Crystal structure of the simian virus 40 large T-antigen origin-binding domain

Abstract

The origins of replication of DNA tumor viruses have a highly conserved feature, namely, multiple binding sites for their respective initiator proteins arranged as inverted repeats. In the 1.45-angstroms crystal structure of the simian virus 40 large T-antigen (T-ag) origin-binding domain (obd) reported herein, T-ag obd monomers form a left-handed spiral with an inner channel of 30 angstroms having six monomers per turn. The inner surface of the spiral is positively charged and includes residues known to bind DNA. Residues implicated in hexamerization of full-length T-ag are located at the interface between adjacent T-ag obd monomers. These data provide a high-resolution model of the hexamer of origin-binding domains observed in electron microscopy studies and allow the obd's to be oriented relative to the hexamer of T-ag helicase domains to which they are connected.

FIG. 1.

The SV40 core origin and structure of SV40 large T-antigen origin binding domain. (A) DNA sequence of the SV40 core origin. The 64-base-pair SV40 core origin is depicted as B-form DNA, with the central, site II region colored the same in the sequence and in the DNA double helix. The GAGGC pentameric sequences labeled P1 to P4 are indicated in cyan and their complements in magenta. The cyan arrows indicate the 5′→3′ orientation of the pentameric sequence. The arrangement of the pentamers positions the major grooves of P1 and P3 on approximately the same face of the DNA. The same is true for P2 and P4. In addition, P1 and P2 (like P3 and P4) are on opposite faces of the DNA. The AT-rich and early palindrome (EP) sequences are gray; the mononucleotide spacer between the pentamers is yellow. (B) The amino acid sequence of T-ag obd with corresponding secondary-structure assignments (using the program DSSP) indicated by cylinders (alpha helices) or arrows (beta sheets); the A1 and B2 motifs are indicated. Residues closer than 4 Å to the protein-protein interface are indicated by boxes, magenta in one subunit and green in the adjacent subunit. (C) Ribbon diagram of the T-ag obd monomer with secondary structure elements labeled. Residues at the protein-protein interface that generates the spiral are magenta and green as in panel B. (D) Ribbon diagram of six spirally arranged obd molecules. The sixfold screw axis is perpendicular to the page. Each monomer is colored differently. The inner diameter of the channel is ∼30 Å. These six obd's form an “open ring” or spiral, and the position of the gap is shown. (E) Ribbon diagram of the spiral shown in side view to visualize the gap and the screw translation. Each monomer is colored as in panel D. The black line through the center of the spiral represents the sixfold screw axis (panel D rotated by ∼90°). The pitch of the spiral is 35.8 Å. The opening or gap between the first molecule (yellow) and the last molecule (red) is indicated. Figures were made with programs NUCCYL and PYMOL unless otherwise indicated.

FIG. 2.

Structure of spirally arranged T-ag obd's. (A) Surface representation of the T-ag obd hexamer. The surface is colored according to the electrostatic potential. Blue surfaces indicate positive potential and red negative (+/−10 kT/e). Three views showing each face of the hexamer, as well as a side view to better illustrate the spiral, are included. A cartoon schematic of the spirally arranged T-ag obd's is presented below each view to indicate the direction of rotation. A yellow triangle is placed at the same position on each hexamer to aid in orientation. This figure was made using the SwissPDB viewer and rendered using POVRAY. (B) Residues involved in the protein-protein interface between each adjacent monomer. Residues listed are within 4 Å of the interface. Magenta ovals indicate residues from one subunit, and green squares indicate residues from the adjacent subunit. An asterisk indicates that mutagenesis resulted in a T-ag molecule that was defective in the initiation of replication (see text). (C) Surface representation displaying the protein-protein interface described in panel B. The contact surfaces of adjacent subunits are depicted in magenta and green. (D) Close-up of the protein-protein interface. Side chains of residues of adjacent subunits at the protein-protein interface are magenta and green. Residues known to be important for hexamerization by mutagenesis (see text) are cyan. The van der Waals surface for Phe 151 (magenta), Phe 183, and Ser 185 (both cyan) shows the tight packing of the protein interface. All other protein residues are yellow.

FIG. 3.

Surface representation of the T-ag obd hexamer and the modeled DNA complex. (A) The surface representation of the T-ag obd hexamer (yellow) shows that the A1 (amino acids 147 to 159) motif (red) and B2 (amino acids 203 to 207) motif (purple) map onto the inner channel and one surface of the T-ag obd hexamer. The subunits are labeled A to F. This view is the same as the left panel of Fig. 2A. (B) Model of the T-ag obd hexamer with duplex DNA running through the central channel of the T-ag obd hexamer; the DNA is colored as in Fig. 1A. The protein is shown as a surface representation colored as in panel A. The view is rotated ∼45° relative to the view of the model presented in panel A. The spiral positions two of the six T-ag obd's proximal to the repeating GAGGC sequences in P1 and P2, and the duplex DNA easily fits in the positively charged central channel. (C) Cartoon of a T-ag obd spiral with DNA along the central channel. The T-ag obd's are represented as orange spheres and the DNA as a cylinder. The positions of pentamers P1 and P2 are indicated on the DNA. This schematic shows how subunit E is poised for interaction with P1 while subunit B (180 degrees away from the first) is positioned to interact with P2.

FIG. 4.

Model of the T-ag obd double hexamer with DNA. (A) The T-ag obd double-hexamer model is shown as a ribbon diagram with B-form DNA along the sixfold screw axis. The DNA is colored as in Fig. 1A. The T-ag obd hexamers are green and yellow. The A1 (red) and B2 (purple) loops are oriented away from the double-hexamer interface and proximal to the expected position of the C-terminal helicase domains. The C termini point away from the hexamer-hexamer interface, while the N termini point toward the interface and are situated on the periphery of the model. Mutagenesis of amino acid residues 213, 215, 217, and 220 (shown as orange van der Waals spheres) impairs double-hexamer formation (see text). (B) Mutants which impair double-hexamer formation map to one face of the T-ag obd hexamer. Shown is a view of a surface representation of the T-ag obd hexamer (yellow) which displays the putative double-hexamer interface. Amino acid residues 213, 215, 217, and 220 (orange) are solvent accessible.