Sindbis Macrodomain Poly-ADP-Ribose Hydrolase Activity Is Important for Viral RNA Synthesis

Abstract

ADP-ribosylation is a highly dynamic posttranslational modification frequently studied in stress response pathways with recent attention given to its role in response to viral infection. Notably, the alphaviruses encode catalytically active macrodomains capable of ADP-ribosylhydrolase (ARH) activities, implying a role in remodeling the cellular ADP-ribosylome. This report decouples mono- and poly-ARH contributions to macrodomain function using a newly engineered Sindbis virus (SINV) mutant with attenuated poly-ARH activity. Our findings indicate that viral poly-ARH activity is uniquely required for high titer replication in mammalian systems. Despite translating incoming genomic RNA as efficiently as WT virus, mutant viruses have a reduced capacity to establish productive infection, offering a more complete understanding of the kinetics and role of the alphavirus macrodomain with important implications for broader ADP-ribosyltransferase biology. IMPORTANCE Viral macrodomains have drawn attention in recent years due to their high degree of conservation in several virus families (e.g., coronaviruses and alphaviruses) and their potential druggability. These domains erase mono- or poly-ADP-ribose, posttranslational modifications written by host poly-ADP-ribose polymerase (PARP) proteins, from undetermined host or viral proteins to enhance replication. Prior work determined that efficient alphavirus replication requires catalytically active macrodomains; however, which form of the modification requires removal and from which protein(s) had not been determined. Here, we present evidence for the specific requirement of poly-ARH activity to ensure efficient productive infection and virus replication.

Keywords: ADP-ribosylation; PARP/ARTD; RNA synthesis; Sindbis virus; alphavirus; virus replication.

FIG 1

Designing a poly-ARH deficient SINV macrodomain mutant. (A) Sequence alignments of viral and host macrodomains showing the first 45 amino acids of nsP3 (SINV numbering). Catalytic core residues are highlighted in green and the PAR binding motif (residues 37–41) is highlighted in magenta. (B) LOGO plots of a section of host and viral macrodomains with the catalytic core and PAR binding motif highlighted as in (A). (C) Crystal structures of the CHIKV macrodomain in complex with MAR or RNA are depicted with electrostatic potential maps overlaid on the ribbon structure and select residues labeled.

FIG 2

Biochemical validation of APH SINV. (A) Thermal shift analysis of recombinant macrodomains (WT, N10A, APH). Purified protein (10 μM) was incubated over the indicated temperature range in the presence or absence of 2 mM mono-ADPr, fluorescence absorbance was measured, and the first derivative of the fitted model plotted. Triplicate samples were measured and averaged, n = 1 experiment. (B) The indicated macrodomain (20 μM) was incubated (with greek symbol obviously) with 300 μM mono-ADPr and ITC data were plotted (top) and fit to a 1:1 stochiometric binding curve model (middle) for calculation of K_d values (bottom) for WT and APH SINV macrodomains. Data are representative of n = 2 independent experiments. (C-D) PARP12 (mono, C) and PARP1 (poly, D) were auto-ADP-ribosylated in the presence or absence of NAD⁺[³²P] for 30 min. After desalting, 500 nM the indicated macrodomains were incubated with the auto-ADP-ribosylated proteins and MAR and PAR removal were quantified using radiodensitometry. Relative removal of ADP-ribose was calculated for mutant SINV macrodomains normalized to WT. Data represents mean values from n = 2 independent experiments performed in singlicate that were pooled. SARS-1, SARS-CoV-1; ns, not significant. Error bars correspond to SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001, one-way ANOVA.

FIG 3

SINV poly-ARH activity is critical for macrodomain function. (A) M0 cells were infected with WT, N10A, and APH SINV (multiplicity of infection [MOI] 0.1 as determined on the respective cell line) and virus titers in the supernatants were determined by plaque assay on BHK-21 cells. 3AB (1 mM) or DMSO vehicle control was included during and after the infection. Mean titers from representative experiments performed in duplicate are plotted. (B) RNA was harvested from M0 infected with the indicated SINV variants or mock infected for 24 h. After cDNA synthesis, relative mRNA levels for the indicated inflammatory genes were determined, normalizing with RPS11. Mean fold change from mock values from triplicate biological samples in technical duplicates are plotted. Data are representative of n = 2 experiments. (C) Triplicate samples of V5HF cells were infected (MOI 0.5) with WT, N10A, and APH viruses. Medium and cells were harvested at the indicated times and viral titers in the medium were determined by plaque assay using BHK-21 cells (left). Cells were subjected to staining for dsRNA using J2 antibody and the percentage of dsRNA+ cells was determined by flow cytometry (right). Mean values are plotted and are representative of n = 3 experiments. (D) V5HF and STAT1^−/− HF cells were infected with WT, N10A and APH SINV (MOI 0.5) and virion production was determined after 16 h (left). Cells were harvested and the percentage of infected cells determined by staining for the viral dsRNA followed by flow cytometry (right). Mean values of triplicate samples are plotted. Data are representative of n = 3 experiments. (E) Infectious center assay in V5HF cells. Cells were infected (MOI 0.5) and cocultured with BHK-21 cells. After 2 d the monolayers were fixed, and plaques were enumerated. Mean values of triplicate samples are plotted. Data are representative of n = 3 experiments. In all graphs, error bars correspond to SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001, one-way ANOVA for panels B, D & E; two-way-ANOVA for panel C.

FIG 4

APH attenuation occurs after genome translation. (A) SINV virus and replicon schematics showing Toto1101 temperature sensitive NanoLuc virus (tsNLuc, top) and SINrep5 mNeonGreen (SR5mNG, bottom) The orange coloration indicates nsP4 that contains mutations conferring temperature sensitivity to RNA replication. The APH macrodomain mutations were introduced into these constructs. V5HF cells were infected (MOI 10, as determined on the respective cell line) (B) or transfected (250 ng) (C) with WT or APH tsNLuc virus or tsNLuc RNA, respectively, and incubated at 40°C. Luciferase activity in lysates collected at the indicated time points is shown in B. Luciferase activity in lysates collected 24 h after transfection is shown in C. Mean values from quadruplicate samples are plotted; error bars represent the SD. Data are representative of n = 3 experiments. (D) V5HF cells were transfected with WT, N10A, or APH SR5mNG and harvested for flow cytometric analysis at the indicated times. The percentage of cells expressing mNG is plotted (left); the median fluorescence intensity of the 12 h sample is shown in the bar graph (right). Error bars represent the SD of triplicate samples. Data are representative of n = 3 experiments. *, P < 0.05; **, P < 0.01; ****, P < 0.0001, Student's t test for C, and two-way-ANOVA for the left panel of D, and one-way-ANOVA for the right panel of D.

FIG 5

Attenuated poly-ARH activity depresses viral RNA synthesis. (A) immunoFISH schematic. (B) V5HF cells were infected with WT or APH SINV (MOI 20, as determined on V5HF cells) and at the indicated times the cells were fixed and subjected to smFISH and immunostaining for quantification of the genomic plus-strand and dsRNA, respectively. The MFI of the signals in individual cells is plotted; data are representative of n = 2 experiments with 25 automated images captured per well. (C) Dot plot linking dsRNA and plus-strand levels (arbitrary fluorescence units) within single cells. For each cell at each time point, colored as indicated, the plus-strand and dsRNA fluorescent signals are plotted.