Structure of herpes simplex virus pUL7:pUL51, a conserved complex required for efficient herpesvirus assembly

Abstract

Herpesviruses are an ancient family of highly-prevalent human and animal pathogens that acquire their membrane envelopes in the cytoplasm of infected cells. While multiple conserved viral proteins are known to be required for efficient herpesvirus production, many of these proteins lack identifiable structural homologues and the molecular details of herpesvirus assembly remain unclear. We have characterized the complex of assembly proteins pUL7 and pUL51 from herpes simplex virus (HSV)-1, an α-herpesvirus, using multi-angle light scattering and small-angle X-ray scattering with chemical crosslinking. HSV-1 pUL7 and pUL51 form a stable 1:2 complex that is capable of higher-order oligomerization in solution. We solved the crystal structure of this complex, revealing a core heterodimer comprising pUL7 bound to residues 41-125 of pUL51. While pUL7 adopts a previously-unseen compact fold, the extended helix-turn-helix conformation of pUL51 resembles the cellular endosomal complex required for transport (ESCRT)-III component CHMP4B, suggesting a direct role for pUL51 in promoting membrane scission during virus assembly. We demonstrate that the interaction between pUL7 and pUL51 homologues is conserved across human α-, β-and γ-herpesviruses, as is their association with trans-Golgi membranes in cultured cells. However, pUL7 and pUL51 homologues do not form complexes with their respective partners from different virus families, suggesting that the molecular details of the interaction interface have diverged. Our results demonstrate that the pUL7:pUL51 complex is conserved across the herpesviruses and provide a structural framework for understanding its role in herpesvirus assembly. Significance StatementHerpesviruses are extremely common human pathogens that cause diseases ranging from cold sores to cancer. Herpesvirus acquire their membrane envelope in the cytoplasm via a conserved pathway, the molecular details of which remain unclear. We have solved the structure of a complex between herpes simplex virus (HSV)-1 proteins pUL7 and pUL51, two proteins that are required for efficient HSV-1 assembly. We show that formation of this complex is conserved across distantly-related human herpesviruses, as is the association of these homologues with cellular membranes that are used for virion assembly. While pUL7 adopts a previously-unseen fold, pUL51 resembles key cellular membrane-remodeling complex components, suggesting that the pUL7:pUL51 complex may play a direct role in deforming membranes to promote virion assembly.

Structure Overview of Herpesvirus Tegument Proteins

“…An interesting issue to note is the oligomerization tendency of UL7-UL51 and BBRF2-BSRF1 complexes. More than the crystallographic multimers mentioned in the previous section, UL7 and UL51 were found to form hetero-complexes with different stoichiometric ratios with the potential for further oligomerization 23 , and near-full-length BSRF1 with intact C-terminus is a dimer rather than a monomer as BSRF1Δ 35 . The BSRF1 homolog in HCMV, UL71, contains a basic leucine zipper (bZIP)-like domain, which accounts for the oligomerization of UL71 and is essential for viral envelopment 36 .…”

“…The UL7-UL51 interaction was found first in HSV 23 and later proved a conserved event for all three herpesvirus subfamilies, including EBV BBRF2-BSRF1 12 and HCMV UL103-UL71 24 . Loss of either UL7 or UL51 reduces the virus replication of HSV-1 and affects viral secondary envelopment [25][26][27][28][29][30] .…”

“…The crystal structure of BBRF2 was determined at high resolution as a monomer. In comparison, the solo structure of UL7 was not available, because UL7 tends to aggregate in solution 23 . The globular BBRF2 contains a central 6-stranded β-sheet surrounded by 10 helices as UL7, representing a new fold (named herpesvirus tegument fold 1, or HTF1; also mentioned as 'conserved UL7 tegument assembly/release domain', or CUSTARD fold, for HSV-1 UL7) that has not been observed for other proteins in the structural database.…”