Fig. 1. Structure of the NaeI monomer. (A) Ribbon diagram: α-helices in cyan, 3₁₀ helices in blue and β-strands in green. The N-terminal (Endo) domain has topology similar to other restriction endonucleases. The C-terminal (Topo) domain contains a CAP or helix–turn–helix motif (H7, H8, 3₁₀ helix before H9, B10, B11) for DNA binding. The hinge loop is around Gln170. L43 is shown in ball-and-stick representation. (B) Sequence and secondary structure of NaeI. Cyan cylinders and green arrows represent α-helices and β-strands, respectively. The helices H4a and H4b are bent at Met65.

Fig. 3. C_α trace for the superposition of two monomers (red and green) in the NaeI dimer. The transformational matrix of the Endo domain was used for the superposition of the entire molecule. The superposition showed ∼5 Å displacement for the backbone atoms of the Topo domain, indicating that the Endo and Topo domains have significantly different orientation of their 2-fold axes.

Fig. 5. Two patterns of DNA recognition of the endonucleases. (A) The β-strand recognition for subgroup I endonucleases consists of two elements: a β-strand (NaeI 1 in this figure and R1 in Figure 2A) and a β-like turn (NaeI 2 in this figure and R2 in Figure 2A). The color codes are red for NaeI (100–113 and 135–158), gold for PvuII (75–88 and 129–152), green for BglI (149–162 and 256–279) and blue for EcoRV (101–113 and 175–195). The 8 bp double helix DNA is adopted from the PvuII–DNA structure (Cheng et al., 1994). (B) The helix recognition for subgroup II endonucleases of BamHI (gold; BamHI 1 for 113–126 and BamHI 2 for 145–165), Cfr10I (blue; 192–206), EcoRI (magenta; 137–152) and MunI (red; 113–128). The double helix DNA is adopted from the BamHI–DNA structure (Newman et al., 1995).

Fig. 7. Modeling of a DNA fragment onto the Topo domain of NaeI. The NaeI dimer is shown in red and gold ribbons, whereas DNA in the ball-and-stick model is adopted from the CAP–DNA structure. Ser234 and Tyr44 (magenta balls) are two potential candidates to form a covalent phosphoester intermediate for the topoisomerase activity of the L43K mutant. Leu43 is also shown as magenta balls.

Fig. 2. Structure of the NaeI dimer. (A) Two monomers in red and yellow ribbons are related by a molecular 2-fold axis located vertically. The tentative active site residues of Glu70, Asp86, Asp95 and Lys97 for the endonuclease activity are shown in green ball-and-stick representation. Two putative recognition loops are located at the very top of the picture, and are marked R1 (residues 102–109, β-strand B4) and R2 (residues 141–149, a β-like turn). (B) A view of the NaeI dimer with a 90° rotation about the vertical axis from (A). The CAP DNA-binding motif in the Topo domain of NaeI, labeled helix–turn–helix and H8 and 3₁₀ (before H9), is common for many of the DNA-binding proteins such as topoisomerases and transcription factors. This site is presumably a second site of DNA binding and responsible for the topoisomerase activity of the NaeI L43K mutant. (C) Surface presentation of dimeric NaeI: red for negatively charged residues and blue for positively charged residues. Two DNA fragments are modeled into the endonuclease (top) and topoisomerase (bottom) sites on the basis of the superposition of NaeI over the PvuII–DNA and CAP–DNA structures, respectively.

Fig. 4. Structure-based sequence alignment of the NaeI Endo domain with partial sequences of eight restriction endonucleases, two DNA repair endonucleases and λ-exonuclease. Amino acids highlighted in yellow are used in the calculation of the r.m.s. deviation of the superposition. Three negatively charged residues for divalent metal binding and the catalytic lysine are marked with asterisks and colored red. Hydrophobic residues of the molecule core are colored green. The alignment suggests that the restriction endonucleases are divided into subgroup I (NaeI, EcoRV, PvuII and BglI) and subgroup II (BamHI, EcoRI, MunI, Cfr10I and FokI). Subgroup II shows a β-strand running in an opposite direction to B6 in NaeI, and also a helix in place of β-strand B4 in NaeI. B4 and B7 in NaeI are, respectively, contained in recognition loops R1 and R2 (Figure 2A).

Fig. 6. Superposition of the CAP motifs of NaeI (red) and topoisomerase IA (green). The active site Tyr319 of topoisomerase IA, which forms a covalent intermediate with DNA substrate, is shown as gold balls. The 3₁₀ helix before H9 in NaeI corresponds to the helix that binds the major groove of DNA in CAP DNA-binding proteins.