Aza-proline effectively mimics l-proline stereochemistry in triple helical collagen† †Electronic supplementary information (ESI) available: A detailed explanation of all procedures for synthesis, purification, characterization, crystallography, and computational analysis. See DOI: 10.1039/c9sc02211b

Chenoweth and co-workers provide an atomic resolution crystal structure and computational analysis illustrating that aza-proline mimics l-proline stereochemistry in collagen.


Table of Contents
, 250 x 10 mm, 5 μm particle size, 100 Å pore size; Analytical: Luna Omega PS C18, 250 x 4.6 mm, 5 μm particle size, 100 Å pore size). MALDI-TOF MS was performed using a Bruker MALDI-TOF Ultraflex III Mass Spectrometer. CHCA was used as the matrix for all MALDI-TOF MS measurements. Peptide was lyophilized using a Labconco FreeZone Plus 12 Liter Cascade Console Freeze Dry System. CD measurements were performed using a JASCO J-1500 Circular Dichroism Spectrometer. UV-vis measurements were performed using a JASCO V-650 UV-vis Spectrophotometer. XRD was performed using the 24-ID-C undulator beamline operated by the Northeastern Collaborative Access Team (NE-CAT) at the Advanced Photon Source (APS) (Argonne National Laboratory, Argonne, IL). Proton nuclear magnetic resonance spectroscopy ( 1 H NMR) and carbon nuclear magnetic resonance spectroscopy ( 13 C NMR) spectra were recorded on a Bruker UNI 500 1 H NMR. High-resolution mass spectra were obtained at the University of Pennsylvania's Mass Spectrometry Service Center on a Micromass AutoSpec electrospray/chemical ionization spectrometer. Flash chromatography was performed using a Teledyne ISCO CombiFlash Rf chromatography system.

Computational Details:
All DFT calculations were carried with the Gaussian 09 software package 5 using the M06-2X 6, 7 DFT functional with the 6-31+G(d,p) basis set for all atoms. The implicit SMD 8 solvation model was used to simulate the effects of water throughout the calculated structures. Frequency calculations were carried out for all structures to confirm them as either a minimum or a TS. Three-dimensional structures were produced with UCSF Chimera. 9 RMSD values reported in the manuscript and SI were calculated using UCSF Chimera. For the RMSD reported in Figure 3 comparing the full backbone and sidechain conformations of CMPs 1 & 2, models were aligned using the MatchMaker tool and solvent molecules and hydrogen atoms were removed. The two peptides (468 atom pairs) were then selected and compared using the rmsd sel command. For Figure 4A and for RMSD Analysis of CMPs 1 & 2 in the SI, the RMSD values were calculated in UCSF Chimera using the match command. The match command performs least-squares fit root-mean-square deviations of specified atoms, moving the first set of atoms (by default, the entire models containing them) onto the second. Figure S1. Extent of pyramidalization in N-amidourea-containing structures. A substructure search of the Cambridge Structural Database (CSD) 10 was performed using the search structure shown. The search results were then categorized according to the extent of substitution of the Nx, Ny, and Nz atoms in each structure (see Figure S2). In the  Figure S1 above (i.e. multiple Namidourea moieties), each instance of the substructure motif was treated as a separate entry.    Table S2. Twisted nature of amide in azPro residue in each strand of collagen triple helix. Data taken from PDB 6M80 for azPro7. The twist angle (τ) 13,14 describes the magnitude of rotation around the N-CO amide bond. A twist angle (τ) of 0° corresponds to a planar amide and a twist angle of 90° corresponds to a fully orthogonal twisted amide.  Table S3. Pyramidalization parameters and main-chain dihedrals of azPro residue in each strand of CMP (1). Data taken from PDB 6M80 for azPro7. Φ (CNNC) and Ψ (NCNN) main-chain dihedrals are in degrees. δ is a measurement of pyramidal character (in degrees) and is defined as δ = S -360°, where S is the sum of the valence angles around the atom of interest (αN). If δ = 0, then the site is fully planar. As δ becomes more negative, the magnitude of pyramidalization increases (sp 3 character). 15 The hinge angle (α) is a measurement of pyramidal character (in degrees) and is defined as the angle from the plane of the αN and the two adjacent atoms with respect to the carbonyl carbon atom. The hinge angle (α) is between 180° (pure sp 2 ) and 125° (pure sp 3 ). 13,14 Another measurement of pyramidalization (d) is defined using a triangular pyramid with the αN at the apex and substituent atoms positioned at the remaining vertices of the base. The distance from the apex normal to the base plane is measured as the d value, in Å.

