Arylboronic acids, but not the corresponding deboronated arenes, recently have been found to be weakly mutagenic in microbial assays [1].  Hence arylboronic acids may be considered potentially genotoxic impurities, and controlling the levels of residual arylboronic acids in APIs could become a regulatory requirement.  The issues should be decided by toxicology studies for the specific arylboronic acids in question.

Several approaches have been successful in removing boronic acids.  Diethanolaminomethyl polystyrene (DEAM-PS) [2],[3] and immobilized catechol [4] have been used to scavenge boronic acids.  Complex formation with diethanolamine may solubilize residual boronic acids in mother liquors.  Since arylboronic acids ionize similarly to phenols, basic washes of an API solution may remove arylboronic acids.  A selective crystallization can purge an arylboronic acid from the API.

The best means to control residual aryl boronic acids in APIs at the ppm level may be to decompose them through deboronation.  Sterically hindered, electron-rich aryl boronates and heteroaromatic boronates are especially prone to deboronation [5],[6],[7].  Deboronation of a hindered difluorobenzeneboronic acid was accelerated in the presence of Pd, and increased in the presence of H₂O [8].  Kuivila and co-workers found that protodeboronation of 2,6-dimethoxybenzeneboronic acid was slowest about pH 5, and rapid under more acidic or basic conditions [9].  Butters and co-workers found that protodeboronation occurred for an arylboronate anion, but not for the corresponding arylboronic acid [10].  Amgen researchers found that deboronation of an ortho-substituted arylboronic acid was fastest in the presence of K₂CO₃ [11].  The Snieckus group found that deboronation occurred readily when pinacol was added to 4-pyridylboronic acid [12].  Percec and co-workers found that deboronation of neopentylglycol boronates, especially an ortho-substituted arylboronic acid ester, was catalyzed by nickel species [13].  Kuivila and co-workers found that CuCl₂ catalyzed the deboronation of 2,6-dimethoxybenzeneboronic acid and other arylboronic acids, with formation of the corresponding aryl chlorides [14].  Unfortunately, adding reagents to a reaction mixture increases the burdens of analysis and impurity removal, but additives such as these may accelerate deboronation in difficult cases.  Simply extending the reaction conditions, which are generally basic for efficient Suzuki coupling, or heating with some amount of aqueous hydroxide are probably the preferred treatments to decompose an arylboronic acid.  By knowing the kinetics of the decomposition of the arylboronic acid it may be possible to show by QbD that analyses for the residual arylboronic acid in an API are not necessary.

---

[1] O’Donovan, M. R.; Mee, C. D.; Fenner, S.; Teasdale, A.; Phillips, D. H. Mutat. Res.: Genet. Toxicol. Environ. Mutagen. 2011, 724(1-2), 1.

[2] Antonow, D.; Cooper, N.; Howard, P. W.; Thurston, D. E. J. Comb. Chem. 2007, 9, 437.

[3] Hall, D. G.; Tailor, J.; Gravel, M. Angew. Chem. Int. Ed. 1999, 38, 3064.

[4] Yang, W.; Gao, X.; Springsteen, G.; Wang, B. Tetrahedron Lett. 2002, 43, 6339.

[5] Goodson, F. E.; Wallow, T. I.; Novak, B. M. Organic Syntheses; Vol. 74; Smith, A. B., III, Ed.; Organic Syntheses, Inc.; 1997; p. 64.

[6] Baudoin, O.; Cesario, M.; Guénard, D.; Guéritte, F. J. Org. Chem. 2002, 67, 1199.

[7] Dai, Q.; Xu, D.; Lim, K.; Harvey, R. G. J. Org. Chem. 2007, 72, 4856.

[8] Cammidge, A. N.; Crépy, K. V. L. J. Org. Chem. 2003, 68, 6832.

[9] Kuivila, H. G.; Reuwer, J. F., Jr.; Mangranite, J. A. Can. J. Chem. 1963, 41, 3081.

[10] Butters, M.; Harvey, J. N.; Jover, J.; Lennox, A. J. J.; Lloyd-Jones, G. C.; Murray, P. M. Angew. Chem. Int. Ed. 2010, 49, 5156.

[11] Achmatowicz, M.; Thiel, O. R.; Wheeler, P.; Bernard, C.; Huang, J.; Larsen, R. D.; Faul, M. M. J. Org. Chem. 2009, 74, 795.

