Manufacturing Chemist, April 2016
As the structures of small-molecule active pharmaceutical ingredients (APIs) become increasingly complex, the development of synthetic routes to these drug substances that are practical at commercial scale grows more challenging. Adding to the difficulty is the need for cost-efficient, atom-economical processes that have minimal impact on the environment (green chemistry).
Catalysis is the key enabler to achieving these goals. Transition metal catalysts are widely used in the pharmaceutical industry today to mediate numerous asymmetric transformations that result in the formation of multiple bonds and chiral centres in one step. As technology for the development of biocatalysts has advanced, enzymes have increasingly been used for commercial-scale pharmaceutical manufacturing.
While biocatalysts do offer some advantages over chemocatalysts (metal-free reactions that often proceed under mild conditions in water), it is highly unlikely that they will completely replace transition metal complexes in small-molecule API synthesis. There are, in fact, often instances where enzyme-based catalyst systems either do not perform as well or are not available for transformations that are effectively mediated by chemocatalysts.
Indeed, the two technologies are often complementary in their chemistry or their applicability at different stages in the manufacturing lifecycle of small-molecule drugs. Access to a growing toolbox containing both enzyme and transition metal catalysts is therefore enabling more rapid development of more cost-efficient and environmentally friendly synthetic routes to novel medicines – all of which benefits patients in need.
A Note about Chirality
Chirality relates to the structure of a molecule and the fact that certain atoms, when bonded to different substituents, can adopt different physical structures. For instance, a carbon atom substituted with four different groups can adopt two different structures that are non-imposable mirror reflections of one another — like right and left hands. Although these enantiomers have the same chemical formulas and many of the same physicochemical properties, they rotate polarised light in opposite directions and generally exhibit very different interactions with other chiral molecules, including proteins and enzymes.
As a result, the pharmacokinetic and pharmacodynamic behaviours of two enantiomers of a chiral API can be very different, leading to different efficacies and toxicities. In some cases, one enantiomer exhibits higher efficacy than its isomer, while in others one enantiomer accounts for all of the efficacy and the other is inactive.
It is also possible for the enantiomers to function via different modes of action and be effective for the treatment of different diseases. In the worst-case scenario, the second enantiomer is toxic and results in harm to the patient. In addition, the solubilities and stabilities of the two enantiomers may be different under physiological conditions, and they may interact differently with any chiral excipients that are used in the formulation.
Neither the US Food and Drug Administration (FDA) nor the European Medicines Agency (EMA) requires that chiral drugs be developed as single enantiomer products. However, they do require a comprehensive evaluation of the properties of the individual enantiomers and racemic (50:50) mixtures of chiral APIs to determine, based on scientific evidence, whether a chiral drug should be marketed as a single isomer or a mixture.
More than half of the small-molecule drugs on the market today are chiral compounds. Many of these APIs are now offered as single-enantiomer products. Some were originally developed that way, and others are chiral switches — products originally offered as racemic mixtures but then redeveloped as pure-isomer drugs, typically to extend patent life. Indeed, advances in chiral synthesis and separation technologies have made the chiral switch approach quite attractive over the past decade.
Going from a racemic mixture to a single-enantiomer product can offer many medical benefits as well – such as lower doses to achieve the same result, elimination of undesirable side-effects, reduced drug interactions and cleaner pharmacokinetic / pharmacodynamic profiles (greater predictability) for more accurate data collection and monitoring.
Chiral Catalyst Selection Factors
The choice of chiral catalyst technology best suited to the commercial production of pharmaceutical intermediates and APIs depends on a number of factors. The phase of development is perhaps the first consideration. In the early development phases when speed is paramount, drug manufacturers may in fact choose to produce small quantities of a racemic mixture of the desired compound using established chemistry and then separate the two enantiomers, either using chiral chromatography or more traditional resolution techniques. An enantioselective catalytic route is developed in parallel for more economical large-scale production.
The stage of development can also affect the choice of a chemocatalyst versus a biocatalyst. At least until recently, the number of enzymes available for commercial-scale manufacturing was limited, and typically engineering of an enzyme with the desired reactivity – which was a lengthy process – was required. As a result, enzymes have most often been used during the later phases of drug development and/or after commercial launch when drug manufacturers have sought to lower costs to maintain competitiveness.
With respect to selecting chiral chemocatalysts, for some transformations, only one type of catalytic reaction is available, but for most compounds there is a choice of methods. The yields and selectivities for each catalyst, the necessary catalyst loadings and reaction conditions (temperature, solvent, workup) and the availability and costs of the metal and ligand, as well as any potential intellectual property issues with the ligand, all must be considered when selecting a transition metal catalyst.
