The N-terminal K Homology Domain of the Poly(rC)-binding Protein Is a Major Determinant for Binding to the Poliovirus 5′-Untranslated Region and Acts as an Inhibitor of Viral Translation

The poly(rC)-binding proteins (PCBP1 and PCBP2) are RNA-binding proteins whose RNA recognition motifs are composed of three K homology (KH) domains. These proteins are involved in both the stabilization and translational regulation of several cellular and viral RNAs. PCBP1 and PCBP2 specifically interact with both the 5′-element known as the cloverleaf structure and the large stem-loop IV RNA of the poliovirus 5′-untranslated region. We have found that the first KH domain of PCBP2 (KH1) specifically interacts with the viral RNAs, and together with viral protein 3CD, KH1 forms a high affinity ternary ribonucleoprotein complex with the cloverleaf RNA, resembling the full-length PCBP protein. Furthermore, KH1 acts as a dominant-negative mutant to inhibit translation from a poliovirus reporter gene in both Xenopus laevis oocytes and HeLa cellin vitro translation extracts. The poly(rC)-binding proteins (PCBP1 and PCBP2) are RNA-binding proteins whose RNA recognition motifs are composed of three K homology (KH) domains. These proteins are involved in both the stabilization and translational regulation of several cellular and viral RNAs. PCBP1 and PCBP2 specifically interact with both the 5′-element known as the cloverleaf structure and the large stem-loop IV RNA of the poliovirus 5′-untranslated region. We have found that the first KH domain of PCBP2 (KH1) specifically interacts with the viral RNAs, and together with viral protein 3CD, KH1 forms a high affinity ternary ribonucleoprotein complex with the cloverleaf RNA, resembling the full-length PCBP protein. Furthermore, KH1 acts as a dominant-negative mutant to inhibit translation from a poliovirus reporter gene in both Xenopus laevis oocytes and HeLa cellin vitro translation extracts. untranslated region internal ribosomal entry site poly(rC)-binding protein heterogeneous nuclear ribonucleoprotein, KH, K homology maltose-binding protein Translation in eukaryotic cells is a highly regulated process involving a complex protein machinery. There is increasing evidence that translation of several mRNAs is determined by the specific and regulated interaction of certain proteins with RNA elements in the 3′- and 5′-untranslated regions (for reviews, see Refs. 1Sachs A.B. Sarnow P. Hentze M.W. Cell. 1997; 89: 831-838Abstract Full Text Full Text PDF PubMed Scopus (587) Google Scholar and 2Jackson R.J. Wickens M. Curr. Opin. Genet. Dev. 1997; 7: 233-241Crossref PubMed Scopus (69) Google Scholar). Although many of these cis-acting RNA elements have been defined, only a few trans-acting regulatory proteins are known, and the mechanisms by which they regulate translation are poorly understood.As with cellular messages, translation of viral RNA is also subjected to complex regulation. For example, for positive strand RNA viruses, the genomic RNA is utilized as a template for translation, RNA replication, and formation of new virions; hence, the usage of the RNA must be regulated. For poliovirus, this regulation seems to be dependent on signals within the 5′-UTR.1 Whereas the majority of cellular mRNAs depend on the 5′-cap structure to initiate translation, poliovirus initiates translation internally via a cap-independent mechanism from an RNA element termed the internal ribosomal entry site (IRES) located within the 5′-UTR. Computer modeling and biochemical analysis suggest that the secondary structure of the poliovirus 5′-UTR is composed of six distinct domains, stem-loops I–VI (see Fig. 1 A). In addition to the canonical initiation factors, poliovirus translation initiation requires additional host cell factors, some of which appear to interact directly with the viral RNA (reviewed in Refs. 3Andino R. Böddeker N. Silvera D. Gamarnik A.V. Trends Microbiol. 1999; 76: 76-82Abstract Full Text Full Text PDF Scopus (66) Google Scholar and 4Belsham G.J. Sonenberg N. Microbiol. Rev. 1996; 60: 499-511Crossref PubMed Google Scholar). Among these essential factors are PCBP1 and PCBP2, two closely related poly(rC)-binding proteins, (also referred to as hnRNPs E1 and 2 or αCP1 and αCP2). These proteins facilitate viral translation through the interaction with both the first stem-loop domain (which folds into a cloverleaf-like structure) and stem-loop IV of the poliovirus 5′-UTR (5Blyn L.B. Swiderek K.M. Richards O. Stahl D.C. Semler B.L. Ehrenfeld E. