US 2013 OO29385A1 (19) United States (12) Patent Application Publication (10) Pub. No.: US 2013/0029385 A1 Cabirol et al. (43) Pub. Date: Jan. 31, 2013

(54) PROCESSESUSING AMNO ACID Related U.S. Application Data DEHYDROGENASES AND KETOREDUCTASE-BASED COFACTOR (60) Provisional application No. 61/303,179, filed on Feb. REGENERATING SYSTEM 10, 2010. (75) Inventors: Fabien L. Cabirol, Essen (DE); Steven J. Collier, Singapore (SG); Thomas Publication Classification Daussmann, Annweiler (DE); Naga Modukuru, Singapore (SG) (51) Int. Cl. CI2PI3/04 (2006.01) (73) Assignee: CODEXIS, INC., Redwood City, CA (52) U.S. Cl...... 435/106 (US) (21) Appl. No.: 13/577,772 (57) ABSTRACT (22) PCT Filed: Feb. 8, 2011 The present disclosure relates to the use of an amino acid dehydrogenase in combination with a cofactor regenerating (86) PCT NO.: PCT/US11A24102 system comprising a ketoreductase. In particular embodi S371 (c)(1), ments, the process can be used to prepare L-tert-leucine using (2), (4) Date: Oct. 16, 2012 a leucine dehydrogenase. US 2013/0029385 A1 Jan. 31, 2013

PROCESSES USING AMNO ACID 0006. In certain embodiments, the present disclosure pro DEHYDROGENASES AND vides a process for converting a 2-oxo acid compound of KETOREDUCTASE-BASED COFACTOR formula I which is a Substrate for an amino acid dehydroge REGENERATING SYSTEM nase to a chiral amino acid of formula IIa.

1. TECHNICAL FIELD R CO2H CO2H 0001. The present disclosure relates to biocatalysts and s" - “S- processes for preparing chiral amino acids using the biocata O NH2 lysts. I IIa 2. REFERENCE TO SEQUENCE LISTING, TABLE OR COMPUTER PROGRAM comprising contacting the compound of formula I with a reaction medium comprising an amino acid dehydrogenase, 0002 The official copy of the Sequence Listing is submit an ammonium ion donor, NAD"/NADH or NADP/NADPH, ted concurrently with the specification as an ASCII formatted and a cofactor regenerating system comprising a ketoreduc text file via EFS-Web, with a file name of “CX2-038USP1 tase and a lower secondary alcohol, under conditions where ST25.txt, a creation date of Feb. 9, 2010, and a size of 48 the compound of formula I is converted to the chiral amino kilobytes. The sequence listing filed via EFS-Web is part of acid of formula IIa and the lower secondary alcohol is con the specification and is hereby incorporated in its entirety by verted to a ketone. In certain embodiments of the process, R reference herein. is a Substituted or unsubstituted (C-Co.)alkyl, —(C-C) alkenyl, —(C-C)alkynyl. —(C-C)cycloalkyl, heterocy 3. BACKGROUND cloalkyl, aryl, or heteroaryl. In certain embodiments, R is a 0003) Amino acid dehydrogenases comprise a group of Substituted or unsubstituted group selected from those groups coenzyme-dependent enzymes that catalyze the reversible shown in either of Table 1 or Table 2 disclosed herein. oxidative deamination of an amino acid to its keto acid and 0007. In certain embodiments of the process for convert ammonia with the concomitant reduction of either cofactor ing a compound of formula I which is a Substrate for an amino NAD+, NADP+ or FAD. The enzyme with dehydrogenase acid dehydrogenase to a chiral amino acid of formula IIa, the properties is distributed in a number of diverse prokaryotic amino acid dehydrogenase is an L-amino acid dehydrogenase and eukaryotic organisms. Amino acid dehydrogenases have and the chiral amino acid of formula IIa is IIb, been studied widely because of their potential applications in IIb biosensors or diagnostic kits, synthesis of L- and D-amino R CO2H acids for use in production pharmaceutical peptides, herbi cides and insecticides (Brunhuber et al., 1994, Crit. Rev. 1 Biochem. Mol. Biol. 29(6):415-467: Hummel et al., 1989, Eur. J. Biochem. 184:1-13: Krix et al., 1997, J. Biotech. 53:29-39: Ohshima et al., 1990, Adv. Biochem. Eng. Biotech wherein the product of formula IIb is formed in enantiomeric nol. 42:181-209; U.S. Pat. No. 7,550,277). For example, the excess. In certain embodiments, the L-amino acid dehydro anti-hypertensives ramipril, enalapril, benaZapril, and prini genase is from Bacillus, Clostridium, Corynebacterium, vil are prepared using L-homophenylalanine, and certain sec Geobacillus, Natronobacterium, Synechocystis, Thermoacti ond generation pril analogs are synthesized from p-substi nomyces, Thermos, Thermonicrobium, or Carderia. In some tuted-Lhomophenylalanine. Certain B-lactamantibiotics use embodiments, the L-amino acid dehydrogenase is selected substituted D-phenylglycine side chains, and while other from L-alanine dehydrogenase, L-aspartate dehydrogenase, antibiotics are based on aminoadipic acid and other unnatural L-erythro-3,5-diaminohexanoate dehydrogenase, L-leucine amino acids. The unnatural amino acids L-tert-leucine, dehydrogenase, L-glutamate dehydrogenase, lysine dehydro L-nor-valine, L-nor-leucine, L-2-amino-5-1.3dioxolan-2- genase, L-phenylalanine dehydrogenase, L-serine dehydro yl-pentanoic acid have been used as a precursor in the Syn genase, L-valine dehydrogenase, L-2,4-diaminopentanoate thesis of a number of different developmental drugs. The dehydrogenase, L-glutamate synthase, L-diaminopimelate enzyme leucine dehydrogenase and mutants thereof have dehydrogenase, L-N-methylalanine dehydrogenase, L-lysine been shown to be capable of catalyzing the reductive amina 6-dehydrogenase, and L-tryptophan dehydrogenase. tion of the corresponding 2-ketoacids of alkyl and branched 0008. In certain embodiments of the process for convert chain amino acids, and L-tert-leucine has been produced ing a compound of formula I which is a Substrate for an amino commercially with Such an enzyme. acid dehydrogenase to a chiral amino acid of formula IIa, the 0004 Given the industrial utility of L- and D-amino acid amino acid dehydrogenase is a D-amino acid dehydrogenase dehydrogenases, it is desirable to develop processes and sys and the chiral amino acid of formula IIa is IIc, tems that can enhance the biocatalytic reactions carried out by amino acid dehydrogenases. IIc RN-CO2H 4. SUMMARY 0005. The present disclosure provides coupled systems for s H the efficient biosynthesis of chiral amino acid compounds using an L- or D-amino acid dehydrogenase (AADH') wherein the compound of formula IIc is formed in enantio coupled with a cofactor regenerating system comprising a meric excess. In certain embodiments of the process, the ketoreductase (“KRED). D-amino acid dehydrogenase is from Halobacterium, Metha US 2013/0029385 A1 Jan. 31, 2013

nosarcina, Pseudomonas, Pyrobaculum, Salmonella, 0012 where the process comprises contacting the com Corynebacterium, and Escherichia. In certain embodiments, pound mixture of formula IId with an enantioselective amino the D-amino acid dehydrogenase is selected from a D-alanine acid dehydrogenase in a reaction medium comprising NAD/ dehydrogenase, D-threonine dehydrogenase, and D-proline NADH or NADP/NADPH and a cofactor recycling system dehydrogenase. comprising a ketoreductase and a lower alkyl ketone, under 0009. In certain embodiments of the process for convert conditions where the compound mixture of formula IId is ing a compound of formula I which is a Substrate for an amino converted to the composition of formula I and a chiral amino acid dehydrogenase to a chiral amino acid of formula IIa, the acid of formula IIa, and the lower alkyl ketone is converted to amino acid dehydrogenase comprises a L-leucine dehydro a lower secondary alcohol. In some embodiments of the pro genase, the compound of formula I is 3.3-dimethyl-2-oxobu cess, the compound mixture of IId is a racemic mixture of tanoic acid, the product of formula IIa is (S)-2-amino-3,3- formula IIe: dimethylbutanoic acid. In some embodiments of the process, the leucine dehydrogenase is a wild type leucine dehydroge nase or an engineered leucine dehydrogenase. In some IIe embodiments, the leucine dehydrogenase is from Bacillus, R CO2H Clostridium, Corynebacterium, Geobacillus, Natronobacte y rium, Thermoactinomyces, Thermos, Thermomicrobium, or Carderia. In some embodiments, the leucine dehydrogenase is from Bacillus acidokaludarius, Bacillus brevis, Bacillus caldolyticus, Bacillus cereus, Bacillus megaterium, Bacillus 0013. In some embodiments of the process, the amino acid mesentericus, Bacillus mycoides, Bacillus natto, Bacillus dehydrogenase comprises an L-amino acid dehydrogenase pumilus, Bacillus sp., Bacillus sphaericus, Bacillus Stearo and the chiral amino acid of formula IIa is IIc thermophilus, Bacillus subtilis, Clostridium thermoaceti cum, Corynebacterium pseudodiphtheriticum, Geobacillus Stearothermophilus, Natronobacterium magadii, or Thermo IIc actinomyces intermedius. In some embodiments, the leucine Rn-CO2H dehydrogenase comprises the amino acid sequence of SEQ ID NO: 26. s H 0010. In certain embodiments, the present disclosure pro vides a process for producing (S)-2-amino-3,3-dimethylbu tanoic acid, comprising: contacting 3.3-dimethyl-2-oxobu wherein the process results in chiral amino acid of formula IIc tanoic acid with a leucine dehydrogenase in a reaction in enantiomeric excess. In some embodiments of the process, medium comprising an ammonium ion donor, cofactor the amino acid dehydrogenase comprises a D-amino acid NAD"/NADH or NADP"/NADPH, and a cofactor recycling dehydrogenase and the chiral amino acid of formula IIa is IIb system comprising a ketoreductase and a lower secondary alcohol, under conditions where the 3.3-dimethyl-2-oxobu IIb tanoic acid is converted to (S)-2-amino-3,3-dimethylbutanoic R CO2H acid, wherein the3.3-dimethyl-2-oxobutanoic acid is at about 75 g/L to 125 g/L, the cofactor is at about 0.30g/L to 0.70 g/L, 1 and the leucine dehydrogenase and ketoreductase are each independently at about 0.5 to about 1.0 g/L. In certain embodiments of the process, the secondary alcohol is present wherein the process results in a chiral amino acid of formula in at least 1.5 fold stoichiometric excess of substrate. In some IIb in enantiomeric excess. embodiments of the process, the secondary alcohol is isopro 0014. In certain embodiments, the present disclosure pro panol, wherein the isopropanol is at about 7% to 12% volume vides a process for preparing an N-protected amino acid of the reaction medium by (weight/volume). compound, wherein the method comprises: (i) contacting a 0011. In certain embodiments, the present disclosure pro compound of formula I with a reaction medium comprising vides a process for converting a compound mixture of for an amino acid dehydrogenase, an ammonium ion donor, mula IId which comprises a Substrate for an amino acid dehy NAD"/NADH or NADP"/NADPH, and a cofactor regenerat drogenase to a composition of formula I and a chiral amino ing system comprising a ketoreductase and a lower secondary acid of formula IIa: alcohol under suitable conditions where the compound of formula I is converted to the chiral amino acid compound of formula IIa and the lower secondary alcohol is converted to a R CO2H R CO2H ketone; and (ii) contacting the amino acid compound of for r" . mula IIa with a compound comprising an N-protecting group NH2 O under conditions, where the N-protecting group reacts with IId I the compound of formula IIa to form an N-protected amino R CO2H acid compound. 0015. In certain embodiments of the process for preparing N an N-protected amino acid compound, the biocatalytic step NH2 comprises contacting 3.3-dimethyl-2-oxobutanoic acid with IIa a leucine dehydrogenase in a reaction medium comprising an ammonium ion donor, cofactor NAD"/NADH or NADP/ NADPH, and a cofactor recycling system comprising a US 2013/0029385 A1 Jan. 31, 2013

