Innovations in API Synthesis

February 2, 2007
Patricia Van Arnum
Pharmaceutical Technology
Volume 31, Issue 2

As custom manufacturers gather for InformexUSA this month, strategies in asymmetric synthesis and catalysis prevail.

As custom manufacturers gather for InformexUSA this month, they are confronted with the ongoing task of meeting the pharmaceutical industry's demand for cost-effective routes to manufacturing increasingly complex molecules. This challenge is particularly keen in enantioselective reactions.

Synthesis of a chiral aminolactam

To illustrate the process for optimizing such a synthesis, Michael Schwarm, director, research and development, exclusive synthesis at Degussa GmbH (Düsseldorf, Germany), outlines the route the company took to synthesize an enantiomerically pure aminolactam ((S)-CAP), a chiral building block for a new antithrombotic drug.

"Initial route-scouting experiments started from readily available L-aspartic acid," he says, "but they rather quickly turned out to be less promising from both technical and economic points of view."

More progress was achieved with a new route based on L-homoserine. "Surprisingly, it was found that under appropriate conditions selective O-acetylation of L-homoserine was possible without significant N-acetylation or cyclization to the lactone," says Schwarm. After Z- or Boc-protection of the amino function, the carboxyl group was activated as an anhydride and then reacted with p-cyanoaniline to form the corresponding amide. The acetyl group then was removed by treating it with aqueous ammonia in methanol. Following the activation of the alcohol function by mesylation or tosylation, the amide then was cyclized to the lactone under alkaline conditions. N-deprotection finally afforded the desired aminolactam, (S)-CAP, in high yield and excellent purity.

Further optimization to reduce the isolation of intermediates and improve productivity resulted in an eight-stage process with only three product-isolation steps. This was shown successfully on a kilogram scale and was fully feasible for transfer to production scale, says Schwarm. Because of limited availability of larger quantities of L-homoserine at that time, however, this route had to be abandoned.

L-methionine, therefore, was selected as the most suitable starting material (see Figure 1). First, the amino acid was N-protected and then converted into the p-cyanoanilide using common peptide chemistry methods, explains Schwarm. S-alkylation then transformed the methylthio function into a leaving group. For future large-scale production, special care had to be taken to avoid the formation and release of foul-smelling volatile organosulfur compounds. That task was achieved by selecting halocarboxylic acids as alkylating agents. The leaving group resulting from that process is a methylthiocarboxylic acid, which remains in the reaction solution as a salt after the subsequent basic cyclization to the protected lactone. The formation of any unpleasant odor, therefore, could be well controlled.

Figure 1: Degussa´s L-methionine route toward (S)-CAP.

As the final step, the N-protective group was removed. As the Z-group had been selected for economic reasons, careful adjustment of the reaction conditions for the catalytic hydrogenation was mandatory to limit the formation of an aminomethylanilide by undesired parallel reduction of the cyano function. After the optimization of this six-stage process, only three product-isolation steps were finally required.

Although the overall yields were largely comparable with the first process based on L-homoserine, the new route starting from L-methionine had the advantages of a reduced number of steps and established commercial availability of the starting material. This process was selected for further scale-up and successfully transferred first to pilot and then to full production scale.

In a subsequent research program, the same alkylation procedure also was applied for converting L-methionine into L-homoserine.

"As we had experienced ourselves initially, this amino acid had not been readily available on the market till now. This was due to its high water solubility and its inherent sensitivity to cyclization to the lactone under acidic conditions, which makes isolation difficult and laborious," says Schwarm.

In the Degussa process (as outlined in Figure 2), L-methionine is alkylated with bromoacetic acid. Hydrolysis under slightly acidic conditions affords a mixture of methylthioacetic acid and the desired L-homoserine. This is finally isolated using a series of different ion exchange resins and subsequent crystallization from a salt-free aqueous solution by the addition of alcohol. Strongly acidic conditions are carefully avoided to prevent undesired lactonization. Again, this process was successfully introduced on a large scale.

Figure 2: Degussa´s route to enantiomerically pure homoserine.

