Structure of the dsDNA bacteriophage HK97 and implications for capsid maturation

William R Wikoff, Purdue University

Abstract

The structure of the HK97 head II empty particle was determined to 7 A resolution. The structure determination used molecular replacement averaging and phase extension. Averaging was initiated at very low (200-57 A) resolution, and phases were extended to 7 A. This is the first crystal structure of a dsDNA bacteriophage, and at 650 Ain diameter, one of the largest virus structures solved to date. The head II particle displays icosahedral (532) symmetry with strict T = 7 quasi symmetry. There are 420 copies of the capsid protein (gp5) arranged on the icosahedral lattice as pentamers and hexamers. The particle is shaped like an icosahedron with flat, triangular faces. The capsid shell is unusually thin, with an average dimension of 25 A. The gp5 monomer is composed largely of alpha helices, with two domains, one adjacent to the pentamer and hexamer axes (I) and the other forming a compact unit near the quasi threefold axes (II). There are a number of unusual features of the structure, particularly the alpha-helical fold of gp5. The arrangement of the subunits in the capsid shell conforms closely to a T = 7 surface lattice. This is the first crystal structure of a T = 7 virus with hexameric morphological units at hexavalent lattice points. dsDNA viruses assemble in a highly coordinated set of events. These events involve large conformational changes in secondary, tertiary and quaternary structures. The central feature of bacteriophage assembly occurs when dsDNA is packaged into the empty prohead shell, termed "expansion". Expansion involves an increase in the capsid diameter and a dramatic conformational change in the capsid. The HK97 head II structure was used in combination with the cryo-EM reconstruction of the prohead II particle to understand the structural transitions during expansion. A model was built by "docking" the partial atomic model of head II into the prohead II cryo-EM reconstruction. The model shows that the central region of the hexamers are "skewed" by 22 A in the prohead, but the central region of the pentamer is unchanged. This central region is formed by domain I of head II gp5. Domains I and II are in different relative orientations in prohead and head particles. The prohead model and head II structure suggest a mechanism for particle maturation. The hexamers in the prohead are skewed into a dimer of trimers (domain I); during expansion the trimers slide relative to each other to form a 6-fold symmetric hexamer. The hexamers and pentamers in the prohead protrude from the capsid surface, and in the process of expansion, flatten out. Spectroscopy of the prohead and head particles indicates that there is a partial refolding, a 50% increase in the alpha-helical content of the capsid during the transition. The refolding may occur in domain II, which cannot be accommodated as a rigid body in prohead II. The model of particle maturation includes this large secondary structure change and the two domain motions: the "un-skewing" of the hexamer, the hinge motion of domain I relative to domain II.

Degree

Ph.D.

Advisors

Johnson, Purdue University.

Subject Area

Biophysics|Biochemistry|Molecular biology|Microbiology

COinS