An understanding of the catalysis mechanism of a given transformation and the species that actually mediates the transformation (active catalytic species) is also necessary for identifying the optimum metal-ligand-additive/activator combination that provides the most robust and reliable commercial-scale process. Most drug companies and contract development and manufacturing organisations (CDMOs) use a wide variety of techniques, such as high throughput screening, combinatorial ligand discovery, computational tools, and modern analytical methods, to gain the knowledge needed.
For some reactions, there may also be a choice of heterogeneous and homogeneous catalysts for a given reaction. Unlike homogeneous catalysts, heterogeneous catalysts do not dissolve in reaction solutions and thus are easier to recover and can often be reused. However, homogeneous catalysts tend to provide higher yields and selectivities than comparable heterogeneous catalysts; because very high selectivities are required for chiral transformations, homogeneous catalysts are most often employed for these reactions.
Finally, the choice of catalyst technology often depends on the capabilities of the drug company and/or its contract development and manufacturing service providers. Some companies specialise in certain types of catalysis — biocatalysis or a specific type of chemocatalysis, such as asymmetric hydrogenation or oxygenation – and therefore will be more likely to choose a method that fits with their chiral catalyst toolbox.
Spotlight on Chemocatalysts
Asymmetric hydrogenation is the most widely use chiral catalytic method in the pharmaceutical industry. Other important chemocatalytic transformations for large-scale small-molecule API manufacturing include the Jacobsen epoxidation chemistry, Buchwald-Hartwig amination reactions, Suzuki and other cross-coupling reactions, and transfer hydrogenation processes.1
There are many more potential reactions, however, and researchers in both academia and industry are constantly developing new chiral transformations with potential application to pharmaceutical manufacturing. For instance, chiral Grignard addition reactions are particularly appealing because this chemistry is widely used at large scale for the preparation of achiral compounds. However, the high reactivity and basicity of Grignard reagents makes it challenging to design highly enantioselective catalytic systems.2
Chao-Shan Da of the State Key Laboratory of Applied Organic Chemistry at Lanzhou University has used a chelating ligand to deactivate the Grignard reagent and a titanium / chiral ligand (commercially available) system to catalyse the asymmetric transfer of a Grignard substituent to aldehydes. The reaction is effective for the preparation of secondary aryl propanols that are intermediates for chiral drug synthesis and can be scaled to at least a 2.0g scale without loss of yield or enantioselectivity.
Other notable developments in chiral chemistry include new methods for the synthesis of chiral natural and non-natural amino acids and their derivatives, chiral fluorinated compounds, the asymmetric hydrogenation of a wider range of substrates, asymmetric C-H insertion reactions to functionalise chiral intermediates, and chiral carbon-carbon coupling reactions that form asymmetric quaternary carbon centres and a broad array of other useful chiral building blocks.3 Chiral multicomponent and cascade reactions that afford highly complex intermediates from basic starting materials in an atom economical manner have also been reported.
Even with the significant advances described above, there are limitations to chiral chemocatalysis. Researchers continue to seek chiral catalysts that can achieve the synthesis of challenging structural motifs and enable the chiral transformation of compound types that have proven resistant to selective chiral modification. Considerable attention is being paid to the development of novel ligands as a means for addressing these issues. The development of ligands that are also simpler, more readily synthesised, and cheaper but still provide very high stereoselectivities in enantiomeric reactions are also a focus of many industrial and academic research groups.
For homogeneous reactions, practical issues, such as the reclamation and reuse of the catalysts and the reduction of catalyst loadings, remain important areas of study. Immobilisation of catalysts on a solid support and encapsulation of toxic complexes without loss of selectivity and reactivity are two approaches that have been shown to be effective in some cases (e.g. Rh-Duphos on phosphotungstic acid for asymmetric hydrogenation and OsO4 for Sharpless asymmetric dihydroxylation, respectively).1 The replacement of transition metal catalysts based on precious metals with alternative catalyst systems based on cheap, earth-abundant metals is another area where researchers are making notable progress. The development of catalysts that provide scalable performance is also a need for commercial-scale chiral catalysis.
Biocatalysts are attractive for pharmaceutical manufacturing because they typically catalyse reactions under mild conditions, and thus these reactions do not consume much energy and generate limited greenhouse gas emissions. The lack of a need for precious, heavy, or toxic metals is a considerable advantage over transition metal catalysts. In addition, reactions catalysed by enzymes can often be performed in water rather than organic solvents and typically generate less waste than those mediated by chemocatalysts.
Enzyme-catalysed reactions are also ideal for the synthesis of chiral pharmaceutical compounds because of their high efficiencies and stereo- and regioselectivities. In many cases they can also mediate complex transformations that allow the reduction of the number of steps required for the synthesis of complicated molecules with multiple chiral centres.
