1 Department of Chemistry, Technical University of Denmark2 Organic Chemistry, Department of Chemistry, Technical University of Denmark
This thesis describes two different projects. The first project deals with the design, synthesis and biological activity of novel reversible peptidyl FVIIa inhibitors (Chapter 1–3). FVIIa was launced as NovoSeven R over a decade ago by Novo Nordisk for the treatment of hemophilia A and B complicated by antibodies. FVIIa is a serine protease and hence liquid formulations are not stable due to autoproteolysis. A reversible inhibitor would stabilize FVIIa making a liquid formulation possible, representing an important follow-up product for Novo Nordisk. Peptidyl benzyl ketones was chosen as a new class of potential inhibitors, whose sequence was rationally selected from a previously reported FVIIa-TF specificity profile. Since arginine was found to be the most active P1-amino acid, a mild and efficient synthesis of the corresponding arginyl benzyl ketone building block was sought. Two stategies were proposed, the one involving a tetramic acid key intermediate being the most straightforward and with less protective group manipulation. For introduction of the benzyl functionality, a palladium-catalyzed -arylation was developed. This transformation occurs under mild conditions showing high functional group tolerance. Unfortunately, these -aryl tetramic acids were too unreactive and ring opening toward the synthesis of the building block did not succeed. However, -aryl tetramic acids are still interesting compounds due to their potential biological activity. The building block 3.15 (P1) was instead synthesized via a Weinreb amide and a small library of peptides were prepaired by solution-phase Boc/Bn-synthesis. Different P3-P2 sequences (tyrosine, threonine, phenylalanine, leucine) and N-terminals (P4 = H-, Ac-, BnSO2- and Cbz-) were investigated. Unfortunately, O-debenzylation was found to be very difficult which restricted the number of peptides containing tyrosine and threonine. Cbz-D-Phe-Phe-Arg-bk (3.50) was identified as the most potent FVIIa inhibitor with a Ki = 8 μM (IC50 = 16 μM) and with a 35- and 28-fold selectivity against thrombin and FXa, respectively, but with a poor solubility in aqueous media. A SAR study revealed that especially a bulky aromatic Cbz-terminal was crucial for potency. In the view of potency and selectivity, 3.50 seems to be a promising candidate for future development of liquid formulations of NovoSeven R . The second project deals with the rhodium-catalyzed enantioselective synthesis of diaryl amines, which is an important class of compounds (Chapter 4). For example it is found in the third generation anti-histaminic agent levocetirizine. Development of efficient synthetic routes is therefore of considerably interest. The rhodium-catalyzed enantioselective synthesis employing -carbamoyl sulfones and arylboronic acids was therefore investigated using the chiral ligand (R,R)-deguPHOS. Rh(acac)(coe)2 was originally utilized requiring the use of a glovebox, but through catalyst screening [RhCl(cod)]2 was found to be equally efficient. Contrary to Rh(acac)(coe)2, this new catalyst is air-stable, commercially available and inexpensive. [RhCl(cod)]2 and (R,R)-deguPHOS was preincubated prior to use to secure excellent enantioselectivity. A cannulation technique was implemented for application outside the glovebox. A low content of boroxine in the batch of arylboronic acid was found to be crucial for a satisfactory outcome. The highly functionalized diaryl amine 4.13 was synthesized in good yield and excellent enantioselectivity in gram-scale. The absolute configuration was determined by X-ray crystallography to be the (S)-enantiomer. The improvements make the reaction very usable and efficient for synthesis of important amine drug candidates.