Achieving Enantioselectivity in Synthesizing APIs and Intermediates

October 3, 2006
Patricia Van Arnum

Patricia Van Arnum was executive editor of Pharmaceutical Technology.

Pharmaceutical Technology, Pharmaceutical Technology-10-03-2006, Volume 2006 Supplement, Issue 6

As single-enantiomer drugs dominate the pharmaceutical market (1), custom manufacturers and technology providers are developing novel routes using biocatalysis and chemocatalysis, alone and combined, to improve enantioselectivity, yield, reaction conditions, and the economics of manufacture.

AS SINGLE-ENANTIOMER drugs dominate the pharmaceutical market (1), custom manufacturers and technology providers are developing novel routes using biocatalysis and chemocatalysis, alone and combined, to improve enantioselectivity, yield, reaction conditions, and the economics of manufacture.

Biocatalysis in pharmaceutical development

Biocatalysis is increasingly becoming an important tool in pharmaceutical chemical development. "After years of being almost ignored by chemists, biocatalysis is moving into the mainstream," says David Rozzell, president and CEO of BioCatalytics, Inc. (Pasadena, CA, "Lipases have been used for resolution reactions for years, and many people still identify biocatalysis with lipase-catalyzed resolutions. But the combination of high-throughput screening methods and bioinformatics have stimulated enzyme-discovery research, leading to new enzymatic alternatives for a broad range of chemical reactions."

These choices include dehydrogenases, transaminases, and reductive aminases to catalyze different amino acids and amines, outlines Rozzell. Other examples include ketoreductases, which may by used in stereoselective ketone reduction to produce chiral alcohols. And, amidases and nitrilases are used for stereoselective hydrolytic reactions of amides and nitriles.

Biocatalytic routes to Pfizer's atorvastatin

The versatility of biocatalysis in asymmetric synthesis can be seen in two approaches to synthesizing an intermediate in Pfizer's "Lipitor" (atorvastatin), the top-selling single-enantiomer drug (1).

Dowpharma (Midland, MI, uses a nitrilase developed by Diversa Corporation (San Diego, CA, for the asymmetric hydrolysis of 3-hydroxyglutaronitrile to make ethyl (R)-4-cyano-3-hydroxybutyric acid, which is then converted to ethyl (R)-4-cyano-3-hydroxybutyrate, a key chiral building block in atorvastatin.

"We believe that our route to ethyl (R)-4-cyano-3-hydroxybutyrate has several advantages as a manufacturing route," says Karen Holt-Tiffin, head of biocatalysis for Dowpharma. "It is a three-step process starting with low-cost epichlorohydrin, a product that Dow makes at a multiton level, which then is subjected to a cyanide reaction. These steps then are followed by an enzyme desymmetrization (100% theoretical yield) using a nitrilase engineered by Diversa to work at 3 M substrate concentration to give 99% conversion and 99% enantiomeric excess." The reaction scheme is outlined in Figure 1 (2).

Figure 1: Dowpharma´s route to ethyl (R)-4-cyano-3-hydroxybutyrate, a key chiral building block in Pfizer´s Lipitor (atorvastatin).

A key part of the process is the scale-up of the production of the nitralase, which Dowpharma does via its "Pfenex Expression Technology" to give soluble, active enzyme in titres in excess of 25 g/L fermentation broth, explains Holt-Tiffin. The final step is then a simple esterification.

Codexis, Inc. (Redwood City, also has developed a biocatalytic process for making ethyl (R)-4-cyano-3-hydroxybutyrate. The process involves two enzymes that catalyze the enantioselective reduction of ethyl 4-chloroacetoacetate by glucose to form an enantiopure chlorohydrin. In the second step, a third evolved enzyme catalyzes the biocatalytic cyanation of the chlorohydrin to cyanohydrin under neutral conditions, explains John Grate, Codexis's senior vice-president of research and development and chief technical officer.

