Characterization, Directed Evolution, and Targeting of DNA Virus-Encoded RNA Capping Enzymes Using Phenotypic Yeast Platforms

Abstract

The constant and the sudden emergence of zoonotic human and animal viruses is a significant threat to human health, the world economy, and the world food supply. This has necessitated the development of broad-spectrum therapeutic strategies to combat these emerging pathogens. Mechanisms that are essential for viral replication and propagation have been successfully targeted in the past to develop broad-spectrum therapeutics that can be readily repurposed to combat new zoonotic pathogens. Because of the importance of viral RNA capping enzymes to viral replication and pathogenesis, as well as their presence in both DNA and RNA viruses, these viral proteins have been a long-standing therapeutic target. Here, we use genome sequencing information and yeast-based platforms (YeRC0M) to identify, characterize, and target viral genome-encoded essential RNA capping enzymes from emerging strains of DNA viruses, i.e., Monkeypox virus and African Swine Fever Virus, which are a significant threat to human and domestic animal health. We first identified and biochemically characterized these viral RNA capping enzymes and their necessary protein domains. We observed significant differences in functional protein domains and organization for RNA capping enzymes from emerging DNA viruses in comparison to emerging RNA viruses. We also observed several differences in the biochemical properties of these viral RNA capping enzymes using our phenotypic yeast-based approaches (YeRC0M) as compared to the previous in vitro studies. Further, using directed evolution, we were able to identify inactivation and attenuation mutations in these essential viral RNA capping enzymes; these data could have implications on virus biocontainment as well as live attenuated vaccine development. We also developed methods that would facilitate high-throughput phenotypic screening to identify broad-spectrum inhibitors that selectively target viral RNA capping enzymes over host RNA capping enzymes. As demonstrated here, our approaches to identify, characterize, and target viral genome-encoded essential RNA capping enzymes are highly modular and can be readily adapted for targeting emerging viral pathogens as well as their variants that emerge in the future.

Figure 1.

(A) Several emerging viruses encoded their own RNA-capping enzymes. These enzymes are involved in the maturation of the cap 0 (7-methylguanosine-5′-triphosphate), cap 1 (2’O-methylation), and cap 2 (2’-O-methylation) structure. (B) Conventionally, in eukaryotic host cells, including mammalian cells, the cap-0 maturation process proceeds as follows: (1) hydrolysis of the 5′ γ-phosphate of RNA by an RNA triphosphatase, (2) transfer of a GMP molecule onto the 5′-end of RNA by a guanylyl transferase, and (3) methylation of this guanosine by a guanine-N7-methyltransferase. Viruses often use distinct mechanisms and proteins architecture for the maturation of their own RNA cap-0 structure. Outlined here are the strategies by which African Swine Fever virus, SARS-CoV-2, Monkeypox virus, human, and yeast cells synthesize cap 0 structures. Despite significant differences between these systems, we have shown that our yeast derived platforms (YeRC0M) are compatible with several viral RNA-capping enzymes. Figure made using Biorender.

Figure 2.

Using YeRC0M to test the N7-MTase activity of various truncations of MPV vD1. (A) Schematic description of YeRC0M platform. (B) Ability of WT MPV vD1 and truncations to catalyze the N7-methylation of yeast mRNAs were studied by transforming their respective plasmid construct in S. cerevisiae abd1::kanMX4 pMO1 pMO42 haploid and monitoring their ability to recover growth in the presence of 5-FOA. Data represent three biological replicates. Plasmid name and details are listed. (C) Representative strains listed in panel (A) were grown in liquid selection medium in the presence of 5-FOA and the total growth after 48 h was measured (OD600). Each circle represents an individual replicate. Bars represent the average of four replicates and error bars represent mean ± SD. (D) SWISS-MODEL structure of the MPVvD1 + vD12 complex. vD12 is shown in gray ribbons. vD1 is shown in color: green ribbon corresponds to amino acids 783–844, wheat ribbon corresponds to amino acids 560–590, red ribbon corresponds to amino acids 498–560, and blue ribbon is the rest of vD1. SAH is shown in orange sticks in the vD1 active site.

Figure 3.

Using YeRC0M for functional characterization of MPV vD1 N7-MTase activity and the identification of inactive variants. (A) Ability of WT MPV vD1 and a series of vD1 mutants to catalyze the N7-methylation of yeast mRNAs were studied by transforming their respective plasmid construct in S. cerevisiae abd1::kanMX4 pMO1 pMO42 haploid and monitoring their ability to recover growth in the presence of 5-FOA. Data represent three biological replicates. (B) Representative strains listed in panel (A) were grown in liquid selection medium in the presence of 5-FOA and the total growth after 48 h was measured (OD600). Each circle represents an individual replicate. Bars represent the average of four replicates and error bars represent mean ± SD. (C) Active site images of an overlay of SWISS-MODEL MPV vD1/vD12 complex (blue) and VACV vD1/vD12 complex crystal structure (black) (pdb: 2VDW) with ADOMET coenzyme (shown in orange). Residues targeted for MPV vD1 mutagenesis in this study are labeled: Y683, R548, R655, D620, and D598.

