His-tagged Unr was also purified using Ni2+-NTA agarose (Qiagen).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
), in addition to the full-length 5'-UTR, were synthesized in vitro and used as radiolabelled probes in UV cross-linking and gel-shift assays. UV cross-linking assays were carried out using 200 nM recombinant Unr, expressed in E. coli, and 20 % (v/v) HeLa cytoplasmic extract (final concentration of Unr approximately 10 nM). In cross-linking HeLa cell extract to the full-length HRV-2 5'-UTR and its subdomains, endogenous Unr was seen to give a strong cross-linking signal with the full-length probe (Fig. 2, lane 1) and a weak signal with the subdomain 2 probe (lane 5). Recombinant Unr gave a very strong signal with the full-length IRES probe (Fig. 2, lane 2) and also gave a clear signal with the subdomain 2 probe (lane 6). A weak signal was also detected with the subdomain 1, 5P (subdomain 5 and the polypyrimidine tract) and 5+6 (subdomains 5 and 6, including the polypyrimidine tract) probes (lanes 4, 14 and 16, respectively). No cross-linking was detected between Unr and the other subdomain probes.
|
. Unr demonstrated a high affinity for the full-length 5'-UTR (Fig. 3a, lanes 1, 2), which can be seen more clearly on a lower percentage acrylamide gel (Fig. 3b), completely shifting all of the probe into a slower migrating complex. However, the affinity of Unr for the individual subdomains was much lower than for the full-length 5'-UTR. Of the subdomain probes, a shift in the mobility of the probe was seen for subdomains 2 and 5+6 (lanes 5, 6 and 13, 14). No shift was detected with the other subdomain probes. The results of the UV cross-linking and gel-shift assays suggest that Unr binds to sites in subdomains 2 and 5+6 of the HRV-2 5'-UTR. Given the weak binding of Unr to the individual subdomains compared to the full-length 5'-UTR, it is likely that these subdomains (2 and 5+6) need to be presented to Unr in the context of the whole 5'-UTR for efficient binding. The results presented here are representative of many (n > 6) RNA-binding assays using different protein and RNA preparations.
|
Subdomain 2 of the HRV-2 5'-UTR was initially used in the interference assay as it gave the best results in the gel-shift and UV cross-linking assays of individual subdomains. Recombinant Unr expressed in insect cells was used in this experiment as this protein preparation had slightly higher RNA-binding activity than that expressed in E. coli. 5' end-labelled subdomain 2 RNA was modified by DMS and hydrazine. The RNA was also subjected to RNase T1 cleavage under conditions in which the RNA would be denatured, in order to aid identification of nucleotides. A representative autoradiograph of four independent experiments is shown in Fig. 4(a). Lanes 1 and 3 show the pattern of bands from total subdomain 2 RNA modified by DMS and subsequent reduction by NaBH4, and lane 2 shows the pattern of bands from Unr-bound RNA modified by DMS and NaBH4. There were several bands whose signal was invariably reduced in the Unr–RNA complex compared with the total RNA input, indicating that modification of these nucleotides interfered with the binding of Unr to the RNA. These nucleotides are marked with an asterisk. The nucleotide marked with an open circle indicates that its modification did not always interfere with Unr binding. Similarly, lanes 4 and 6 show the pattern of bands from total subdomain 2 RNA modified by hydrazine, and lane 5 shows the pattern of bands from Unr-bound RNA modified by hydrazine. Fig. 4(b) shows a diagram of the secondary structure of subdomain 2 with the nucleotides identified in the modification interference experiment marked, again with asterisks and open circles. The adenosine nucleotides marked on this diagram were confirmed using modification interference with DEPC (data not shown). It can be seen that the nucleotides involved in Unr-binding are situated on both sides of the large bulge of subdomain 2, and near a small bulge in the lower stem of the subdomain.