[12] Alessi, M.; Larkin, A. L.; Ogilvie, K. A.; Green, L. A.; Lai, S.; Lopez, S.; Snieckus, V. J. Org. Chem. 2007, 72, 1588.

[13] Moldoveanu, C.; Wilson, D. A.; Wilson, C. J.; Leowanawat, P.; Resmerita, A.-M.; Liu, C.; Rosen, B. M.; Percec, J. Org. Chem. 2010, 75, 5438.

[14] Kuivila, H. G.; Reuwer, J. F., Jr.; Mangravite, J. A. J. Am. Chem. Soc. 1964, 86, 2666.

In January 2011 the FDA issued guidance entitled “Process Validation: General Principles and Practices,” which replaces the 1987 guidance [1].  The guidance stresses lifecycle management, with the ability to store and retrieve data being important.  The document describes three basic stages of process validation: process design, process qualification, and continued process verification.  Design of experiments (DoE) and statistical process control (SPC) are urged.  The document states that operations controlled by process analytical technology (PAT) may be preferred over non-PAT systems.  The FDA urges readers to contact them with specific questions in areas not covered by this and related guidances.

In December 2010 Organic Process Research & Development posted a manuscript I wrote with two colleagues, entitled “Current Practices of Process Validation for Drug Substances and Intermediates.” [2].  This manuscript is consistent with the current FDA guidance.  Described in the manuscript are a number of previously unpublished examples of challenges and successes of implementing processes.  More experienced process chemists and engineers may groan at some of these “war stories,” which can serve as valuable lessons.

---

[1] Guidance for Industry. Process Validation: General Principles and Practices” issued 1/24/2011 http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM070336.pdf

[2] Anderson, N. G.; Burdick, D. C.; Reeve, M. M. Org. Process Res. Dev. 2011, 15, 162.

The attractiveness of using biocatalysis can be assessed by considering the benefits, drawbacks, and cost of biocatalytic systems. Benefits may include a route with fewer intermediates, or intermediates with higher yield and higher quality; these changes could result in fewer unit operations and fewer intermediates to assay, hence lower cost of goods. Other benefits may include decreased emissions of volatile organic compounds (solvents), decreased cost of remediating solvent-laden waste streams, and decreased solvent recovery operations. A significant benefit may be increased flexibility in selecting equipment to prepare the API.

Drawbacks of biocatalytic systems can include a potentially significant time to develop a biocatalytic process, unless a commercially available biocatalyst can be suitably employed. Ownership of intellectual property and licensing should be clearly addressed before a contract research organization develops a biocatalyst specifically for the API. If only one vendor can supply the biocatalyst, contracts must be carefully composed to ensure reliable delivery at an attractive price. Conditions and compounds that inhibit the biotransformation must be identified and controlled, similar to other catalytic processing. If the biocatalyst must be shipped to the site of API manufacturing, stability of the biocatalyst should be considered.

Recently authors have reviewed in detail the factors for estimating the cost of a biocatalyst [Tufvesson, P.; Lima-Ramos, J.; Nordblad, M.; Woodley, J.M. Org. Process Res. Dev. 2011, 15, 266]. (These authors did not estimate the cost of developing a biocatalyst.) With high fermentation yield the cost of the biocatalyst decreases; not all enzymes may be produced in high titer. To increase the turnover number an enzyme may be purified or even immobilized. The extent of purification has a great impact, with costs increasing as the processing evolves from whole cells to crude enzymes to purified enzymes to immobilized enzymes. As production volume increases the cost of the biocatalyst decreases, as expected through economy on scale. Costs associated with equipment may be detailed for an in-house COG estimate, or may not be mentioned in a proposal by a contract manufacturing organization.

Many factors should be considered to assess the desirability of a biocatalytic process. The “greenness” of a biocatalytic process must include the operations and waste both in the process and in generating the biocatalyst. Ultimately the acceptability of the cost of a biocatalyst must be judged relative to the value of the product. The value of increased flexibility in equipment choice and improved ruggedness in quality of a product may not be quantifiable before routine manufacture of an API. Nonetheless biocatalytic processes may provide clear advantages even early in the development of API candidates.

Many congratulations to Professors Heck, Negishi, and Suzuki on receiving the Nobel Prize for their pioneering research into metal-catalyzed cross-coupling reactions.  Their influential work has led to processes for APIs, intermediates, drug candidates, and more.