Entry
(1)  Table S4. Pyramidalization parameters and main-chain dihedrals of Pro residue in each strand of CMP (2). Data taken from PDB 5K86 for Pro7. Φ (CCNC) and Ψ (NCCN) main-chain dihedrals are in degrees. δ is a measurement of pyramidal character (in degrees) and is defined as δ = S -360°, where S is the sum of the valence angles around the atom of interest (αC). If δ = 0, then the site is fully planar. As δ becomes more negative, the magnitude of pyramidalization increases (sp 3 character). 15 The hinge angle (α) is a measurement of pyramidal character (in degrees) and is defined as the angle from the plane of the αC and the two adjacent atoms with respect to the carbonyl carbon atom. The hinge angle (α) is between 180° (pure sp 2 ) and 125° (pure sp 3 ). 13,14 Another measurement of pyramidalization (d) is defined using a triangular pyramid with the αC at the apex and substituent atoms positioned at the remaining vertices of the base. The distance from the apex normal to the base plane is measured as the d value, in Å.   Table S5. N-CO distances of urea moiety in CMP (1). Data taken from PDB 6M80, illustrating divergent N-CO distances.

Synthesis
Peptide 1 was synthesized using manual SPPS on Rink Amide resin (0.54 mmol/g) using Fmoc as the primary protecting group. When it was necessary to stop the synthesis after completing a coupling, the post-coupling wash step was modified such that the resin was washed with DMF (3x) and DCM (3x); following the Arg coupling, additional washes with EtOH were also performed (see step 4 below). When beginning from a dry resin at any point during the synthesis, the resin was swelled for at least 30 min in DMF prior to initial deprotection. The synthesis of peptide 2 was reported previously. 17

Fmoc-ProHyp(tBu)Gly-OH synthon coupling
Following initial deprotection, the resin was washed with DMF (6x). A stock solution of HATU was prepared by dissolving 690 mg (1.81 mmol) HATU in 20 mL DMF. This solution was then used to prepare coupling solutions as noted. A coupling solution of 34 mg (0.06 mmol, 3 eq.) Fmoc-ProHyp(tBu)Gly-OH, 0.67 mL (0.06 mmol, 3 eq.) HATU in DMF, and 20 μL DIEA (0.12 mmol, 6 eq.) was prepared in a 4-mL vial and allowed to activate for ~10 min at room temperature. This solution was then added to the resin in the SPPS vessel and stirred for 40-70 min. The coupling solution was drained from the vessel and the resin was washed with DMF (6x). These steps were repeated 4 times to couple a total of 4 ProHyp(tBu)Gly trimers onto the resin.

Aza-glycine (azGly) coupling
The Fmoc protecting group was removed as described above, and the resin was washed with DMF (6x). A coupling solution of 10 mg (0.06 mmol, 3 eq.) CDI, 0.4 mL DMF, and 15 mg (0.06 mmol, 3 eq.) Fmoc-hydrazine (Fmoc-NH-NH2) was mixed in a vial and allowed to activate for 5-10 min at room temperature. The coupling solution was then added to the resin and stirred overnight. The following day, a fresh coupling solution was prepared and allowed to activate as described above. The original coupling solution was drained from the vessel and the fresh solution was added to the resin and stirred for an additional 4 h. This second coupling solution was drained from the vessel and the resin was washed with DMF (6x).