|
shows a representative autoradiograph of four experiments. Lanes 1 and 4 show the pattern of bands from total subdomain 5+6 RNA modified by DEPC, and lanes 2 and 3 show the pattern of bands from Unr-bound RNA. Lanes 5 and 8 show the pattern of bands from total subdomain 5+6 RNA modified by hydrazine, and lanes 6 and 7 show the pattern of bands from Unr-bound RNA. The secondary structure of subdomain 5 is shown in Fig. 5(b), with the nucleotides identified in the modification experiments marked with asterisks and open circles. The guanosine nucleotide marked on this diagram was identified using modification interference with DMS (data not shown). These experiments with subdomain 5+6 showed that the nucleotides involved in Unr binding are situated around the large loop near the top of subdomain 5. None were identified in the polypyrimidine tract or subdomain 6, and this part of the probe is not shown in Fig. 5(b).
|
shows the results of gel-shift assays with recombinant Unr, expressed in insect cells. The wild-type HRV-2 5'-UTR probe on its own is shown in lane 1, with lanes 2 and 3 containing 50 and 100 nM Unr, respectively. Unr (50 nM) gave a shifted band corresponding to a Unr–5'-UTR complex, and increasing the concentration of Unr gave rise to more of the RNA probe being incorporated into this complex. The subdomain 2 mutant 5'-UTR probe with 0, 50 and 100 nM Unr is shown in lanes 4–6. A weak shifted band could be detected at a Unr concentration of 100 nM, but could not be seen at 50 nM Unr, indicating that the affinity of Unr for the mutant 5'-UTR was lower than for the wild-type 5'-UTR. The subdomain 5 mutant 5'-UTR (lanes 7–9) similarly gave a weak shifted band at 100 nM Unr only, and the double mutant 5'-UTR (lanes 10–12) was barely able to form any Unr–RNA complex at 100 nM Unr. The autoradiographs shown are representative of three independent experiments.
|
). Unr gave a strong cross-linking signal with the wild-type HRV-2 5'-UTR (lane 1), but a much weaker signal with the subdomain 2 mutant 5'-UTR (lane 2). The cross-linking signal with the subdomain 5 mutant 5'-UTR (lane 3) was weaker than wild-type, although it was not reduced as much as with the subdomain 2 mutant. Cross-linking of Unr to the double mutant 5'-UTR was very weak (lane 4). The cross-linking signals of other proteins, including PTB, were not reduced by mutation of either subdomain 2 or 5 alone. However, the ability of PTB to cross-link to the 5'-UTR was reduced by the double mutation. This suggests that there may be a requirement for Unr binding to the 5'-UTR for efficient PTB binding.
Effect of subdomain 2 and 5 mutations on translation from the HRV-2 IRES
Translation from the HRV-2 IRES is stimulated by the presence of Unr and PTB (Hunt et al., 1999). Therefore, the effect of the subdomain 2 and 5 mutations on the ability of Unr to stimulate translation from the HRV-2 IRES was tested. The subdomain 2 and 5 substitution mutations were introduced into a dicistronic construct (pXLJHRV10-611; Borman & Jackson, 1992), consisting of a cyclin B2 open reading frame (cyclin), the HRV-2 5'-UTR and a truncated influenza NS open reading frame (NS'), all under the control of a T7 promoter. Dicistronic mRNA was transcribed in vitro from the plasmid and the mRNA was used to programme RRL for in vitro translation. Translation of the upstream cyclin cistron is 5'-end dependent, whereas translation of NS' is dependent on activity of the (wild-type or mutant) HRV-2 IRES. Translation from the mutant HRV-2 IRESs was compared with the translation from the wild-type HRV-2 IRES. Fig. 7(a) shows the autoradiographs resulting from SDS-PAGE of the in vitro translation reactions. These autoradiographs are representative of two independent experiments. The activity of the wild-type (lanes 1–5), subdomain 2 mutant (lanes 6–10), subdomain 5 mutant (lanes 11–15) and double mutant (lanes 16–20) IRESs was tested in unsupplemented RRL (lanes 1, 6, 11 and 16), in the presence of 20 % (v/v) HeLa cytoplasmic extract (lanes 2, 7, 12 and 17), with PTB alone (lanes 3, 8, 13 and 18), with Unr alone (lanes 4, 9, 14 and 19), and with both Unr and PTB (lanes 5, 10, 15 and 20). Densitometric analysis of the autoradiograph bands was carried out, and the fold stimulation of translation compared to the control reaction is shown graphically in Fig. 7(b).