Stereoselective reduction of ketones

Catalysis is another important tool in enantioselective reactions, with recent reports of enzymatic systems for synthesizing key pharmaceutical intermediates (1–6). BioCatalytics, Inc. (Pasadena, CA), for example, recently developed a range of ketoreductases for the stereoselective reduction of ketones, explains David Rozzell, president and CEO of BioCatalytics. He points to two recent examples in biocatalytic ketone reduction.

In work by researchers at Merck & Co. (Whitehouse Station, NJ), (S)-3,5-bis-trifluoromethyl phenylethanol was produced at commercial scale in greater than 99.9% enantiomeric excess by asymmetric enzymatic reduction of the corresponding ketone. Isolated yields were greater than 90%, with volumetric productivities exceeding 260 g/L day. The nicotinamide cofactor could be recycled using either formate dehydrogenase or glucose dehydrogenase (5). BioCatalytics developed variants of both recycling enzymes that improved catalytic activity and stability for commercial use, he explains.

The synthetic liver : biocatalysis in synthesizing drug metabolites

Another recently published example of ketoreductases involves producing the natural insect pheromone (+)-sitophilure by the enzyme-catalyzed reduction of 4-methyl-3,5-heptadione. The natural pheromone has been identified as (4S, 5R)-4-methyl-5-hydroxy-3-heptanone. Researchers at BioCatalytics screened a group of roughly 100 ketoreductases and identified the best enzyme for the stereoselective monoreduction of the diketone 4-methyl-3,5-heptadione to produce the desired monoalcohol as a single diastereomer. The starting ketone is easily prepared in a single step by the methylation of 3,5-heptadione. The synthesis is accomplished in a two-step process, producing (+)-sitophilure in high yield (6).

Ketone reduction is one application of biocatalysis, explains Rozzell, who outlines other recent advances in enzymatic reactions. Over the past year, BioCatalytics has launched several new enzyme products to catalyze different reactions products including nitrilases, transaminases for synthesizing both amines and amino acids, reductive aminases, alkene reductases, and epoxidases. Late in 2006, BioCatalytics and Novozymes A/S (Bagsvaerd, Denmark) formed a pact under which BioCatalytics will bring to market a group of enzymes used in chemical synthesis. "The enzymes will be offered in screening sets containing new enzymes as well as improved versions of existing enzymes, all of which have been specially formulated for use in chemical synthesis," explains Rozzell.

Companies target biocatalysis

Company Web sites

Biocatalysis also is useful in enantioselective reactions, particularly with molecules with more than one chiral center.

"The biggest benefit provided by biocatalysis is its selectivity," explains Jean-Marie Sonet, general manager, PCAS Biosolution SAS, the recently formed joint venture between the custom manufacturer PCAS (Longjumeau, France) and Proteus (Nimes, France), which specializes in bioprocessing and biotransformations.

"Because biocatalysts feature an extremely high chemo-, regio-, and stereo-selectivity, they give access to difficult-to-produce fine chemicals, such as complex chiral molecules, reduce the number of steps of processes by avoiding multiple protection and deprotection steps, hence increasing the conversion rate, reducing the amount of by-products and improving atom economy, overall productivity, and energy consumption," says Sonet.

One approach by PCAS Biosolution is to develop biocatalysts from microbially diverse pools from extreme environments (7). Sonet explains the microbial biodiversity enables the design of robust biocatalysts, including biocatalysts adapted to high temperature. Examples include esterases with optimal temperatures at roughly 100 °C for accelerated processing of low-solubility material without solvents. The biocatalysts can be used several days at 100 °C (8).

Other parameters, such as production yields, enantiomeric excess, biocatalysts' stability, and catalytic properties, may be improved using molecular technology. PCAS Biosolution has proprietary platforms ("EvoSight" and "L-Shuffling") in directed molecular evolution technology based on gene shuffling (9–13). Sonet says directed molecular evolution technology overcomes limits of prior art—directed evolution technologies based on mutagenesis or polymerase chain reaction-like recombination. In one example, the company achieved a 50-fold increase in catalytic activity versus the wild type natural enzyme by using a combination of its EvoSight technology and three subsequent rounds of gene shuffling (14).