Process understanding of cross-couplings is deepening and mild reaction conditions have been developed.  After extensive DoE screening GSK workers developed a Pd-catalyzed coupling that proceeds under very mild conditions: KHCO₃, iPrOH – H₂O (4:1), 60 ºC / 2 hours, 82% isolated yield [[1]].  A group from AstraZeneca and the University of Bristol has shown through incisive NMR studies that in a Pd-catalyzed cross-coupling of trifluoroborates water provides sustained release of the corresponding arylboronic acid and leads to improved yields; THF – H₂O (10 : 1) was the optimal solvent mixture [[2]].  In the case of a Pd-catalyzed amidation of o-bromotoluene 2.5 eq. of H₂O was optimal when the reaction was run with Cs₂CO₃ in toluene; in dioxane or with PhONa or NaOtBu as base the amidation was less sensitive to H₂O.  The authors posed that H₂O increased the solubility of the base in the reaction solvent [[3]].  Snieckus and coworkers have shown that 10 mol% H₂O is optimal in a study of Ni-catalyzed couplings with boroxines; increased charges of H₂O lead to the formation of NiO / Ni(OH)₂ [[4]].  The Burke group found that in Pd-catalyzed cross-couplings anhydrous conditions are necessary to prevent hydrolysis of arylboronic acids protected with the MIDA group [[5]]; MIDA boronates effectively cross-couple in Pd-mediated processes wherein the boronic acid is generated under alkaline conditions [[6]].  Lipshutz’s group has extended their work on reactions in aqueous microemulsions, and have shown that Fujiwara – Moritani reactions can be conducted at room temperature in water [[7]].  Workers at Syncom and DSM have written an extensive review on cross-couplings of heteroarenes [[8]].  Consider the possible benefits and limitations of water in your cross-coupling processes.

---

[1]. Bullock, K. M.; Mitchell, M. B.; Toczko, J. F. Org. Process Res. Dev. 2008, 12, 896.

[2]. Butters, M.; Harvey, J. N.; Jover, J.; Lennox, A. J. J.; Lloyd-Jones, G. C.; Murray, P. M. “Aryl Trifluoroborates in Suzuki–Miyaura Coupling: The Roles of Endogenous Aryl Boronic Acid and Fluoride.” Angew. Chem. Int. Ed. 2010, 49, 5156.

[3]. Dallas, A. S.; Gothelf, K. V., “Effect of Water on the Palladium-Catalyzed Amidation of Aryl Bromides.” J. Org. Chem. 2005, 70, 3321.

[4]. Antoft-Finch, A.; Blackburn, T.; Snieckus, V. “N,N-Diethyl O-Carbamate: Directed Metalation Group and Orthogonal Suzuki−Miyaura Cross-Coupling Partner.” J. Am. Chem. Soc. 2009, 131, 17750.

[5]. Ballmer, S. G..; Gillis, E. P.; Burke, M. D. Org. Synth. 2009, 86, 344.

[6]. Knapp, D. M.; Gillis, E. P.; Burke, M. D. J. Am. Chem. Soc. 2009, 131, 6961.

[7]. Nishikata, T.; Lipshutz, B. H. “Cationic Pd(II)-Catalyzed Fujiwara-Moritani Reactions at Room Temperature in Water” Org. Lett. 2010, 12, 1972.

[8]. Slagt, V. F.; de Vries, A. H. M.; de Vries, J. G.; Kellogg, R. M. “Practical Aspects of Carbon-Carbon Cross-Coupling Reactions Using Heteroarenes” Org. Process Res. Dev. 2010, 14, 30.

Biocatalysis can provide alternative routes and starting materials to the process chemist, especially for chiral chemistry.   Early incorporation of a biocatalytic process may afford significant savings during the lifetime of a drug or product.  The time needed to develop a biocatalyst is becoming less and less.  Below is a list of some biocatalytic transformations that may prove useful.