|
suggests that binding of PTB may be affected by the ability of Unr to bind to the RNA. Hence the mutations had a small effect on the ability of PTB to stimulate translation. Stimulation of translation from the mutant IRESs in the presence of HeLa extract was reduced by around 25 %. This small reduction may reflect the presence of other factors in the extract (such as PCBP-2, Hunt et al., 1999) whose binding to the 5'-UTR, and potential stimulatory activity, is not affected by the mutations introduced. The subdomain 2+5 double mutation resulted in further reduction in translation stimulation (the effect of Unr and PTB was reduced by more than 75 %), since Unr was unable to bind the 5'-UTR at either site. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Le & Zuker, 1990; Nicholson et al., 1991). However, it is not known how these secondary structure elements come together in the complex tertiary structure that almost certainly exists to form an IRES.
In this study, the multiple cold-shock domain (CSD)-containing protein Unr, one of two activities identified so far as stimulating translation from the HRV-2 IRES (the other being PTB), was shown to bind to subdomains 2 and 5+6 of the HRV-2 5'-UTR in gel-shift analysis and UV cross-linking assays. Increasing the concentration of recombinant Unr in these assays did result in binding of Unr to subdomain 5 alone (data not shown), but the preference for subdomain 5+6 suggests that subdomain 5 attains a more favourable structure for Unr binding in the presence of the downstream polypyrimidine tract and stem–loop. Subdomain 5 in the closely related poliovirus IRES has previously been shown to contain neurovirulence determinants (Campbell et al., 2005; Malnou et al., 2002; Svitkin et al., 1985, 1988) and protein-binding sites important for translation (Hellen et al., 1994; Gutiérrez et al., 1997). In addition, HRV-2 subdomain 5 has been reported to contain a binding site for a heterodimer of double-stranded RNA-binding protein 76 and nuclear factor of activated T cells, 45 kDa, which represses translation from the HRV-2 IRES in neuronal cells (Merrill and Gromeier, 2006). However, this is the first report of a protein-binding site in subdomain 2 important for translation from an entero- or rhinovirus IRES.
As Unr has five RNA-binding domains, it is possible that the protein binds to the 5'-UTR at more than one site simultaneously, and acts as an RNA chaperone to aid folding of the IRES into a structure competent for ribosome-binding. Indeed, we have shown that all five CSDs are required for maximum activation of the HRV-2 IRES (Brown & Jackson, 2004). Despite subdomains 2 and 5 being widely separated in the primary sequence and secondary structure of the 5'-UTR, optimal binding of Unr depends on these subdomains being presented to Unr in the context of the whole 5'-UTR, rather than as discrete elements, since Unr binds the full-length 5'-UTR much better than the individual subdomains. If one molecule of Unr binds these two sites simultaneously on the full-length 5'-UTR, it is understandable that Unr could have a much higher affinity for the full-length 5'-UTR than for the individual subdomains which contain Unr-binding sites. If two molecules of Unr bind to the two binding sites independently, then it would be expected that Unr would have a higher affinity for the individual subdomains than it does. Thus, the data would suggest that subdomains 2 and 5 are brought close together in the tertiary structure of the IRES by the binding of Unr.
Unr and the neuronal isoform of PTB (nPTB) have also been shown to bind to and activate translation from the cellular Apaf-1 IRES (Mitchell et al., 2001). The function of these proteins as RNA chaperones has been reported (Mitchell et al., 2003), whereby binding of Unr to the Apaf-1 IRES alters the RNA structure such that nPTB can bind, which in turn opens up the ribosome-binding site. The results of our mutational analysis of the HRV-2 IRES also suggest that the binding of Unr may be required for efficient PTB binding to the HRV-2 IRES [Figs 6(b) and 7]. The structural changes to the Apaf-1 IRES effected by Unr and nPTB (Mitchell et al., 2003) demonstrate the impact that these factors may have on the folding of complex RNA structures, such as the HRV-2 IRES. Further evidence for the ability of Unr to stimulate IRES-dependent translation comes from studies of translation of the PITSLRE mRNA (Tinton et al., 2005).
