Matthias Urmann

Evidence for CCl/CBr⋅⋅⋅π Interactions as an Important Contribution to Protein–Ligand Binding Affinity†

Attractive chlorine: Noncovalent interactions between chlorine or bromine atoms and aromatic rings in proteins open up a new method for the manipulation of molecular recognition. Substitution at distinct positions of two factor Xa inhibitors improves the free energy of binding by interaction with a tyrosine unit. The generality of this motif was underscored by multiple crystal structures as well as high-level quantum chemical calculations (see picture).

A General and Mild Palladium‐Catalyzed Domino Reaction for the Synthesis of 2H‐Indazoles

Indazoles play an increasingly important role in drug discovery. They act as an efficient isostere for privileged structures such as indoles and benzimidazoles. Furthermore, this important scaffold is able to interact with a variety of diverse targets, as highlighted by the growing number of reports of biologically active indazole derivatives. However, only a limited number of approaches for the regioselective synthesis of N-substituted indazoles are available today. Most approaches afford the thermodynamically favored 1H-indazole or mixtures of 1Hand 2H-indazoles, whereas the regioselective formation of 2H-indazoles remains a very challenging task. The lack of a direct, efficient, and regioselective synthetic procedure for the construction of 2Hindazoles prevents their broader application in, for example, medicinal chemistry. Thus, there is an unmet need for the development of a simple and general synthesis of 2Hindazoles from readily available precursors. Herein we report a straightforward domino reaction sequence consisting of a regioselective coupling of monosubstituted hydrazines 2 with 2-halophenylacetylenes 1, followed by an intramolecular hydroamination through a 5-exo-dig cyclization and subsequent isomerization of the exocyclic double bond to give the aromatic 2H-indazole (Scheme 1). The first challenge in this strategy was the development of a regioselective transition-metal-catalyzed coupling of monosubstituted hydrazines 2 with 2-halophenylacetylenes 1 to afford the required N,N’-disubstituted hydrazines 4. Although a number of transition-metal-catalyzed coupling reactions of aryl halides with amides, amines, hydrazides, and hydrazones are known, only a few coupling reactions of hydrazines have been reported, and only one of the reported hydrazine couplings, for the formation of N,N-diaryl hydrazines, is regioselective. A second challenge in the development of our proposed strategy was the control of the hydroamination/ cyclization step to form the 1,2-dihydroindazole 5, as other possible cyclization pathways lead to other products, such as 1,2-dihydrocinnolines and N-azaindoles. The isomerization of the exocyclic double bond in dihydro-2H-indazole 5 to give the aromatic 2H-indazole 3 was expected to occur spontaneously under the reaction conditions and thus not to pose any problems. We initiated our investigation by screening for reaction conditions under which the coupling of 1-chloro-2-phenylethynylbenzene (1a) and phenylhydrazine (2a) would proceed efficiently to give the N,N’-diaryl hydrazine 4. Upon optimization of the reaction parameters (the transition metal, metal salt, ligand, base, solvent, and temperature), the desired coupling was found to proceed cleanly within just a few hours and with complete regioselectivity when the catalyst system [Pd2(dba)3]/PtBu3 (1:2) was used in toluene at 80 8C with NaOtBu as the base (dba = dibenzylideneacetone). This reaction is to our knowledge the first regioselective transition-metal-catalyzed coupling of monosubstituted hydrazines to give N,N’-disubstituted hydrazine products. We further optimized the reaction parameters to identify conditions that would promote the complete domino reaction of 1 a with 2a as a one-pot reaction (Table 1). We found that the use of polar solvents, such as DMF, NMP, or DMA, in combination with Cs2CO3 led to the formation of the desired 2H-indazole 3a in good yield (Table 1, entries 2–4), whereas Scheme 1. Proposed synthesis of 2H-indazoles.

Modified bile acids as carriers for peptides and drugs

Abstract For the development of future drugs two aspects are of major importance, a site-specific drug action without adverse side-effects and a preferably oral applicability. The liver has a central role in drug action and many disorders are unique to the liver demanding a liver-specific drug action. In oral drug therapy the small intestine is often the limiting barrier of drug absorption. Bile acids are natural substrates undergoing an enterohepatic circulation involving the liver and the small intestine. This organotropism of bile acids is achieved by specific Na + -dependent transport systems in the plasma membrane of hepatocytes and ileocytes. Di- and tripeptides as well as orally active α -amino- β -lactam antibiotics are intestinally absorbed by a H + /oligopeptide cotransport system of high transport capacity. We, therefore, investigated whether the hepatic and the intestinal bile acid transport systems as well as the intestinal H + /oligopeptide transporter can be used in drug therapy to improve the membrane permeability and intestinal absorption of peptide drugs, to target a drug to the liver and the biliary system and to obtain liver-specific drugs. For this, modified bile acids with linkers of varying structure, length, position and stereochemistry at the steroid nucleus were synthesized and covalently linked to drugs or peptides or alternatively bile acid structural elements were introduced into drugs. To investigate the H + /oligopeptide transporter as a putative peptide delivery system, peptides were covalently attached to the 3′-position of the tripeptide-analogue d -cephalexin. The interaction of these bile acid and cephalexin conjugates with the hepatic and intestinal bile acid and peptide transport systems as well as their pharmacokinetic and pharmacodynamic behaviour was investigated by transport measurements and photoaffinity labeling techniques using membrane vesicles, isolated hepatocytes and in vivo models.