Functional Group Conversions by Biocatalysts

hydrolysis of esters (lipases and esterases)

reduction of ketones to secondary alcohols (ketoreductases, aka KREDs)

reduction of polarized double bonds (ene reductases)

epoxidation of olefins (monooxygenases, peroxidases)

oxidation of ArH to ArOH (monooxygenases)

oxidation of arenes to the corresponding dihydrocatechols (dioxygenases)

Baeyer – Villager oxidation of ketones to lactones (monooxygenases)

oxidation of RCH₂OH to RCHO and allylic alcohols to enones (alcohol oxidases)

removal of the p-methoxyphenyl group (PMP) from amines or imines (laccases)

halogenations, even with fluoride (halogenases)

hydration of nitriles to primary amides (nitrile hydratases)

hydrolysis of nitriles to RCO₂H (nitrilases)

hydrolysis of epoxides (epoxide hydrolases)

epoxide formation from halohydrins (dehalogenases)

formation or hydrolysis of amides (peptidases)

transamination of primary amines and ketones or aldehydes (transaminases)

aldol reaction (aldolases)

cyanide addition to RCHO (hydroxynitrile lyases)

glycosidation without protecting groups (glycosidases)

References

Meyer, H.-P.; Ghisalba, O.; Leresche, J. E. Biotransformations and the Pharma Industry; Chapter 7 in Green Catalysis; Vol. 3; Anastas, P. T., Ed.; Wiley – VCH: Weinheim; 2009; pp. 171-212.

Tao, J.; Xu, J.-H. Enzymes and Their Synthetic Applications: An Overview, Chapter 1 in Biocatalysis for the Pharmaceutical Industry: Discovery, Development, and Manufacturing; Tao, J.; Lin, G.-Q.; Liese, A., Eds.; John Wiley; 2009; pp. 1-19.

Kudos to Codexis and Merck for their 2010 Presidential Green Chemistry Challenge Award!  Codexis developed a transaminase to convert a ß-ketoamide into a chiral ß-amino amide, and Codexis and Merck implemented this biocatalytic process to manufacture sitagliptin.  The benefits of this step include eliminating one intermediate, improved yields and productivity, decreased waste, and more flexibility in selecting process equipment.  [(Savile, C. K.; Janey, J. M., et al., Science, 2010, 329, 305 (16 July)]  By superseding the previous manufacturing route with a biocatalytic route clearly it is possible to improve upon even optimized, creative manufacturing processes.

Recent publications by Codexis show it is feasible to select and develop enzymes for productive processing, e.g., operations at 0.5 M or higher in substrate at above ambient temperatures with up to 50% organic solvents and loadings of <1 wt% enzyme.  Previously researchers have adapted conditions to best utilize existing enzymes, but now biocatalysts can be developed to meet the demands for desired operations.

The correct addition sequence can be crucial to the success of a process, including the sequence of adding solids and solvents.  Charging solids to a reactor containing a flammable solvent can build up an electric charge, and subsequent discharge can ignite vapors above the solvent; for this reason solvents are usually added to an inerted vessel after solids have been charged.  Sometimes when solvent is added second the solids clump and are poorly suspended, changing the availability of  the solids for subsequent reaction.  When conducting a use-test make sure to follow the addition sequence that will be used on scale-up.

Processing may be easier with reduced cost of goods if that catalyst is not needed.  For instance, with the introduction of each additional component come not only increased costs to purchase and qualify the material, but also an increased burden of removing the component and possibly analyzing for the impurity in the product.  Of course increased reaction rates and improved impurity profiles with a catalyst can readily compensate for these increased costs.  Screening may demonstrate that a catalyst is actually detrimental (for an example of adding a thiol to an epoxide, see Schwartz, A.; Madan, P. B.; Mohacsi, E.; O’Brien, J. P.; Todaro, L. J.; Coffen, D. L. J. Org. Chem. 1992, 57, 851).

Screen crystallization solvents for H₂O, as a small amount of H₂O in solvents or reagents can affect not only the formation of hydrates but also polymorph formation.   For an example of how small amounts of water in a starting material affected polymorph formation, see Gross, T. D.; Schaab, K.; Ouellette, M.; Zook, S.; Reddy, J. P.; Shurtleff, A.; Sacaan, A. I.; Alebic-Kolbah, T.; Bozigian, H. Org. Process Res. Dev. 2007, 11, 365.

When designing routes and selecting reagents, carefully consider the toxicity and safe handling of reagents.  Incorporating a potentially genotoxic intermediate into a route may be necessary to reduce the preparation time and cost of goods, thus outweighing the costs of handling and removing this intermediate from the product.  On the other hand, if pilot plant personnel are unwilling to sample a drum of a dangerous reagent prior to using this lot in pilot plant scale-up, this raw material should be avoided: use-tests of individual lots are essential before committing valuable time and materials to scale-up.  Consult with others on the toxicity and safe handling of reagents.