Table 1.
Statistics of X-ray data reduction and refinement
| Complex | Pre-reactive complex | Binary complex | Post-reactive complexa | |
|---|---|---|---|---|
| HinP1I/DNA/Ca2+ | HinP1I/DNA/Ca2+ | HinP1I/DNA | HinP1I/DNA/Mg2+ | |
| PDB ID | 2FKC | 2FKH | 2FL3 | 2FLC |
| Beamline (wavelength Å) | APS 22-ID (0.97179) | APS 22-BM (0.99997) | APS 22-ID (1.0) | APS 22-ID (1.0) |
| Space group | P3221 | P6522 | P6522 | P6522 |
| Unit cell dimensions (Å) | a = b = 105.1, c = 130.4 | a = b = 109.1, c = 113.3 | a = b = 100.9, c = 139.4 | a = b = 100.4, c = 139.4 |
| Resolution range (Å) (highest resolution shell) | 19.89–2.39 (2.48–2.39) | 31.51–3.09 (3.20–3.09) | 34.17–2.39 (2.48–2.39) | 34.10–2.59 (2.65–2.59) |
| Measured reflections | 225 720 | 130 658 | 325 661 | 106 424 |
| Unique reflections | 31 659 | 7757 | 15 524 | 13 460 |
| <I/σ> | 19.0 | 17.1 | 15.2 | 17.6 |
| Completeness (%) | 94.7 (97.2) | 99.9 (99.9) | 89.0 (50.2) | 99.2 (95.7) |
| Rlinear = Σ|I − 〈I〉|/Σ〈I〉 | 0.108 (0.368) | 0.071 (0.278) | 0.079 (0.374) | 0.070 (0.401) |
| R-factor = Σ|Fo − Fc|/Σ|Fc| | 0.222 (0.309) | 0.220 (0.255) | 0.246 (0.427) | 0.275 (0.475)a |
| R-free (5% data) | 0.260 (0.362) | 0.313 (0.338) | 0.284 (0.419) | 0.324 (0.504)a |
| Non-hydrogen atoms | ||||
| Protein | 3981 (two molecules) | 1958 (one molecule) | 1939 (one molecule) | 1925 (one molecule) |
| DNA | 808 (two duplexes) | 404 (one duplex) | 404 (one duplex) | 404 (one duplex) |
| Metal ions | 4 (Ca2+) | 2 (Ca2+) | – | 2 (Mg2+) |
| Water molecules | 45 | 12 | 10 | 1 |
| Root-mean-square deviation from ideality | ||||
| Bond lengths (Å) | 0.008 | 0.007 | 0.008 | 0.011 |
| Bond angles (°) | 1.4 | 1.3 | 1.4 | 1.6 |
| Dihedral (°) | 23.0 | 22.6 | 22.7 | 22.7 |
| Improper (°) | 2.9 | 2.9 | 2.8 | 2.8 |
| Estimated coordinate error | ||||
| From Luzzati plot (Å) | 0.33 | 0.38 | 0.44 | 0.49 |
| From SIGMAA (Å) | 0.29 | 0.17 | 0.51 | 0.55 |
aThe refinement of the post-reactive complex was a challenge.
Analysis of reflection data indicated that Mg2+ soaking may lead the crystal to form a superlattice, which probably contains two alternating sublattices: a short one similar to the original native crystal and a long one with doubled a and b axes (see Supplementary Data). Efforts were made to separate the contributions from the two sublattices and the dominant short sublattice was finally used for the refinement of the post-reactive complex. Due to the error introduced by the contributions from the long sublattice (∼12% in the crystal), the R-factor and R-free values are relatively high although electron density maps appear suitable. Assignment of solvent molecules to discontinued densities would further reduce the R-factor, but not the R-free. We took a conservative approach without including such solvent molecules, and carefully examined omit difference electron density maps particularly in the area of the cleaved scissile bond and the Mg2+ binding sites.