Expanding the toolbox in biocatalysis

Codexis recently strengthened its technology position with two acquisitions. It acquired the assets of Enzis BV in May 2006 and Jülich Fine Chemicals, now Jülich Chiral Solutions (Jülich, Germany, in 2005. Enzis's technology is based on the conversion of epoxides to yield chiral intermediates by using epoxide hydrolases and haloalcohol dehalogenases. Epoxide hydrolases are used to synthesize optically active epoxides or diols. Enantioselective conversions of haloalcohols and epoxides catalyzed by haloalcohol dehalogenase are used to obtain optically pure epoxides and substituted chiral alcohols, explains Peter Seufer-Wasserthal, vice-president and general manager of Codexis's enzymes and intermediates business unit.

Jülich is partnered with CMS Chemicals Ltd. (Abingdon, UK. to produce N-acetylneuraminic acid at a multiton scale. Acetylneuraminic acid is an intermediate in the synthesis of GlaxoSmithKline's "Relenza" (zanamivir). The acetylneuraminic acid is produced in high enantiomeric excess (>99%) and purity (>99%) using an enzymatic production route developed and optimized by Jülich and its parent company Codexis, notes Thomas Daussmann, Jülich's managing director.

Reductive amination at work

Another example of biocatalysis in stereochemical reactions is in reductive amination. Reductive amination is used in synthesizing chiral amino acids, which are important pharmaceutical intermediates.

Amino-acid dehydrogenases are used to convert 2-ketoacids to the corresponding α-amino acid. While enzymes for producing L-amino acids are well developed, enzymes for D-amino acids have been more limited. Biocatalytics recently developed D-selective amino acid dehydrogenases for producing natural and unnatural D-amino acids (3), as outlined in Figure 2.

Figure 2: Enzymatic reductive amination to produce unnatural amino acids.

Meanwhile, Degussa AG (Düsseldorf, Germany, is advancing a chemocatalytic route for reductive amination. The company developed a novel method for direct enantioselective reductive amination using hydrogen and a hydride donor. Degussa is applying this approach in the enantioselective reductive amination of α-keto acids to enable a more efficient route to chiral α-amino acids (4, 5).

"A common method for preparing amines is the reductive amination of carbonyl compounds, and we successfully used rhodium-catalyzed asymmetric reductive amination with hydrogen as the reducing agent," explains Renat Kadyrov, manager of research and development at Degussa Homogeneous Catalysts. During an extensive systematic survey of multiple-reaction conditions using hydrogen as a reducing agent, Kadyrov's team first observed that the reaction of phenylypyruvic acid in methanolic ammonia in the presence of a rhodium catalyst efficiently produced N-(phenylacetyl)phenylalaninamide (see Figure 3), even in the absence of hydrogen. In this reaction, the substrate acts as a reducing agent as well (4).

Figure 3: The reaction of phenylpyruvic acid in methanolic ammonia in the presence of a rhodium catalyst to produce N-(phenylacetyl)phenylalaninamide.

"That was a key observation. It made us examine alternative hydride sources," explains Kadyrov. His team used high-throughput screening to identify the enantioselective homogeneous catalysts and reaction conditions for the reductive amination of keto compounds under transfer hydrogenation conditions. This approach is important because it demonstrated the first highly enantioselective reduction of keto compounds to the primary amines using ruthenium-based catalysts and ammonium formate as the amine source and reducing agent (see Figure 4).

Figure 4: Ruthenium-catalyzed enantioselective Leuckart-Wallach reaction.

In addition, Kadyrov's team succeeded in the direct enantioselective reductive amination using hydrogen as a reducing agent by using a cationic Rh-catASium D ((cycloocta-1,5-diene)((3R,4R)-1-benzyl-3,4-bis(diphenylphosphino)pyrrolidine)rhodium(I) terafluoroborate) catalyst. Several chiral α-amino acids were produced in good yield and up to 98% enantiomeric excess (5).

The approach allows for a one-step asymmetric reductive amination of ketones to enantiomerically pure amines through reactions with amines or ammonia. An example of this method is outlined in Figure 5. Degussa developed this approach in partnership with Prof. A. Börner of the Leibniz Institute for Organic Catalysis (Rostock, Germany).

Figure 5: Enantioselective reductive amination of of phenyl pyruvic acid with benzylamine.