Modification interference analysis of the binding of Unr to subdomains 2 and 5 identified the nucleotides whose chemical modification prevented the binding of Unr to the RNA. The large bulge of subdomain 2 (and a small bulge in the lower stem) and the large loop in the middle of subdomain 5 appear to be critical for Unr binding. The importance of the CAC trinucleotide in the middle of the subdomain 2 bulge and the UAA trinucleotide in the middle of the subdomain 5 loop for both Unr binding and IRES translation activity was confirmed by site-directed mutagenesis studies. These showed that mutation of just these nucleotides decreased the ability of Unr to bind to the IRES and reduced the translational competency of the IRES in the presence of its stimulatory trans-acting factors Unr and PTB.
It is significant that the proposed Unr binding sites are situated in predicted single-stranded regions, as the CSD is a single-stranded nucleic acid-binding domain and would be expected to bind single-stranded regions of the 5'-UTR. The short sequences of these binding sites can be compared to the SELEX sequences that Triqueneaux et al. (1999) reported as the preferred RNA sequences for Unr binding (5'-N5AAGUAN-3' and 5'-N8AACGN2-3'). Adenosine residues feature strongly in the SELEX results, although the unstructured SELEX sequences identified as binding to Unr do not match the highly structured HRV-2 5'-UTR sites. Unr has also been reported to bind to the major protein-coding-region determinant of instability of c-fos mRNA (Chang et al., 2004), the instability element of parathyroid hormone (PTH) mRNA (Dinur et al., 2006) and the autoregulatory sequence of poly(A)-binding protein (PABP) mRNA (Patel et al., 2005). In each of these cases, Unr has been shown to bind to A-rich sequences in complex with either poly(A)-binding protein (c-fos and PABP mRNAs) or AU-rich-binding factor 1 (PTH mRNA).
It seems likely then that Unr binds to single-stranded A-rich regions of RNA, with selectivity being imposed by its protein-binding partners and the secondary structure of the RNA. On binding the HRV-2 5'-UTR, Unr is most likely recognizing a combination of short sequence and structural motifs. In terms of the effect of protein-binding partners on binding to the HRV-2 5'-UTR, while PTB also binds to the 5'-UTR, it does not bind to Unr, and doesn't affect the binding of Unr to the HRV-2 5'-UTR. Unrip, the protein that co-purified alongside Unr as part of the B-type translation activity, does not bind to RNA. Although it binds to Unr, Unrip does not alter the position of the binding site of Unr on the 5'-UTR (E. C. Anderson and R. J. Jackson, unpublished data).
To conclude, future studies on the structure of the HRV-2 IRES and IRES–protein complexes may be aided by the knowledge that subdomains 2 and 5, situated widely apart in the primary sequence, may be brought together by the binding of the RNA-binding protein Unr.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Andino, R., Rieckhof, G. E. & Baltimore, D. (1990). A functional ribonucleoprotein complex forms around the 5' end of poliovirus RNA. Cell 63, 369–380.[CrossRef][Medline]
Bailly, J.-L., Borman, A. M., Peigue-Lafeuille, H. & Kean, K. M. (1996). Natural isolates of ECHO virus type-25 with extensive variations in IRES sequences and different translational efficiencies. Virology 215, 83–96.[CrossRef][Medline]
Belsham, G. J. & Jackson, R. J. (2000). Translation initiation on picornavirus RNA. In: Translational Control of Gene Expression, pp. 869–900. Edited by N. Sonenberg, J. W. B. Hershey & M. B. Mathews. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
Borman, A. & Jackson, R. J. (1992). Initiation of translation of human rhinovirus RNA: mapping the internal ribosome entry site. Virology 188, 685–696.[CrossRef][Medline]
Boussadia, O., Jacquemin-Sablon, H. & Dautry, F. (1993). Exon skipping in the expression of the gene immediately upstream of N-ras (unr/NRU). Biochim Biophys Acta 1172, 64–72.[Medline]
Boussadia, O., Amiot, F., Cases, S., Triqueneaux, G., Jacquemin-Sablon, H. & Dautry, F. (1997). Transcription of unr (upstream of N-ras) down-modulates N-ras expression in vivo. FEBS Lett 420, 20–24.[CrossRef][Medline]
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. & Ehrenfeld, E. (1979). Translation of poliovirus in vitro: changes in cleavage pattern and initiation sites by ribosomal salt wash. Virology 97, 396–405.[CrossRef][Medline]
Brown, E. C. & Jackson, R. J. (2004). All five cold shock domains of unr (upstream of N-ras) are required for stimulation of human rhinovirus RNA translation. J Gen Virol 85, 2279–2287.
