1. N,N'-Carbonyldiimidazole (CDI) was purchased from Sigma-Aldrich (reagent grade) and used as received.
2. Dichloromethane was purchased from Sigma-Aldrich (certified 99.8%, containing 50-150 ppm amylene as stabilizer) and used as supplied. Caution should be taken with the CH2Cl2 quality in order to prevent the formation of undesired side compounds. CH2Cl2 stabilized with EtOH should be avoided, otherwise the formation of ethyl 1H-imidazole-1-carboxylate could also be detected on the first step.
3. Cooling is necessary in order to avoid the formation of symmetrical urea (Note 22).
4. The mixture is stirred at 500 rpm throughout the reaction.
5. L-Alanine methyl ester hydrochloride, 99% from Sigma-Aldrich, was used as received. This compound is very hygroscopic.
6. Diisopropylethylamine (DIPEA) (99.5%) was purchased from Sigma-Aldrich and used without further purification.
7. For small scale synthesis (<15 mmol) triethylamine (Et3N) was used as base. For operational facility on large scale DIPEA was used, which, in contrast to Et3N, generates a soluble salt in the presence of HCl•H2N-Ala-OMe in CH2Cl2. This soluble salt is easier to add dropwise into the CDI pre-cooled suspension.
8. The flask containing the mixture of L-alanine methyl ester hydrochloride and DIPEA was rinsed twice with CH2Cl2 (2 x 5 mL). Then, the combined 10 mL were added to the dropping funnel.
9. A dropping funnel with a Teflon tap was used for the addition of the amino acid to the CDI suspension.
10. Calcium chloride (anhydrous, granular, ≤ 7.0 mm, ≥93.0%) from Sigma-Aldrich was used as received. The drying tube containing a sintered glass filter was used.
11. The dropping funnel containing the mixture of L-alanine methyl ester hydrochloride and DIPEA was rinsed twice with CH2Cl2 (2 x 5 mL).
12. The slow addition of L-alanine methyl ester solution is necessary in order to avoid the symmetric urea formation (Note 22).
13. A second reaction was allowed to proceed for 22 hours.
14. The formation of the desired product was observed by TLC (on Merck silica gel 60 F254 TLC aluminum plates) and visualized with UV light and ninhydrin staining solution (Note 35). Rf product: 0.20, eluent: EtOAc 100%. Visualization was difficult with UV light when dilute samples were used.
15. Ethyl acetate (≥99.8% from Sigma-Aldrich) was used as received.
16. Magnesium sulfate (≥99.5% from Sigma-Aldrich) was used as received.
17. Additional EtOAc (2 x 20 mL) is used during filtration to assist transfer. An M grade glass filter was used for the filtration.
18. The crude reaction product (14.5 g) was adsorbed on silica (40 g of silica with 150 mL of CH2Cl2 followed by evaporation) and then loaded onto a column (diameter: 7 cm, height: 70 cm) packed with silica gel (300 g of silica, pore size 60Å, 230-400 mesh, 40-63 µm particle size, Fluka Analytical) slurry in EtOAc 100%. After 500 mL of initial elution, fraction collection (250 mL fractions) is begun, and elution is continued with 4.5 L of pure EtOAc. The desired α-activated amino ester is obtained in fractions 4-18, which are concentrated by rotary evaporation (30 °C, 10 mmHg) and then dried at 0.05 mmHg. Residual ethyl acetate was difficult to remove and could be observed by 1H NMR after 1 week under house vacuum (<1%). Ethyl acetate did not hinder the formation and isolation of product during the next step.
19. Yields were adjusted for wt% of product as determined by QNMR (97.2% wt% and 98.2% wt% for Run 1 and 2, respectively).
20. The chemical yield of this step can be improved to 86% by using 1.5 equiv of CDI.
21. A second reaction at the same scale provided 10.52 g (73.2%, 98.2% wt%) of a white solid. Yields were adjusted for potency of product. Weight percent was determined by QNMR with benzyl benzoate as an internal standard. Characterization as follows: [α]D26 +31.3 (c 0.7, CHCl3); 1H NMR pdf(400 MHz, CDCl3) δ: 1.54 (d, J = 7.2 Hz, 3 H), 3.81 (s, 3 H), 4.66 (quint, J = 7.2 Hz, 1 H), 7.01 (br d, J = 6.9 Hz, 1 H), 7.08 (s, 1 H), 7.41 (t, J = 1.4 Hz, 1 H), 8.18 (s, 1 H); 13C NMR pdf(100 MHz, CDCl3) δ: 17.3, 49.5, 52.5, 116.4, 129.5, 136.1, 148.7, 172.8; IR (neat): 3138.9, 2969.5, 2878.8, 2809.5, 1740.3, 1711.7, 1553.7, 1484.4, 1454.5, 1376.9, 1288.4, 1256.4, 1213.2, 1150.6, 1104.2, 1072.7, 754.8 cm-1; HRMS (ESI)+ [M+H]+ calcd for C8H12N3O3: 198.0879. Found: 198.0873. mp 90.0-92.5 °C; the melting point was lower than that reported by the submitters (mp 95.6-97.9 °C) due to residual solvent and impurities.
