Biochemical characterization of enzyme fidelity of influenza A virus RNA polymerase complex

Abstract

Background: It is widely accepted that the highly error prone replication process of influenza A virus (IAV), together with viral genome assortment, facilitates the efficient evolutionary capacity of IAV. Therefore, it has been logically assumed that the enzyme responsible for viral RNA replication process, influenza virus type A RNA polymerase (IAV Pol), is a highly error-prone polymerase which provides the genomic mutations necessary for viral evolution and host adaptation. Importantly, however, the actual enzyme fidelity of IAV RNA polymerase has never been characterized.

Principal findings: Here we established new biochemical assay conditions that enabled us to assess both polymerase activity with physiological NTP pools and enzyme fidelity of IAV Pol. We report that IAV Pol displays highly active RNA-dependent RNA polymerase activity at unbiased physiological NTP substrate concentrations. With this robust enzyme activity, for the first time, we were able to compare the enzyme fidelity of IAV Pol complex with that of bacterial phage T7 RNA polymerase and the reverse transcriptases (RT) of human immunodeficiency virus (HIV-1) and murine leukemia virus (MuLV), which are known to be low and high fidelity enzymes, respectively. We observed that IAV Pol displayed significantly higher fidelity than HIV-1 RT and T7 RNA polymerase and equivalent or higher fidelity than MuLV RT. In addition, the IAV Pol complex showed increased fidelity at lower temperatures. Moreover, upon replacement of Mg(++) with Mn(++), IAV Pol displayed increased polymerase activity, but with significantly reduced processivity, and misincorporation was slightly elevated in the presence of Mn(++). Finally, when the IAV nucleoprotein (NP) was included in the reactions, the IAV Pol complex exhibited enhanced polymerase activity with increased fidelity.

Significance: Our study indicates that IAV Pol is a high fidelity enzyme. We envision that the high fidelity nature of IAV Pol may be important to counter-balance the multiple rounds of IAV genome amplification per infection cycle, which provides IAV Pol with ample opportunities to generate and amplify genomic founder mutations, and thus achieve optimal viral mutagenesis for its evolution.

Figure 1. NTP concentration dependent ApG-primed RNA polymerase activity of IAV Pol complex.

ApG primer (0.3 mM) was extended by IAV Pol using RNA templates encoding the first 14 (A), 50 (B) and 137 (C) nucleotides (nt) 3′ end sequences of the IAV PA gene. RNA polymerization was initiated by adding 1 mM ATP, 1 mM UTP, 1 mM CTP, 0.08 µM α-³²P-GTP, and varying concentrations of nonradioactive GTP (lanes 1-5; 0.1, 1.0, 10, 100 and 1000 µM). M: Size markers C: Negative control without polymerase. F: Fully extended products. The fully extended products observed in the reaction with 50-nt (D) and 137-nt (E) were determined by densitometric analysis (observed activity) and these values were plotted against the actual GTP concentrations used in each reaction. The activity at 0.1 µM GTP was set arbitrarily to 100%.

Figure 2. Misincorporation assay with IAV Pol, bacterial phage T7 RNA polymerase, HIV-1 RT and MuLV RT with biased nucleotide substrate pools.

(A) Template sequences used for the misincorporation assay with IAV Pol, T7 RNA polymerase and RTs of HIV-1 and MuLV. The IAV Pol template used is a 30-nt sequence from the 3′ end of the viral PA sequence with a ApG primer binding site (P:primer, T:template,). The first UTP or TTP incorporation site of each template was marked with “X”, and the first stop site in the (−) UTP or TTP reaction was marked with “1*” under each template sequence (the second and third stop sites were marked “2*” and “3*” for the IAV template). The RNA synthesis initiation sites were marked in blue. (B) ApG-initiated RNA-dependent RNA polymerization by IAV Pol: ApG primer was extended with a 30-nt RNA template and three different amounts of H3N2 IAV Pol protein (1x, 2x and 3x) with 500 µM four NTPs (“4”) and 0.16 µM α-³²P-GTP, and the same reactions were repeated except using two biased NTP pools, minus UTP (“- U”) or minus ATP (“- A”). The sequence of the incorporated nucleotides near the three UTP stop sites (red) is shown at the side. The extended products in the (−) UTP reaction, which was used for calculating the misincorporation efficiency, was marked as “]”.Dotted lines refer to RNAs. (C) DNA dependent RNA polymerization of bacterial phage T7 RNA polymerase: A 47 bp ds DNA (box) encoding T7 promoter (grey box) and 29 bp sequence (white box) was used for RNA synthesis with T7 RNA polymerase at 37°C for 60 mins. “]”: Fully extended misincorporated product. (D) and (E) RNA dependent DNA polymerization reaction by HIV-1 RT (D) and MuLV RT (E). A 48 mer RNA template annealed to a 18-mer single stranded DNA primer was used for the DNA polymerization by HIV-1 and MuLV RTs at 37°C for 60 mins. 500 µM dNTPs mixed with α-³²P-dNTPs (0.16 µM), which is the same nucleotide concentration and ratio as used in the reactions with IAV Pol and T7 RNA polymerase, was used for DNA synthesis. “]”: Fully extended misincorporated product. Dotted lines refer to RNA and solid lines refer to DNA. (F) and (G) Comparison of the misincorporation efficiency of the four polymerases. For calculation of the misincorporation efficiency, the fully extended misincorporated product in (−) U/(−) T reactions in all four polymerases was normalized to the total extended product in the lanes with all four NTPs (dNTPs). The fold differences of the calculated misincorporation percentages between three activities of IAV Pol (F) and between IAV Pol and three other polymerases (G) were determined. At least five repeats of the assay were conducted in this analysis.

Figure 3. Effect of temperature on polymerase activity and misincorporation efficiency of IAV Pol complex.

(A) IAV polymerization at different temperatures. LC: loading control (B) Fully extended products for the polymerase activity of IAV Pol at the four different temperatures were compared after being normalized with the value obtained at 30°C (100%). (C) The misincorporation efficiency of IAV Pol at four temperatures were calculated as described in Figure 2F, and compared with the value obtained at 30°C (100%). C: no polymerase control.

Figure 4. Effect of manganese on polymerase activity and misincorporation efficiency of IAV Pol complex.

(A) IAV polymerization with 5 mM Mg(OAC)₂ and varying concentration of MnCl₂ (lanes 1 to 4, 0.5, 1, 5, 7.5 mM). LC: loading control (B) Fully extended products for the IAV Pol polymerase activity with different metal concentrations were compared after being normalized with the value obtained with 5 mM Mg(OAC)₂ (100%). (C) The misincorporation efficiency of IAV Pol at 0.5 mM MnCl₂ was calculated as described in Figure 2F, and compared with the value obtained with 5 mM Mg(OAC)₂ (100%). C: no polymerase control.

Figure 5. Effect of H3N2 NP on polymerase activity and the misincorporation efficiency of the IAV Pol complex.

(A) IAV polymerization in the presence and absence of varying concentrations of H3N2 NP with the 50-mer template used in Figure 1B. Molar ratios between the RNA template and NP molecule were altered between 0 and 2. The RNA template was preincubated with NP before adding the polymerase complex. LC: loading control; (B) Fully extended products for the IAV Pol activity with different molar ratios of RNA/NP were compared after being normalized with the value obtained with no NP (100%). (C) The misincorporation efficiency of IAV Pol was calculated as described in Figure 2F, and compared using the value obtained with no NP (100%). C: no polymerase control.