The Architecture and Chemical Stability of the Archaeal Sulfolobus Turreted Icosahedral Virus

Viruses utilize a diverse array of mechanisms to deliver their genomes into hosts. While great strides have been made in understanding the genome delivery of eukaryotic and prokaryotic viruses, little is known about archaeal virus genome delivery and the associated particle changes. The Sulfolobus turreted icosahedral virus (STIV) is a double-stranded DNA (dsDNA) archaeal virus that contains a host-derived membrane sandwiched between the genome and the proteinaceous capsid shell. Using cryo-electron microscopy (cryo-EM) and different biochemical treatments, we identified three viral morphologies that may correspond to biochemical disassembly states of STIV. One of these morphologies was subtly different from the previously published 27-Å-resolution electron density that was interpreted with the crystal structure of the major capsid protein (MCP). However, these particles could be analyzed at 12.5-Å resolution by cryo-EM. Comparing these two structures, we identified the location of multiple proteins forming the large turret-like appendages at the icosahedral vertices, observed heterogeneous glycosylation of the capsid shell, and identified mobile MCP C-terminal arms responsible for tethering and releasing the underlying viral membrane to and from the capsid shell. Collectively, our studies allow us to propose a fusogenic mechanism of genome delivery by STIV, in which the dismantled capsid shell allows for the fusion of the viral and host membranes and the internalization of the viral genome.Viruses are valuable biological tools for manipulating the cellular processes of their hosts, and they can also serve as model systems for describing macromolecular interactions through the analysis of their architecture. The Sulfolobus turreted icosahedral virus (STIV) is an archaeal virus that infects Sulfolobus solfataricus (phylum Crenarchaeota). STIV is a lytic virus that was isolated from an acidic hot spring (>80°C and pH of <3) in Yellowstone National Park (27). Hence, STIV is an important model for studying the biochemical requirements to sustain life in extreme physicochemical conditions and has the potential to become a tool for the biochemical and genetic manipulation of its host—much like bacteriophages lambda, P22, and phi29 have done for their respective hosts.Prior structural studies of STIV using cryo-electron microscopy (cryo-EM), X-ray crystallography, and proteomics have described large pentameric turret-like structures, with petal-like protrusions emanating from their central shafts (27). The T=31d capsid shell is composed of trimeric capsomers exhibiting pseudo-hexagonal symmetry, in which each of the three capsomer subunits donates two viral jelly rolls with its β-sheets normal to the capsid surface (15, 27). Capsomers surrounding the icosahedral 3-fold axes, and their neighboring subunits, make direct contact with the viral membrane via a highly basic C-terminal helix of each subunit (15, 23). Surrounding the base of the turrets are proteins that make contact with the capsid shell and a host-derived viral membrane (15). The viral membrane and the enclosed viral genome are referred to as the lipid core.The capsid architecture of STIV and the crystal structure of its major capsid protein (MCP) are strikingly similar to those of the bacteriophages PRD1, Bam35, and PM2, the alga virus PBCV-1, and the mammalian adenovirus. This similarity suggests that these viruses share an ancestral virus (2, 4, 7, 15, 25). Given the evolutionary relationship shared between STIV and PRD1, we postulated that the large turret-like vertices of STIV were used to inject the viral genome into the Sulfolobus host—a genome delivery mechanism employed by PRD1 (27).A recent report by Brumfield et al. (5) describes gross cellular ultrastructural changes induced in the Sulfolobus host during STIV infection and release. The authors identified distinct particles that appear to be assembly intermediates of STIV en route to maturation. From these intermediates the authors proposed a general mechanism of capsid assembly, in which MCP subunits and minor capsid proteins (mCPs) coassemble with the lipid membrane to form a lipid-enclosed protein vesicle. These vesicles are spherical and lack the double-stranded DNA (dsDNA) genome and turret-like appendages at the vertices.While these studies confirm an empty procapsid intermediate, the corresponding molecular mechanism associated with assembly and disassembly remains to be understood. Moreover, little is known about STIV or other archaeal virus genome delivery into the host. To obtain a better understanding of the molecular mechanism of STIV architecture and its role in genome delivery, we characterized three distinct morphologies of STIV particles using cryo-EM. An image reconstruction of one of these revealed the absence of a number of constituents decorating the STIV capsid. Hence, for simplicity, we refer to the previously reported image reconstruction (27) as “decorated” and the new image reconstruction reported here as “undecorated.” Reference-free two-dimensional (2D) class averages of the second identified morphology reveal a partially decorated STIV lipid core. The third identified morphology corresponds to the isolated STIV lipid core. Taken together, our analyses indicate that these morphologies correspond to different disassembly intermediates of STIV that can be isolated in vitro and help provide a picture of the STIV capsid architecture. Additionally, these morphologies allow us to propose an alternative possible mechanism of genome delivery.