Characterization and crystal structure of the type IIG restriction endonuclease RM.BpuSI

Abstract

A type IIG restriction endonuclease, RM.BpuSI from Bacillus pumilus, has been characterized and its X-ray crystal structure determined at 2.35Å resolution. The enzyme is comprised of an array of 5-folded domains that couple the enzyme's N-terminal endonuclease domain to its C-terminal target recognition and methylation activities. The REase domain contains a PD-x(15)-ExK motif, is closely superimposable against the FokI endonuclease domain, and coordinates a single metal ion. A helical bundle domain connects the endonuclease and methyltransferase (MTase) domains. The MTase domain is similar to the N6-adenine MTase M.TaqI, while the target recognition domain (TRD or specificity domain) resembles a truncated S subunit of Type I R-M system. A final structural domain, that may form additional DNA contacts, interrupts the TRD. DNA binding and cleavage must involve large movements of the endonuclease and TRD domains, that are probably tightly coordinated and coupled to target site methylation status.

Figure 1.

(a) The structure of full-length BpuSI. The domains are labeled and color coded according to individual-folded elements with known functions based on sequence and structural homology. The N-terminal endonuclease domain displays a PD-(D/E)xK nuclease fold and is packed against the MTase domain in a manner that buries its active site and precludes DNA-cleavage activity in the modeled conformation. These two functional domains are connected via a small, helical ‘connector’ domain that displays structural similarity to similar modules found in a wide variety of multi-domain, flexible protein scaffolds. The C-terminal TRD is interrupted by an additional α/β domain that may play a role in establishing additional DNA contacts as discussed in the text. (b) Multi-sequence alignment of BpuSI and two closely related isoschizomers BsmFI and BspLU11III (two frame-shift mutations were introduced into the published sequence in order to correct the reading frame at the N-terminus of BspLU11III). The catalytic residues E21, D60, E76 and K78 and the neighboring histidine triad (H26, 50 and 51) in the REase domain are indicated with red arrows; the NPPY catalytic motif in the MTase domain is indicated with a red box.

Figure 2.

(a)Isolated endonuclease domains, shown side-by-side and directly superimposed, from BpuSI (green) and FokI (red). (b) Location of putative BpuSI active site residues (left panel) and superposition of the active-site residues and α-carbon backbones of the catalytic domain of BpuSI (blue), FokI (PDB 1fok; orange) and the HsdR domain of the type I R–M system from Vibrio vulnificus (PDB 3h1t, cyan) (right panel).

Figure 3.

Comparison of structures of domain 2 (the ‘m*’ domain; colored in light green) and MTase domains (dark green) from four type I restriction/modification enzymes and from BpuSI (center panel). The structure of these domains from EcoKI (PDB ID 27YH) is taken from an EM reconstruction of the DNA-bound enzyme, using a high-resolution structure of that subunit bound to the Ocr inhibitor as the initial model.

Figure 4.

MTase domains. (a) MTase domains from BpuSI (left) and M.TaqI (right) with bound cofactor analogues. (b) Unbiased electron density (green mesh) for the S-adenosylhomocysteine cofactor/product (green carbond atoms) bound in the MTase domain of BpuSI. The surrounding grey electrondensity corresponds to a 2F_o − F_c omit map contoured at 2σ around the cofactor binding site.

Figure 5.

(a) MTase and TRD domains from DNA-bound M.TaqI (left) and from unbound BpuSI (right). The bound SAM and S-adenosylhomocysteine moieties in each MTase domain are shown as sticks. The MTase domains are shown in the same orientation; relative to this superposition the corresponding TRDs are in dramatically different positions and orientations as described in the text. (b) Rotation of the TRD in BpuSI from its initial location in the unbound crystal structure [as shown in (a), right] into a position corresponding to its DNA-bound position in M.TaqI [(a), left) is easily accomplished by a single-backbone rotation at a discrete hinge point as described in the text, leading to the superposition with M.TaqI as shown.

Figure 6.

Model of DNA-bound BpuSI, generated using the coordinates of DNA-bound M.TaqI as a reference (see Figure 5B). (a) Model of DNA-bound BpuSI with the bound DNA substrate previously visualized for M.TaqI. The left and right panels show two different orientations of this same model. The endonuclease domain (light green, easily seen in right panel) is positioned exactly as observed in the unbound BpuSI structure. (b) The left panel shows a model of DNA-bound BpuSI with a longer, canonical B-form DNA (encompassing the GGGAC target site and the downstream N10/N14 bases, corresponding to the cleavage pattern produced by the enzyme). The endonuclease domain has been rotated as described in the text, to bring its active site into appropriate proximity with the N10 phosphate group. The right panel shows the same model, with the original position of the endonuclease domain (yellow ribbon) superimposed.

Figure 7.

(a) Design of the one-site and two-site substrates and the expected cleavage products. Black horizontal arrows represent the orientation of the BpuSI recognition site (GGGAC) from 5′ to 3′. Red vertical arrows denote the approximate locations of the BpuSI cleavage sites (GGGAC 10/14). The numbers on the DNA are the coordinates on pACYC184 DNA. (b)A cleavage reaction of the single- and double-site substrates as analyzed on a 2.5% agarose gel. The identity of the bands is shown. (c) A plot of the single-site (black squares) and double-site (red circles) substrates as fraction of total DNA versus time. Error bars represent 1 standard deviation.