Supplementary of “NMR Conformational analysis in solution of a potent class of cysteine proteases inhibitors”
Archimede Rotondo 1,*, Roberta Ettari2, Silvana Grasso 1 and
Maria Zappalà 2
S.1 NMR Experiments
NMR studies were all performed running 2D homonuclear spectra aimed to: a) have a complete unambiguous and consistent assignment of the signals; b) after the comlete attribution the same information are evaluated not only as qualitative data but also by a careful evaluation/integration as in the case of the chloroform sample of 3 whose following figures show the 2D-COSY (correlation spectroscopy) and 2D-NOSY (nuclear Ovehouser effect spectroscopy)
Figure S1. 2D-COSY Spectrum of 3 (@500MHz) dissolved in CDCl3. Every off-diagonal spot indicates a “through the bond” correlation.
Figure S2. 2D-NOESY Spectrum (@500MHz 500 ms mixing time) of 3 dissolved in CDCl3 (expansion of the aliphatic region). Every anti-phase (blue) off-diagonal spot indicates a close proximity between connected resonances; these precious information evaluated also by integration of the 1D version experiment lead to the average distances which are the most accurate data for the 3D conformational analysis.
The assignment of the 13C parent atoms was contemporary possible through the 2D-HSQC gradient-edited experiment; as shown in the following figure it is easy to connect any 1H resonance to the parent atom whereas the color of the spots indicate whether the chemical group is a methylene (blue) or a C-H residue.
Figure S3. 2D-13C-HSQC spectrum of 3 dissolved in CDCl3. It is possible to assign any CH2 (cold tones) or CH (hot tones) 13C resonances, moreover the 1H assignments are unambiguously confirmed
Figure S4. 2D-13C-HMBC spectrum of 3 dissolved in CDCl3. A careful analysis based on logical considerations allows to assign all of the 13C resonances.
Figure S5. 2D-15N-HMBC spectrum (@300MHz) of 1 dissolved in CDCl3. A careful analysis based on logical considerations allows to assign all of the three 15N resonances.
S.2. Results and Discussion
Table S1. Assignment of quaternary 13C resonances, and tertiary 15N resonances. Chemical shifts in ppm are referred to the experimental section and ref. 34, 35, 36.
Compounds\Moiety / BDZ / Ph / IOX1-N / 2-C / 4-N / 5-C / 6-C / 7-C / ipso / 2-N / 3-C
1 in CDCl3 / 104 / 163.8 / 137 / 136.6 / 102.1 / 138.4 / 137.2 / 363 / 139.9
2 in CDCl3 / 102 / 163.8 / 139 / 136.6 / 102.1 / 138.5 / 137.2 / 365 / 139.8
3 in CDCl3 / 104 / 163.6 / 139 / 136.6 / 102.3 / 138.8 / 137.1 / 362 / 139.5
3 in CD3OD / 102 / 165.4 / 141 / 137.9 / 102.9 / 139.8 / 138.7 / 361 / 140.7
Most of the structural considerations have been made by 1H-NOESY spectra of 1-3 samples which are consistent with the 3J couplings analysis. Specifically 3 was analyzed also by the 1D and 2D-NOE experiments referred to any single proton resonance at different mixing times (200, 300 and 500ms) to extract average distances based on the two spin system approximation with the interpolation method referred to the germinal protonic distance as suggested by the references 34, 35 together with many related papers. In Table S2 the close proximities in red are taken as fixed reference and these rough values extracted from CD3OD experiments account for: a) the two main conformations around the N-aC bond; b) apparently long-range distances are well detected data demonstrating a certain degree of freedom (conformational changes).
Table S2. Measured average distances in Å for 3 (referred to the SP isomer).
Calculated average distances for 3 (BDZBuIOX)proton 1 / proton 2 / d(Å) in CDCl3 / d(Å) in CD3OD
IOX-5-CH / IOX-4-CH2proS / 2.5 / 2.4
IOX-5-CH / IOX-4-CH2proR / >13 / 7.5
IOX-5-CH / d-CH2 / 3.4 / 3.5
5-IOX-CH / g-CH2 / 4.2 / 5.4
4-IOX-CH2proR / 4-IOX-CH2proS / 1.8 / 1.8
4-IOX-CH2proR / g-CH2 / 6.4 / 7.5
4-IOX-CH2proR / d-CH2 / 4.3 / 4.3
a-CH2proR / a-CH2proS / 1.8 / 1.8
a-CH2proR / g-CH2 / 4.1 / 5.0
a-CH2proR / b-CH2proS / 2.8 / 2.8
a-CH2proR / BDZ-11-CH / 5.9 / 5.7
a-CH2proS / b-CH2proR / 3.3 / 3.5
a-CH2proS / BDZ-11-CH / 2.3 / 2.0
a-CH2proS / g-CH2 / 4.6 / 4.0
b-CH2proS / b-CH2proR / 1.8 / 1.8
b-CH2proR / d-CH2 / 4.0 / 6.0
b-CH2proR / g-CH2 / 4.6 / 4.0
b-CH2proS / g-CH2 / 4.6 / 7.5
b-CH2proS / d-CH2 / 4.0 / 5.2
b-CH2proS / BDZ-11-CH / 3.8 / 4.2
AS-g and AS-d protons are considered undistiguishable and therefore calculated as couples whose average distance is referred to both geminal protons
S.3 Problems concerning the diasterotopic mixtures
The 1H spectra of 1-3 samples evidenced broadening of the signals around the stereogenic 5-IOX center; this sometimes leads also to the line splitting which is not a possible coupling as it is more pronounced working at 500MHz. The parent 13C appear consistently slightly split and according to the brilliant papers referenced in the text we do have a real diasteromeric mixture well within the NMR timescales. From a biological point of view we do not know which of the four isomer is active, however, from the energetical point of view, semiempirical and MM calculations asses that the geometric conformations are very close in stability, whereas the twist-chair minimum (assumed on the halfway toward the transition state) is around 30KJ/mol higher. This is consistent with references 22-26 of the manuscript.
S.4 Structural data
To seek for completeness these supplementary data contain the two optimized diasteromeric isomers of 3 conventionally called S-P and S-M (assuming that their mirror images are easily obtainable as well). These different twin structures can be used as possible model-substrate to analyze the interaction with the biological target through molecular docking. Considerations concerning the rigid conformations of 1 and 2 (able to split just into pseudo axial and pseudo equatorial arrangements) because of the bumping between BDZ and IOX are also supported by the enclosed 3D-structures of the S-P isomer 1 in its pseudo-axial and pseudo equatorial arrangements again fitting the NMR data (see also Figure 5S).
Figure 5S. 3D-Tube model for 1 S-P isomer: (a) in pseudo axial conformation; (b) pseudo equatorial conformation.