Other technology providers, such as Codexis, Inc. (Redwood City, CA), and custom manufacturers , such as Kaneka (New York, NY), DSM (Heerlen, Netherlands), Dowpharma (Midland, MI), BASF (Ludwigshafen, Germany), and Wacker (Müchen, Germany) are examples of other companies advancing biocatalysis in synthesizing APIs (15).


1. S. Kambourakis and J.D. Rozzell, "Chemo-Enzymatic Method for the Synthesis of Statine, Phenylstative and Analogues," Adv. Synth. Catal. 345 (6–7), 699–705 (2003).

2. S. Kambourakis and J.D. Rozzell, "Ketoreductases in the Synthesis of Valuable Chiral Intermediates: Application in the Synthesis of α-hydroxy β-amino and β-hydroxy γ-Amino Acids," Tetrahedron60 (3), 663–669 (2004).

3. J.D. Rozzell et al., "A Recombinant Ketoreductase Tool-box. Assessing the Substrate Selectivity and Stereoselectivity Toward the Reduction of β-Ketoesters," Tetrahedron62 (5), 901–905 (2006).

4. I.A. Kaluzna, J.D. Rozzell, and S. Kambourakis, "Ketoreductases: Stereoselective Catalysts for the Facile Synthesis of Chiral Alcohols," Tetrahedron: Asymmetry16 (22), 3682–3689 (2005).

5. D. Pollard et al., "Effective Synthesis of (S)-3,5-Bistrifluoromethylphenyl Ethanol by Asymmetric Enzymatic Reduction," Tetrahedron: Asymmetry17 (4), 554–559 (2006).

6. D Kalaitzakis et al., "Highly Stereoselective Reductions of Alpha-Alkyl-1,3-Diketones and Alpha-Alkyl-Beta-Keto Esters Catalyzed by Isolated NADPH-Dependent Ketoreductases," Org Lett.7 (22), 4799–4801 (2005).

7. G. Ravot, J.M. Masson, and F. Lefevre, "Applications of Extremophiles: The Industrial Screening of Extremophiles for Valuable Biomolecules," in Methods in Microbiology, F. Rainey and A. Oren, Eds. (Elsevier Science, New York, NY, 2006), pp. 785–813.

8. G. Ravot et al., "Screening for Thermostable Esterases: From Deep Sea to Industry, "Engineering in Life Sciences, 4 (6), 533–538 (2004).

9. M. Montezin, J.M. Sonet, and F. Lefevre, "Rapid Development of Novel Biocatalysts for the Manufacturing of Chiral Compounds," PharmaChem, 3 (9), 7–10 (2004).

10 . D. Dupret et al., "Process for Obtaining Recombined Nucleotide Sequences In Vitro Libraries of Sequences and Sequences Thus Obtained," US Patent 6,951,719, Oct. 4, 2005.

11. D. Dupret et al., "Process for In Vitro Creation of Recombinant Polynucleotide Sequences by Oriented Ligation," US Patent 6,991,922, Jan. 31, 2006.

12. D. Dupret et al., "In Vitro Recombination of Polynucleotide Fragments to Obtain Sequences with Improved Properties Involves Ligation on an Assembly Matrix," European Patent 1104457, Mar. 27, 2006.

13. D. Dupret et al., "Method for Obtaining (In Vitro) Recombined Polynucleotide Sequences, Sequence Banks and Resulting Sequences," Australian Patent 769516, Jan. 29, 2004.

14. L. Fourage and J.M. Sonet, "Biocatalysis from Biodiversity for Competitive Green Chemistry," Specialty Chemicals Magazine 26 (2), 42–43 (2006).

15. P. Van Arnum, "Achieving Enantioselectivity in Synthesizing APIs and Intermediates," Pharm. Technol. 30 Technology Outlook: APIs, Intermediates, and Formulation," suppl. s20–s25 (2006).