Structures and Nucleotide Analog Inhibition of the CCHFV Polymerase Reveal Sofosbuvir-Like Compounds as Potent Chain Terminators
核心洞察
Full-length structures of the Crimean–Congo hemorrhagic fever (搜索) virus (CCHFV) L polymerase, the largest known viral polymerase at ~4000 residues, have been resolved including a 3.0-Å elongation complex.
The structures reveal large insertions in all three major functional regions—endonuclease, RdRP, and cap-binding domains—that create critical interaction networks for virus replication.
Nucleotide analogs with ribose-2′-modifications identical to sofosbuvir specifically and efficiently inhibit CCHFV RdRP through an immediate chain termination mechanism.
A landmark structural study has unveiled the full-length architecture of the Crimean–Congo hemorrhagic fever (搜索) virus (CCHFV) L polymerase—the largest known viral polymerase at approximately 4,000 residues—and demonstrated that nucleotide analogs bearing the same ribose-2′-modifications as the hepatitis C drug sofosbuvir potently inhibit the enzyme through immediate chain termination. The findings, published in Nature, provide the first global structural information for any Nairoviridae polymerase and open a potential therapeutic avenue for a pathogen designated by the World Health Organization as a priority disease requiring urgent research attention.
The CCHFV L protein belongs to the Nairoviridae family and has long resisted structural characterization due to its extraordinary size and complexity. Researchers now report structures of the full-length polymerase, including a 3.0-Å-resolution polymerase elongation complex that captures both early and late elongation stages of RNA synthesis.
Structural Architecture Reveals Unique Features
The CCHFV polymerase is organized into three major functional regions: the PA-like region containing the endonuclease domain, the PB1-like region housing the RNA-dependent RNA polymerase (RdRP) core, and the PB2-like region containing the cap-binding domain. Compared with other bunyavirus and influenza polymerases—including those from SFTSV, LACV, TSWV, LASV, HTNV, and influenza A—the CCHFV L protein is noticeably larger in all three regions.
The PA- and PB1-like domains contain additional structural elements in the endonuclease and α-ribbon-like regions, respectively, while the PB2-like region adopts a distinct modular architecture. These large additions and insertions extend RNA binding paths on both sides of the RdRP active site and create interaction networks that are critical for virus replication, as confirmed by CCHFV minigenome assay data.
Conformational Dynamics During RNA Synthesis
Binding of the 5′ viral RNA (vRNA) induces ordering of residues L1233–1257, while 3′ vRNA binding triggers structural changes in the vRBL domain, notably ordering of residues L1092–1104. Upon template-product RNA duplex formation and UpNHpp binding, extensive rearrangements occur throughout the RdRP core. Motifs A–D in the palm domain undergo conformational changes, motifs G–H in the fingers domain become ordered, and the priming loop of the thumb domain is fully extruded. The lid and bridge domains, which interact with the upstream end of the RNA duplex, undergo marked structural rearrangements that pull on thumb-ring-1, driving its conformational remodeling.
These core rearrangements create surfaces for peripheral domain docking. Hydrophobic interactions between the lid domain and linkers from peripheral domains, along with interactions between the thumb-ring-1 domain and the mid-link and 627-like domains, assemble the complete elongation complex.
Mutagenesis Confirms Functional Importance
Mutational analysis targeting 5′ vRNA-interacting residues demonstrated the functional significance of these structural features. Four mutant polymerases—LM4 (R1265A/N1381A/Y1399A/R1551A), LM5 (N1259A/S1025A/R1257A/K1261A), LM6 (W2444A/F1255A), and LM7 (R1252A/S1248A)—all showed severely impaired RNA synthesis relative to wild-type, despite comparable purity as confirmed by SDS-PAGE.
Cap-Binding Domain Characterization
The cap-binding domain (CBD), subdivided into the β-platform, α-linker, and C-flank, was shown to specifically bind m7GTP cap structures. Microscale thermophoresis (MST) measurements quantified the binding affinity, and m7GTP pull-down assays confirmed specificity. This domain architecture supports the cap-snatching mechanism used by the virus to initiate transcription.
Sofosbuvir-Like Nucleotide Analogs as Potent Inhibitors
A critical finding of the study is that nucleotide analogs with ribose-2′-modifications identical to sofosbuvir—the approved hepatitis C virus drug—specifically and efficiently inhibit CCHFV RdRP. These compounds act through an immediate chain termination mechanism, halting RNA synthesis once incorporated into the growing RNA strand.
Using the sofosbuvir–hepatitis C virus RdRP system as a reference, the potency of these nucleotide analogs was further demonstrated in competition assays conducted in the presence of corresponding natural NTPs. The results indicate that these compounds can compete effectively with natural substrates at the polymerase active site.
Endonuclease Domain Comparisons
Structural comparisons of the CCHFV endonuclease domain with those of influenza A and KASV polymerases revealed both conserved and distinct features in RNA substrate recognition and inhibitor binding. The influenza A PA endonuclease–baloxavir complex structure highlights how existing influenza therapeutics target this domain, while the KASV endonuclease–WXSH0208 complex provides a template for understanding inhibitor interactions in related viruses. These comparisons may inform structure-based drug design targeting the CCHFV endonuclease.
The study represents a significant advance in understanding the molecular mechanisms of Nairoviridae replication and establishes a structural framework for the rational development of antiviral therapies against CCHFV, for which no approved treatments currently exist.