Tetracistronic Minigenomes Elucidate a Functional Promoter for Ghana Virus and Unveils Cedar Virus Replicase Promiscuity for all Henipaviruses

Abstract

Batborne henipaviruses, such as Nipah virus and Hendra virus, represent a major threat to global health due to their propensity for spillover, severe pathogenicity, and high mortality rate in human hosts. Coupled with the absence of approved vaccines or therapeutics, work with the prototypical species and uncharacterized, emergent species is restricted to high biocontainment facilities. There is a scarcity of such specialized spaces for research, and often the scope and capacity of research which can be conducted at BSL-4 is limited. Therefore, there is a pressing need for innovative life-cycle modeling systems to enable comprehensive research within lower biocontainment settings. This work showcases tetracistronic, transcription and replication competent minigenomes for Nipah virus, Hendra virus, Cedar virus, and Ghana virus, which encode viral proteins facilitating budding, fusion, and receptor binding. We validate the functionality of all encoded viral proteins and demonstrate a variety of applications to interrogate the viral life cycle. Notably, we found that the Cedar virus replicase exhibits remarkable promiscuity, efficiently rescuing minigenomes from all tested henipaviruses. We also apply this technology to GhV, an emergent species which has so far not been isolated in culture. We demonstrate that the reported sequence of GhV is incomplete, but that this missing sequence can be substituted with analogous sequences from other henipaviruses. Use of our GhV system establishes the functionality of the GhV replicase and identifies two antivirals which are highly efficacious against the GhV polymerase.

Figure 1.. Establishment and rescue of tetracistronic, transcription and replication competent (TC-tr) minigenomes.

(A) Design of rHNV TC-tr minigenomes which encode a HiBiT-mCherry reporter gene in addition to HNV-M, -F, and -RBP. Full-length virus genome structure is shown above for reference. Detailed descriptions of each minigenome design are available in supplementary figure 1. (B) Schematic describing the rescue approach for TC-tr minigenomes in which BSRT7 cells are co-transfected with plasmid encoding the rHNV minigenome, T7-driven HNV-N, -P, -L, and codon-optimized T7 polymerase. A more detailed depiction of the rescue process is available in supplementary figure 2. (C) Microscopy demonstrating rescue of rNiV, rHeV, and rCedV TC-tr minigenomes in the presence (+) or absence (−) of their respective HNV-L proteins. For all images, Hoechst is shown in blue while mCherry is shown in red. (D) Quantification of mCherry positive events produced by rHNV TC-tr minigenome rescue and (E) matched quantification of normalized nanoluciferase signal. To measure vRdRp activity, RLUs from rescues in the presence of HNV-L were normalized to RLUs from rescues in the absence of HNV-L. Statistical significance was assessed by multiple unpaired t tests in GraphPad Prism to compare counts or normalized RLUs in the presence of HNV-L with counts or normalized RLUs in the absence of HNV-L for each respective minigenome. Rescues were conducted in quadruplicate. Error bars depict standard deviation. For all graphs (ns) = P > 0.05 ; (*) = P ≤ 0.05 ; (**) = P ≤ 0.01 ; (***) = P ≤ 0.001 ; (****) = P ≤ 0.0001.

Figure 2.. Matrix protein encoded by rHNV TC-tr VLPs is functionally competent.

(A) Diagram depicting the experimental approach used for the BiFC assay. HeLa cells transfected with VN173-fused NHP2L1 or VN173-fused NOP56 were co-cultured with BSRT7 cells infected with rNiV TC-tr minigenomes encoding WT NiV-M or NiV-M fused with VC155. Fusion mediated by NiV-F and NiV-RBP facilitates cytoplasmic mixing and results in BiFC within mCherry-positive syncytia. Inset cartoon demonstrates reconstitution of Venus fluorescent protein if both VC155-fused NiV-M and VN173-fused host box C/D snoRNP proteins interact. (B) Microscopic images capturing instances of BiFC within mCherry-positive syncytia. For all images, mCherry is colored in red, BiFC (venus reconstitution) is colored in yellow, and Hoechst stain is colored in blue. Scale bars represent 100 micrometers. (C) Detection of TC-tr minigenome vRNA copies in supernatant from rescue cells. (D) Iterative passaging of rNiV and rHeV TC-tr VLPs captured by quantification of mCherry positive events at each passage (P0 = rescue; P1 = passage 1; P2 = passage 2). Rescues and passaging experiments were conducted in triplicate. Error bars depict standard deviation.