Campbell, S. A., Lin, J., Dobrikova, E. Y. & Gromeier, M. (2005). Genetic determinants of cell type-specific poliovirus propagation in HEK 293 cells. J Virol 79, 6281–6290.
Chang, T. C., Yamashita, A., Chen, C. Y., Yamashita, Y., Zhu, W., Durdan, S., Kahvejian, A., Sonenberg, N. & Shyu, A. B. (2004). UNR, a new partner of poly(A) binding protein, plays a key role in translationally coupled mRNA turnover mediated by the c-fos major coding-region determinant. Genes Dev 18, 2010–2023.
Dasso, M. C. & Jackson, R. J. (1989). On the fidelity of mRNA translation in the nuclease-treated rabbit reticulocyte lysate system. Nucleic Acids Res 17, 3129–3144.
Dinur, M., Kilav, R., Sela-Brown, A., Jacquemin-Sablon, H. & Naveh-Many, T. (2006). In vitro evidence that upstream of N-ras participates in the regulation of parathyroid hormone messenger ribonucleic acid stability. Mol Endocrinol 20, 1652–1660.
Doniger, J., Landsman, D., Gonda, M. A. & Wistow, G. (1992). The product of unr, a highly conserved gene upstream of N-ras, contains multiple repeats similar to the cold-shock domain (CSD), a putative DNA-binding motif. New Biol 4, 389–395.[Medline]
Dorner, A. J., Semler, B. L., Jackson, R. J., Hanecak, R., Duprey, E. & Wimmer, E. (1984). In vitro translation of poliovirus RNA: utilization of internal initiation sites in reticulocyte lysate. J Virol 50, 507–514.
Gutiérrez, A. L., Denova-Ocampo, M., Racaniello, V. R. & del Angel, R. M. (1997). Attenuating mutations in the poliovirus 5' untranslated region alter its interaction with polypyrimidine tract-binding protein. J Virol 71, 3826–3833.[Abstract]
Hellen, C. U., Pestova, T. V., Litterst, M. & Wimmer, E. (1994). The cellular polypeptide p57 (pyrimidine tract-binding protein) binds to multiple sites in the poliovirus 5' nontranslated region. J Virol 68, 941–950.
Hunt, S. L. & Jackson, R. J. (1999). Polypyrimidine-tract binding protein (PTB) is necessary, but not sufficient, for efficient internal initiation of translation of human rhinovirus-2 RNA. RNA 5, 344–359.[Abstract]
Hunt, S. L., Hsuan, J. J., Totty, N. & Jackson, R. J. (1999). Unr, a cellular cytoplasmic RNA-binding protein with five cold-shock domains, is required for internal initiation of translation of human rhinovirus RNA. Genes Dev 13, 437–448.
Jackson, R. J. & Hunt, T. (1983). Preparation and use of nuclease-treated rabbit reticulocyte lysates for the translation of eukaryotic messenger RNA. Methods Enzymol 96, 50–74.[Medline]
Jackson, R. J. & Kaminski, A. (1995). Internal initiation of translation in eukaryotes: the picornavirus paradigm and beyond. RNA 1, 985–1000.[Medline]
Jacquemin-Sablon, H., Triqueneaux, G., Deschamps, S., le Maire, M., Doniger, J. & Dautry, F. (1994). Nucleic acid binding and intracellular localisation of Unr, a protein with five cold shock domains. Nucleic Acids Res 22, 2643–2650.
Jang, S. K., Krausslich, H.-G., Nicklin, M. J. H., Duke, G. M., Palmenberg, A. C. & Wimmer, E. (1988). A segment of the 5' non-translated region of EMC virus RNA directs internal entry of ribosomes during in vitro translation. J Virol 62, 2636–2643.
Kaminski, A., Howell, M. T. & Jackson, R. J. (1990). Initiation of encephalomyocarditis virus RNA translation: the authentic initiation site is not selected by a scanning mechanism. EMBO J 9, 3753–3759.[Medline]
Le, S. Y. & Zuker, M. (1990). Common structures of the 5' non-coding RNA in enteroviruses and rhinoviruses. Thermodynamical stability and statistical significance. J Mol Biol 216, 729–741.[CrossRef][Medline]
Malnou, C. E., Pöyry, T. A., Jackson, R. J. & Kean, K. M. (2002). Poliovirus internal ribosome entry segment structure alterations that specifically affect function in neuronal cells: molecular genetic analysis. J Virol 76, 10617–10626.
