Step 1
Ac2O |
|
EtOH |
BnBr, K2CO3 |
|
Acetone, Reflux |
HCl(g) |
|
MeOH, Reflux |
Methyl Anthranilate |
|
toluene |
Cbz-L-Ala, DCC |
|
CH2Cl2 |
PPh3, I2, EtN(i-Pr)2 |
|
CH2Cl2 |
20% piperidine |
|
EtOAc |
1) 45% HBr/HOAc |
|
2) SiO2, EtN(i-Pr)2/EtOAc |
The starting reagent was 2-amino-5-hydroxybenzoic acid methyl ester. This was N-acetylated with acetic anhydride in ethanol at about 40-50 degrees. The acetylation was utilized to protect the reactive amine group to set up for O-benzylation of the hydroxyl group in the next step.
The N-acetyl derivative was O-benzylated with benzyl bromide and potassium carbonate in acetone.
The O-benzyl derivative was treated with HCl in methanol at reflux to give a methyl anthranilate. This compound is the reagent for the next step (4).
Sulfinamide anhydride, this starting reagent, was prepared from anthranilic acid and thionyl chloride. Next, methyl anthranilate, the product from our last step, was dissolved in toluene and reacted with the starting reagent. The resulting compound was a 2-(2-amino-benzoylamino)-benzoic acid methyl ester.
The reactive amine group was then condensed with N-Cbz-L-alanine and DCC in chloroform to obtain a tripeptide derivative. There was a second synthetic approach that attached the Cbz-L-Ala to the lower amine group first and then attached the methyl anthranilate to the upper amine group. Unfortunately, tripeptide coupling agents like DCC were ineffective on this intermediate and an extra step was necessary, resulting in lower yields than the method presented.
Cyclization of the tripeptide derivative was achieved by treatment with PPh3 and I2 in the presence of EtN(i-Pr)2 in CH2Cl2 at room temperature. Separation via chromatography with silica gel yielded the desired product. However, there was about 20% of a similar but not as pure compound that was thought to contain some starting material. Interestingly, it was still used directly in the synthesis and only needed purification on the final step.
The benzoxazine derivative was treated with 20% piperidine in EtOAc for 5 hours to obtain an amidine carboxamide.
The amidine carboxamide was immediately deprotected by the use of 45% HBr in HOAc at 50 degrees C. This is thought to remove both the Cbz and the Bz groups. Secondly, in SiO2 and triethylamine the compound is thought to cyclize and quickly eliminate methanol yielding circumdatin C. Specific rotations, and NMR spectroscopy were utilized to verify the product. It should be noted that circumdatin C had a poor solubility, so CDCl3 and DMSO-d6 were both used as solvents.