Figure 3.. Constrained sequence alignment of the GhV genome reveals that GhV is missing 28 nucleotides from its genomic promoter.

(A) Sequence alignment of the reported genomic vRNA sequences of NiV, HeV, CedV, and GhV with constraint to proper phasing of the genomic promoter element II (PrE-II) sequence. (B) Sequence alignment of the reported antigenomic vRNA sequences of NiV, HeV, CedV, and GhV with constraint to proper phasing of the antigenomic PrE-II sequence. All alignments are shown in a 3’ to 5’ orientation, reflective of the biologically relevant sequences. Bases that are outwards facing towards solvent are shown in full case, while bases buried within the nucleocapsid protein are shown in superscript text. Dashes (−) are used to denote unmapped nucleotides missing in the GhV sequence. The consensus sequence is shown below each hexamer, with absolutely conserved residues colored in red. All nucleotides follow IUPAC nomenclature. (C) Cartoon depicting a vRNP with properly phased bipartite promoters on each end. The inset panel shows the relative localization of PrE-I and PrE-II within the genomic promoter, with numbers detailing each hexamer position relative to the 3’ end of the vRNA. This model reflects phasing observed in the cryoEM structure of the NiV helical nucleocapsid assembly (PDB 7NT5). (D) Cartoon depicting hexamers and their relative positions in the genomic promoter of the GhV vRNP which correspond to incompletely mapped GhV sequence. Missing hexamers/sequence are shown as a transparent outline. For all figures, monomers of nucleocapsid containing a hexamer of vRNA are denoted in brown, except for hexamers 1–3 and 14–16; hexamers encoding elements of the bipartite promoter elements are colored in teal. The antigenomic bipartite promoter elements are depicted with additional shaded pattern.

Figure 4.. Restoration of a functional PrE-I facilitates rescue of a rGhV TC-tr minigenome.

(A) Cartoon depicting the design of chimeric rGhV TC-tr minigenomes in which the unmapped terminal 28 nucleotides of the GhV 3’ Ldr sequence are replaced with equivalent sequences derived from NiV (NiV Ldr28), HeV (HeV Ldr28), CedV (CedV Ldr28), or the GhV antigenomic promoter (GhV Tr28). The minigenome design of rGhV is further detailed in supplementary figure 1D. (B) Quantification of mCherry positive events and (C) matched vRdRp activity resulting from the rescue of each rGhV TC-tr minigenome in the presence (+) or absence (−) of GhV-L. (D) Microscopy depicting rGhV (HeV Ldr28) TC-tr minigenome rescue in the presence or absence of GhV-L. Red depicts mCherry signal while blue depicts Hoechst stain. (E) Comparison of GhV vRdRp activity resulting from the rescue of rGhV TC-tr minigenomes encoding either the HeV Ldr28 sequence or the GhV Tr28 sequence in the presence or absence of GhV-L. (F) Quantification of HeV vRdRp activity resulting from the rescue of rHeV TC-tr minigenomes encoding either the WT (HeV Ldr28) sequence or the HeV Tr28 sequence in the presence or absence of HeV-L. To calculate vRdRp activity, RLUs from rescues in the presence of HNV-L were normalized to respective RLUs generated by each minigenome in the absence of HNV-L. Statistical significance was assessed using a 2way ANOVA analysis in GraphPad Prism to compare counts or normalized RLUs in the presence of HNV-L with respective counts or normalized RLUs in the absence of HNV-L. For (E) and (F), additional comparisons were conducted to determine significance between respective minigenome mutants rescued in the presence of HNV-L. All rescues were conducted in triplicate. Error bars depict standard deviation. For all graphs (ns) = P > 0.05 ; (*) = P ≤ 0.05 ; (**) = P ≤ 0.01 ; (***) = P ≤ 0.001 ; (****) = P ≤ 0.0001.

