Crystal structure of native and Cd/Cd-substituted Dioclea guianensis seed lectin. A novel manganese-binding site and structural basis of dimer-tetramer association

Diocleinae legume lectins are a group of oligomeric proteins whose subunits display a high degree of primary structure and tertiary fold conservation but exhibit considerable diversity in their oligomerisation modes. To elucidate the structural determinants underlaying Diocleinae lectin oligomerisation, we have determined the crystal structures of native and cadmium-substituted Dioclea guianensis (Dguia) seed lectin. These structures have been solved by molecular replacement using concanavalin (ConA) coordinates as the starting model, and refined against data to 2.0 A resolution. In the native (Mn/Ca-Dguia) crystal form (P4(3)2(1)2), the asymmetric unit contains two monomers arranged into a canonical legume lectin dimer, and the tetramer is formed with a symmetry-related dimer. In the Cd/Cd-substituted form (I4(1)22), the asymmetric unit is occupied by a monomer. In both crystal forms, the tetrameric association is achieved by the corresponding symmetry operators. Like other legume lectins, native D. guianensis lectin contains manganese and calcium ions bound in the vicinity of the saccharide-combining site. The architecture of these metal-binding sites (S1 and S2) changed only slightly in the cadmium/cadmium-substituted form. A highly ordered calcium (native lectin) or cadmium (Cd/Cd-substituted lectin) ion is coordinated at the interface between dimers that are not tetrameric partners in a similar manner as the previously identified Cd(2+) in site S3 of a Cd/Ca-ConA. An additional Mn(2+) coordination site (called S5), whose presence has not been reported in crystal structures of any other homologous lectin, is present in both, the Mn/Ca and the Cd/Cd-substituted D. guianensis lectin forms. On the other hand, comparison of the primary and quaternary crystal structures of seed lectins from D. guianensis and Dioclea grandiflora (1DGL) indicates that the loop comprising residues 117-123 is ordered to make interdimer contacts in the D. grandiflora lectin structure, while this loop is disordered in the D. guianensis lectin structure. A single amino acid difference at position 131 (histidine in D. grandiflora and asparagine in D. guianensis) drastically reduces interdimer contacts, accounting for the disordered loop. Further, this amino acid change yields a conformation that may explain why a pH-dependent dimer-tetramer equilibrium exists for the D. guianensis lectin but not for the D. grandiflora lectin.

FokI dimerization is required for DNA cleavage

FokI is a type IIs restriction endonuclease comprised of a DNA recognition domain and a catalytic domain. The structural similarity of the FokI catalytic domain to the type II restriction endonuclease BamHI monomer suggested that the FokI catalytic domains may dimerize. In addition, the FokI structure, presented in an accompanying paper in this issue of Proceedings, reveals a dimerization interface between catalytic domains. We provide evidence here that FokI catalytic domain must dimerize for DNA cleavage to occur. First, we show that the rate of DNA cleavage catalyzed by various concentrations of FokI are not directly proportional to the protein concentration, suggesting a cooperative effect for DNA cleavage. Second, we constructed a FokI variant, FokN13Y, which is unable to bind the FokI recognition sequence but when mixed with wild-type FokI increases the rate of DNA cleavage. Additionally, the FokI catalytic domain that lacks the DNA binding domain was shown to increase the rate of wild-type FokI cleavage of DNA. We also constructed an FokI variant, FokD483A, R487A, which should be defective for dimerization because the altered residues reside at the putative dimerization interface. Consistent with the FokI dimerization model, the variant FokD483A, R487A revealed greatly impaired DNA cleavage. Based on our work and previous reports, we discuss a pathway of DNA binding, dimerization, and cleavage by FokI endonuclease.

Structure of FokI has implications for DNA cleavage

FokI is a member an unusual class of restriction enzymes that recognize a specific DNA sequence and cleave nonspecifically a short distance away from that sequence. FokI consists of an N-terminal DNA recognition domain and a C-terminal cleavage domain. The bipartite nature of FokI has led to the development of artificial enzymes with novel specificities. We have solved the structure of FokI to 2.3 A resolution. The structure reveals a dimer, in which the dimerization interface is mediated by the cleavage domain. Each monomer has an overall conformation similar to that found in the FokI-DNA complex, with the cleavage domain packing alongside the DNA recognition domain. In corroboration with the cleavage data presented in the accompanying paper in this issue of Proceedings, we propose a model for FokI DNA cleavage that requires the dimerization of FokI on DNA to cleave both DNA strands.

Novel site-specific DNA endonucleases

Site-specific hydrolysis of DNA is common to many biological processes. Three new structures, FokI, I-CreI and PI-SceI, were reported in the past year, providing the first view of type IIs endonucleases and homing endonucleases. Together, they reveal an extraordinary set of new mechanisms by which endonucleases target the hydrolysis of specific DNA sequences.

Structure of the multimodular endonuclease FokI bound to DNA

FokI is a member of an unusual class of bipartite restriction enzymes that recognize a specific DNA sequence and cleave DNA nonspecifically a short distance away from that sequence. Because of its unusual bipartite nature, FokI has been used to create artificial enzymes with new specificities. We have determined the crystal structure at 2.8A resolution of the complete FokI enzyme bound to DNA. As anticipated, the enzyme contains amino- and carboxy-terminal domains corresponding to the DNA-recognition and cleavage functions, respectively. The recognition domain is made of three smaller subdomains (D1, D2 and D3) which are evolutionarily related to the helix-turn-helix-containing DNA-binding domain of the catabolite gene activator protein CAP. The CAP core has been extensively embellished in the first two subdomains, whereas in the third subdomain it has been co-opted for protein-protein interactions. Surprisingly, the cleavage domain contains only a single catalytic centre, raising the question of how monomeric FokI manages to cleave both DNA strands. Unexpectedly, the cleavage domain is sequestered in a 'piggyback' fashion by the recognition domain. The structure suggests a new mechanism for nuclease activation and provides a framework for the design of chimaeric enzymes with altered specificities.

Crystallization and preliminary X-ray analysis of restriction endonuclease FokI bound to DNA

FokI is a type IIs restriction endonuclease which recognizes an asymmetric DNA sequence and cleaves DNA a short distance away from the sequence. The enzyme is bipartite in nature with its DNA recognition and cleavage functions located on distinct domains. We report here cocrystals of the complete FokI enzyme (579 amino acids) bound to a 20-bp DNA fragment containing its recognition sequence. The complex is amongst the largest protein-DNA complexes to be crystallized, and required macroseeding techniques for optimal crystal growth. The cocrystals diffract to at least 2.8 A in resolution and belong to space group P2(1) with unit cell dimensions of a=67.9 A, b=119.8 A, c=69.1 A, beta = 96.6 degrees. Using specific amino acid analysis we show that asymmetric unit contains a single FokI molecule bound to the 20-bp DNA fragment. This paper reports the first cocrystals of a type IIs restriction endonuclease.