[ removed from website 18.12.03, replaced by newtargs.html ] Apart from the values for carboxylates (asp and glu) these target distances are based entirely on distances in accurately determined structures of appropriately chosen small molecules compounds found in the Cambridge Structural Database( CSD). The values agree reasonably well with distances found in the PDB for structures determined at resolutions of 1.6 Å or better. The values given for carboxylates of Ca, Mg, Mn and Zn have been adjusted slightly to give a better fit with the (more numerous) PDB observations. They are for monodentate carboxylates; for bidentate carboxylates it is a good approximation to increase the monodentate values by 0.11 Å. Note, too, that for Zn a range of intermediate states between mono- and bidentate carboxylate is possible ( see Acta Cryst. D57, 401-411 (2001)). All the values for Cu(II) are based on the short bonds found in 4-coordinate Cu(II) and in the basal plane of 5- and 6- coordinate Cu(II). Values in brackets are estimates only, there is insufficient data in the CSD.
O O O O N S water asp or glu ser or tyr his cys monodentate thrCa 2.39 2.36 2.43 [2.20] [2.38] [2.56]
Mg 2.07 2.26* 2.10 [1.87] [2.05] [2.23]
Mn 2.21 2.21 2.25 1.88 2.19 2.35
Fe 2.09 2.01 2.13 1.93 2.08 2.27
Cu(II) 1.97 1.96 2.00 1.90 1.99 [2.17]
Zn 2.09 2.04 2.14 1.95 2.00 2.29
Na 2.42 2.42 2.42 [2.20] [2.38] [2.56]
K 2.84 2.84 2.84 [2.62] [2.80] [2.98]
For Ca, Mg, Mn, Fe, Cu, Zn a standard deviation of 0.10 Å should be reasonable for use in association with these. For Na and K a larger standard deviation, ~0.3 Å, ahould be used since these interactions are predominantly electrostatic, and thus more flexible in distance.
For metal to main chain carbonyl O the M - O asp (monodentate) distance above should be used.
*Mg - carboxylate distances present more uncertainty than any others, no doubt partly on account of the low scattering power of the Mg cation. The target distance for Mg - monodentate carboxylate derived from the CSD was 2.11 Å, based on 15 observations; this was revised upwards, to give a better fit with 28 observations in the PDB, but these PDB observations are likely to be less accurate.
Full details are given in Harding,Acta Cryst. D57, 401-411 (2001)