Enzymes: the key to cleaner synthesis?

Enzymes are useful synthetic tools as they can give chirally pure products cleanly and in good yields, and they are increasingly being applied to industrial-scale syntheses

The growing number of single isomer pharmaceutical products reaching the market has led to pressure on chemical synthesis to create efficient methods for their manufacture. One of the most powerful techniques available is the use of enzymes, as they are invariably enantiospecific.

Nearly 3,000 enzymes are known that can catalyse chemical reactions in a highly stereo-, regio- and chemoselective way under mild reaction conditions. However, they are often not very stable, unmodified enzymes generally give only 'natural' selectivity, they are expensive, and some need cofactors to be added to the reaction mixture for the reaction to proceed.

Of the commercially-available enzymes, only a limited number are of any great practical use for synthesis, and the big challenge is selecting the best one for a particular process. If an available enzyme with the necessary activity cannot be found, the alternative is to identify an enzyme by a screening programme that covers a range of microbes. This latter method can be extremely successful, but the activity of enzymes is not predictable in advance.

Hydrolytic enzymes are used most often for organic synthesis, and can can be employed in the hydrolysis of amide, ester or glycosidic bonds. Many are substrate-specific, but a number have wide-ranging activities. Enzymes can also be used to catalyse resolutions, where a racemic mixture is enriched in one enantiomer by the selective conversion of the other.

One of the core competencies of Ascot Fine Chemicals is industrial biocatalysis, and the cloning of enzymes. Active enzymes are discovered using a screening programme, but these are generally only available in very limited amounts. So cloning the organism to produce overexpression of the enzyme means commercial-scale quantities can be made readily available.

An example is a γ-lactamase, used to resolve a racemic bicyclic γ-lactam, giving a chiral intermediate for carbocyclic nucleosides (Scheme 1). Screening identified an enzyme from a strain of Comoamonas acidovorans. The cloned lactamase is very cheap, and only needs to be semi-purified before use.

DSM, in collaboration with four Dutch universities, has been investigating the use of enzyme catalysis in the synthesis of fine chemicals. A process for the biocatalytic production of penicillins and cephalosporins was developed that uses enzymes to catalyse the coupling of side chains with β-lactams. It was initially developed to make the cephalosporin cephalexin (Scheme 2), and has subsequently been applied to other cephalosporins, and the penicillins amoxicillin and ampicillin. Advantages of the process include reduced waste products, and no halogenated solvents are needed. DSM is now using the process to manufacture cephalexin at a plant in Barcelona. The company has also developed a process for the manufacture of antibiotic intermediates like 7-aminodeacetoxycephalosporanic acid (7-ADCA), as used in the cephalexin synthesis, which uses fewer steps than the conventional route.

DSM also uses amidases to manufacture chirally-pure amino acids, both natural and unnatural. The amidase of Pseudomonas putida is able to hydrolyse racemic α-amino acids with one hydrogen at the α-carbon, but as it is inactive towards disubstituted amino acids, the company screened for an active amidase. The company found that the amidase from Ochrobactrum antropi has the desired effect, and by genetically modifying the organisms, the activity could be increased.

US company BioCatalytics has a number of patents covering its processes for the enzymatic production of amino acids by transamination. It starts with 2-ketoacid precursors, and can make high yields of both D- and L-amino acids. The fundamental principal is the use of aspartic acid as the amino group donor. The ketoacid is reacted with aspartic acid, and this is then simple to decarboxylate to give the pyruvate (Scheme 3). It can be carried out either in solution or with immobilised transaminases. The company has developed similar processes for the preparation of L-homophenylalanine, a key intermediate in a number of ACE inhibitors, including enalapril.

Chiral vicinal aminoalcohols are key building blocks for a number of pharmaceutical actives, including the protease inhibitors saquinavir, ritonavir, nelfinavir and indinavir. Biocatalytics has patented a process that gives a broad range of cyclic and acyclic chiral vicinal aminoalcohols. The method gives control of the absolute stereochemistry at both the amino and alcohol-bearing centres, so any of the four possible diasteroisomers can be produced selectively.

The starting materials are commercially-available β-ketoesters, such as acetoacetic ester and β-ketophenylpropionic acid esters. The key step in the synthesis is the stereoselective reduction of the keto group to a hydroxyester, catalysed by an alcohol dehydrogenase (Scheme 4). This reaction requires the presence of a nicotinamide cofactor. The two enantiomers of the 2-substituted β-ketoester are in equilibrium in aqueous solution, and the enzyme reacts with only one of these; this, added to the fact that the ketone reduction itself is stereoselective, means that two chiral centres are generated simultaneously.

The ester is then converted to the hydrazide, prior to Curtius rearrangement, which proceeds with retention of stereochemistry to give the amino alcohol.

The company has developed a number of enzymatic methods for the manufacture of a range of different chiral aminoalcohols. Chiral 1,3-aminoalcohols are made by initially oxidising stereoselsectively a 1,4-diol, subsituted at C2 or C2 and C3 using an alcohol dehydrogenase to give a chiral lactone. The lactone is then converted it its amide, hydrazide or hydroxamic derivative, which can then undergo stereospecific rearrangement to the chiral 1,3-aminoalcohol.

BioCatalytics has also licensed a technology for preparing enzymes whose catalytic activity in organic solvents is greatly enhanced, useful because of the limited aqueous solubility of many organic substrates. Its salt-immobilised enzymes are lyophilisates of an enzyme in a salt matrix. Optimising the type of salt and the proportional mix have led to enzyme activities up to 3,000 times greater in organic solvents than the enzyme alone.

Altus Biologics has created a range of enzyme catalysts that are much easier to handle than simple enzymes. Its cross-linked enzyme crystals (CLECs) are made by crystallising the target enzyme, and then chemically cross-linking them, forming insoluble, porous solids. The porosity is important as it allows substrates to access the enzyme's active sites, and the resulting CLECs are pure, stable and heterogeneous.

The enzyme protein molecules are linked together to form a 3D lattice with channels running between the molecules, the channel size being of the order of 20–50Ã…. The lattice helps to prevent protein denaturing, because the lattice strength adds to the energy required to denature the proteins.

Altus claims its technology is applicable to almost any enzyme or protein, and it has already produced over 15 different enzymes in this form. These are shown in Table 1. As the catalysts are heterogeneous, it makes them easy to recycle and reuse. Their improved stability also means reactions can be performed at higher temperatures and greater enzyme concentrations, either in aqueous or organic solution.

The cross-linking confers an innate stability on the catalysts, greatly increasing shelf-life, and moving them firmly into the realms of the standard organic reagent because they are much easier to handle than the pure enzymes.