ketoreductase and a lower secondary alcohol, under condi Some embodiments of the process, the ketoreductase used in tions where the 3.3-dimethyl-2-oxobutanoic acid is converted the cofactor recycling system has no activity with the com to (S)-2-amino-3,3-dimethylbutanoic acid. In some embodi pound of formula I. ments, the3.3-dimethyl-2-oxobutanoic acid is at about 75 g/L 0020. In certain embodiments of the various processes for to 125 g/L, the cofactor is at about 0.30 g/L to 0.70 g/L, and preparing chiral amino acid compounds disclosed herein, the the leucine dehydrogenase and ketoreductase are each inde ketoreductase used in the cofactor recycling system is an pendently at about 0.5 to about 1.0 g/L. In some embodi engineered ketoreductase is capable of recycling cofactor by ments, the secondary alcohol is present in at least 1.5 fold converting isopropanol (IPA) to acetone in a reaction medium Stoichiometric excess of substrate. In some embodiments, the of 3 to 20% IPA at a pH of about 9.0 to 10.5 with an activity secondary alcohol is isopropanol, and the isopropanol is at at least 1.5-fold greater than the reference ketoreductase of SEQID NO: 2. about 7% to 12% volume of the reaction medium by (weight/ 0021. In certain embodiments of the various processes for Volume). In some embodiments, the N-protecting group is preparing chiral amino acid compounds disclosed herein, the selected from Cbz, FMOC, BOC and MOC. process further comprises removing from the reaction 0016. In certain embodiments of the various processes for medium the ketone formed from the lower secondary alcohol, preparing chiral amino acid compounds disclosed herein, the and in certain embodiments the lower secondary alcohol is process is carried out in a cell free system. In some embodi isopropanol and the ketone removed is acetone. In some ments of the various processes, the amino acid dehydroge embodiments, the secondary alcohol is present in at least 1.5 nase is present as a crude extract, and in Some embodiments, fold stoichiometric excess of substrate. the amino acid dehydrogenase is Substantially purified. 0022. In certain embodiments of the various processes for preparing chiral amino acid compounds disclosed herein, the 0017. In certain embodiments of the various processes for preparing chiral amino acid compounds disclosed herein, the reaction medium is at a pH of about 8.5 to about 10.5, or a pH ketoreductase is a wild type ketoreductase or an engineered of about 8.5 to about 9.5, or a pH of about 9.0. ketoreductase. In some embodiments, the ketoreductase is 0023. In certain embodiments of the various processes for from Lactobacillus, Candida, Novosphingobium, or Saccha preparing chiral amino acid compounds disclosed herein, the romyces, and in Some embodiments, the ketoreductase is reaction medium is at a temperature of about 25°C. to about from an organism selected from Lactobacillus kefir, Lactoba 45° C., or about 35° C. to about 40° C. cillus brevis, Lactobacillus minor; Candida magnoliae, Sac 5. DETAILED DESCRIPTION charomyces cerevisiae, and Novosphingobium aromati civorans. In some embodiments, the ketoreductase is an 5.1 Definitions engineered ketoreductase derived from the wild-type ketore ductase of Novosphingobium aromaticivorans, wherein the 0024. As used herein, the following terms are intended to engineered ketoreductase comprising an amino acid have the following meanings. sequence having at least 80%, 85%, 90%, 95%, 98%, 99%, or (0025 “Protein,” “polypeptide,” “oligopeptide,” and “pep more identity to a sequence selected from SEQID NO: 2, 4, tide' are used interchangeably to denote a polymer of at least 6, 8, 10, 12, 14, 16, 18, 20, 22, and 24. two amino acids covalently linked by an amide bond, regard less of length or post-translational modification (e.g., glyco 0018. In certain embodiments of the various processes for Sylation, phosphorylation, lipidation, myristilation, ubiquiti preparing chiral amino acid compounds disclosed herein, the nation, etc.). Included within this definition are D- and ketoreductase is characterized by increased thermostability, L-amino acids, and mixtures of D- and L-amino acids. increased solvent stability, and/or increased enzymatic activ 0026 “Amino acid dehydrogenase' or 'AADH are used ity relative to a reference ketoreductase. In some embodi interchangeably herein to refer to a polypeptide capable of ments, the ketoreductase used in the cofactor recycling sys carrying out the conversion of an amino acid, in the presence tem has an improved property over a reference ketoreductase of an electron acceptor, to a 2-oxoacid, NH, and reduced of increased activity in the conversion of the lower secondary acceptor. In some embodiments, amino acid dehydrogenases alcohol (e.g., isopropanol) of the recycling system to the are also capable of carrying out the reverse reaction of con corresponding lower alkylketone. In some embodiments, the Verting the 2-oxoacid, in the presence of an ammonium ion ketoreductase having the increased activity in the conversion donor and an electron donor, to an amino acid and oxidized of the lower secondary alcohol is at least 2.0 fold, 2.5 fold, 5.0 electron donor. L-amino acid dehydrogenase refers to an fold, 7.5 fold, 10-fold, or more improved relative to a refer amino acid dehydrogenase that is stereospecific or Stereose ence ketoreductase (e.g., a reference ketoreductase of SEQID lective for an L-amino acid. D-amino acid dehydrogenase NO: 2). refers to anamino acid dehydrogenase that is stereospecific or 0019. In certain embodiments of the various processes for stereoselective for a D-amino acid. Generally, the electron preparing chiral amino acid compounds disclosed herein, the acceptor? donor for the amino acid dehydrogenase is nicoti ketoreductase used in the cofactor recycling system has an namide adenine dinucleotide in oxidized/reduced form (i.e., improved property over a reference ketoreductase of NAD+/NADH) or nicotinamide adenine dinucleotide phos decreased or no activity with the compound of formula I (e.g., phate in oxidized/reduced form (i.e., NADP+/NADPH) trimethylpyruvic acid) which is a substrate for the amino acid Amino acid dehydrogenase as used herein include naturally dehydrogenase used in the process. In some embodiments of occurring (wild type) amino acid dehydrogenases as well as the process, the activity of the ketoreductase used in the non-naturally occurring polypeptides generated by human cofactor recycling system with the compound of formula I is manipulation (e.g., recombinant or engineered enzymes). less than about 5%, 2%, 1%, 0.5%, 0.1%, 0.05%, 0.01%, oran 0027 “Ketoreductase' and “KRED are used inter even smaller 96, of the activity of the amino acid dehydroge changeably herein to refer to a polypeptide that is capable of nase used in the process with the compound of formula I. In reducing a ketone to an alcohol product and/or oxidizing an US 2013/0029385 A1 Jan. 31, 2013 alcohol to a ketone product. The polypeptide typically utilizes the Substrate to the product (e.g., percent conversion of start a cofactor reduced nicotinamide adenine dinucleotide ing amount of Substrate to product in a specified time period (NADH) or reduced nicotinamide adenine dinucleotide phos using a specified amount of KRED) as compared to a refer phate (NADPH) as the reducing agent; or may use the corre ence enzyme. Exemplary methods to determine enzyme sponding NAD" or NADPH as oxidizing agent. Ketoreduc activity are provided in the Examples. Any property relating tases as used herein include naturally occurring (wild type) to enzyme activity may be affected, including the classical ketoreductases as well as non-naturally occurring polypep enzyme properties of Km, Vmax or kcat, changes of which tides generated by human manipulation (e.g., recombinant or can lead to increased enzymatic activity. Improvements in engineered enzymes). They may be used to effect one or more enzyme activity can be from about 1.5 times the enzymatic chemical transformations, including the regeneration of activity of the corresponding wild-type ketoreductase cofactors such as NAD(P)H or NAD(P)". enzyme, to as much as 2 times. 5 times, 10 times, 20 times, 25 0028 “Naturally-occurring” or “wild-type” refers to the times, 50 times, 75 times, 100 times, or more enzymatic form found in nature. For example, a naturally occurring or activity than the naturally occurring ketoreductase or another wild-type polypeptide or polynucleotide sequence is a engineered ketoreductase from which the ketoreductase sequence present in an organism that can be isolated from a polypeptides were derived. In specific embodiments, the Source in nature and which has not been intentionally modi engineered enzyme exhibits improved enzymatic activity in fied by human manipulation. the range of 1.5 to 50 times, 1.5 to 100 times greater than that 0029) “Recombinant’ or “engineered’ or “non-naturally of the parent enzyme. It is understood by the skilled artisan occurring when used with reference to, e.g., a cell, nucleic that the activity of any enzyme is diffusion limited such that acid, or polypeptide, refers to a material, or a material corre the catalytic turnover rate cannot exceed the diffusion rate of sponding to the natural or native form of the material, that has the Substrate, including any required cofactors. The theoreti been modified in a manner that would not otherwise exist in cal maximum of the diffusion limit, or kcat/Kim, is generally nature, or is identical thereto but produced or derived from about 10 to 10 (M-1s-1). Hence, any improvements in the synthetic materials and/or by manipulation using recombi enzyme activity of an enzyme will have an upper limit related nant techniques (e.g., genetic engineering). Non-limiting to the diffusion rate of the substrates acted on by the enzyme. examples include, among others, recombinant cells express Comparisons of enzyme activities are made using a defined ing genes that are not found within the native (non-recombi preparation of enzyme, a defined assay under a set condition, nant) form of the cell or express native genes that are other and one or more defined substrates, as further described in wise expressed at a different level. detail herein. Generally, when lysates are compared, the num 0030) “Derived from identifies the originating polypep bers of cells and the amount of proteinassayed are determined tide, and/or the gene encoding Such polypeptide, upon which as well as use of identical expression systems and identical the engineering was based. host cells to minimize variations in amount of enzyme pro 0031 “Substrate” refers to a substance or compound that duced by the host cells and present in the lysates. is converted or meant to be converted into another compound 0035 “Conversion” refers to the enzymatic transforma by the action of an enzyme. The term includes aromatic and tion of the substrate to the corresponding product. “Percent aliphatic compounds, and includes not only a single com conversion” refers to the percent of the substrate that is con pound, but also combinations of compounds, such as solu verted to the product within a period of time under specified tions, mixtures and other materials which contain at least one conditions. Thus, the “enzymatic activity” or “activity” of an substrate. enzyme(s) can be expressed as “percent conversion of the 0032 “Stereoselectivity” refers to the preferential forma substrate to the product. tion in a chemical or enzymatic reaction of one stereoisomer 0036 “Improved thermostability” and “improved thermal over another. Stereoselectivity can be partial, where the for stability” are used interchangeably herein to refer to a prop mation of one stereoisomer is favored over the other, or it may erty of increased resistance to inactivation when exposed to a be complete where only one stereoisomer is formed. When set temperature or set oftemperatures in defined conditions as the Stereoisomers are enantiomers, the Stereoselectivity is compared to the resistance to inactivation of a reference referred to as enantioselectivity, the fraction (typically enzyme. Activity of the enzyme pre- and post treatment are reported as a percentage) of one enantiomer in the Sum of measured under the same defined assay condition. Thermo both. It is commonly reported in the art (typically as a per stability can also be compared and expressed as the tempera centage) as the enantiomeric excess calculated therefrom ture at which half of the initial activity is retained after a according to the formula major enantiomer-minor enanti defined incubation time after an increase from one tempera omer/major enantiomer+minor enantiomer. Where the Ste ture to another, i.e., from X C. to Y C. “Residual activity” or reoisomers are diastereoisomers, the stereoselectivity is “residual enzyme activity” refers to the activity that remains referred to as diastereoselectivity, the fraction (typically following exposure to the set temperature in a defined condi reported as a percentage) of one diastereomer in the Sum with tion. others. 0033 “Stereospecific” refers to the preferential conver 0037 “Solvent stable” refers to a polypeptide that main sion in a chemical or enzymatic reaction of one stereoisomer tains similar activity (more than e.g., 60% to 80%) after over another. Stereospecificity can be partial, where the con exposure to varying concentrations (e.g., 5-99%) of Solvent version of one stereoisomer is favored over the other, or it may (isopropyl alcohol, tetrahydrofuran, 2-methyltetrahydrofu be complete where only one stereoisomer is converted. ran, acetone, toluene, butylacetate, methyl tert-butylether, 0034) “Increased enzymatic activity” refers to an etc.) for a period of time (e.g., 0.5-24 hrs) compared to the improved property of an enzyme, which can be represented untreated polypeptide. by an increase in specific activity (e.g., product produced/ 0038 “pH stable' refers to a polypeptide that maintains time/weight protein) or an increase in percent conversion of similar activity (more than e.g., 60% to 80%) after exposure US 2013/0029385 A1 Jan. 31, 2013 to high or low pH (e.g., 4.5-6 or 8 to 12) for a period of time groups include, but are not limited to, ethenyl; propenyls such (e.g., 0.5-24 hrs) compared to the untreated polypeptide. as prop-1-en-1-yl, prop-1-en-2-yl, prop-2-en-1-yl, prop-2- 0039. “Thermo- and solvent stable' refers to a polypeptide en-2-yl, cycloprop-1-en-1-yl; cycloprop-2-en-1-yl; butenyls that is both thermostable and solvent stable, particularly with Such as but-1-en-1-yl, but-1-en-2-yl, 2-methyl prop-1-en-1- respect to its biological function, e.g., enzymatic activity. y1, but-2-en-1-yl, but-2-en-2-yl, buta-1,3-dien-1-yl, buta-1,3- 0040 “Cofactor” refers a substance that is necessary or dien-2-yl, cyclobut-1-en-1-yl, cyclobut-1-en-3-yl, cyclobuta beneficial to the activity of an enzyme. In the context of an 1,3-dien-1-yl, etc.; and the like. In some embodiments, the amino acid dehydrogenase, the cofactor is generally a nico alkenyl group is (C-C)alkenyl. tinamide cofactor. 0045 “Alkynyl' by itself or as part of another substituent 0041. “Nicotinamide cofactor refers to any type of the refers to an unsaturated branched, straight-chain or cyclic oxidized and reduced forms of nicotinamide adenine dinucle alkyl having at least one carbon-carbon triple bond derived by otide (NAD+ and NADH, respectively) and the oxidized and the removal of one hydrogenatom from a single carbonatom reduced forms of nicotinamide adenine dinucleotide phos of a parent alkyne. Typical alkynyl groups include, but are not phate (NADP+ and NADPH, respectively) and derivatives limited to, ethynyl; propynyls such as prop-1-yn-1-yl, prop and analogs thereof. With regard to a nicotinamide cofactor, 2-yn-1-yl, etc.; butynyls such as but-1-yn-1-yl, but-1-yn-3-yl, the term "derivative' means any compound containing a pyri but-3-yn-1-yl, etc.; and the like. In some embodiments, the dine structural element, including nicotinamides that have alkynyl group is (C-C)alkynyl. been chemically modified by attachment to soluble or 0046) “Heteroalkyl.” Heteroalkanyl.” Heteroalkenyl.” insoluble polymeric materials. Some examples of derivatives Heteroalkynyl.” Heteroalkyldiyl and “Heteroalkyleno” by of nicotinamide cofactors are described in U.S. Pat. No. themselves or as part of another substituent refer to alkyl, 5,106,740, and Mansson and Mosbach, 1987, Methods in alkanyl, alkenyl, alkynyl, alkyldiyl and alkyleno groups, Enzymology 136:345, the disclosures of which are incorpo respectively, in which one or more of the carbon atoms are rated herein by reference. The term “analogs.” as used herein, each independently replaced with the same or different het refers to materials that undergo a formal hydride transfer in a eratoms or heteroatomic groups. Heteroatoms and/or hetero redox reaction similar to that undergone by nicotinamide atomic groups which can replace the carbon atoms include, cofactors. Examples of analogs of nicotinamide cofactors but are not limited to. —O— —S —S O— —NR' . useful in the practice of the present process include com PH-, -S(O) , S(O) , S(O)NR' , S(O) pounds described in U.S. Pat. No. 5,801,006, the disclosure of NR' , and the like, including combinations thereof, where which is incorporated herein by reference. Other suitable each R' is independently hydrogen or (C-C)alkyl. cofactors, as defined herein, can be used in the practice of the 0047. “Cycloalkyl and “Heterocycloalkyl” by them invention, as would be recognized by those skilled in the art. selves or as part of another substituent refer to cyclic versions 0042 “Cofactor regenerating system’’ and "cofactor recy of “alkyl and "heteroalkyl groups, respectively. For het cling system” are used interchangeably herein to refer to a set eroalkyl groups, a heteroatom can occupy the position that is of reactants that participate in a reaction that reduces the attached to the remainder of the molecule. Typical cycloalkyl oxidized form of the cofactor (e.g., NADP+ to NADPH). In groups include, but are not limited to, cyclopropyl, cyclobu the embodiments herein, cofactors oxidized by the amino tyls. Such as cyclobutanyl and cyclobutenyl; cyclopentyls. Such acid dehydrogenase-catalyzed reaction are regenerated in as cyclopentanyl and cyclopentenyl, cyclohexyls such as reduced form by the cofactor regenerating system. Cofactor cyclohexanyl and cyclohexenyl; and the like. Typical hetero regenerating systems comprise a stoichiometric reductant cycloalkyl groups include, but are not limited to, tetrahydro that is a source of reducing hydrogen equivalents and is furanyl (e.g., tetrahydrofuran-2-yl, tetrahydrofuran-3-yl, capable of reducing the oxidized form of the cofactor. The etc.), piperidinyl (e.g., piperidin-1-yl, piperidin-2-yl, etc.), cofactor regenerating system may further comprise a catalyst, morpholinyl (e.g., morpholin-3-yl, morpholin-4-yl, etc.), for example an enzyme catalyst, that catalyzes the reduction piperazinyl (e.g., piperazin-1-yl, piperazin-2-yl, etc.), and the of the oxidized form of the cofactor by the reductant. like. 0043 “Alkyl by itself or as part of another substituent 0048 Aryl by itself or as part of another substituent refers to a saturated branched or straight hydrocarbon chain refers to a monovalent aromatic hydrocarbon group having derived by the removal of one hydrogen atom from a single the stated number of carbonatoms (i.e., Cs-Cs means from 5 carbon atom of a parent alkane. Alkyl groups include, but are to 15 carbon atoms) derived by the removal of one hydrogen not limited to, methyl, ethyl; propyls, such as propan-1-yl, atom from a single carbon atom of a parent aromatic ring propan-2-yl (isopropyl), etc.; butyls. Such as butan-1-yl, system. Aryl groups include, but are not limited to, groups butan-2-yl (sec-butyl), 2-methyl propan-1-yl (isobutyl), derived from aceanthrylene, acenaphthylene, acephenan 2-methyl propan-2-yl (t-butyl), etc.; and the like. In some thrylene, anthracene, aZulene, benzene, chrysene, coronene, embodiments, the alkyl groups are (C-C)alkyl. “Lower fluoranthene, fluorene, hexacene, hexaphene, hexylene, as alkyl refers to a straight-chain or branched saturated ali indacene, S-indacene, indane, indene, naphthalene, octacene, phatic hydrocarbon having 1 to 6, preferably 1 to 4, carbon octaphene, octalene, ovalene, penta-2,4-diene, pentacene, atoms. Typical lower alkyl groups include methyl, ethyl, pro pentalene, pentaphene, perylene, phenalene, phenanthrene, pyl, isopropyl, butyl, t-butyl, 2-butyl, pentyl, hexyl and the picene, pleiadene, pyrene, pyranthrene, rubicene, triph like. enylene, trinaphthalene, and the like, as well as the various 0044 Alkenyl refers to by itself or as part of another hydro isomers thereof. In some embodiments, the aryl group Substituent refers to an unsaturated branched, straight chain is (Cs-Cs) aryl, with (Cs-Co) being even more preferred. In or cyclic alkyl having at least one carbon-carbon double bond Some embodiments, the aryls are cyclopentadienyl, phenyl derived by the removal of one hydrogen atom from a single and naphthyl. carbonatom of a parentalkene. The group may be in either the 0049) “Heteroaryl” by itself or as part of another substitu cis or trans conformation about the double bond(s). Alkenyl ent refers to a monovalent heteroaromatic group having the US 2013/0029385 A1 Jan. 31, 2013