Proline (Pro) coupling
The Fmoc protecting group was removed as described above, and the resin was washed with DMF (6x). A coupling solution of 20.2 mg (0.06 mmol, 3 eq.) Fmoc-L-Pro-OH, 0.67 mL (0.06 mmol, 3 eq.) HATU, and 20 μL DIEA (0.12 mmol, 6 eq.) was prepared and allowed to activate for ~10 min at room temperature. This solution was then added to the resin and stirred for 45 min. The co upling solution was drained from the vessel and the resin was washed with DMF (6x).

Glycine (Gly) coupling
The Fmoc protecting group was removed as described above, and the resin was washed with DMF (6x). A coupling solution of 18 mg (0.06 mmol, 3 eq.) Fmoc-Gly-OH, 0.67 mL (0.06 mmol, 3 eq.) HATU, and 20 μL DIEA (0.12 mmol, 6 eq.) was prepared and allowed to activate for ~10 min at room temperature. This solution was then added to the resin and stirred for 1 h 30 min. The coupling solution was drained from the vessel and the resin was washed with DMF (6x).

Fmoc-GlyAzProHyp(tBu)-OH synthon coupling
The Fmoc protecting group was removed as described above, and the resin was washed with DMF (6x). A coupling solution of 34.0 mg (0.06 mmol, 3 eq.) Fmoc-GlyAzProHyp(tBu)-OH, 0.67 mL (0.06 mmol, 3 eq.) HATU, and 20 μL DIEA (0.12 mmol, 6 eq.) was prepared in a 4-mL vial and allowed to activate for ~10 min at room temperature. This solution was then added to the resin and stirred for 50 min. The coupling solution was drained from the vessel and the resin was washed with DMF (6x).

Hydroxyproline [Hyp(tBu)] coupling
The Fmoc protecting group was removed as described above, and the resin was washed with DMF (6x). A coupling solution of 24.6 mg (0.06 mmol, 3 eq.) Fmoc-L-Hyp(tBu)-OH, 0.67 mL (0.06 mmol, 3 eq.) HATU, and 20 μL DIEA (0.12 mmol, 6 eq.) was prepared and allowed to activate for ~10 min at room temperature. This solution was then added to the resin and stirred for 40 min. The coupling solution was drained from the vessel and the resin was washed with DMF (6x).

Proline (Pro) coupling
The Fmoc protecting group was removed as described above, and the resin was washed with DMF (6x). A coupling solution of 20.2 mg (0.06 mmol, 3 eq.) Fmoc-L-Pro-OH, 0.67 mL (0.06 mmol, 3 eq.) HATU, and 20 μL DIEA (0.12 mmol, 6 eq.) was prepared and allowed to activate for ~10 min at room temperature. This solution was then added to the resin and stirred for 45 min. The co upling solution was drained from the vessel and the resin was washed with DMF (6x).

Fmoc-ProHyp(tBu)Gly-OH synthon coupling
The Fmoc protecting group was removed as described above, and the resin was washed with DMF (6x). A coupling solution of 34.0 mg (0.06 mmol, 3 eq.) Fmoc-ProHyp(tBu)Gly-OH, 0.67 mL (0.06 mmol, 3 eq.) HATU, and 20 μL DIEA (0.12 mmol, 6 eq.) was prepared in a 4-mL vial and allowed to activate for ~10 min at room temperature. This solution was then added to the resin and stirred for 1 h. The coupling solution was drained from the vessel and the resin was washed with DMF (6x).

Final deprotection and cleavage
The Fmoc protecting group was removed as described above, and the resin was washed thoroughly with DMF (6x). The completed peptide was then cleaved from the resin by mixing with a cleavage cocktail of TFA, phenol, H2O, and EDT in an 87:5:5:3 (v/v) ratio for 1 h 42 min.

