Christian Wenzel Tornøe

Palladium-Catalyzed Three-Component Approach to Promazine with Formation of One Carbon–Sulfur and Two Carbon–Nitrogen Bonds

The formation of aromatic carbon–heteroatom bonds has traditionally been achieved by nucleophilic aromatic substitution or by the copper-mediated Ullman reaction. The former type of chemistry is generally limited to activated substrates, whereas the latter often requires prolonged heating in the presence of excess copper salts. The palladium-catalyzed formation of aromatic C N bonds, extensively developed by the groups of Hartwig and Buchwald, has provided a powerful alternative. Whereas aryl amination has been optimized so that it is even applicable to aryl chlorides and activated phenols, the analogous formation of C O and C S bonds has attracted less attention. For a medicinal chemistry project, we have identified conditions that enable the formation of C S bonds from thiophenols and aryl iodides, and C N bonds from amines and aryl bromides using the same catalyst in a one-pot reaction. We now report the application of this discovery to the synthesis of the promazine class of antipsychotics. Our attention was drawn to the phenothiazine backbone of the promazine series 1a–e as a suitable model system for the controlled construction of three carbon–heteroatom bonds in a single synthetic operation (Scheme 1). This synthesis would require that either C S or C N bond formation occur initially with subsequent cyclization to the phenothiazine nucleus. This disconnection of the promazines leads to the precursors 2-bromothiophenol (2), primary amine 3, and an appropriately substituted 1-bromo-2-iodobenzene 4a–e. From reported literature and our experience with reactions involving dppf, binap, dpephos, and xantphos, the formation of the C S bond was expected to precede the amination steps. Indeed, clean formation of diaryl sulfide 6 was observed when a mixture of 2, 3, 4a, and NaOtBu was treated with [Pd2dba3] and dppf at 60 8C for 20 minutes, and 1a was formed in high yield after 2 hours at 160 8C under microwave (MW) irradiation. These conditions were mimicked in a ligand optimization study using oil-bath heating (Figure 1). Figure 1 summarizes results obtained with commercially available and easily handled ligands reported for C S and C N coupling reactions. Ferrocene ligands such as dppf gave mainly the desired product (1a) in addition to the noncyclized intermediate 7. Significant amounts of aniline 5 were observed with davephos, x-phos, and binap. Noncyclized promazine 7 was the major product with dpephos and xantphos. Trace amounts of the desired product were formed with PPh3, P(o-tol)3, P(tBu)3, [12,13] or the carbene ligand. Low conversion of 4a occurred with diaryl sulfide 6 as the only detectable product in the absence of palladium and ligand, whereas small amounts of 6 and bromobenzene were formed without the ligand. The reaction appears to proceed in a stepwise fashion from diaryl sulfide 6 (only product with one equivalent of NaOtBu), to aniline 7 (only product with two equivalents of NaOtBu), to 1a (74% yield; Table 1) under MW conditions. The three-component reaction worked well for the parent promazines 1a–e with yields of the isolated products ranging from 50% to 76%, and an average yield of greater than 75% for each of the three bonds formed (Table 1). The reaction with allyl amine gave a complex mixture of unidentified products. Benzyl amines were good substrates and the scope of the reaction was extended to include anilines; 2,6disubstituted anilines participated in the reaction, albeit with reduced yields as the steric hindrance around the nitrogen atom increased. Themicrowave method was relatively slow as the reagents had to be mixed immediately before starting the reactions to avoid catalyst deactivation. Conveniently, the reaction was performed with conventional heating, warming from room temperature to 160 8C over approximately 0.5 hours, with subsequent stirring at 160 8C overnight (reaction times have not been optimized). Scheme 1. Retrosynthetic analysis for the promazines.

Peptidotriazoles: Copper(I)-Catalyzed 1,3-Dipolar Cycloadditions on Solid-Phase

[1,2,3]-Triazoles are important pharmacophores in medicinal chemistry and it is therefore important to develop synthetic methods for triazoles that are both mild, high yielding and applicable to combinatorial synthesis. [1,2,3]-Triazoles are typically prepared by refluxing an alkyne and an azide in toluene (110 °C), but the 1,3-dipolar cycloaddition has also been performed at lower temperatures by using sodium acetylide [1], lithium and magnesium acetylide [2,3] with varying success. The present work describes the preparation of [1,2,3]-triazoles at 25 °C on solid-phase with high yields (80–95%), and is compatible with Fmoc chemistry.

Ralph F. Hirschmann award address 2009: Merger of organic chemistry with peptide diversity

A huge unleashed potential lies hidden in the large and diverse pool of encoded and particularly nonencoded chiral α‐, β‐, and γ‐amino acids available today. Although these have been extensively exploited in peptide science, the community of organic chemistry has only used this source of diversity in a quite focused and targeted manner. The properties and behavior of peptides as functional molecules in biology are well documented and based on the ability of peptides to adapt a range of discrete conformers at a minimal entropic penalty and therefore ideally fitting their endogenous targets. The development of new organic reactions and chemistries that in a general and quantitative way transform peptides into new functional molecules, preferably on solid support, is a source of completely new classes of molecules with important and advantageous functional properties. The peptide diversity and the ability to perform chemistry on solid support add tremendously to the combinatorial scope of such reactions in pharmaceutical and materials screening scenario. In recent years, the need for “click” reactions to shape complex molecular architecture has been realized mainly with a basis in the world of peptides and DNA, and in polymer chemistry where connection of highly functionalized biologically active substances or property bearing fragments are assembled as molecular LEGO® using quantitative and orthogonal click chemistries. In this article, three such new reactions originating in the Carlsberg Laboratory over the last decade taking advantage of organic transformations in the peptide framework is presented. Initially, the click reaction between azide and terminal alkynes catalyzed by Cu(1) (CuAAC‐reaction) is described. This CuAAC “click” reaction was observed first at Carlsberg Laboratory in reactions of azido acid chlorides with alkynes on solid support. Second, the Electrophilic Aromatic Substitution Cyclization–Intramolecular Click‐Cascade (EASCy‐ICC) reaction will be presented. This quantitative stereo‐selective cascade reaction provides a highly diverse set of interesting novel scaffolds from peptides. Finally, we describe the preparation of solid phase peptide phosphine‐ and carbene‐based green catalysts (organozymes), which upon complex formation with transition metal perform with high turnovers under aqueous conditions. These catalysts thrive from the peptide folding and diversity, while phosphines and carbenes in the backbone provide for bidental complex formation with transition metals in a format providing an excellent entry into combinatorial catalyst chemistry.

Solid-phase Synthesis of Chemotactic Peptides Using α-Azido Acids

Erratum has been published for this article in Journal of Peptide Science 6 (10) 2000, 497–540.

EXPO3000—a new expandable polymer for synthesis and enzymatic assays

A new polymer for synthesis and enzymatic assays is presented which combines moderate loading with the biocompatibility of poly(ethylene glycol)-based resins. The polymer displays low swelling in all solvents until selective cleavage of a silyl based crosslinker expands the polar resin to render it penetratable for enzymes (an example with a 27 kDa protease is given). An efficient alkylation procedure for derivatization of long PEG-chains is also presented.

Enzymatic and chiral HPLC resolution of α-azido acids and amides

Abstract For the first time, enzymatic resolution of α-azido acid amides has been successfully demonstrated with high yields and enantiomeric excess. In one case dynamic kinetic resolution was achieved leading to more than 50% yield of the enantiomerically pure azido acid. Chiral HPLC was also used to separate racemic α-azido acids and the separation process was automated. Two routes to enantiopure α-azido acid building blocks for solid-phase peptide synthesis have, therefore, been established.