stated number of ring atoms (e.g., '5-14 membered” means alcohol refers to an alcohol in which the –OH group is from 5 to 14 ring atoms) derived by the removal of one bonded to a carbon atom that is bonded to one hydrogenatom hydrogenatom from a single atom of a parent heteroaromatic and to two other carbon atoms, such as in 2-propanol (isopro ring system. Typical heteroaryl groups include, but are not panol), 2-butanol, 2-hexanol and the like. A "lower secondary limited to, groups derived from acridine, benzimidazole, ben alcohol refers to a secondary alcohol in which the alkyl Zisoxazole, benzodioxan, benzodiaxole, benzofuran, ben group is of 3 to about 6 carbon atoms. Zopyrone, benzothiadiazole, benzothiazole, benzotriazole, 0053 “Ketone” refers to a carbonyl compound of general benzoxazine, benzoxazole, benzoxazoline, carbazole, 3-car formula R' C(O)—R" in which the carbonyl carbon is boline, chromane, chromene, cinnoline, furan, imidazole, bonded to two carbon atoms. In some embodiments, R' and indazole, indole, indoline, indolizine, isobenzofuran, isoch R" are the same and in some embodiments, RandR" are each romene, isoindole, isolindoline, isoquinoline, isothiazole, independently an optionally substituted alkyl or aryl. A lower isoxazole, naphthyridine, oxadiazole, oxazole, perimidine, alkyl ketone refers to a carbonyl compound of general for phenanthridine, phenanthroline, phenazine, phthalazine, pte mula R' C(O)—R" in which RandR" is each an alkyl of C ridine, purine, pyran, pyrazine, pyrazole, pyridazine, pyri to Cs carbon atoms, where the total number of carbon atoms dine, pyrimidine, pyrrole, pyrrolizine, quinazoline, quino in the ketone is 3 to 6 carbon atoms. line, quinolizine, quinoxaline, tetrazole, thiadiazole, thiazole, 0054 “Protecting group' refers to a group of atoms that, thiophene, triazole, Xanthene, and the like, as well as the when attached to a reactive functional group in a molecule, various hydro isomers thereof. In some embodiments, the mask, reduce or prevent the reactivity of the functional group. heteroaryl group is a 5-14 membered heteroaryl. In some Typically, a protecting group may be selectively removed as embodiments, the heteroaryl group is a 5-10 membered het desired during the course of a synthesis. Examples of protect eroaryl. ing groups can be found in P. G. M. Wuts and T. W. Greene, 0050 “Substituted” when used to modify a specified “Greene's Protective Groups in Organic Synthesis Fourth group or radical, means that one or more hydrogen atoms of Edition.” John Wiley and Sons, New York, N.Y., 2007, Chap the specified group or radical are each, independently of one ter 7 (“Greene') which chapter is hereby incorporated by another, replaced with the same or different substituent(s). reference in its entirety, and Harrison et al., Compendium of Each substituent can be the same or different. Examples of Synthetic Organic Methods, Vols. 1-8, 1971-1996, John suitable substituents include, but are not limited to, alkyl, Wiley & Sons, NY. alkenyl, alkynyl, cycloalkyl, aryl, aralkyl, cycloheteroalkyl, 0055 “N-protecting group” (or nitrogen protecting group) heteroaryl, OR (e.g., hydroxyl, alkoxy (e.g., methoxy, means a substituent commonly employed to block or protect ethoxy, and propoxy), aryloxy, heteroaryloxy, aralkyloxy, a nitrogen functionality (e.g., the amine nitrogen of an amino ether, ester, carbamate, etc.), hydroxyalkyl, alkoxycarbonyl, acid) while carrying out a reaction with other functional alkoxyalkoxy, perhaloalkyl, perfluoroalkyl (e.g., CF, CF, groups on a compound. Accordingly, an "N-protected com CF), perfluoroalkoxy (e.g., OCF. OCFCF), alkoxyalkyl, pound refers to a modified form of the compound where an SR" (e.g., thiol, alkylthio, arylthio, heteroarylthio, aralky N-protecting group is blocking a nitrogen functionality on the lthio, etc.), SR, S(O)R', SOR, NR'R' (e.g., primary compound from undergoing reaction. amine (i.e., NH), secondary amine, tertiary amine, amide, 0056 "Ammonium source' or “ammonium ion donor carbamate, urea, etc.), hydrazide, halide, nitrile, nitro, Sulfide, refers to a compound or composition that forms ammonia or Sulfoxide, Sulfone, Sulfonamide, thiol, carboxy, aldehyde, ammonium ion in a reaction medium. keto, carboxylic acid, ester, amide, imine, and imide, includ 0057. “Reaction medium” refers to a solution comprising ing seleno and thio derivatives thereof, wherein each of the a mixture of two or more components (e.g., enzyme, Sub substituents can be optionally further substituted. In embodi strate, cofactor) which can undergo reaction in the solution. ments in which a functional group with an aromatic carbon For the enzymatic reactions described herein the reaction ring is substituted, such substitutions will typically number medium typically is an at least partially aqueous solution. In less than about 10 substitutions, more preferably about 1 to 5, Some embodiments, the reaction medium comprises aqueous with about 1 or 2 substitutions being preferred. and organic solvents (e.g., isopropanol), one or more phases. 0051 “Amino acid refers to a molecule having the gen 0.058 “Isolated polypeptide refers to a polypeptide which eral formula NHR' CHR COOH (wherein R is H, and is separated from other contaminants that naturally accom R” is an amino acid side chain, or RandR together with the pany it, e.g., protein, lipids, and polynucleotides. The term carbon and nitrogen to which they are bonded form a ring, embraces polypeptides which have been removed or purified e.g., proline) which is capable of forming a peptide bond with from their naturally-occurring environment or expression one or more other molecules having the same general for system (e.g., host cell or in vitro synthesis). mula. The term embraces both Land Damino acids. A “chiral 0059 "Substantially pure polypeptide' refers to a compo amino acid refers to an amino acid in which the C-carbon is sition in which the polypeptide species is the predominant an asymmetric carbon atom, which is a carbon atom bonded species present (i.e., on a molar or weight basis it is more to four different entities, such that an interchanging of any abundant than any other individual macromolecular species two groups gives rise to an enantiomer. In the context of an in the composition), and is generally a Substantially purified amino acid, a chiral amino acid of general formula NHR composition when the object species comprises at least about CHR' COOH, the R'group is an amino acid side chain other 50 percent of the macromolecular species present by mole or than H. A chiral amino acid can be an L-amino acid or a % weight. Generally, a Substantially pure ketoreductase com D-amino acid. position will comprise about 60% or more, about 70% or 0052 Alcohol refers to an alkyl group in which one or more, about 80% or more, about 90% or more, about 95% or more of the hydrogen atoms has been replaced by an —OH more, and about 98% or more of all macromolecular species group. A “lower alcohol refers to an alcohol in which the by mole or % weight present in the composition. In some alkyl group is about 1 to about 6 carbon atoms. A 'secondary embodiments, the object species is purified to essential homo US 2013/0029385 A1 Jan. 31, 2013

