Search

Reactions that make C C bonds using Grignard reagents or organolithium reagents and related compounds were discussed in Chapter 11 . Aliphatic substitution reactions were also discussed, using a cyanide ion as a reagent, as well as alkyne anions. Chapter 11 concluded with the so-called metal-hydrogen exchange reaction, which is simply an acid-base reaction. Organosulfur compounds and organophosphorus compounds can be converted to α-lithio derivatives using this reaction, and these new reagents react as carbanions. In addition, organocopper reagents, specifically organocuprates, react with alkyl halides, epoxides, acid chlorides, as well as aldehydes and ketones. Finally, phosphorus and sulfur carbanions known as ylids will be used to generate alkenes or epoxides, respectively. This chapter will therefore focus on acid-base reactions that generate carbanions or carbanion-like reagents, which in all cases exhibit nucleophilic reactivity.

Glycosylation is presented in terms of a continuum of mechanisms spanning the full range from S N 1 to S N 2, and factors affecting the stereoselectivity of glycosylation are discussed in the context of these mechanisms. Recent advances in stereoselective glycosylation are surveyed with particular emphasis placed on the reputedly difficult classes of glycosidic linkage, namely, the 2-deoxyglycosides, the 1,2- cis axial and equatorial glycosides, the uronic acid glycosides, the α-sialosides, and the β-arabinofuranosides.

Over the past decade, rapid progress has been made in elucidating the structures and mechanisms of the enzymes involved in the synthesis and degradation of glycosides. At present the CAZy sequence-based classification reveals a total of over 100 families of glycosidases and close to 100 glycosyltransferases. Although the glycosidases are known for their structural diversity, the glycosyltransferases are largely categorized as two folds: GTA and GTB, both of which feature Rossman folds that accommodate the nucleoside phosphate moiety of the donor sugar. However, variable folds are seen in the recently determined structures of two lipid phosphosugar-dependent transferases, suggesting that solving more structures may lead to more diversity. The vast majority of glycosidases employ single or double displacement mechanisms that involve acid/base catalysis and oxocarbenium ion-like transition states. The carboxylic acid side chains of Glu and Asp residues play prominent roles in catalysis, though in the case of hexosaminidases the substrate acetamide functions as the catalytic nucleophile, while in the sialidases and trans-sialidases a tyrosine residue plays this role. A major deviation in mechanism was recently discovered for GH4 and 109 glycosidases, which hydrolyze the glycosidic bond via an NAD-assisted redox elimination/addition mechanism. The inverting glycosyltransferases also appear to be mechanistically simple, employing a single displacement with acid/base assistance via oxocarbenium ion-like transition states. The retaining glycosyltransferases, by contrast, are more complex and appear to exploit a mechanistic continuum from a short-lived ion-pair intermediate in most cases to a covalent intermediate in others.

In forming the glycosidic linkage, close attention must be paid to the orientation of the hydroxyl group at C2 of the glycosyl donor; there are several common scenarios. Indeed, the glycosidic linkage is formed from a glycosyl donor and a glycosyl acceptor. A glycosyl donor, of either the α- or the β-configuration, is treated with a glycosyl acceptor to form, by the elimination of HX, the disaccharide containing the new glycosidic linkage, of either the α- or the β-configuration at C1´. In the process, there is no change in configuration at C4 in the glycosyl acceptor. In less common circumstances, the glycosyl acceptor may react through the hydroxyl group of the anomeric (hemiacetal) center. In this case the formation of the glycosidic linkages results in the α- or β-configuration at both C1 and Cl´; the product is a nonreducing disaccharide. An example of such a disaccharide is trehalose. The formation of a glycosidic linkage will not be an easy task. Apart from the activation of the glycosyl donor, there are problems of the stereoselectivity (α- or β-) of the process and the access to just the desired hydroxyl group of the glycosyl acceptor (protecting group chemistry). Nature, of course, circumvents all of these problems with the use of enzymes, but for the synthetic carbohydrate chemist, much ingenuity, creativity, and hard work are necessary to match the rewards of evolution. The majority of methods available for the formation of the glycosidic linkage use a glycosyl donor that is a precursor of either an intermediate oxocarbenium ion (as part of an ion pair) or, at least, a species that has a significant positive charge at the anomeric carbon atom.

Publisher Summary Amphidinolides are a series of unique cytotoxic macrolides, many of which have shown significant antitumor properties against a variety of NCI tumor cell lines. Amphidinolides A through W exhibit remarkable structural diversity, the range is from 12-membered to 29-membered ring systems. More than half of the amphidinolides possess an odd-numbered macrocyclic lactone ring, different from the general macrolide natural products that possess even-numbered rings. The absolute configurations of some amphidinolides have been determined by X-ray analysis and interconversion. This synthesis represents a well-convergent and stereocontrolled route to the first total synthesis of natural product amphidinolide T1. A number of synthetic technologies have been developed during the course of this endeavor. The functional group compatibility of the oxocarbenium ion-mediated alkylation reaction, efficient cross metathesis, diastereoselective aldol reactions, and the use of a cyclic bromoether as a novel protection of the exo-methylene group are noteworthy features. The first total synthesis of amphidinolide T1 was accomplished in 16 (longest linear sequence) steps and 5.6% overall yield, starting with a syn-aldol reaction developed in the laboratory. This chapter gives the chronology of events that led to its first total synthesis.

Publisher Summary The metabolism of cyclic ADP-ribose (cADPR) and 2'-phospho-cADPR (P-cADPR) in mammals is apparently achieved by enzymes known as nicotinate adenine dinucleotide phosphate (NADP) glycohydrolases. NAD(P)ases catalyze the release of nicotinamide from NAD or NADP with the formation of an enzyme-bound oxocarbenium ion intermediate. This intermediate has three possible fates—(1) it can react via an intermolecular reaction with nicotinamide to regenerate the original dinucleotide or with a variety of other nucleophiles with retention of configuration to form analogs of either NAD or NADP; (2) it can undergo an intramolecular reaction to form cADPR or P-cADPR; or (3) it can react with water to form ADPR or P-ADPR. The methodology described in the chapter uses an NAD(P)ase from the marine mollusk Aplysia californica . The strategy described in the chapter achieves the synthesis of cyclic nucleotides by simple passage of NAD, NADP, or a structural analog through a column of immobilized ADPR cyclase. Conversely, to prepare NAD or NADP analogs, cADPR or P-cADPR can be mixed with the nucleophile of choice and passed through a column of immobilized enzyme. Because the approach involves few reactants and products, purification of the products can be achieved easily.

Glycosidases which hydrolyses their substrates with net retention of anomeric configuration (retaining enzymes) do so via a double-displacement mechanism in which a covalent glycosyl-enzyme is formed and hydrolysed with acid/base catalytic assistance via oxocarbenium ion-like transition states. Strategies for identification of the nucleophilic residue and the acid/base catalytic residue have been devised. By use of mechanism-based inhibitors to trap the glycosyl-enzyme intermediate and affinity labels to derivatise the acid/base catalyst, in combination with novel mass spectrometric techniques, the identities of these residues have been determined in several glycosidases. In all cases identified to date these have turned out to be the carboxylic amino acids glutamate and aspartate. Armed with this knowledge, a retaining glycosidase has been converted into an inverter through replacement of the nucleophilic glutamate by alanine, along with the addition of alternative anionic nucleophiles such as azide and formate to the reaction medium. This provides the first example in which such a change of mechanism in a glycosidase has been effected.