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 11115-11120Crossref PubMed Scopus (169) Google Scholar, 6Gamarnik A.V. Andino R. RNA. 1997; 3: 882-892PubMed Google Scholar, 7Parsley T.B. Towner J.S. Blyn L.B. Ehrenfeld E. Semler B.L. RNA. 1997; 3: 1124-1134PubMed Google Scholar, 8Blyn L.B. Towner J.S. Semler B.L. Ehrenfeld E. J. Virol. 1997; 71: 6243-6246Crossref PubMed Google Scholar). Both PCBP1 and PCBP2 form a low affinity complex with the cloverleaf RNA, but together with viral protein 3CD (the precursor of the viral polymerase 3D and the viral protease 3C), they are incorporated into a high affinity ternary ribonucleoprotein complex (9Andino R. Rieckhof G.E. Baltimore D. Cell. 1990; 63: 369-380Abstract Full Text PDF PubMed Scopus (378) Google Scholar,10Andino R. Rieckhof G.E. Achacoso P.L. Baltimore D. EMBO J. 1993; 12: 3587-3598Crossref PubMed Scopus (407) Google Scholar). Ternary complex formation is required for positive strand RNA synthesis (9Andino R. Rieckhof G.E. Baltimore D. Cell. 1990; 63: 369-380Abstract Full Text PDF PubMed Scopus (378) Google Scholar). Moreover, the interactions of the PCBPs and 3CD with the cloverleaf RNA seem to determine whether the genomic RNA is used as a template for protein synthesis or RNA replication. Binding of PCBPs to the cloverleaf stimulates viral translation, whereas binding of 3CD down-regulates translation (11Gamarnik A.V. Andino R. Genes Dev. 1998; 12: 2293-2304Crossref PubMed Scopus (402) Google Scholar).Recent evidence indicates that association of PCBPs with a number of cellular mRNAs is important for their post-transcriptional regulation, either by stabilizing or by directly influencing the translational state of the mRNA (reviewed in Ref. 12Ostareck-Lederer A. Ostareck D.H. Hentze M.W. Trends Biochem. Sci. 1998; 23: 409-411Abstract Full Text Full Text PDF PubMed Scopus (156) Google Scholar). For instance, PCBPs interact with the highly stable 15-lipoxygenase mRNA (13Holcik M. Liebhaber S.A. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 2410-2414Crossref PubMed Scopus (160) Google Scholar), inducing translational silencing presumably by inhibition of 80 S ribosome assembly (14Ostareck D.H. Ostareck-Lederer A. Wilm M. Thiele B.J. Mann M. Hentze M.W. Cell. 1997; 89: 597-606Abstract Full Text Full Text PDF PubMed Scopus (426) Google Scholar). In addition to their role in poliovirus translation, PCBPs have been implicated in the regulation of translation of other viral RNAs. Depletion of PCBPs from in vitro translation extracts leads to a reduction of protein synthesis from hepatitis A virus (15Graff J. Cha J. Blyn L.B. Ehrenfeld E. J. Virol. 1998; 72: 9668-9675Crossref PubMed Google Scholar); and addition of PCBP1, PCBP2, and hnRNP K to in vitro translation reactions induces translational silencing of the human papilloma virus type 16 L2 mRNA (16Collier B. Goobar-Larsson L. Sokolowski M. Schwartz S. J. Biol. Chem. 1998; 273: 22648-22656Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar).PCBP1 and PCBP2 contain three copies of the RNA-binding K homologous (KH) motif, first described for hnRNP K (17Siomi H. Matunis M.J. Michael W.M. Dreyfuss G. Nucleic Acids Res. 1993; 21: 1193-1198Crossref PubMed Scopus (453) Google Scholar). The arrangement of these three motifs within hnRNP K, PCBP1, and PCBP2 is similar: two consecutive KH domains at the amino terminus followed by a region of variable sequence and length preceding a third KH motif. The degree of homology within the corresponding domain in all of these proteins is higher than that shared by KH motifs within the same polypeptide (18Leffers H. Dejgaard K. Celis J.E. Eur. J. Biochem. 1995; 230: 447-453Crossref PubMed Scopus (191) Google Scholar). No other known RNA-binding motif is found within the PCBPs, and the KH domains are able to function as discreet and independent nucleic acid-binding units: expression of each of the three domains individually demonstrated that the KH1 and KH3 domains of PCBP1 and PCBP2 specifically bind to poly(rC) homopolymers (19Dejgaard K. Leffers H. Eur. J. Biochem. 