geneity (i.e., contaminant species cannot be detected in the 0064. As provided in the present disclosure, it has been composition by conventional detection methods) wherein the found that processes using amino acid dehydrogenases when composition consists essentially of a single macromolecular carried out in the presence of a cofactor regenerating system species. Solvent species, small molecules (<500 Daltons), comprising a ketoreductase and an alcohol can be used for the and elementalion species are not considered macromolecular efficient conversion of an oxoacid to its corresponding amino species. acid. In particular, use of a lower secondary alcohol for the 0060 5.2 Processes Using Amino Acid Dehydrogenases ketoreductase-based cofactor regenerating system can Coupled to Ketoreductase-Based Cofactor Regenerating Sys increase the conversion of oxo acid to the amino acid in the tem amino acid dehydrogenase catalyzed reaction, avoiding the 0061 The present disclosure provides a process for the production of acetaldehyde which can react with and inacti conversion of a 2-oxoacid (i.e., a keto acid) to an amino acid vate enzymes. Moreover, the ketone product formed by the in the presence of an ammonium source in a reaction medi ketoreductase catalyzed reaction, Such as acetone, is less ated by an amino acid dehydrogenase (AADH) and a cofac Volatile than acetaldehyde, thereby providing greater control tor recycling system comprising a ketoreductase (“KRED), over the reaction, particularly in larger scale processes (e.g., as generally depicted in Scheme 1. reaction medium of 50 L, 100 L, 300 L,500 L, or even greater Volume). 0065. Further, the loss of process control due to the vola Scheme 1 tility of acetaldehyde would create greater difficulty when using an amino acid dehydrogenase in the reverse reaction O AADH NH' (e.g., converting amino acid to the corresponding oxoacid) as + Ammonium Ion Donor described in greater detail herein. In particular, pushing the O (e.g., "NH3") O equilibrium of the reverse reaction would be facilitated with a R R1 is less Volatile lower alkyl ketone such as acetone. O 7 O 0.066 Although advantageously less volatile than acetal NADHANADPH NADANADP dehyde, the ketone product formed by the ketoreductase reac tion of a lower secondary alcohol (e.g., acetone) is Sufficiently O N ? OH volatile to allow its facile removal from the reaction medium ls KRED 1. thereby shifting the equilibrium of the ketoreductase medi R R" R R" ated process towards further cofactor reduction, and further conversion of oxo acid to the amino acid by the amino acid 0062. The stereoselectivity of the amino acid dehydroge dehydrogenase. Consequently, the combination of a ketore nase can be exploited to carry out the reverse reaction, i.e., ductase and lower secondary alcohol provides the advantages conversion of an amino acid to the 2-oxoacid, and permit greater reaction control (and increased safety) and enhanced chiral resolution of L- and D-amino acids. ability to drive the desired amino acid dehydrogenase reac 0063. In the conversion of the 2-oxo acid to the amino tion to completion. acid, amino acid dehydrogenases typically use a cofactor, 0067 Significant benefit can be further obtained when generally nicotinamide adenine dinucleotide (NAD+/ engineered ketoreductases that have improved enzyme prop NADH) or nicotinamide adenine dinucleotide phosphate erties, including among others, increased enzymatic activity, (NADP+/NADPH). To enhance the amino acid dehydroge increased thermostability, increased solvent stability and/or nase-mediated process, a cofactor regenerating system of increased pH stability are used in the process. Engineered formate dehydrogenase, glucose dehydrogenase, or phos ketoreductases with Such improved properties can allow use phite dehydrogenase have been used to convert the oxidized of conditions not well tolerated by the naturally occurring NAD+/NADP+ to the reduced form NADH/NADPH. See, enzymes, including conditions such as, for example, high oxo e.g., US patent publication 2009087995: US patent publica acid concentration, high alcohol concentration, high ammo tion 200901 17627; EP1925674; Johannes et al., 2005, Appl nium ion donor concentration, elevated incubation tempera Environ Microbiol. 71 (10):5728-5734; Johannes et al., 2007, tures, and increased incubation times. Use of engineered Biotechnol Bioeng.96(1):18-26; McLachan et al., 2008, Bio ketoreductases also can reduce the amount of enzyme needed technol Bioeng. 99(2):268-274). By continual replenishment in the process. of the reduced NADH or NADPH, the equilibrium of the 0068 Accordingly, the present disclosure provides a pro amino acid dehydrogenase mediated process can be shifted cess for converting a 2-oxoacid compound of formula I that towards product formation, thereby increasing the conversion is a Substrate for an amino acid dehydrogenase to a chiral of the oxoacid to the amino acid product. A whole cell-based amino acid compound of formula IIa, system for conversion of D-amino acid to L-amino acid using amino oxidase, amino acid dehydrogenase, and a cofactor regenerating system is described in U.S. Pat. No. 7,217,544. R CO2H R CO2H The patent publication describes the use of formate dehydro r" - Y genase, malate dehydrogenase or alcohol dehydrogenase O NH2 activities present in whole cells for regeneration of the cofac tor. A coupled enzyme system of a phenylalanine dehydroge I IIa nase and a cofactor regenerating system using an alcohol dehydrogenase is described in Paradisi et al., 2007, J. Bio in a reaction medium comprising NAD"/NADH or NADP"/ tech. 128:408-411. However, in Paradisi et al., ethanol was NADPH, and a cofactor regenerating system, where the employed as the Substrate for the alcohol dehydrogenase, cofactor regenerating system comprises a ketoreductase and a thereby forming acetaldehyde as the product. secondary alcohol. In particular, the secondary alcohol is a US 2013/0029385 A1 Jan. 31, 2013

lower secondary alcohol. Such as isopropanol, 2-butanol, TABLE 1-continued 3-methyl-2-butanol, 2-pentanol, 3-pentanol, or 3.3-dimethyl 2-butanol. rN1N1 NH2 0069. In some embodiments, the process comprises con tacting the 2-oxoacid compound of formula I with a reaction medium comprising an amino acid dehydrogenase, an ammo nium ion donor, NAD"/NADH or NADP/NADPH, and a al cofactor regenerating system comprising a ketoreductase and a lower secondary alcohol, under Suitable conditions where the compound of formula I is converted to the chiral amino acid compound of formula IIa and the lower secondary alco hol is converted to a ketone. 0070. In the processes herein, the 2-oxoacid compound of 1ns1 formula I is a Substrate for the amino acid dehydrogenase. Accordingly, the R group in the compound of formula I can be a Substituted or unsubstituted: (C-Co.)alkyl, —(C-C)alk enyl, —(C-C)alkynyl, heteroalkyl, —(C-C)cycloalkyl, ~K heterocycloalkyl, aryl, or heteroaryl. Since amino acid dehy drogenases are known to recognize naturally occurring amino acids, the R group can be any side chain attached to the alpha carbon of an amino acid of a naturally occurring amino acid. r These include, among others, the following side chain struc tures shown in Table 1 (where Squiggly line denotes point of connection of R group to rest of molecule):

TABLE 1. ri'Y rol -R) 0071. Other R groups recognized by amino acid dehydro genases include, among others, the following structures shown in Table 2 (where Squiggly line denotes point of con nection of R group to rest of molecule):

TABLE 2

n OH

rN-1N1 | YOH US 2013/0029385 A1 Jan. 31, 2013

TABLE 2-continued TABLE 2-continued D Dam C > CrsO r O)- C) (r. O r O)- Br ) O)-) C S US 2013/0029385 A1 Jan. 31, 2013

TABLE 2-continued Brevibacterium sp., Clostridium symbiosum, Clostridium dif ficile, Geobacillus Stearothermophilus, Natronobacterium magadii, Synechocystis sp. PCC 6803, Thermoactinomyces intermedius, Citrobacter sp., Proteus sp., and Pseudomonas Sp. s 0075 L-amino acid dehydrogenases useful in the process of the present disclosure capable of carrying out the conver sion of an L-amino acid, in the presence of an electron accep tor, to a 2-oxo acid, NH, and reduced acceptor. Suitable L-amino acid dehydrogenases have been described in e.g., or Oshima et al., International Industrial Biotechnology 9 (1989) 5-11: Ohsima et al., European Journal of Biochemis try 191 (1990) 715-720; Khan et al., Bioscience, Biotechnol ogy and Biochemistry 69 (2005) 1861-1870: Hummel et al., Applied Microbiology and Biotechnology 26 (1987) 409-416 R and Bommarius in Enzyme Catalysis in Organic Synthesis, 2nd Edition (2002), ed. Drauz and Waldmann, Wiley-VCH Weinheim. O 0076. The polynucleotide and/or amino acid sequences of various L-amino acid dehydrogenases are known in the art and are available from known public databases e.g., the Gen 0072. In some embodiments, the amino acid dehydroge Bank (located at www.ncbi.nlm.nih.gov). In some embodi nase can be an L- or D-amino acid dehydrogenase. In other ments, L-enantioselective amino acid dehydrogenases useful words, in the process for conversion of a pro-chiral 2-oxoacid with the process of the present disclosure have been described of formula I to the amino acid of formula IIa, the amino acid with respect to the type of amino acid acted upon/formed in dehydrogenase can be enantioselective for the L- or D-amino the enzyme catalyzed process. Accordingly, in Some embodi acid. Thus, by selection of the appropriate amino acid dehy ments, useful L-amino acid dehydrogenase references and drogenase, the process of the present disclosure can be used to sequence information can be obtained by entering into the produce the L- or D-amino acid in enantiomeric excess from indexed and searchable public databases any one of the fol the prochiral 2-oxoacid. The amino acid dehydrogenase can lowing enzyme classifications: L-alanine dehydrogenase (EC be a naturally occurring, i.e., wild type, amino acid dehydro 1.4.1.1) (see e.g., Ohashima et al., 1979, Eur J. Biochem. genase or an engineered amino acid dehydrogenase. The 100(1):29-30; Grimshaw et al., 1981, Biochemistry. Septem engineered amino acid dehydrogenase can be selected for ber 29, 20020):5650-5), L-aspartate dehydrogenase (EC 1.4. improved properties, such as increased enzymatic activity, 1.21), L-erythro-3,5-diaminohexanoate dehydrogenase (EC thermostability, solvent stability, pH stability, co-factor pref 1.4.1.11), L-leucine dehydrogenase (EC 1.4.1.9), erence, and/or altered Substrate specificity. L-glutamate dehydrogenase (EC 1.4.1.2), glutamate dehy 0073. In some embodiments of the process, the amino acid drogenase (NAD(P)") (EC 1.4.1.3), glutamate dehydroge dehydrogenase comprises a L-amino acid dehydrogenase and nase (NADP+) (EC 1.4.1.4), glycine dehydrogenase (EC 1.4. the chiral amino acid compound of formula IIa is IIb, 1.10), lysine dehydrogenase (EC 1.4.1.15), L-phenylalanine dehydrogenase (EC 1.4.1.20), L-serine dehydrogenase (EC 1.4.1.7), L-valine dehydrogenase (EC 1.4.1.8), L-2,4-diami IIb nopentanoate dehydrogenase, L-glutamate synthase, L-di R CO2H aminopimelate dehydrogenase (EC 1.4.1.12), L-N-methyla lanine dehydrogenase, L-lysine 6-dehydrogenase, and 1 L-tryptophan dehydrogenase, glutamate synthase (NADPH) (EC 1.4.1.13); glutamate synthase (NADHD) (EC 1.4.1.14), diaminopimelate dehydrogenase (EC 1.4.1.16); N-methyla having the indicated chirality. In this process, the chiralamino lanine dehydrogenase (EC 1.4.1.17), lysine 6-dehydrogenase acid of formula IIb is formed in enantiomeric excess. In some (EC 1.4.1.18), and tryptophan dehydrogenase (EC 1.4.1.19). embodiments, the chiral amino acid of formula IIb can be The choice of the amino acid dehydrogenase can be based on formed in at least 25%, 50%, 55%, 60%. 65%, 70%, 75%, the type of oxoacid Substrate recognized by the enzyme. 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% or more in enantiomeric excess. In some embodiments, the chiral amino 0077. In some embodiments, the L-amino acid dehydro acid of formula IIb is formed in greater than 99% enantio genases are engineered L-amino acid dehydrogenases in meric excess. which mutations have been introduced into the naturally 0074 Various L-amino acid dehydrogenase can be occurring polypeptide to generate an enzyme with altered obtained from organisms of the genus Bacillus, Clostridium, properties. Engineered L-amino acid dehydrogenases are Corynebacterium, Geobacillus, Natronobacterium, Syn described for, by way of example and not limitation, L-phe echocystis, Thermoactinomyces, Thermonicrobium, Card nylalanine dehydrogenase (see e.g., Seah et al., 1995 FEBS eria, Citrobacter; Proteus, and Pseudomonas, as well as from Lett. 370(1-2): 93–96: Busca et al., 2004, Org. Biomol. Chem. mammalian Sources (e.g., beef liver). Specific organisms 2, 2684-2691.) where the L-amino acid dehydrogenase can be obtained 0078. In some embodiments, the amino acid dehydroge include, among others, Bacillus subtilis, Bacillus sphaericus, nase comprises a D-amino acid dehydrogenase, and the chiral Bacillus Stearothermophilus, Bacillus thermoproteoliticus, amino acid of formula IIa is IIc, US 2013/0029385 A1 Jan. 31, 2013 11