Merrill, M. K. & Gromeier, M. (2006). The double-stranded RNA binding protein 76:NF45 heterodimer inhibits translation initiation at the rhinovirus type 2 internal ribosome entry site. J Virol 80, 6936–6942.
Mitchell, S. A., Brown, E. C., Coldwell, M. J., Jackson, R. J. & Willis, A. E. (2001). Protein factor requirements of the Apaf-1 internal ribosome entry segment: roles of polypyrimidine tract binding protein and upstream of N-ras. Mol Cell Biol 21, 3364–3374.
Mitchell, S. A., Spriggs, K. A., Coldwell, M. J., Jackson, R. J. & Willis, A. E. (2003). The Apaf-1 internal ribosome entry segment attains the correct structural conformation for function via interactions with PTB and Unr. Mol Cell 11, 757–771.[CrossRef][Medline]
Nicholson, R., Pelletier, J., Le, S.-Y. & Sonenberg, N. (1991). Structural and functional analysis of the ribosome landing pad of poliovirus type 2: in vivo translation studies. J Virol 65, 5886–5894.
Ohlmann, T., Rau, M., Pain, V. M. & Morley, S. J. (1996). The C-terminal domain of eukaryotic protein synthesis initiation factor eIF4G is sufficient to support cap-independent translation in the absence of eIF4E. EMBO J 15, 1371–1382.[Medline]
Patel, G. P., Ma, S. & Bag, J. (2005). The autoregulatory translational control element of poly(A) binding protein mRNA forms a heteromeric ribonucleoprotein complex. Nucleic Acids Res 33, 7074–7089.
Pelletier, J. & Sonenberg, N. (1988). Internal initiation of translation of eukaryotic mRNA directed by a sequence derived from poliovirus RNA. Nature 334, 320–325.[CrossRef][Medline]
Pestova, T. V., Shatsky, I. N. & Hellen, C. U. T. (1996). Functional dissection of eukaryotic initiation factor 4F: the 4A subunit and the central domain of the 4G subunit are sufficient to mediate internal entry of 43S preinitiation complexes. Mol Cell Biol 16, 6870–6878.[Abstract]
Skinner, M. A., Racaniello, V. R., Dunn, G., Cooper, J., Minor, P. D. & Almond, J. W. (1989). New model for the secondary structure of the 5' non-coding RNA of poliovirus is supported by biochemical and genetic data that also show that RNA secondary structure is important in neurovirulence. J Mol Biol 207, 379–392.[CrossRef][Medline]
Svitkin, Y. V. & Agol, V. I. (1978). Complete translation of encephalomyocarditis virus RNA and faithful cleavage of virus-specific proteins in a cell-free system from Krebs-2 cells. FEBS Lett 87, 7–11.[CrossRef][Medline]
Svitkin, Y. V., Maslova, S. V. & Agol, V. I. (1985). The genomes of attenuated and virulent poliovirus strains differ in their in vitro translation efficiencies. Virology 147, 243–252.[CrossRef][Medline]
Svitkin, Y. V., Pestova, T. V., Maslova, S. V. & Agol, V. I. (1988). Point mutations modify the response of poliovirus RNA to a translation initiation factor: a comparison of neurovirulent and attenuated strains. Virology 166, 394–404.[CrossRef][Medline]
Tinton, S. A., Schepens, B., Bruynooghe, Y., Beyaert, R. & Cornelis, S. (2005). Regulation of the cell-cycle-dependent internal ribosome entry site of the PITSLRE protein kinase: roles of Unr (upstream of N-ras) protein and phosphorylated translation initiation factor eIF-2alpha. Biochem J 385, 155–163.[CrossRef][Medline]
Triqueneaux, G., Velten, M., Franzon, P., Dautry, F. & Jacquemin-Sablon, H. (1999). RNA binding specificity of Unr, a protein with five cold shock domains. Nucleic Acids Res 27, 1926–1934.
Received 11 August 2006;
accepted 23 July 2007.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