1996; 241: 425-431Crossref PubMed Scopus (102) Google Scholar). However, the basis for the association of these KH domain-containing proteins with their specific RNA targets, such as the cloverleaf and stem-loop IV of the poliovirus 5′-UTR, remains to be established.We have examined the determinants for poliovirus RNA recognition in PCBP. By expression of individual KH domains, we have found that KH1, the first KH motif of PCBP, is able to specifically recognize both the cloverleaf and stem-loop IV structures of the poliovirus 5′-UTR. In addition, the KH1 domain retains the ability to form a high affinity ternary complex with the cloverleaf and viral protein 3CD. Finally, we show that the KH1 protein specifically inhibits translation from a poliovirus IRES and not from a capped mRNA and that this effect appears to be mediated by a direct competition of KH1 with endogenous PCBPs for the binding sites within the poliovirus 5′-UTR. Translation in eukaryotic cells is a highly regulated process involving a complex protein machinery. There is increasing evidence that translation of several mRNAs is determined by the specific and regulated interaction of certain proteins with RNA elements in the 3′- and 5′-untranslated regions (for reviews, see Refs. 1Sachs A.B. Sarnow P. Hentze M.W. Cell. 1997; 89: 831-838Abstract Full Text Full Text PDF PubMed Scopus (587) Google Scholar and 2Jackson R.J. Wickens M. Curr. Opin. Genet. Dev. 1997; 7: 233-241Crossref PubMed Scopus (69) Google Scholar). Although many of these cis-acting RNA elements have been defined, only a few trans-acting regulatory proteins are known, and the mechanisms by which they regulate translation are poorly understood. As with cellular messages, translation of viral RNA is also subjected to complex regulation. For example, for positive strand RNA viruses, the genomic RNA is utilized as a template for translation, RNA replication, and formation of new virions; hence, the usage of the RNA must be regulated. For poliovirus, this regulation seems to be dependent on signals within the 5′-UTR.1 Whereas the majority of cellular mRNAs depend on the 5′-cap structure to initiate translation, poliovirus initiates translation internally via a cap-independent mechanism from an RNA element termed the internal ribosomal entry site (IRES) located within the 5′-UTR. Computer modeling and biochemical analysis suggest that the secondary structure of the poliovirus 5′-UTR is composed of six distinct domains, stem-loops I–VI (see Fig. 1 A). In addition to the canonical initiation factors, poliovirus translation initiation requires additional host cell factors, some of which appear to interact directly with the viral RNA (reviewed in Refs. 3Andino R. Böddeker N. Silvera D. Gamarnik A.V. Trends Microbiol. 1999; 76: 76-82Abstract Full Text Full Text PDF Scopus (66) Google Scholar and 4Belsham G.J. Sonenberg N. Microbiol. Rev. 1996; 60: 499-511Crossref PubMed Google Scholar). Among these essential factors are PCBP1 and PCBP2, two closely related poly(rC)-binding proteins, (also referred to as hnRNPs E1 and 2 or αCP1 and αCP2). These proteins facilitate viral translation through the interaction with both the first stem-loop domain (which folds into a cloverleaf-like structure) and stem-loop IV of the poliovirus 5′-UTR (5Blyn L.B. Swiderek K.M. Richards O. Stahl D.C. Semler B.L. Ehrenfeld E. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 11115-11120Crossref PubMed Scopus (169) Google Scholar, 6Gamarnik A.V. Andino R. RNA. 1997; 3: 882-892PubMed Google Scholar, 7Parsley T.B. Towner J.S. Blyn L.B. Ehrenfeld E. Semler B.L. RNA. 1997; 3: 1124-1134PubMed Google Scholar, 8Blyn L.B. Towner J.S. Semler B.L. Ehrenfeld E. J. Virol. 