Transamination reactions

Amine transfer reactions are an alternative to reductive amination for introducing an amino group. "An advantage of transamination is that by using the appropriate enzyme, either amino acids or amines may be produced," explains Rozzell. Inexpensive amine donors such as aspartate, glutamate, or simple amines are used. Amine transaminases also may be used to resolve a racemic mixture of a chiral amine (see Figure 6).

Figure 6: Enzyme-catalyzed transamination can produce optically active amino acids and amines.

Integrating the technologies

In asymmetric synthesis, biocatalysis may be combined with chemocatalysis and classical resolution.

"Having a portfolio of technologies allows you to select the optimum solution for a specific molecule," explains Holt-Tiffin. "In addition, it can prove extremely powerful to combine technologies, for example by using biocatalysis and asymmetric hydrogenation," she adds.

Dowpharma used a three-pronged approach in the synthesis of travoprost, a synthetic prostaglandin F2a analogue and active ingredient for an antiglaucoma product (6). "Here, we used classical resolution to make one key intermediate, biocatalysis for another, then used our process development and manufacturing skills to produce the prostaglandin on a multikilogram scale," says Holt-Tiffin.

In addition, Dowpharma has used the combination of asymmetric hydrogenation and biocatalysis to manufacture a number of unnatural phenylalanine derivatives (7). This technology can be used to manufacture either (R)- or (S)-amino acids.

"The types of products that we have manufactured include pyridylalanines, naphthylalanines, 3-chlorophenylalanine, and 4-cyanophenylalanine, plus many more," says Holt-Tiffin. "Currently, we are manufacturing several such amino acids for pharmaceutical compounds."

That marriage of different tools also is evident in another project. Working in collaboration with Pfizer, Dowpharma recently developed a process whereby a combination of asymmetric hydrogenation and in situ chiral salt upgrade provided the optimal solution for a particular process as outlined in Figure 7 (8) that was used to provide an advanced intermediate for an activated thrombin-activatable fibrinolysis inhibitor for the potential treatment of thrombosis. Dowpharma first developed a classical resolution of the acid using quinidine and a separate asymmetric hydrogenation. For the hydrogenation, it achieved high activity (catalyst loading of substrate/catalyst of 5000), but not high enough enantiomeric excess. "By running the asymmetric hydrogenation on the quinidine salt, we were able to upgrade the enantiomeric excess in a single recrystallization and double the throughput as compared with the classical resolution process alone," says Ian Lennon, head of chemocatalysis for Dowpharma. This process was operated on a large scale by Pfizer.

Figure 7: A process route codeveloped by Pfizer and Dowpharma involving a combination of asymmetric hydrogenation and in situ chiral salt upgrade.

Other advances in biocatalysis

Dowpharma recently entered into a collaboration with Aquapharm Bio-Discovery, Ltd. (Oban, Scotland, to investigate Aquapharm's marine bioorganisms as sources of novel biocatalysts. Aquapharm specializes in the discovery and development of bioactive natural products from the isolation and culture of new marine microbes from off the coast of the United Kingdom. Dowpharma has had previous success in developing biocatalysts from marine sources: the L-acylase it uses in the synthesis of t-amino acids is of marine origin (9,10).

BASF targets chiral synthesis

BASF provides another example of using multiple approaches in asymmetric synthesis. In February 2006, the company launched a new production technology to manufacture chiral alcohols and chiral epoxides using an epoxide hydrolase. The new technology adds to its toolbox in producing chiral alcohols and chiral epoxides that use other biocatalytic routes based on lipases and dehydrogenases and chemoselective routes using asymmetric hydrogenation and Corey-Bakshi-Shibata reduction. The new technology uses a hydrolase for selective epoxide ring-opening to produce various chiral intermediates, notes John Banger, manager, new business development of chemicals with BASF's intermediates group. BASF also added two new chiral amines to its toolbox: (R, R)- and (S, S)-bis (1-phenylethyl) amine, which can be used for the asymmetric synthesis of nonnatural amino acids and as starting materials for chiral bases (i.e., bases for deprotonation according to Simpkins' reactions).