The limited number of robust enzymes available off-the-shelf for the commercial-scale production of pharmaceutical intermediates and APIs has been an issue, however. Often engineering of a custom-designed enzyme is required to achieve a given reaction with the desired yield and selectivity.
As a result, while interest in biocatalysis is increasing, the use of enzymes in the pharmaceutical industry remains limited compared with the use of chemocatalysts, and they have generally found the widest application in the development of lower-cost alternative routes for late-stage manufacturing where development times are longer.
Advances in molecular biology, metabolic engineering, and bioinformatics are leading to significant gains in knowledge about a much greater number of organisms.4 For instance, techniques are now available for the rapid sequencing of genomes, providing access to a much wider array of enzymes that can serve as potential biocatalysts. Greater understanding of biosynthetic routes, the physical structures of enzymes and the relationships between those structures and enzyme functionalities are reducing the time required to engineer effective biocatalysts for specific applications. Site-specific mutagenesis by protein crystal structure-based rational design, directed evolution, and computational de novo design methods have all been explored to obtain enzymes with desired substrate preferences, stability, activity, expression levels and solubilities.5
As a result of these advances, it is now possible to custom engineer enzymes as biocatalysts in a few months at a fraction of the cost required just 10–15 years ago, and often at a price less than that for the expensive chiral ligands required for chemocatalysts.4 Consequently, biocatalysts are now more frequently considered throughout all pharmaceutical development phases, including for combinatorial synthesis for drug discovery efforts and initial route development, in addition to commercial-scale manufacturing.6
Multiple Enzyme Use
One area of particular interest to the pharmaceutical industry is the use of multiple enzymes to achieve numerous biocatalysed reactions in one step. The ultimate goal of synthetic biology is the production of highly complex pharmaceutical intermediates and/or APIs in engineered whole cells. In these systems, chiral compounds are selectively produced through the action of enzymes that work simultaneously and/or sequentially.
This approach is possible because for many enzymes, the presence of activated donor substrates is necessary, and these compounds can often be generated via the enzyme-catalysed reaction of simple starting materials.5
While most biocatalysed reactions in the pharmaceutical industry are one-step transformations, there have been a few multi-enzyme systems applied to pharmaceutical synthesis. One example is the use of an engineered N-acetyl amino acid racemase and an L- or D-acylase enzyme at the ChiroTech Technology Centre, which is part of Dr. Reddy's Custom Pharmaceutical Services business, to achieve the dynamic kinetic resolution of non-natural a-amino acids with high enantiomeric purities.5
Academic groups are also making advances with multi-enzyme transformations. Professor John Ward of University College London has used protein engineering and directed evolution to increase the range of substrates for transketolases, which mediate C-C bond forming reactions. His group has also engineered numerous transaminases for the synthesis of chiral amides and other derivatives via the insertion of NH2 into ketones and aldehydes. When combined, the transketolases and transaminases catalyse cascade reactions resulting in the formation of chiral enantiomers of amino diols, allowing the preparation of all four possible enantiomers of each amino diol.5
Meanwhile, Xi Chen at the University of California-Davis is using one-pot multi-enzyme (OPME) reactions to produce carbohydrates with post-glycosylational modifications (PGMs) to be able to study how they regulate the important recognition roles of carbohydrates. Examples include one-pot, three-enzyme systems for the synthesis of structurally defined fucose-containing oligosaccharides and structurally defined naturally and non-naturally occurring sialosides.
The systems are designed to use inexpensive starting materials, proceed under mild reaction conditions, and tolerate a wide range of substrates for the preparation of diverse products. Some of the oligosaccharides synthesised using these sequential one-pot multi-enzyme strategies may in fact be potential therapeutics.5
- Challener, C. Asymmetric Chemocatalysis: Going for the Lowest Loadings, Pharm. Tech., March 27, 2013. http://www.pharmtech.com/asymmetric-chemocatalysis-going-lowest-loadings.
- Challener, C. Advances in chiral Grignard and organolithium chemistry, Speciality Chemicals, June 18, 2012. http://www.specchemonline.com/featuredarticles/advances-in-chiral-grignard-and-organolithium-chemistry
- Challener, C. Asymmetric Chemistry Continues to Advance, Pharm. Tech. 38 (8), September 2, 2014.
- Challener, C. Synthetic Biology: The Next Frontier in Chiral Chemistry for API Synthesis, Pharm. Tech. 38 (3), March 02, 2014.
- Challener, C. Mimicking Nature, Speciality Chemicals, September 2014.http://www.specchemonline.com/featuredarticles/mimicking-nature
- Challener, C. DSM, Almac Partner in Biocatalysis, Pharm. Tech., March 13, 2013.http://www.pharmtech.com/dsm-almac-partner-biocatalysis.