Poliovirus cis-acting replication element-dependent VPg Uridylylation lowers the Km of the initiating nucleoside triphosphate for viral RNA replication

Abstract

Initiation of RNA synthesis by RNA-dependent RNA polymerases occurs when a phosphodiester bond is formed between the first two nucleotides in the 5' terminus of product RNA. The concentration of initiating nucleoside triphosphates (NTPi) required for RNA synthesis is typically greater than the concentration of NTPs required for elongation. VPg, a small viral protein, is covalently attached to the 5' end of picornavirus negative- and positive-strand RNAs. A cis-acting replication element (CRE) within picornavirus RNAs serves as a template for the uridylylation of VPg, resulting in the synthesis of VPgpUpU(OH). Mutations within the CRE RNA structure prevent VPg uridylylation. While the tyrosine hydroxyl of VPg can prime negative-strand RNA synthesis in a CRE- and VPgpUpU(OH)-independent manner, CRE-dependent VPgpUpU(OH) synthesis is absolutely required for positive-strand RNA synthesis. As reported herein, low concentrations of UTP did not support negative-strand RNA synthesis when CRE-disrupting mutations prevented VPg uridylylation, whereas correspondingly low concentrations of CTP or GTP had no negative effects on the magnitude of CRE-independent negative-strand RNA synthesis. The experimental data indicate that CRE-dependent VPg uridylylation lowers the K(m) of UTP required for viral RNA replication and that CRE-dependent VPgpUpU(OH) synthesis was required for efficient negative-strand RNA synthesis, especially when UTP concentrations were limiting. By lowering the concentration of UTP needed for the initiation of RNA replication, CRE-dependent VPg uridylylation provides a mechanism for a more robust initiation of RNA replication.

FIG. 1.

CRE mutation inhibited nascent negative-strand RNA synthesis within PIRCs. (A) Diagram of PV RNAs with either a WT or mutant (KO CRE) CRE. Eight silent mutations (PV nucleotides C⁴⁴⁵⁶U, A⁴⁴⁵⁹C, G⁴⁴⁶²A, C⁴⁴⁶⁵U, G⁴⁴⁶⁸A, A⁴⁴⁶⁹U, G⁴⁴⁷⁰C, and A⁴⁴⁷⁴G) were engineered to completely disrupt the CRE RNA structure and prevent any detectable VPgpUpU_OH synthesis (29). PV RNA2 templates make only negative-strand RNA due to two nonviral 5′ guanosine residues (10, 19). ORF, open reading frame. (B and C) PIRCs with PV RNA2 WT CRE (lanes 1 to 6) or PV RNA2 KO CRE (lanes 7 to 12) templates were incubated in [α-³²P]UTP labeling reaction mix containing 1 mM ATP, 250 μM GTP, 250 μM CTP, 10 μM unlabeled UTP, and 1.25 μM [α-³²P]UTP (40 μCi per 40-μl reaction mixture). RNA replication reaction mixtures were incubated from 4 to 20 min in the absence of guanidine (lanes 1 to 5 and 7 to 11, as indicated in panel C). A total of 2 mM of guanidine was included in two reaction mixtures (lanes 6 and 12). RNA products from the reaction mixtures were separated by electrophoresis in 1% agarose and detected by ethidium bromide and UV light (B) or by phosphorimaging (C). a, RNA ladder; b and c, PV RNA2 WT CRE and KO CRE template RNAs. Note mobility of 28S and 18S rRNAs relative to that of PV RNAs.

FIG. 2.

Elevated UTP concentrations compensated for the defect in negative-strand RNA synthesis associated with CRE mutations. PIRCs with PV RNA2 KO CRE (lanes 1 to 6) or PV RNA2 WT CRE (lanes 7 to 12) templates were incubated for 60 min in the absence (−) or presence (+) of guanidine in reaction mixtures containing either [α-³²P]UTP (A) or [α-³²P]CTP (B). Concentrations of exogenous UTP and exogenous CTP in the reaction mixtures varied from 10 to 250 μM as indicated. A total of 1 mM ATP, 250 μM GTP, and 250 μM of either CTP (A) or UTP (B) were also present in the reaction mixtures. RNA products from the reaction mixtures were separated by electrophoresis in 1% agarose, and radiolabeled RNAs were detected by phosphorimaging. GuHCl, guanidine HCl.

FIG. 3.