22. The symmetrical urea depicted below is generated from auto-condensation of the free α-aminoester and its activated form. Its formation is observed either when the reaction is run at higher temperatures (> 0 °C) or when the reagents are added at once. It might be noted that on small-scale reactions, only small amounts (<7%) of this compound were observed in the crude material.6 A peak in 1H NMR was observed after column chromatography that is consistent with the 6H singlet expected at ~3.7 ppm. If this resonance is assigned correctly, then 1.7% and 1% of the symmetrical urea impurity was observed, respectively, in Runs 1 and 2 after column chromatography. The impurity was removed during the subsequent isolation in Step 2.
23. Charges were not adjusted for wt% of methyl (1H-imidazole-1-carbonyl)-L-alaninate.
24. The checkers performed this chemistry at ambient temperature (20-22 °C) in an unjacketed flask without temperature control. The submitters report that control of temperature is very important to ensure good reaction yields. Temperatures lower than 25 °C lead to the formation of dipeptides, with erosion of isolated yields.
25. The following reagents were purchased from commercial sources and used without further purification: Boc-Phe-OH, ≥99% from Aldrich; HOBt hydrate, wetted with not less than 14 wt.% water, 97% from Aldrich; CuBr2, 99% from Aldrich.
26. The rubber septum is equipped with a needle (22G x 1½" 0.7 x 40 mm) in order to allow the removal of CO2 that is formed during the reaction.
27. The mixture is stirred at 400 rpm throughout the reaction.
28. The formation of the dipeptide was monitored by TLC analysis on Merck silica gel 60 F254 TLC aluminum plates and visualized with UV light and ninhydrin staining solution (Note 35). Rf dipeptide: 0.3, eluent: pentane/EtOAc 7:3. Pentane (≥99% from Sigma-Aldrich) was used as received.
29. The mixture is swirled vigorously for 10 min while deep blue color faded to very pale light blue.
30. The addition of a saturated solution of NaHCO3 to the organic layer was followed by the formation of an emulsion, which disappears after standing in the separatory funnel for approximately 15 minutes.
31. Additional CH2Cl2 (2 x 20 mL) is used during filtration to assist transfer.
32. Yields are based on two full scale runs and are adjusted for wt% of product as determined by QNMR (99.6% wt% and 100% wt% for Run 1 and 2, respectively).
33. The crude product (14.0 g) was adsorbed on silica (40 g of silica with 120 mL of CH2Cl2 followed by evaporation) and then was loaded onto a column (diameter: 7 cm, height: 70 cm) packed with a short pad of silica gel (200 g of silica pore size 60Å, 230-400 mesh, 40-63 µm particle size, Fluka Analytical) slurry in pentane:EtOAc 7:3. After 500 mL of initial elution, fraction collection (250 mL fractions) is begun, and elution is continued with 3.25 L of additional solvent. The desired dipeptide is obtained in fractions 3-13, which are concentrated by rotary evaporation (30 °C, 10 mm Hg) and then dried at 0.05 mmHg.
34. A second reaction at the same scale provided 11.86 g (83%, 99.6% wt%) of a white foam. Yields were adjusted for potency of product. Weight percent was determined by QNMR with benzyl benzoate as an internal standard. Characterization as follows: mp 108.0-110.5 °C; [α]D26 +0.40 (c 0.99, CHCl3); 1H NMR pdf(400 MHz, CDCl3) δ: 1.34 (d, J = 7.0 Hz, 3 H), 1.40 (s, 9 H), 3.02-3.12 (m, 2 H), 3.71 (s, 3 H), 4.26-4.44 (m, 1 H), 4.52 (quint, J = 7.1 Hz, 1 H), 5.08 (br. S, 1 H), 6.46-6.66 (m, 1 H), 7.18-7.32 (m, 5 H); 13C NMR pdf(100 MHz, CDCl3) δ: 18.0, 28.1, 38.3, 47.9, 52.2, 55.4, 79.9, 126.7, 128.4, 129.3, 136.6, 155.3, 170.9, 172.8. IR (neat): 3323.4, 2984.1, 2946.6, 1751.7, 1692.8, 1655.5, 1522.4, 1445.9, 1385.8, 1366.2, 1250.5, 1159.4, 1049.5, 988.9, 859.6, 664.4 cm-1; HRMS (ESI)+ [M+H]+ calcd for C18H27N2O5: 351.1920. Found: 351.1915.
35. The ninhydrin stain was prepared using 1.5 g of ninhydrin dissolved in 100 mL of n-butanol and 3.0 mL of AcOH.