rium str. LT2 gil 16420334|gb|AAL20718.116420334: IIc AAL53615 d-amino acid dehydrogenase small subunit Bru N-CO2H cella melitensis bv. 1 str. 16Migil 179845291gb|AAL53615. 1|gnlinteggen IBMEII0373.17984529; AAL73201 putative s H D-amino acid dehydrogenase Agrobacterium sp. IP I-671 gil 18478564|gb|AAL73201.1|AF335479 5 18478564: AAM38531 D-amino acid dehydrogenase subunit Xanth having the indicated chirality. In this process, the chiralamino OFiOS axonopodis pV. Citri Str. 306 acid of formula IIc is formed in enantiomeric excess. In some gi2111.0075 lgb|AAM38531.1||gnlunicamp XAC3688 embodiments, the chiral amino acid of formula IIc can be 2111.0075; AAM42918 D-amino acid dehydrogenase sub formed in at least 25%, 50%, 55%, 60%. 65%, 70%, 75%, unit Xanthomonas campestris pv. Campestris Str. ATCC 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% or more in 33.913 gi21114929Igb|AAM42918. enantiomeric excess. In some embodiments, the chiral amino 1|gnlunicamp XCC3648.21114929): AAM85736 D-amino acid of formula IIc is formed in greater than 99% enantio acid dehydrogenase subunit Yersinia pestis KIM 10 meric excess. gi21959016Igb|AAM85736.1|AE013821 221959016: 0079 D-amino acid dehydrogenases can be obtained from AAN34096 D-alanine dehydrogenase, small subunit Bru Halobacterium, Methanosarcina, Pseudomonas, Pyrobacu cella Suis 1330 lum, Salmonella, Corynebacterium, and Escherichia. Spe gi234.642961gnlitigr|BRAO924 Igb|AAN34096.1 cific species where the D-amino acid dehydrogenase can be 23464296; AAN42793 D-amino acid dehydrogenase sub obtained include, among others, Pseudomonas aeruginosa, unit Shigella flexneri 2a Str. 301 Pseudomonas fluorescens, Pyrobaculum islandicum, Salmo gi56383395Igb|AAN42793.211gnllmgcchina|SF0001178 nella typhimurium, Corynebacterium glutamicum, and 56383395); and AAN69891 D-amino acid dehydrogenase, small subunit, putative Pseudomonas putida KT2440 Escherichia coli. gi24986030gb|AAN69891.1|AE016628 0080. Like the L-amino acid dehydrogenases, references and sequences of wild-type D-amino acid dehydrogenases 4gnlitigr|PP4311 24986030). are publicly available, for example from the GenBank data I0081. In some embodiments, the D-amino acid dehydro base available at the NCBI web-site. A few exemplary wild genase is selected from D-alanine dehydrogenase (e.g., type D-amino acid dehydrogenase sequences listed by Gen gi234.64296 lgb|AAN34096.1D-alanine dehydrogenase, Bank accession include: AAC36880 D-amino acid small subunit Brucella suis 1330), D-threonine dehydroge dehydrogenase Escherichia coli gil 145703gb|AAC36880. nase (e.g., gi3845577db|BAA34184.1D-threonine dehy 1145703: AAC36881 catabolic alanine racemase Escheri drogenase Pseudomonas cruciviae), D-proline dehydroge chia coligil 145704 Igb|AAC36881. 1145704: AAC38139 aSC (e.g., gil 145283.977 Igb|ABP51559.1|D-proline D-amino acid dehydrogenase Klebsiella aerogenes dehydrogenase Pyrobaculum arsenaticum DSM 13514). gi2360965 Igb|AAC38139.1||2360965); AACT4273 I0082 In some embodiments, the D-amino acid dehydro D-amino acid dehydrogenase Escherichia coli str. K-12 sub genases are engineered D-amino acid dehydrogenases in str. MG 1655 gi1787438Igb|AAC74273.11787438: which mutations have been introduced into the naturally AAD06449 D-Amino acid dehydrogenase Helicobacter occurring polypeptide to generate an enzyme with altered pylori J99 gil 4155445 lgb|AAD06449.14155445: enzyme properties. Engineered D-amino acid dehydrogena AAF40633 D-amino acid dehydrogenase, small subunit ses are described in e.g., Vedha-Peters et al., “Creation of a Neisseria meningitidis MC58 Broad-Range and Highly Stereoselective D-Amino Acid gi7225395gnlitigr|NMB01761gb|AAF40633.1||7225395: Dehydrogenase for the One-Step Synthesis of D-Amino AAF83661 D-amino acid dehydrogenase subunit LXylella Acids' J. Am. Chem. Soc. 2006, 128, 10923-10929, or in U.S. fastidiosa 9a5cl gi91057591gb|AAF83661.1|AE003925 1 Pat. No. 7,550,277, which is hereby incorporated by refer 9105759: AAF93951 D-amino acid dehydrogenase, small ence herein. subunit Vibrio cholerae O1 biovar El Tor str. N16961 I0083. In the process mediated by the amino acid dehydro gi96552351gb|AAF93951.1||gnl ITIGRIVC07869655235); genases, the reaction medium contains an ammonium source, AAF951412,4-dienoyl-CoA reductase Vibrio cholerae O1 which provides the NH group for formation of the amino biovar El Tor str. N16961) gi9656535 Igb|AAF95141. acid of formula IIa from the 2-oxo acid of formula I. Any 1||gnl ITIGRIVC1993.9656535: AAG08469 probable oxi compound which is Suitable for this purpose can be used as doreductase Pseudomonas aeruginosa PAO1 the ammonium source. Exemplary compounds include, gi99513781gb|AAG08469.1|AE004921 among others, ammonium salts, such as ammonium halide 61gnl PseudoCAPPA50849951378: AAG08689 D-amino (e.g., ammonium chloride), ammonium formate, ammonium acid dehydrogenase, Small Subunit Pseudomonas aerugi Sulfate, ammonium phosphate, ammonium nitrate, ammo nosa PAO1 gi995.1620gb|AAG08689.1|AE004943 nium tartrate, and ammonium acetate. 5gnl|PseudoCAPPA5304,995.1620: AAG56040 D-amino I0084. While the process described herein is generally used acid dehydrogenase subunit Escherichia coli O157:H7 for the preparation of L- or D-amino acids, the amino acid EDL933 gil 125148891gb|AAG56040.1|AE005336 1 dehydrogenase can also carry out the reverse reaction, i.e., 12514889; AAK90026 D-amino acid dehydrogenase conversion of an amino acid to its corresponding 2-oxoacid. Agrobacterium tumefaciens Str. C58 When the substrate is an enantiomerically pure L- or D-amino gil 1515.9999.gb|AAK90026.11515.9999; AAK90097 acid, an appropriate amino acid dehydrogenase can be used D-amino acid dehydrogenase, Small subunit Agrobacterium for the conversion of the amino acid to the 2-oxoacid. For tumefaciens str. C58 gil 151600861gb|AAK90097.1 instance, an L-amino acid dehydrogenase is selected for con 15160086); AAL20718 D-amino acid dehydrogenase sub version of L-amino acid preparations to the corresponding unit Salmonella enterica Subsp. enterica serovar Tiphimu 2-oxoacid. US 2013/0029385 A1 Jan. 31, 2013 12

0085. In some embodiments, the stereospecificity of amino acid dehydrogenases can be exploited for the chiral IIc resolution of mixtures of L- and D-amino acids. For example, RS-CO2H. an L-amino acid dehydrogenase can be used to stereospecifi cally convert the L-amino acid in a mixture of L- and D-amino s H acids to the corresponding 2-oxoacid, thereby resulting in a composition that has the D-amino acid in enantiomeric excess. Similarly, a D-amino acid dehydrogenase can be used having the indicated chirality in enantiomeric excess. In these to stereospecifically convert the D-amino acid in a mixture of embodiments, the process can comprise contacting the com L- and D-amino acids to the corresponding 2-oxo acid, pound mixture of formula IId (which can include a racemic thereby resulting in a composition that has the L-amino acid mixture of formula IIe) with a reaction medium comprising in enantiomeric excess. When desired, the amino acid present an L-amino acid dehydrogenase, NAD"/NADH or NADP"/ in enantiomeric excess can be isolated from the product mix NADPH and a cofactor recycling system comprising a ture. ketoreductase and a lower alkyl ketone, under conditions I0086 Accordingly, in some embodiments, the present dis where the chiral amino acid compound of formula IIb in the closure provides a process for converting a compound mix compound mixture of formula IId is converted to the 2-oxo ture of formula IId, which comprises a substrate for an amino acid compound of formula I thereby resulting in an enantio acid dehydrogenase, to a composition of a 2 oxo acid com meric excess of the chiral amino acid of formula IIc, and the pound of formula I and a chiral amino acid of formula IIa: lower alkyl ketone is converted to the corresponding lower secondary alcohol. In some embodiments, the chiral amino acid of formula IIc can beformed in at least 25%, 50%, 55%, R CO2H R CO2H 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% or more in enantiomeric excess. In some r" - r embodiments, the chiral amino acid of formula IIc is formed NH2 O in greater than 99% enantiomeric excess. IId I I0089. In the process for chiral resolution of a mixture of L and D-amino acids to form the D-amino acid in enantiomeric excess, the L-amino acid dehydrogenase can be from Bacil slot lus, Clostridium, Corynebacterium, Geobacillus, Natrono NH2 bacterium, Synechocystis, Thermoactinomyces, Thermos, IIa Thermonicrobium, or Carderia. 0090. In some embodiments, the L-amino acid dehydro having a chiral carbon marked withan, where the compound genase can be selected from L-alanine dehydrogenase, L-as mixture of formula IId comprises L- and D-amino acid com partate dehydrogenase, L-erythro-3,5-diaminohexanoate pounds of formula IIb and IIc and the chiral amino acid of dehydrogenase, L-leucine dehydrogenase, L-glutamate formula IIa is an L- or D-amino acid. In these embodiments, dehydrogenase, lysine dehydrogenase, L-phenylalanine the process for chiral resolution of a mixture of L- and dehydrogenase, L-serine dehydrogenase. L-valine dehydro D-amino acids can comprise contacting the compound mix genase, L-2,4-diaminopentanoate dehydrogenase, ture of formula IId with a reaction medium comprising an L-glutamate synthase, L-diaminopimelate dehydrogenase, enantioselective amino acid dehydrogenase, NAD"/NADH L-N-methylalanine dehydrogenase, L-lysine 6-dehydroge or NADP/NADPH, and a cofactor recycling system com nase, and L-tryptophan dehydrogenase, as described above. prising a ketoreductase and a lower alkyl ketone, under con 0091. In some embodiments, the process can be used for ditions where either the L-amino acid or D-amino acid of the the chiral resolution of a mixture of L- and D-amino acids to compound mixture of formula IId is converted to a compound form the chiral amino acid of formula IIb, of formula I thereby resulting in an enantiomeric excess of the amino acid of formula IIa (which the chiral amino acid not converted by the amino acid dehydrogenase), and the lower IIb alkyl ketone is converted to a lower secondary alcohol. R CO2H 0087. In some embodiments of the process, the compound 1 mixture of IId is a racemic mixture of L- and D-amino acid compounds of formulas IIb and IIc, as represented by formula IIe: having the indicated chirality in enantiomeric excess. In these embodiments, the process can comprise contacting the com pound mixture of formula IId (which can include a racemic R CO2H. mixture of formula IIe) with a reaction medium comprising a y D-amino acid dehydrogenase, NAD"/NADH or NADP/ NADPH and a cofactor recycling system comprising a ketoreductase and a lower alkyl ketone, under conditions where the chiral amino acid of formula IIc in the compound mixture of formula IId is converted to the 2-oxoacid com IIe pound of formula I thereby resulting in an enantiomeric 0088. In some embodiments, the process can be used for excess of the chiral amino acid of formula IIb, and the lower the chiral resolution of a mixture of L- and D-amino acids to alkyl ketone is converted to the corresponding lower second form a chiral amino acid of formula IIc: ary alcohol. In some embodiments, the chiral amino acid of US 2013/0029385 A1 Jan. 31, 2013