CRE-independent negative-strand RNA synthesis was sensitive to low concentrations of UTP. (A) VPg uridylylation. PIRCs with PV RNA2 WT CRE (lanes 1 and 2), PV RNA2 KO CRE (lanes 3 and 4), PV ribozyme KO CRE (lanes 5 and 6), or PV ribozyme WT CRE (lanes 7 and 8) RNA templates were incubated for 60 min in the absence (−) or presence (+) of guanidine in reaction mixtures containing [α-³²P]UTP, 100 μM UTP, and 250 μM CTP. Reaction products were separated by electrophoresis, and VPgpUpU_OH was detected as described in Materials and Methods (25, 29). (B) RNA synthesis in reaction mixtures containing 10 μM UTP (lanes 1 to 5) or 500 μM UTP (lanes 6 to 10), 250 μM CTP, and [α-³²P]CTP. PIRCs with PV RNA2 WT CRE (lanes 1, 2, 6, and 7), PV RNA2 KO CRE (lanes 3 and 8), PV ribozyme KO CRE (lanes 4 and 9), or PV ribozyme WT CRE (lanes 5 and 10) RNA templates were incubated for 60 min in the absence (lanes 2 to 5 and 7 to 10) or presence (lanes 1 and 6) of guanidine. (C) RNA synthesis in reaction mixtures containing 10 μM CTP (lanes 1 to 5) or 250 μM CTP (lanes 6 to 10), 500 μM UTP, and [α-³²P]GTP. PIRCs with PV RNA2 WT CRE (lanes 1, 2, 6, and 7), PV RNA2 KO CRE (lanes 3 and 8), PV ribozyme KO CRE (lanes 4 and 9), or PV ribozyme WT CRE (lanes 5 and 10) RNA templates were incubated for 60 min in the absence (lanes 2 to 5 and 7 to 10) or presence (lanes 1 and 6) of guanidine. Reaction products were separated by electrophoresis in 1% agarose, and radiolabeled RNAs were detected by phosphorimaging. GuHCl, guanidine HCl.

FIG. 4.

Apparent K_m of UTP required for negative-strand RNA synthesis. PIRCs containing PV RNA2 WT CRE (lanes 1 to 9) or PV RNA2 KO CRE (lanes 10 to 18) templates were incubated for 60 min in the presence (lanes 9 and 18) or absence (lanes 1 to 8 and 10 to 17) of guanidine in reaction mixtures containing [α-³²P]CTP, 100 μM CTP, and 0 to 500 μM UTP as indicated. RNA products from the reaction mixtures were separated by electrophoresis in 1% agarose and detected by ethidium bromide and UV light (A) or by phosphorimaging (B). Mobilities of PV RNA, RF RNA, 28S rRNAs, and 18S rRNAs are indicated. (C) Apparent K_m of UTP. The amount of radiolabeled negative-strand RNA within RF bands in panel B was measured by phosphorimaging using Quantity One software with background from the lane 18 guanidine HCl control subtracted. Magnitude of negative-strand RNA synthesis (in arbitrary phosphorimager units) was plotted versus the concentration of UTP in each reaction mixture, and the K_m values for WT (•) and KO CRE (▪) PV RNAs were determined using GraphPad software. Nonlinear regression was used to fit our data using the Michaelis-Menten equation to define the relationship of product formation and substrate concentration, with V_max and K_m as the parameters, as follows: Y = (V_max · X)/(K_m + X), where X is the concentration of UTP. The K_m values of UTP with standard errors for the respective RNAs are presented in the graph. Representative data are from two independent experiments.

FIG. 5.

VPg-linked poly(U) from CRE-dependent and CRE-independent negative-strand RNA synthesis. (A) PV RF RNA fractionated by 1% agarose gel electrophoresis. PIRCs containing WT CRE RNA templates (lanes 1 to 2) or KO CRE RNA templates (lanes 3 to 4) were incubated in reaction mixtures containing 1 mM ATP, 250 μM GTP, 250 μM CTP, 10 μM UTP, [α-³²P]UTP (3 μCi per 1-μl reaction mixture), and 2 mM guanidine HCl (GuHCL; lanes 2 and 4). Reaction products were separated by 1% agarose gel electrophoresis and detected by phosphorimaging. The mobility of PV RF RNA is indicated. (B) RNase T1 oligonucleotides (Oligos) in PV RNAs. WT CRE RNA templates (lane 1) and KO CRE RNA templates (lane 2) were synthesized by T7 RNA transcription in reaction mixtures containing [α-³²P]ATP, digested with RNase T1, and separated by electrophoresis in 7 M urea-18% polyacrylamide (RNase T1; see Materials and Methods) (lanes 1 to 2). [α-³²P]UTP-radiolabeled products of WT CRE (lanes 3 and 4) and KO CRE (lanes 5 and 6) RNA replication, from reaction mixtures with (lanes 4 and 6) and without (lanes 3 and 5) 2 mM guanidine, were digested with RNase T1 and separated by electrophoresis in 7 M urea-18% polyacrylamide (PV RNA replication and RNase T1; see Materials and Methods). Mobilities of specific T1 oligonucleotides and VPg-linked poly(U) are indicated.

FIG. 6.

Model of CRE-independent and CRE-dependent priming of negative-strand RNA synthesis. CRE-independent, VPg-primed, negative-strand RNA synthesis (top) exhibited a K_m of 103 μM UTP, whereas CRE-dependent, VPgpUpU_OH-primed, negative-strand RNA synthesis (bottom) had a K_m of 12 μM UTP. CRE-dependent VPgpUpU_OH must translocate from the CRE RNA template to the 3′ poly(A) template to prime negative-strand RNA initiation as previously suggested by others (36).