Fragment Deconstruction of Small, Potent Factor Xa Inhibitors: Exploring the Superadditivity Energetics of Fragment Linking in Protein-Ligand Complexes.

Predictable thermodynamic additivity is one of the cornerstones of classical covalent chemistry, allowing accurate calculation of energy terms for complete processes by addition of terms for individual components. However this principle breaks down in complex noncovalent systems, such as biological systems, in which the energetics of individual components are not truly independent of each other. This complicates predicting protein structure and folding and, the focus of this work, the prediction of ligand binding to proteins. Molecular recognition in protein–ligand complexes predominantly occurs through multiple noncovalent interactions, whereas their contribution to the total free-energy of binding (DG) is often unevenly distributed over the contact interface. The identification of ligands as “molecular anchors” for high affinity regions in proteins (“hot spots”) is fundamental for fragment-based drug discovery, 3] indicating the similarity of ligandand protein-centric concepts. Often highaffinity ligands encompass more than one fragment in proximal protein sites; in a few cases, individual fragments in two neighboring sites could be linked to result in high binding affinity. Ideally, the DG of linked fragments should be significantly greater than the sum of DG increments from each fragment. This overproportional increase (“superadditivity”) is attributed to the fact that each fragment loses a significant part of its rigid body rotational and translational entropy upon complex formation. Thus, the sum of DG for two fragments includes two unfavorable rigid body entropy barrier terms, whereas the joined molecule is only affected by one of these terms. Any ligand has to overcome this barrier because of entropy loss upon association to its site. The nonadditivity for DG contributions is defined as linker coefficient E corresponding to the difference between the sums of fragment affinity and the final ligand [Eq. (1)]. DGfinal 1⁄4 DGfrag1 þ DGfrag2 þ DGlink with DGlink 1⁄4 R T ln E ð1Þ

Natural product syntheses based on asymmetric Diels-Alder reactions

R)-Pantolactone, (S)-ethyl lactate and other a-hydroxy-car- boxylic acid derivatives are effective chiral auxiliaries for large-scale asymmetric Diels-Alder additions of enoates. Mechanistic aspects and prepa- rations of intermediates for EPC-syntheses of carbocyclic nucleoside ana- logs, prostaglandins and other biologically active compounds are described.

Sulfonylthiadiazoles with an Unusual Binding Mode as Partial Dual Peroxisome Proliferator‐Activated Receptor (PPAR) γ/δ Agonists with High Potency and in vivo Efficacy

Compounds that simultaneously activate the peroxisome proliferator‐activated receptor (PPAR) subtypes PPARγ and PPARδ have the potential to effectively target dyslipidemia and type II diabetes in a single pharmaceutically active molecule. The frequently observed side effects of selective PPARγ agonists, such as edema and weight gain, are expected to be overcome by using partial instead of full agonists for this nuclear receptor family. Herein we report the discovery, synthesis, and optimization of a novel series of sulfonylthiadiazoles that are active as partial agonists. The initial compound 6 was discovered by high‐throughput screening as a moderate partial PPARδ agonist; its optimization was based on the X‐ray crystal structure in complex with PPARδ. In contrast to other PPARδ agonists, this ligand does not interact directly with residues from the activation helix AF‐2, which might be linked to its partial agonistic effect. Interestingly, the thiadiazole moiety fills a novel subpocket, which becomes accessible after moderate conformational rearrangement. The optimization was focused on introducing conformational constraints and replacing intramolecular hydrogen bonding interactions. Highly potent molecules with activity as dual partial PPARγ/δ agonists in the low nanomolar range were then identified. One of the most active members, compound 20 a, displayed EC50 values of 1.6 and 336 nM for PPARδ and γ, respectively. The X‐ray crystal structure of its complex with PPARδ confirms our design hypothesis. Compound 20 a clearly displayed in vivo activity in two chronic mice studies. Lipids were modified in a beneficial way in normolipidemic mice, and the development of overt diabetes could be prevented in pre‐diabetic db/db mice. However, body weight gain was similar to that observed with the PPARγ agonist rosiglitazone. Hence, active compounds from this series can be considered as valuable tools to elucidate the complex roles of dual PPARγ/δ agonists for potential treatment of metabolic syndrome.