formula IIb can be formed in at least 25%, 50%, 55%, 60%, tase of Novosphingobium aromaticivorans (e.g., 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or gil 1453224601gb|ABP64403.1145322460). 99% or more in enantiomeric excess. In some embodiments, 0097. The improved activity of engineered ketoreductases the chiral amino acid of formula IIb is formed in greater than (derived from Novosphingobium aromaticivorans ketoreduc 99% enantiomeric excess. tase of SEQ ID NO: 2) for the conversion of the secondary 0092. In the process for chiral resolution of a mixture of L alcohol, isopropanol (IPA) to its corresponding product, and D-amino acids to form the L-amino acid in enantiomeric acetone was determined relative to the same activity for the excess, the D-amino acid dehydrogenase can be from Halo reference ketoreductase of SEQID NO: 2. Relative IPA activ bacterium, Methanosarcina, Pseudomonas, Pyrobaculum, ity was determined using an assay with the following reaction Salmonella, Corynebacterium, or Escherichia. conditions: 100 ul 10x diluted engineered KRED lysate, 10% IPA (v/v), 0.5 g/L NAD", 100 mM TEA, pH 7.5. Exemplary 0093. In some embodiments of the process herein, the engineered ketoreductases exhibiting at least 2-fold increased amino acid dehydrogenase can be present in the form of activity with IPA relative to SEQID NO: 2 are listed in Table whole cells, including whole cells transformed with poly 3. The fold-improvement in IPA activity relative to SEQ ID nucleotide constructs expressing wild type or engineered NO: 2 was quantified as follows: "+ indicates at least 200% amino acid dehydrogenases. In some embodiments, the to 250% improvement: “++” indicates <250% to 500% amino acid dehydrogenase can be present in the form of cell improvement; and “+++’ indicates >500% to 1000% extracts and/or lysates thereof, and may be employed in a improvement; and “++++’ indicates >1000% to 2000% variety of different forms, including Solid (e.g., lyophilized, improvement. spray-dried, and the like) or semisolid (e.g., a crude paste). In Some embodiments, the amino acid dehydrogenase is iso lated, and can be in a Substantially purified form. In some TABLE 3 embodiments of the process, both the amino acid dehydroge SEQ ID FIOP in IPA nase and the ketoreductase of the regenerating system can be NO: activity present in the form of whole cells, including whole cells 2 control transformed with polynucleotide constructs such that the 4 -- whole cells express both the amino acid dehydrogenase and 6 ---- the ketoreductase. 8 ---- 10 ---- 0094. In the embodiments of the processes disclosure 12 ------herein, the amino acid dehydrogenase is used in combination 14 ---- with cofactor regenerating system comprising: a ketoreduc 16 -- 18 ------tase capable of reducing NAD" and/or NADP" to NADH and 2O ---- NADPH, respectively, and an alcohol (e.g., a lower secondary 24 ------alcohol) that is a substrate for the ketoreductase. In the reduc tion of the cofactor, the ketoreductase converts the alcohol to the corresponding carbonyl compound (e.g., a lower alkyl 0098. In some embodiments, the ketoreductase having the increased activity in the conversion of the lower secondary ketone). alcohol is an engineered ketoreductase comprising an amino 0.095. In the embodiments herein, the ketoreductase can be acid sequence having at least 80%, 85%, 90%. 95%, 98%, a wild type ketoreductase, or an engineered ketoreductase, in 99%, or more identity to a sequence selected from SEQ ID particular an engineered ketoreductase with an improved NO: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, and 24. In some enzyme property. As used herein, a ketoreductase enzyme embodiments, the ketoreductasehaving the increased activity that has an “improved enzyme property” refers to a ketore in the conversion of the lower secondary alcohol is an engi ductase enzyme that exhibits an improvement in any enzyme neered ketoreductase comprising an amino acid sequence property as compared to a reference ketoreductase enzyme. selected from SEQID NO: 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, For the engineered ketoreductase enzymes described herein, and 24. the comparison is generally made to the wild-type ketoreduc 0099. In some embodiments of the process, the ketoreduc tase enzyme, although in some embodiments, the reference tase used in the cofactor recycling system has an improved ketoreductase can be an improved engineered ketoreductase. property over a reference ketoreductase of decreased or no In some embodiments, the ketoreductase is an engineered activity with the 2-oxo acid compound of formula I (e.g., ketoreductase characterized by increased thermostability, trimethylpyruvic acid) which is a substrate for the amino acid increased solvent stability, increased pH stability, and/or dehydrogenase used in the process. In some embodiments of increased enzymatic activity relative to the wild type ketore the process, the activity of the ketoreductase used in the ductase. cofactor recycling system with the compound of formula I is 0096. In some embodiments of the process, the ketoreduc less than about 5%, 2%, 1%, 0.5%, 0.1%, 0.05%, 0.01%, oran tase used in the cofactor recycling system has an improved even Smaller percentage of the activity of the amino acid property over a reference ketoreductase of increased activity dehydrogenase used in the process with the compound of in the conversion of the lower secondary alcohol (e.g., iso formula I. In some embodiments of the process, the ketore propanol) of the recycling system to the corresponding lower ductase used in the cofactor recycling system has no detect alkylketone. In some embodiments, the ketoreductasehaving able activity with the t-oxoacid compound of formula I. the increased activity in the conversion of the lower second 0100. The relative IPA activity of each of the engineered ary alcohol is at least 2.0 fold, 2.5 fold, 5.0 fold, 7.5 fold, ketoreductases shown in Table 3 was measured under the 10-fold, or more improved relative to a reference ketoreduc same conditions in the presence of the compound of formula tase. In some embodiments, the ketoreductase is an engi I, trimethylpyruvic acid (10% IPA (v/v), 0.5 g/L NAD+, 1 neered ketoreductase derived from the wild-type ketoreduc mM TMP, 100 mM TEA pH 7.5, 100 ul 10xKRED diluted US 2013/0029385 A1 Jan. 31, 2013

lysate). Less than 5% increase or decrease was seen in the 0104. In some embodiments, the ketoreductase is from presence of the TMP, indicating that the engineered ketore Lactobacillus, such as, among others, Lactobacillus kefir, ductases listed in Table 3 did not use TMP as a substrate. Lactobacillus brevis, or Lactobacillus minor. Wild type 0101 Various ketoreductases that can be used in the cofac ketoreductase from Lactobacillus kefir is described in Gen tor regenerating system include ketoreductases from, by way bank accession no. AAP94029 GI:33112056. Wild type of example and not limitation, bacteria, Such as the genus ketoreductase from Lactobacillus brevis is described in Escherichia, the genus Bacillus, the genus Pseudomonas, the CAD66648 GI:284OO789. genus Serratia, the genus Brevibacterium, the genus Coryne 0105. In some embodiments, the ketoreductase is an engi bacterium, the genus Streptococcus, the genus Lactobacillus, neered ketoreductase derived from a wild type ketoreductase the genus Novosphingobium; actinomycetes Such as the of Lactobacillus. Engineered ketoreductases of Lactobacil genus Rhodococcus, the genus Streptomyces; yeasts such as lus, for example, L. kefir, L. brevis, and L. minor, with the genus Saccharomyces, the genus Kluyveromyces, the improved enzyme properties are described in US patent pub genus Thermoanerobium, the genus Schizosaccharomyces, lications US20080318295, US20090093031, the genus Sporobolomyces, the genus Zygosaccharomyces, US2009019 1605, US20090155863, US20090162909, U.S. the genus Yarrowia, the genus TrichospOron, the genus Rho Ser. No. 12/545,034, filed Aug. 20, 2009, U.S. Ser. No. dosporidium, the genus Pichia, the genus Candida; and fungi 12/545,761, filed Aug. 21, 2009, U.S. Ser. No. 12/549,154, Such as the genus Neurospora, the genus Aspergillus, the filed Aug. 27, 2009, and U.S. Ser. No. 12/549,293, filed Aug. genus Cephalosporium, the genus Trichoderma. 27, 2009, of which each of the ketoreductase polypeptides 0102. In some embodiments, the ketoreductase can be disclosed therein are hereby incorporated by reference derived from Lactobacillus kefir, Lactobacillus brevis, Lac herein. tobacillus minor, Candida Sonorensis, Candida boidini, Can dida guilliermondi, Candida magnoliae, Candida utilis, Can 0106. In some embodiments, the ketoreductase of Can dida maltosa, Candida kefir, Candida parapslosis, dida is from Candida magnoliae. Wild type ketoreductase of Geotrichum candidum, Rhodococcus erythropolis, Rhodot Candida magnoliae is described in e.g., Wada et al., Biosci. orula glutinis, Hansenula fabiani, Hansenula polymorpha, Biotechnol. Biochem. 62(2): 280-285 (1998). Engineered Hansenula saturnus, Nocardia salmonicolor; Novosphingo ketoreductases with improved enzyme properties derived bium aromaticivorans, Pichia anomala, Pichia capsulata, from Candida magnoliae ketoreductase is described in Pichia membranafaciens, Pichia methanolica, Pichia pinus, US20060195947, of which each of the ketoreductase Pichia silvicola, Pichia supitis, Sphingomonas paucinobilis, polypeptides disclosed therein is hereby incorporated by ref Sporobolomyces salmonicolor, Streptomyces coelicolor; erence herein. Thermoanerobium brockii, or Saccharomyces cerevisiae. 0107. In some embodiments, the ketoreductase of Saccha 0103. In some embodiments, the ketoreductase is a wild romyces is from Saccharomyces cerevisiae. Wild type ketore type ketoreductase listed in Table 4 or an engineered ketore ductase from Saccharomyces cerevisiae is described in ductase derived from a wild-type ketoreductase listed in Table US20080248539. Engineered ketoreductases with improved 4. enzyme properties derived from Saccharomyces cerevisiae TABLE 4 Wild type KRED from various microorganisms Genbank Acc. Microorganism No. GINo. Reference Candida magnoliae ABO36927.1 12657576 SEQID No 2 in US publ. no. 2006O195947A Saccharomyces cerevisiae NP 010159.1 6320079 SEQID NO: 110 in US publ. no. 2009O1916OSA Lactobacilius brevis 1NXQ A 30749782 SEQID NO: 2 in US publ. no. 2009O1916OSA Rhodococcus erythropolis AAN73270.1 34776951 SEQID NO: 112 in US publ. no. 2009O1916OSA Saccharomyces cerevisiae NP O11476 6321399 SEQID NO: 114 in US publ. no. 2009O1916OSA Saccharomyces cerevisiae NP O10656.1 6320576 SEQID NO: 116 in US publ. no. 2009O1916OSA Saccharomyces cerevisiae NP O14490.1 6324421 SEQID NO: 118 in US publ. no. 2009O1916OSA Lactobacilius kefir AAP94O29.1 33112056 SEQID NO. 4 in US publ. no. 2009O1916OSA Sporobotomyces Q9UUN9 30315955 SEQID No 104 in US publ. no. Saimonicolor 2009O1916OSA Streptomyces coelicolor NP 631415.1 21225636 SEQID No 102 in US publ. no. 2009O1916OSA Thermoanaerobium brockii X64841.1 1771790 SEQID No 108 in US publ. no. 2009O1916OSA Candida parapsilosis BAA24528 2815409 Julich Chiral Solutions Cat. No. 03.11 Lactobacilius brevis ABJ63353.1 116098204 Julich Chiral Solutions Cat. No. 8.1 Candida boidinii CAD66648 284.00789 Julich Chiral Solutions Cat. No. 02.10 Lactobacilius leichmannii Fluka Cat. No. 61306 US 2013/0029385 A1 Jan. 31, 2013

ketoreductase are described in US20080248539. Each of the 0113. In some embodiments of the process, the product ketoreductase polypeptides disclosed therein are incorpo formed from the ketoreductase can be removed from the rated by reference herein. reaction medium to improve conversion of the alcohol to the 0108. In some embodiments, the ketoreductase of corresponding carbonyl product, and thereby pushing the Novosphingobium is from Novosphingobium aromati equilibrium of the process to the reduction of NAD+ or civorans. A wild type ketoreductase gene from Novosphin NADP+ to NADH or NADPH. For instance, where the car gobium aromaticivorans is provided as GenBank accession bonyl product is volatile, the product can be removed by CP000677.1, and the encoded polypeptide sequence is acces sparging the reaction medium with an non-reactive gas or by sion no.gil 1453224601gb|ABP64403.1145322460. Engi lowering the vapor pressure of the reaction medium and neered ketoreductases derived from Novosphingobium aro removing the Volatile carbonyl product. In some embodi maticivorans wild type are described in U.S. provisional ments of the process, the alcohol substrate for the ketoreduc application 61/219,162, filed Jun. 22, 2009, which is hereby tase is a lower secondary alcohol, and the corresponding incorporated by reference herein. Exemplary engineered lower alkyl ketone formed from the lower secondary alcohol polynucleotides and the corresponding ketoreductase is removed from the reaction medium. In some embodiments, polypeptides derived from the Novosphingobium aromati where the alcohol is isopropanol, the product acetone can be civorans wild type and having the improved property of removed by sparging the reaction medium with a non-reactive increased activity in converting isopropanol to acetone are gas, such as nitrogen, or by applying a vacuum to the reaction presented in the sequence listing incorporated herein as SEQ medium and removing the acetone by condensation. ID NO: 3-24. In some embodiments, the engineered ketore 0114. Where appropriate for use in the processes, the ductase comprises an amino acid sequence selected from the amino acid dehydrogenases and ketoreductase enzymes group consisting of SEQID NO: 4, 6, 8, 10, 12, 14, 16, 18, 20, present in cells, such as engineered enzymes expressed in 22, and 24. host cells, can be recovered from the cells and or the culture 0109. In some embodiments, the ketoreductase used in the medium using any one or more of the well known techniques process is capable of recycling cofactor by converting isopro for protein purification, including, among others, lysozyme panol (IPA) to acetone in a reaction medium of 3 to 20% IPA treatment, Sonication, filtration, salting-out, ultra-centrifuga at a pH of about 9.0 to 10.5 with an activity at least 1.5-fold tion, and chromatography. Suitable solutions for lysing and greater than the reference ketoreductase of SEQID NO: 2. the high efficiency extraction of proteins from bacteria, Such 0110. In some embodiments, the ketoreductase can be as E. coli, are commercially available under the trade name present in the form of whole cells, including whole cells Cellytic BTM from Sigma-Aldrich of St. Louis Mo. The cell transformed with polynucleotide constructs expressing wild extracts or cell lysates may be partially purified by precipita type or engineered ketoreductases. In some embodiments, the tion (ammonium Sulfate, polyethyleneimine, heat treatment ketoreductase can present as cell extracts and/or lysates or the like, followed by a desalting procedure prior to lyo thereof, and may be employed in a variety of different forms, philization (e.g., ultrafiltration, dialysis, and the like). Any of including Solid (e.g., lyophilized, spray-dried, and the like) or the cell preparations may be stabilized by crosslinking using semisolid (e.g., a crude paste). In some embodiments of the known crosslinking agents, such as, for example, glutaralde process, the ketoreductase is isolated, and can be in Substan hyde or immobilization to a Solid phase (e.g., Eupergit C, and tially purified form. the like). 0111. In the cofactor regenerating process carried out by 0115 The reactions described herein are generally carried the ketoreductase, an alcohol is used as the Substrate reduc out in a solvent. Suitable solvents include water, organic tant for the generation of reduced cofactor NADH or Solvents (e.g., ethyl acetate, butyl acetate, 1-octanol, heptane, NADPH. As noted above, while primary alcohols substrates octane, methyl t-butyl ether (MTBE), toluene, and the like), recognized by the ketoreductase, such as ethanol, can be used, and ionic liquids (e.g., 1-ethyl 4-methylimidazolium tet preferable are secondary alcohols, particularly lower second rafluoroborate, 1-butyl-3-methylimidazolium tetrafluorobo ary alcohols. Suitable secondary alcohols include lower sec rate, 1-butyl-3-methylimidazolium hexafluorophosphate, ondary alkanols and aryl-alkyl carbinols. Examples of lower and the like). In some embodiments, aqueous solvents, secondary alcohols include isopropanol, 2-butanol, 3-me including water and aqueous co-solvent systems, can be used. thyl-2-butanol, 2-pentanol, 3-pentanol, 3.3-dimethyl-2-bu 0116 Exemplary aqueous co-solvent systems have water tanol, and the like. Suitable aryl-alkyl carbinols include and one or more organic solvent. In general, an organic Sol unsubstituted and substituted 1-arylethanols. In some vent component of an aqueous co-solvent system is selected embodiments, the secondary alcohol is isopropanol. Such that it does not significantly inactivate the enzymes (i.e., 0112 The alcohol, particularly a secondary alcohol (e.g., amino acid dehydrogenase and ketoreductase). Appropriate isopropanol), can be present at about 1% to 60% V/V, about co-solvent systems can be readily identified by measuring the 1% to 50% w/v, about 1% to 40% w/v, about 1% to 30% w/v. enzymatic activity of the specified enzyme with a defined about 1% to 20% w/v, or about 1% to 10% w/v of the reaction Substrate of interest in the candidate solvent system, utilizing medium. In some embodiments, the alcohol can be present at an enzyme activity assay, such as those described herein. about 10% to 60% w/v, about 10% to 50% w/v, about 10% to 0117 The organic solvent component of an aqueous co 40% w/v, about 10% to 30% w/v, or about 10% to 20% w/v of Solvent system may be miscible with the aqueous component, the reaction medium. In some embodiments, the alcohol can providing a single liquid phase, or may be partly miscible or be present at about 20% to 60% w/v. about 20% to 50% v/v, immiscible with the aqueous component, providing two liq about 20% to 40% w/v, or about 20% to 30% w/v of the uid phases. Generally, when an aqueous co-solvent system is reaction medium. The amount of alcohol useful in the process employed, it is selected to be biphasic, with water dispersed in can be determined based on the activities of the amino acid an organic solvent, or Vice-versa. Generally, when an aqueous dehydrogenase and the ketoreductase in the presence of a co-solvent system is utilized, it is desirable to select an defined amount of alcohol. organic solvent that can be readily separated from the aque US 2013/0029385 A1 Jan. 31, 2013