BASF has more than 1000 tons of annual commercial-scale capacity (under current good manufacture practices) for chiral amines at its main manufacturing site in Ludwigshafen, Germany. The commercial-scale capabilities complement its R&D and pilot-scale capabilities at Ludwigshafen. The chiral amines produced at Ludwigshafen are for pharmaceutical manufacture and are part of its overall production network in specialty amines.

Excelsyn targets biotransformations

Another addition to BASF's chiral toolbox is a biocatalysis service unit that offers screening, optimization, and biocatalyst fermentation up to 5 m3, and scale-up from kilogram to ton-scale.

Emerging biocatalysts

Oxynitrilases are an emerging set of enzymes used in stereochemical reactions. "For years, the only oxynitrilase that was commercially available was extracted from almonds, but today there are both R-and S-selective oxynitrilases produced microbially," says BioCatalytics' Rozzell. These enzymes enable the efficient synthesis of compounds such as substituted mandelic acids by the combination of stereoselective cyanohydrin formation with hydrolysis of the resulting hydroxynitrile, which can also be catalyzed enzymatically, he explains.

Aldolases are another developing class of enzymes finding interest because of the possibility of catalyzing stereoselective aldol condensations to produce chiral β-hydroxy aldehydes with high stereochemical purity, explains Rozzell. BioCatalytics also notes the recent development and use of isolated enzymes to reduce carbon–carbon double bonds such as alkenes in the presence of nicotinamide cofactors in the same way that ketones are reduced to alcohols. The reduction can be stereoselective when the alkene is substituted.


1. S.Erb, "Single Enantiomer Drugs Poised for Further Market Growth," Pharm. Technol. 30 "Technology Outlook: APIs, Intermediates, and Formulation" supplement, s14–s18 ( 2006).

2. K. Holt-Tiffin et al.,"Nitrilase-Catalyzed Desymmetrization of 3-Hydroxyglutaronitrile: Preparation of a Statin Side-Chain Intermediate," Org. Process Res. Dev. 10 (3), 661–665 (2006).

3. D. Rozzell, et al.,"Creation of a Broad-Range and Highly Stereoselective D-Amino Acid Dehydrogenase for the One-Step Synthesis of D-Amino Acids," J. Am. Chem. Soc., 128 (33), 10923–10929 (2006).

4. R. Kadyrov and T.Riermeir, "Highly Enantioselective Hydrogen-Transfer Reductive Amination: Catalytic Asymmetric Synthesis of Primary Amines," Angew Chem. Int. Ed. 42 (44), 5472–5474 (2003).

5. R.Kadyrov et al.,"The First Highly Enantioselective Homogeneously Catalyzed Asymmetric Reductive Animation: Synthesis of α-N-Benzylamino Acids," J. Org. Chem. 68 (10), 4067–4070 (2003).

6. I.C. Lennon et al., "Synthesis of the Potent Antiglaucoma Agent, Travoprost," Org. Process Res. Dev. 6 (2), 138–145 (2002).

7. I.C. Lennon et al., "The Application of DuPhos Rhodium (I) Catalysts for Commercial Scale Asymmetric Hydrogenation," in Asymmetric Catalysis on Industrial Scales, H. Ulrich Blaser and E. Schmidt, Eds. (Wiley VCH Verlag GmbH, KGaA, Weinheim, Germany, 2004), pp. 269–282.

8. I.C. Lennon et al., "Efficient Synthesis of an Imidazole-Substituted γ-Amino Acid by the Integration of Chiral Technologies," Org. Lett. 7 (10), 1931–1934 (2005).

9. I.N. Taylor et al., "A Thermostable L-Acylase from Thermococcus Litoralis: Cloning, Overexpression, Characterization and Applications in Biotransformations," Extremophiles 6, 111–112 (2002).

10. I.N. Taylor et al., "Immobilization of the Thermostable L-Acylase from Thermococcus Litoralis to Generate a Reusable Industrial Biocatalyst," Biocatalysis and Biotransformations 20 (4), 241–249 (2002).