Purification
Following SPPS, the peptide was precipitated in cold diethyl ether. After initial precipitation, the cleaved peptide solution was centrifuged, the supernatant was decanted, and the solid peptide was resuspended in ether. This process was repeated for a total of

Synthesis of Fmoc-GlyAzProHyp(tBu)-OH (7)
For the solid-phase synthesis of Fmoc-Hyp(tBu)-loaded resin, 2-chlorotrityl chloride resin was used. To a solution of Fmoc-Hyp(tBu)-OH (5 g, 12.20 mmol) in anhydrous DCM (62 mL) in a round-bottom flask, 2-chlorotrityl chloride resin (5 g, 8.5 mmol, 1.7 mmol/g) and DIEA (2.12 mL, 12.20 mmol) were added under nitrogen. After stirring the mixture for 10 min, another portion of DIEA (3.19 mL, 18.36 mmol) was added. After stirring for 2 h, HPLC grade methanol (18 mL) was added to cap any remaining reactive trityl groups. After 20 min, the reaction mixture was filtered through filter paper, and the solid was washed with DCM (6 x 50 mL) followed by air drying. The material was further dried in vacuo at room temperature. The loading was measured according to a reported protocol 19 to be 0.75 mmol/g, and the total mass obtained was 5.27 g (3.94 mmol).
The Fmoc-Hyp(tBu)-loaded 2-chorotrityl chloride resin (2.67 g, 2 mmol, 1 eq.) was suspended in DMF in a SPPS vessel to swell (~40 mL, 15 min, twice). After draining the DMF used to swell the resin, the base-labile Fmoc protecting group was removed with 20% piperidine/DMF at room temperature with stirring (35 mL, 22 min, twice). The resin was washed with DMF (30 mL x 1) and DCM (30 mL x 5). After washing, 20 mL of anhydrous DCM was added to the reaction vessel, and a solution of Fmoc-GlyAzPro-Cl (5) in DCM, prepared as described below, was added and the reaction mixture was allowed to stir at room temperature for 5 h.

Final Cleavage & Purification
After 5 h, the solution was drained and the resin was sequentially washed with DMF (2 x 40 mL) and DCM (5 x 40 mL). The resin was dried under vacuum in the SPPS vessel at room temperature for 6 h. The dried resin was transferred into a round-bottom flask and was treated with cleavage solution consisting of DCM:AcOH:TFE (10:1:1) (131 mL:13 mL:13 mL) at room temperature for 3 h. The mixture was filtered through filter paper, and the filtrate was concentrated in vacuo. AcOH was removed by azeotroping with C6H6 (3 x 100 mL). The resulting foamy solid residue was purified by silica gel column chromatography (3.5-5% MeOH/DCM) (700 mg, 62% overall based on the determined active Fmoc-Hyp(tBu)-chlorotrityl chloride resin substitution level).

Crystallization
Peptide 1 was crystallized using sitting-drop vapor diffusion under conditions adapted from Okuyama et al. 20 Peptide stock solutions were prepared by dissolving the purified solid product in 18 MΩ H2O to a final concentration of 8.4 mg/mL (Using UV-vis, measured A214 and used extinction coefficient of 60 mM -1 cm -1 ). Crystal trials were prepared by combining 1 μL of the peptide solution with 1 μL of a reservoir solution of 0.1 M Tris-HCl, 30% (w/v) PEG4000, and 0.01 M Li2SO4 . H2O (buffer pH = 7.6). Trays were sealed tightly with plastic tape to create a closed system and prevent solvent evaporation. Trays were incubated at 4 °C. Prior to beamline analysis, crystals were dipped in a drop of a cryoprotectant mixture containing equal volumes of the reservoir solution and 30% (w/v) PEG4000 and then frozen in liquid N2.  Table S6. Data collection and refinement statistics. Table 1 was generated using Phenix (version dev-3120). 21 Refinement was performed using Phenix (version dev-3126). Manual modeling was performed using Coot. 22 Data integration was performed using XDS. 23 Space group validation and data reduction were performed using Pointless (version 1.10.29) 24,25 and Scala (version 3.3.22), 24 respectively, in the CCP4 suite. 26 Phasing was performed using Phaser. 27