ous phase. In general, the ratio of water to organic solvent in 0.124 Suitable conditions for carrying out the amino acid the co-solvent system is typically in the range of from about dehydrogenase/ketoreductase mediated process described 90:10 to about 10:90 (v/v) organic solvent to water, and herein include a wide variety of conditions which can be between 80:20 and 20:80 (v/v) organic solvent to water. The optimized by routine experimentation that includes, but is not co-solvent system may be pre-formed prior to addition to the limited to, contacting the enzymes and Substrates at an experi reaction mixture, or it may be formed in situ in the reaction mental pH and temperature and detecting product, for vessel. example, using the methods described in the Examples pro 0118. During the course of the amino acid dehydrogenase vided herein. and ketoreductase mediated process, the pH of the reaction 0.125 Generally, the process can be carried out at a pH of medium may change. The pH of the reaction medium may be about 11 or below, usually in the range of from about 8.0 to maintained at a desired pH or within a desired pH range by the about 11. In some embodiments, the process may be carried addition of an acid or a base during the course of the reaction. out a neutral pH, i.e., about 7.0. The optimal pH of the reac Alternatively, the pH may be controlled by using an aqueous tion medium can be determined based on the pH sensitivities solvent that comprises a buffer. Suitable buffers to maintain of the amino acid dehydrogenase and ketoreductase enzymes. desired pH ranges are known in the art and include, for In some embodiments, the process is carried out at a pH of example, phosphate buffer, triethanolamine buffer, and the about 9.5 or below, usually in the range of from about 8.5 to like. Combinations of buffering and acid or base addition may about 9.5, and in some embodiments at a pH of about 8.75 to also be used. Typically, bases added to unbuffered or partially about 9.25, and in some embodiments the process is carried buffered reaction mixtures over the course of the reduction out at about pH 9. In some embodiments, the process may can are added in aqueous solutions. be carried out at a pH of about 9 to about 11, particularly at 0119. In carrying out embodiments of the process about pH 10 to about 11. described herein, either the oxidized or reduced form of the I0126. In some embodiments, the process described herein cofactor may be provided initially. As described above, the can be carried out at a temperature in the range of from about cofactor regenerating system converts oxidized cofactor to its 15° C. to about 75°C. In some embodiments, the reaction is reduced form (or vice-versa), which is then utilized in the carried out at a temperature in the range of from about 20°C. reduction (or oxidation) of the substrate. to about 55° C. In some embodiments, it is carried out at a 0120. The Solid reactants (e.g., enzyme, salts, etc.) may be temperature in the range of from about 20°C. to about 45° C. provided to the reaction in a variety of different forms, includ In Some embodiments, the process is carried out at a tempera ing powder (e.g., lyophilized, spray dried, and the like), solu ture of about 35° C. to about 40°C. The reaction may also be tion, emulsion, Suspension, and the like. The reactants can be carried out under ambient conditions. readily lyophilized or spray dried using methods and equip I0127. The process is generally allowed to proceed until ment that are known to those having ordinary skill in the art. there is no further conversion or substantial conversion of For example, the protein solution can be frozen at -80°C. in Substrate (e.g., oxoacid) to the product (e.g., amino acid) or Small aliquots, then added to a pre-chilled lyophilization until there is essentially complete, or nearly complete, con chamber, followed by the application of a vacuum. After the version of substrate to product. Conversion of substrate to removal of water from the samples, the temperature is typi product can be monitored using known methods by detecting cally raised to about 4°C. before release of the vacuum and Substrate and/or product. Suitable methods include gas chro retrieval of the lyophilized samples. matography, HPLC, and the like. Conversion yields of the 0121 The quantities of reactants used in the reduction product amino acid generated in the reaction mixture can be reaction will generally vary depending on the quantities of greater than about 50%, greater than about 60%, greater than product desired, and concomitantly, the amount of Substrate about 70%, greater than about 80%, greater than 90%, and are employed. The following guidelines can be used to determine often greater than about 97%. the amounts of amino acid dehydrogenase, ketoreductase, I0128. In some embodiments, the process described herein and cofactor. Those having ordinary skill in the art will can be used in the conversion of a compound of formula I readily understand how to vary these quantities to tailor them which is 3,3-dimethyl-2-oxobutanoic acid (also referred to to the desired level of productivity and scale of production. herein as “trimethylpyruvic acid” or “TMP) to the chiral Appropriate quantities of cofactor regenerating system may amino acid (S)-2-amino-3,3-dimethylbutanoic acid (also be readily determined by routine experimentation based on referred to herein as “L-tert-leucine’), where the amino acid the amount of amino acid dehydrogenase and ketoreductase dehydrogenase comprises a L-leucine dehydrogenase utilized. (“LeuDH), as illustrated by Scheme 2. 0122 The order of addition of reactants is not critical. The Scheme 2 reactants may be added together at the same time to a solvent (e.g., monophasic solvent, biphasic aqueous co-solvent sys Leu)H tem, and the like), or alternatively, Some of the reactants may O + Ammonium Ion NH3" be added separately, and some together at different time S points. (e.g.,OCC NH3) xy" 0123 For improved mixing efficiency when an aqueous X."O 7 Q O co-solvent system is used, the amino acid dehydrogenase and the ketoreductase may be added and mixed into the aqueous O NADH/NADPH NAD/NADP OH phase first. The organic phase may then be added and mixed in, followed by addition of the enzyme substrates. Alterna --- KRED R 1. R" tively, the enzymes may be premixed in the organic phase, prior to addition to the aqueous phase. US 2013/0029385 A1 Jan. 31, 2013

