Quantification of free ligand conformational preferences by NMR and their relationship to the bioactive conformation☆

Graphical abstract

Calculation of error correction for residual dipolar couplings 32 Chemical shift assignments HO6 a Refer to Figure 1 (see text) for atom nomenclature.
b All chemical shifts were determined at a concentration of 50mM and 278.2 K, pH 6.0 in 10% D2O and referenced directly relative to internal d6-DSS. Std. error: 1 H ± 0.001 ppm. c Resonance observed for a hydroxyl proton with no COSY correlations and was therefore assigned to R2 HO3; resonance could correspond to R2 HO31 or HO32. d Hydroxyl proton was not observed due to fast chemical exchange with solvent. e Prochiral stereo-assignment not achieved.  Figure 1 (see text) for atom nomenclature.
b All chemical shifts were determined at a concentration of 50mM and 278.2 K, pH 6.0 in 10% D2O and referenced indirectly relative to internal d6-DSS. Std. error: ± 0.020 -0.050 ppm. c Value not determined. * Chemical shifts labelled with a star indicate nuclei that are degenerate.

Supplementary information
Streptomycin solution structure   Figure 1 (see text) for atom nomenclature.
b All chemical shifts were determined at a concentration of 50mM and 298.2 K, pH* 6.0 in 100% D2O and referenced directly relative to internal d6-DSS. Std. error: 1 H ± 0.001 ppm.
* Chemical shifts labelled with a star indicate nuclei that are degenerate.

Supplementary information
Streptomycin solution structure  Figure 1 (see text) for atom nomenclature.
b All chemical shifts were determined at a concentration of 50mM and 298.2 K, pH* 6.0 in 100% D2O and referenced indirectly relative to internal d6-DSS. Std. error: ± 0.020 -0.050 ppm. c Value not determined. * Chemical shifts labelled with a star indicate nuclei that are degenerate.

Supplementary information
Streptomycin solution structure Coupling constants  Figure 1 (see text) for atom nomenclature.
b All coupling constants shifts were determined by direct measurement from the acquisition dimension of a 13 C-HSQC spectrum recorded without broadband 13 C-decoupling during acquisition on a sample of 50mM streptomycin, pH* 6.0, 100% D2O at 298.2 K,. Std. error: 1 JCH ± 0.5 Hz.
* Chemical shifts labelled with a star indicate nuclei that are degenerate.

Supplementary information
Streptomycin solution structure

Supplementary information Streptomycin solution structure
Goodness of fit of observed and predicted data a Karplus equation in the form 3 J = Acos 2 (θ+ψ) + Bcos(θ+ψ) + C, with ψ in degrees. b Error applied to observed value is that of predictive capability of the Karplus relation, rather than measurement error (see Methods).

Supplementary information
Streptomycin solution structure

Supplementary information
Streptomycin solution structure Density analysis of structural restraints  Figure 1 (see text) for atom nomenclature.
b Hydrogen atoms in streptomycin not included in this list did not have any structural restraints involving them.
* Chemical shifts labelled with a star indicate nuclei that are degenerate.

Supplementary information
Streptomycin solution structure Figure S6. Streptose ring pucker parameters Figure S6. Comparison of the streptose ring (R2) pucker parameters (Kremer & Pople, JACS, 97, 1354-8) for the aqueous solution 4D-structure of streptomycin determined in this work (grey) with published crystal conformations (colours). Values from 1614 lowenergy structures predicted using molecular dynamics simulation are shown in light blue. The distance from the centre of the circle represents the q value (Å) while the polar coordinate represents the ϕ angle (º). The ϕ angle corresponding to each canonical envelope and twist conformation is given. The bioactive conformation (i.e., when streptomycin is bound to the ribosome, PDB code 1FJG) is shown in blue and the free crystal structure of the streptomycin oxime salt (Neidle et al., 1978) is in green. The conformations of streptomycin bound to artificially-selected RNA aptamers (1NTA, 1NTB) are given in yellow and the three conformations measured from the off-target low affinity co-complex with aminoglycoside-6-adenyl-transferase are given in red (3HAV). Calculation of error correction for residual dipolar couplings For residual dipolar couplings (RDCs) the dependence on angle is highly non-linear and thus an extra error correction has to be applied. Correction of the error to take this into can be achieved by appling a scaling the error. The scaling (to produce an effective error εexp′) can be derived in the following way. If θ is the angle between the major axis of alignment in the molecular frame, then starting from the equation defining residual dipolar couplings, equation (1) is obtained, which allows the calculation error to be obtained by differentiations, equation (2). Suitable approximations result in equation (3). (1) (3) Substituting the identity: into (3) and dividing this into the experimental error, results in equations (4) and (5), the latter of which is almost identical to equation (4), but avoids division by zero by having a minimum value of ¼ in the denominator and is therefore used in practice. (4) Using equation (5), it is possible to increase the total experimental error estimate (εexp) to take into account errors associated with predictions of residual dipolar couplings, which can then be used to more-accurately assess the degree of fit with the experimental data.

Supplementary information
Streptomycin solution structure