1997; 71: 6243-6246Crossref PubMed Google Scholar). Both PCBP1 and PCBP2 form a low affinity complex with the cloverleaf RNA, but together with viral protein 3CD (the precursor of the viral polymerase 3D and the viral protease 3C), they are incorporated into a high affinity ternary ribonucleoprotein complex (9Andino R. Rieckhof G.E. Baltimore D. Cell. 1990; 63: 369-380Abstract Full Text PDF PubMed Scopus (378) Google Scholar,10Andino R. Rieckhof G.E. Achacoso P.L. Baltimore D. EMBO J. 1993; 12: 3587-3598Crossref PubMed Scopus (407) Google Scholar). Ternary complex formation is required for positive strand RNA synthesis (9Andino R. Rieckhof G.E. Baltimore D. Cell. 1990; 63: 369-380Abstract Full Text PDF PubMed Scopus (378) Google Scholar). Moreover, the interactions of the PCBPs and 3CD with the cloverleaf RNA seem to determine whether the genomic RNA is used as a template for protein synthesis or RNA replication. Binding of PCBPs to the cloverleaf stimulates viral translation, whereas binding of 3CD down-regulates translation (11Gamarnik A.V. Andino R. Genes Dev. 1998; 12: 2293-2304Crossref PubMed Scopus (402) Google Scholar). Recent evidence indicates that association of PCBPs with a number of cellular mRNAs is important for their post-transcriptional regulation, either by stabilizing or by directly influencing the translational state of the mRNA (reviewed in Ref. 12Ostareck-Lederer A. Ostareck D.H. Hentze M.W. Trends Biochem. Sci. 1998; 23: 409-411Abstract Full Text Full Text PDF PubMed Scopus (156) Google Scholar). For instance, PCBPs interact with the highly stable 15-lipoxygenase mRNA (13Holcik M. Liebhaber S.A. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 2410-2414Crossref PubMed Scopus (160) Google Scholar), inducing translational silencing presumably by inhibition of 80 S ribosome assembly (14Ostareck D.H. Ostareck-Lederer A. Wilm M. Thiele B.J. Mann M. Hentze M.W. Cell. 1997; 89: 597-606Abstract Full Text Full Text PDF PubMed Scopus (426) Google Scholar). In addition to their role in poliovirus translation, PCBPs have been implicated in the regulation of translation of other viral RNAs. Depletion of PCBPs from in vitro translation extracts leads to a reduction of protein synthesis from hepatitis A virus (15Graff J. Cha J. Blyn L.B. Ehrenfeld E. J. Virol. 1998; 72: 9668-9675Crossref PubMed Google Scholar); and addition of PCBP1, PCBP2, and hnRNP K to in vitro translation reactions induces translational silencing of the human papilloma virus type 16 L2 mRNA (16Collier B. Goobar-Larsson L. Sokolowski M. Schwartz S. J. Biol. Chem. 1998; 273: 22648-22656Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar). PCBP1 and PCBP2 contain three copies of the RNA-binding K homologous (KH) motif, first described for hnRNP K (17Siomi H. Matunis M.J. Michael W.M. Dreyfuss G. Nucleic Acids Res. 1993; 21: 1193-1198Crossref PubMed Scopus (453) Google Scholar). The arrangement of these three motifs within hnRNP K, PCBP1, and PCBP2 is similar: two consecutive KH domains at the amino terminus followed by a region of variable sequence and length preceding a third KH motif. The degree of homology within the corresponding domain in all of these proteins is higher than that shared by KH motifs within the same polypeptide (18Leffers H. Dejgaard K. Celis J.E. Eur. J. Biochem. 1995; 230: 447-453Crossref PubMed Scopus (191) Google Scholar). No other known RNA-binding motif is found within the PCBPs, and the KH domains are able to function as discreet and independent nucleic acid-binding units: expression of each of the three domains individually demonstrated that the KH1 and KH3 domains of PCBP1 and PCBP2 specifically bind to poly(rC) homopolymers (19Dejgaard K. Leffers H. Eur. J. Biochem. 1996; 241: 425-431Crossref PubMed Scopus (102) Google Scholar). However, the basis for the association of these KH domain-containing proteins with their specific RNA targets, such as the cloverleaf and stem-loop IV of the poliovirus 5′-UTR, remains to be established. We have examined the determinants for poliovirus RNA recognition in PCBP. By expression of individual KH domains, we have found that KH1, the first KH motif of PCBP, is able to specifically recognize both the cloverleaf and stem-loop IV structures of the poliovirus 5′-UTR. In addition, the KH1 domain retains the ability to form a high affinity ternary complex with the cloverleaf and viral protein 3CD. Finally, we show that the KH1 protein specifically inhibits translation from a poliovirus IRES and not from a capped mRNA and that this effect appears to be mediated by a direct competition of KH1 with endogenous PCBPs for the binding sites within the poliovirus 5′-UTR. We are grateful to Drs. Alan Frankel and Judith Frydman for useful comments on this manuscript, Dr. Nina Boeddeker for comments on the manuscript and stimulating discussions throughout this work, and Dr. Jorge Iñiguez-Lluhı́ for help with calculations.