Figure 4.

Using YeRC0M for the identification of attenuated variants of MPV vD1 N7-MTase. (A) Ability of WT MPV vD1 and a series of mutants of vD1 to catalyze the N7-methylation of yeast mRNAs were studied by transforming their respective plasmid construct in S. cerevisiae abd1::kanMX4 pMO1 pMO42 haploid and monitoring their ability to recover growth in the presence of 5-FOA. Data represent three biological replicates. (B) Representative strains listed in panel (A) were grown in the liquid selection medium in the presence of 5-FOA and the total growth after 48 h was measured (OD600). Each circle represents an individual replicate. Bars represent the average of four replicates and error bars represent mean ± SD. (C) Growth curves of S. cerevisiae abd1::kanMX4 pMO42 complemented by MPV vD1 attenuation mutants identified in this study (D545A (yellow line), R548A (gray line), and R655L (green line), R548K (orange line), and Y683V (blue line)) and WT MPV vD1 (black line). Each data point represents the mean of three biological replicates. Error bars represent mean ± SD. (D) Time to mid-exponential growth of attenuated vD1 mutants identified in this study.

Figure 5.

Using YeRC0M for functional characterization of ASFV NP868R N7-MTase activity and the identification of inactive variants. (A) Ability of WT ASFV NP868R and a series of NP868R mutants to catalyze the N7-methylation of yeast mRNAs were studied by transforming their respective plasmid construct in S. cerevisiae abd1::kanMX4 pMO1 pMO42 haploid and monitoring their ability to recover growth in the presence of 5-FOA. Data represent three biological replicates. (B) Representative strains listed in panel (A) were grown in the liquid selection medium in the presence of 5-FOA and the total growth after 120 h was measured (OD600). Each circle represents an individual replicate. Bars represent the average of three replicates and error bars represent mean ± SD. (C) Active site of ASFV NP868R N7-MTase (pdb: 7D8U) with the ADOMET coenzyme shown in yellow. Residues targeted for MPV vD1 mutagenesis in this study are labeled: Y714, F711, D680, D646, and K647.

Figure 6.

Using YeRC0M for the identification of attenuated variants of ASFV NP868R N7-MTase activity. (A) Ability of WT ASFV NP868R and a series of mutants of NP868R to catalyze the N7-methylation of yeast mRNAs were studied by transforming their respective plasmid construct in S. cerevisiae abd1::kanMX4 pMO1 pMO42 haploid and monitoring their ability to recover growth in the presence of 5-FOA. Data represent three biological replicates. (B) Representative strains listed in panel (A) were grown in liquid selection medium in the presence of 5-FOA and the total growth after 120 h was measured (OD600). Each circle represents an individual replicate. Bars represent the average of three replicates and error bars represent mean ± SD. (C) Growth curves of S. cerevisiae abd1::kanMX4 complemented by ASFV NP868R attenuation mutants identified in this study (K647Y (red line), F711W (blue line), F711L (green line), and D680L (gray line) in comparison to WT ASFV NP868R (black line). Each data point represents the mean of three biological replicates. Error bars represent mean ± SD. (D) Time to mid-exponential growth of attenuated vD1 mutants identified in this study.

Figure 7.

Expanding YeRC0M to identify viral N7-MTase specific inhibitors. (A) Human N7-MTase rnmt plasmid expression is able to recover S. cerevisiae abd1::kanMX4 haploid growth in the presence of 5-FOA. Data represent three biological replicates. (B) This platform can be used to identify small molecule inhibitors of viral N7-MTases. S. cerevisiae abd1::kanMX4 pMO5 haploid can be used as a negative control to ensure small molecule inhibitors only inhibit viral N7 MTase activity, not RNMT activity. (C) Growth rate perturbation of S. cerevisiae abd1::kanMX4 pMO41 pMO42 (YeRC0M-MPV) and S. cerevisiae abd1::kanMX4 pMO5 (YeRC0M-human) in the presence of 25 and 50 μM disulfiram. Cultures were grown in the liquid selection medium, and the total growth after 48 h was measured (OD600). Each circle represents an individual replicate. Bars represent the average of four replicates and error bars represent mean ± SD. (D) Using 96-well plate-based screening method, we calculated the IC₅₀ of disulfiram for the yeast strains expressing viral RNA-capping N7-MTase enzymes and compared it to the yeast strains expressing human RNA cap N7-MTase (9.2 μM IC₅₀ for YeRC0M-MPV vD1/vD12, 9.9 μM IC₅₀ for YeRC0M-ASFV NP868R, and 15.2 μM IC₅₀ for YeRC0M-RNMT).