Figure 5.. Heterotypic cross-rescue of diverse henipaviruses species uncovers a remarkable plasticity in template recognition by the CedV vRdRp.

vRdRp activity above background from the rescue of (A) rNiV, (B) rHeV, (C) rGhV, or (D) rCedV TC-tr minigenomes by NiV-N/-P/-L, HeV-N/-P/-L, CedV-N/-P/-L, or GhV-N/-P/-L, respectively. RLUs from each rescue condition were normalized to RLUs from respective minigenome rescues conducted in parallel but in the absence of HNV-L. (E) Schematic summarizing the capability of each HNV replicase to successfully rescue each rHNV TC-tr minigenome. Statistical significance was determined by ordinary, one-way ANOVA analysis in GraphPad Prism comparing the normalized RLUs yielded by each HNV replicase species to normalized RLUs from the no-L control. All rescues were conducted in triplicate. Error bars demonstrate standard deviation. For all graphs (ns) = P > 0.05 ; (*) = P ≤ 0.05 ; (**) = P ≤ 0.01 ; (***) = P ≤ 0.001 ; (****) = P ≤ 0.0001.

Figure 6.. Incompatibilities exist between the respective replicase, gene starts, and 5’ Tr sequences of GhV and HeV.

(A) Cartoon depicting the rGhV (HeV Ldr28) TC-tr minigenome and the order in which elements required for rescued must be recognized by the vRdRp. (1) The antigenomic promoter in the 5’ Tr must be recognized by the vRdRp to drive synthesis of genomic vRNA; (2) the genomic promoter in the 3’ Ldr of genomic vRNA must be recognized and must prompt the vRdRp to enter scanning mode; and (3) the N gene start must be recognized by the vRdRp to initiate transcription of viral mRNAs. (B) Alignment of the N gene start (GS) sequences from NiV, HeV, CedV, and GhV. Alignments are shown in the genomic vRNA sense, 3’ to 5’. Asterisks depict complete conservation. (C) GhV vRdRp activity above background resulting from the rescue of chimeric rGhV TC-tr minigenomes with systematic replacement of the GhV-N GS or Tr28 sequences with analogous HeV sequences. (D) HeV vRdRp activity above background resulting from the rescue of each chimeric rGhV TC-tr minigenome. RLUs from each rescue condition were normalized to RLUs derived from the rescue of respective minigenomes in the absence of HNV-L. Statistical significance was determined by a 2way ANOVA analysis in GraphPad Prism comparing normalized RLUs from rescues in the presence of HNV-L with normalized RLUs from rescues in the absence of HNV-L. Additional comparisons were conducted to determine significance between the parental rGhV (HeV Ldr28) TC-tr minigenome with the other minigenome mutants rescued in the presence of HNV-L. All rescues were conducted in triplicate. Error bars depict standard deviation. For all graphs (ns) = P > 0.05 ; (*) = P ≤ 0.05 ; (**) = P ≤ 0.01 ; (***) = P ≤ 0.001 ; (****) = P ≤ 0.0001.

Figure 7.. The GhV-L protein is susceptible to two vRdRp inhibitors.

Inhibition curves conducted on a panel of authentic viruses or TC-tr minigenome systems treated with (A) EIDD-2749 or (B) GHP-88309. IC₅₀ values were determined in GraphPad Prism by nonlinear regression of [inhibitor] vs. normalized response, and are listed to the right of each system implemented. The 95% confidence interval estimate for each IC₅₀ is shown in parentheses. Error bars depict standard error of the mean. All inhibition experiments were conducted in at least biological triplicate. (C) Overview of a GhV-L homology model generated using PIV5-L as template (pdb: 6v85).The vRdRP, capping, connector, MTase, and CTD domains are colored blue, green, yellow, orange, and red, respectively. GHP-88309 is shown as orange spheres. The magnified inset depicts the locations where analogous residue mutations are known to induce resistance to GHP-88309. The third inset depicts alignment of a GHP-88309-MeV L complex with the GhV-L model (rms=0.373). Homologous residues to resistance sites in HPIV3 L are shown in red and labeled. GHP-88309 is shown as orange sticks. (D) Sequence alignment of HPIV3, MeV, and members of the HNV genus. All known residues shown to induce resistance (boxed) to GHP-88309 are conserved across genera except for NiV-L and HeV-L at position H1165 (red; H1165).