0129. The chiral amino acid L-tert-leucine is useful in the 2265 lgb|AB1721 11.1||114334729); leucine dehydrogenase synthesis of intermediates, particularly advanced pharmaceu Shewanella sp. MR-7 gill 138886.16|gnlijgil Shewmr7 tical intermediate used in the preparation of drug compounds, 1673 lgb|AB142667.1||1138.88616; leucine dehydrogenase e.g., atazanavir, boceprevir, and telaprevir. Shewanella sp. MR-4 gill 13884623gnlijgil Shewmra 0130. In some embodiments, the process for producing 1598 lgb|AB138675.1||113884623); leucine dehydrogenase L-tert-leucine can comprise contacting the Substrate 3.3-dim Chelativorans sp. BNC1 gill 10286554gnligiMeso ethyl-2-oxobutanoic acid with a reaction medium comprising 3241 lgb|ABG64613.11102865541; leucine dehydrogenase a L-leucine dehydrogenase, an ammonium ion donor, NAD/ Pseudoalteromonas atlantica T6c NADH or NADP/NADPH, and a cofactor regenerating sys gil 109701583gnligiPatl 2995 Igb|ABG41503.1| tem comprising a ketoreductase and an alcohol, under Suit 109701583; leucine dehydrogenase Rubrobacter xylano able reaction conditions to convert the 3,3-dimethyl-2- philus DSM 9941 gil 108766512gnligiRxyl oXobutanoic acid to product (S)-2-amino-3,3- 2467Igb|ABG05394.1108766512; leucine dehydrogenase dimethylbutanoic acid, and the alcohol to the corresponding Chlamydophila pneumoniae TW-183 carbonyl compound. In particular, the alcohol is a secondary gi33236793gb|AAP98880.111gnl Ibyk|CpB0951 alcohol, as described herein. 33236793; Leucine dehydrogenase Bacillus cereus 0131. In some embodiments, the L-leucine dehydroge m1293 gi2291982.93|refZP 04325000.11gnl IWGS:NZ nase used in the process can be a wild type leucine dehydro ACLS01|bcere0001. 382302291982.93: Leucine dehydro genase or an engineered leucine dehydrogenase. L-leucine genase Bacillus CeFeliS ATCC 10876 dehydrogenases can be from genus Bacillus, Clostridium, gi22.9192377 refZP 04319341.1||gnl WGS:NZ Corynebacterium, Geobacillus, Natronobacterium, Thermo ACLT01|bcere0002 403.00229192377; Leucine dehydro actinomyces, Thermos, Thermonicrobium, or Carderia. genase Bacillus cereus BGSC 6E1 gi229186408 |ref ZP 0.132. In some embodiments, the leucine dehydrogenase is 04313572.11gnl WGS:NZ ACLU01|bcere0004 39530 from Bacillus acidokaludarius, Bacillus brevis, Bacillus cal 229 186408; Leucine dehydrogenase Bacillus cereus dolyticus, Bacillus cereus, Bacillus megaterium, Bacillus 172560W gi229180445 refZP 04307788.1|gnl WGS: mesentericus, Bacillus mycoides, Bacillus natto, Bacillus NZACLV01|bcere0005 37900229180445: Leucine pumilus, Bacillus sphaericus, Bacillus Stearothermophilus, dehydrogenase Bacillus CeFeliS MM3 Bacillus subtilis, Clostridium thermoaceticum, Corynebacte gi22.9174841 refZP 04302361.1||gnl WGS:NZ rium pseudodiphtheriticum, Geobacillus Stearothermophi ACLW01|bcere0006 39250229174841); Leucine dehy lus, Natronobacterium magadii, or Thermoactinomyces drogenase Bacillus cereus AH621 gi229168909|ref ZP intermedius. 04296626.1|gnl WGS:NZACLX01|bcere0007 38620 0133. The process disclosed herein could be used with any 229168909; Leucine dehydrogenase Bacillus cereus wild-type L-leucine dehydrogenase. Sequences of such wild R309803 gi229163100 refZP 04291056.1||gnl WGS: NZACLYO1|bcere0009 38690229163100: Leucine type enzymes are publicly available, for example from the dehydrogenase Bacillus cereus ATCC 4342 GenBank database available at the NCBI web-site. gi229157746 refZP 04285821.1||gnl WGS:NZ 0134. A few exemplary wild-type leucine dehydrogenase ACLZ01|bcere0010 39270229157746; Leucine dehydro enzyme sequences listed by GenBank accession include: leu genase Bacillus cereus m1550 gi229152367|ref ZP cine dehydrogenase Geobacillus Stearothermophilus 04280559.1|gnl WGS:NZACMA01|bcere0011 39050 gil34014423db|BAC81829.1; leucine dehydrogenase 229152367; Leucine dehydrogenase Bacillus cereus Geobacillus stearothermophilus gil 143145gb|M22977.1; BDRD-ST24 gi229146739|refZP 04275105. leucine dehydrogenase Geobacillus Stearothermophilus 1||gnl IWGS:NZ ACMB01|bcere0012 38800229146739): gil34014.421|db|AB103384.11; leucine dehydrogenase Ba Leucine dehydrogenase Bacillus cereus BDRD-ST26 cillus licheniformis gil 1477946gb|AAB36205. gi229140899|refZP 042694.44.1||gnl WGS:NZ 1||bbm|385.403bbs|1771711477946); LEUCINE-DEHY ACMC01|bcere0013 39930229140899: Leucine dehy DROGENASE Bacillus cereus drogenase Bacillus CeFeliS BDRD-ST196 gil 674 1939|emb|CAB696.10.116741939; leucine dehydro gi229134979 refZP 04263785.1||gnl WGS:NZ genase Streptosporangium roseum DSM 43021 ACMD01|bcere0014 38860229134979; and Leucine gi27 19701731gnl|REF giSros 8995|refYP dehydrogenase Bacillus CeFeliS BDRD-Cer4 003344369.1||27 1970173: leucine dehydrogenase gi229129445 refZP 04258416.1||gnl WGS:NZ Streptosporangium FOSelli DSM 43021 gi270513348gnlligiSros 8995 Igb|ACZ91626.1 ACME01|bcere0015. 388.80229129445). 270513348; leucine dehydrogenase Natranaerobius ther 0.135 Leucine dehydrogenases useful with the process of mophilus JW/NM-WN-LF gil 179351985 IgnlligiNther the present disclosure include an L-leucine dehydrogenase 2701 lgb|ACB86255.1179351985; leucine dehydrogenase from Bacillus cereus Such as the enzyme disclosed in e.g., Natranaerobius thermophilus JW/NM-WN-LF U.S. Pat. No. 5,854,035, which is hereby incorporated by gil 179351644gnlligiNther 23491gb|ACB85914.1 reference herein. 179351644; leucine dehydrogenase Natranaerobius ther 0.136 Exemplary leucine dehydrogenases useful with pro mophilus JW/NM-WN-LF gil 179350985IgnlligiNther cess of the disclosure can also be obtained from Geobacillus 1681 lgb|ACB85255.1179350985; leucine dehydrogenase Stearothermophilus (e.g., gi34014423db|BAC81829.1; Shewanella amazonensis SB2B gil 143145gb|M422977.1|: gi34014.421|db|AB 103384.10. gil 119767702IgnlligiSama 2067 Igb|ABM00273.1 In one embodiment, an L-leucine dehydrogenase from G. 119767702; leucine dehydrogenase Shewanella sp. ANA Stearothermophilus useful with the present process com 3 gil 11761252.11gnlligil Shewana3 1742gb|ABK47975.1 prises an amino acid sequence of SEQ ID NO: 26. Further 117612521; leucine dehydrogenase Shewanella suitable LeuDH enzymes useful with the present process can frigidimarina NCIMB 400 gill 143347291gnlligiSfri be engineered by site-directed mutagenesis and/or directed US 2013/0029385 A1 Jan. 31, 2013 evolution methods using the polynucleotide of SEQID NO: In some embodiments, the secondary alcohol is present in at 25, which encodes the wild-type LeuDH of SEQID NO: 26. least 1.5 fold stoichiometric excess of substrate. 0137 As noted above, the ammonium ion donor in the 0.142 Generally, the process with the leucine dehydroge process mediated by the leucine dehydrogenase can be any nase can be carried out at a pH of about 11 or below, usually suitable ammonium ion donor, which provides the NH for in the range of from about 8.0 to about 11. In some embodi formation of the amino acid. Exemplary ammonium source ments, the process is carried out at a pH of about 9.0 or below, includes, among others, various ammonium salts, such as usually in the range of from about 8.5 to about 9.0. In some ammonium halide (e.g., ammonium chloride), ammonium embodiments, the process for forming L-tert-leucine with formate, ammonium Sulfate, ammonium phosphate, ammo leucine dehydrogenase, can be carried out at a pH of about 9 nium nitrate, ammonium tartrate, and ammonium acetate. In to about 11, particularly at about pH9 to 10, more particularly particular, the process for forming L-tert-leucine with leucine at about pH 9.5. In some embodiments, the process may be dehydrogenase can use ammonium chloride as the ammo carried out a neutral pH, i.e., about 7.0. It is to be understood nium donor. that the pH of the reaction medium for the formation of 0.138. In the process for formation of L-tert-leucine, the L-tert-leucine can be determined based on the activities of the ketoreductase of the cofactor regenerating system can be any leucine dehydrogenase and ketoreductase at different pHs. Suitable ketoreductase capable of forming reduced cofactor 0143. In some embodiments, the process described herein NADH and/or NADPH utilizing the oxidization of the alco can be carried out at a temperature in the range of from about hol to the corresponding carbonyl compound to drive the 15° C. to about 75°C. In some embodiments, the reaction is reaction towards reduced cofactor formation. As described carried out at a temperature in the range of from about 20°C. herein, the ketoreductase can be a wild type ketoreductase, or to about 55° C. In some embodiments, it is carried out at a an engineered ketoreductase. The engineered ketoreductase temperature in the range of from about 20°C. to about 45° C. can have an improved enzyme property, such as improve In Some embodiments, the process is carried out at a tempera ments in enzymatic activity, thermostability, Solvent stability, ture of about 35°C. to about 45°C. The reaction may also be pH stability, inhibitor resistance, or cofactor preference. carried out under ambient conditions. Ketoreductases found in or derived from any number of 0144. In view of the foregoing, a process for producing organisms, as discussed above, can be used in conjunction (S)-2-amino-3,3-dimethylbutanoic acid (also referred to with the leucine dehydrogenase. herein as 'L-tent-leucine’), can comprise: contacting 3.3- 0.139. In some embodiments, the ketoreductase is an engi dimethyl-2-oxobutanoic acid neered enzyme derived from Candida magnoliae, Lactoba (0145 (also referred to herein as “trimethylpyruvic acid”, cillus kefir, Lactobacillus brevis, Lactobacillus minor; Sac “TMP, or “trimethyl pyruvate”) with a reaction medium charomyces cerevisiae O Novosphingobium comprising a leucine dehydrogenase, an ammonium ion aromaticivorans, as discussed above. Exemplary engineered donor, NAD"/NADH or NADP/NADPH, and a cofactor ketoreductases useful with the leucine dehydrogenase in recycling system comprising a ketoreductase and a lower forming L-tert-leucine can comprise an amino acid sequence secondary alcohol, under conditions where the 3.3-dimethyl selected from the group consisting of SEQID NO: 4, 6, 8, 10. 2-oxobutanoic acid is converted to (S)-2-amino-3,3-dimeth 12, 14, 16, 18, 20, 22, and 24. ylbutanoic acid, wherein the 3.3-dimethyl-2-oxobutanoic 0140. In the leucine dehydrogenase mediated process, the acid is at about 75 g/L to 125g/L, the cofactor is at about 0.30 Substrate alcohol for the ketoreductase can comprise any g/L to 0.70 g/L, and the leucine dehydrogenase and ketore alcohol that is recognized and converted by the ketoreductase ductase are each independently at about 0.5 to about 1.0 g/L. to the corresponding carbonyl compound. The alcohol can be 0146 The process for forming L-tert-leucine with a leu a primary alcohol or a secondary alcohol. An exemplary cine dehydrogenase can be carried out with whole cells or in primary alcohol recognized by ketoreductases is ethanol. In a cell free system. In some embodiments, the leucine dehy Some embodiments, the alcohol is a suitable secondary alco drogenase can be present in whole cells while the ketoreduc hol, including lower secondary alkanols and aryl-alkyl tase is present in a cell free system. Conversely, in some carbinols, as noted above. Exemplary secondary alcohol for embodiments, the leucine dehydrogenase can be present in use in the cofactor regenerating system with the leucine dehy cell free system while the ketoreductase is present in whole drogenase includes, by way of example and not limitation, cells. In some embodiments, the whole process is carried out include isopropanol. 2-butanol, 3-methyl-2-butanol, 2-pen in a cell free system, where the leucine dehydrogenase and the tanol, 3-pentanol. 3.3-dimethyl-2-butanol, and the like, par ketoreductase is present a crude extract or in isolated form. In ticularly isopropanol. Some embodiments, the leucine dehydrogenase and ketore 0141. The alcohol, particularly a secondary alcohol, such ductase is provided in substantially purified form. as isopropanol, can be present at about 1% to 60% V/V, about 0.147. In some embodiments, the process for forming 1% to 50% w/v, about 1% to 40% w/v, about 1% to 30% w/v. L-tert-leucine of the present disclosure can be carried out about 1% to 20% w/v, or about 1% to 10% w/v of the reaction wherein the leucine dehydrogenase polypeptide and/or the medium. In some embodiments, the alcohol can be present at ketoreductase polypeptide is immobilized on a surface, for about 10% to 60% w/v, about 10% to 50% w/v, about 10% to example wherein the enzyme is linked to the surface of a 40% w/v, about 10% to 30% w/v, or about 10% to 20% w/v of Solid-phase particle (e.g., beads) or resin that is present in the the reaction medium. In some embodiments, the alcohol can Solution or otherwise is contacted by the Solution (e.g., a be present at about 20% to 60% w/v. about 20% to 50% v/v, flow-through system in a column) Methods for linking (co about 20% to 40% w/v, or about 20% to 30% w/v of the Valently or non-covalently) enzymes to Solid-phase Supports reaction medium. In some embodiments, the process with the or particles (e.g., porous or non-porous beads, resins, or Solid leucine dehydrogenase has about 5% to 20% isopropanol, Supports) such that they retain activity for use in industrial particularly 5% to 15% isopropanol. An exemplary amount of bioreactors are known in the art (see e.g., Hermanson, G. T., isopropanol is about 8% to 12% of the reaction medium V/v. Bioconjugate Techniques, Second Edition, Academic Press; US 2013/0029385 A1 Jan. 31, 2013

(2008), and “Bioconjugation Protocols: Strategies and Meth 0152 The present disclosure contemplates that the bio ods,” in Methods in Molecular Biology, C. M. Niemeyer ed., catalytic step for converting a compound of formula Ito a Humana Press (2004); Mateo et al., “Epoxy sepabeads: a chiralamino acid compound of formula IIa may be carried out novel epoxy support for stabilization of industrial enzymes using any of the processes disclosed herein wherein the bio via very intense multipoint covalent attachment. Biotechnol catalytic step comprises contacting a compound of formula I Prog. 18(3):629-34 (2002); the disclosures of which are with a reaction medium comprising an amino acid dehydro incorporated herein by reference). Useful solid supports can genase, an ammonium ion donor, NAD/NADH or NADP/ be in the form of beads, spheres, particles, granules, a gel, a NADPH, and a cofactor regenerating system comprising a membrane or a surface, can be porous or non-porous, can ketoreductase and a lower secondary alcohol under Suitable have Swelling or non-Swelling characteristics, can be com conditions where the compound of formula I is converted to posed of organic polymers, such as polystyrene, polyethyl the chiral amino acid compound of formula IIa and the lower ene, polypropylene, polyfluoroethylene, polyethyleneoxy, secondary alcohol is converted to a ketone. and polyacrylamide, as well as co-polymers and grafts 0153. The present disclosure contemplates that the chemi thereof, or can be composed of inorganic Solids, such as glass, cal step where the amino acid compound of formula IIa is silica, controlled pore glass (CPG), reverse phase silica or converted to an N-protected chiral amino acid compound is metal. Such as gold or platinum. A variety of useful solid carried out by contacting the amino acid compound of for Supports for the immobilization of enzymes are commercially mula IIa (produced in the biocatalytic step) with a N-protect available (e.g., SEPABEADS resins; available from Sigma ing group reagent under conditions where the N-protecting Aldrich, USA). group reacts with the amine group nitrogen of the compound 0148. In some embodiments of the process using immo of formula IIa. bilized enzymes, the immobilized Leul)H and KRED 0154 Examples of such N-protecting groups useful in the polypeptides can be on different particles, which can be processes of the disclosure include the formyl group, the trityl mixed in the reaction chamber, or in some embodiments, both group, the methoxytrityl group, the tosyl group, the phthal enzymes can be immobilized on the same particles. In some imido group, the acetyl group, the trichloroacetyl group, the embodiments, the methods using immobilized polypeptides chloroacetyl, bromoacetyl, and iodoacetyl groups, benzy can be carried out wherein the method further comprises a loxycarbonyl (Cbz), methoxycarbonyl (MOC), 9-fluorenyl step of isolating or separating the immobilized enzymes from methoxycarbonyl (FMOC), 2-trimethylsilylethoxycarbonyl the reaction Solution containing the product so that the (Teoc), 1-methyl-1-(4-biphenylyl)ethoxycarbonyl (Bpoc), enzymes can be recycled. In continuous flow-through t-butoxycarbonyl (BOC), allyloxycarbonyl (Alloc), trihalo embodiments, the particles comprising the immobilized acetyl, benzyl, benzoyl, and nitrophenylacetyl and the like. Leu)H and/or KRED are maintained in a reaction chamber Further examples of protecting groups useful with the or column and the reacting Solution flows through at a rate embodiments of the present disclosure can be found in P. G. (and under suitable conditions) to allow for complete conver M. Wuts and T. W. Greene, “Greene's Protective Groups in sion of the substrate the reaction solution to product where Organic Synthesis—Fourth Edition.” John Wiley and Sons, upon it exits the reaction chamber or column. New York, N.Y., 2007, Chapter 7 (“Greene'). 0.155. In one embodiment, the compound of formula I is 0149. As noted above, to facilitate the reaction toward trimethylpyruvic acid (TMP) and the chiral amino acid com formation of product, the process of can further comprise pound of formula IIa is L-tert-leucine. According to the pro removing from the reaction medium the carbonyl product cess of the present disclosure an N-protected L-tert-leucine formed from the alcohol used as the substrate to the ketore (e.g., an L-tert-leucine with its amine group nitrogen pro ductase. In the exemplary embodiments using isopropanol as tected) can be prepared with e.g., a t-butoxycarbonyl (BOC), the alcohol, the corresponding product acetone can be benzyloxycarbonyl (Cbz), 9-fluorenylmethoxycarbonyl removed by sparging the reaction medium with an inert gas, (FMOC), or methoxycarbonyl (MOC) protecting group. Pro Such as nitrogen gas, or by lowering the pressure of the cesses for preparing the BOC and MOC protected L-tert reaction medium and trapping the Volatile acetone in a con Leucine are described in Examples 1-4. denser. 0156 Various features and embodiments of the disclosure 0150. As will be apparent to the skilled artisan, the process are illustrated in the following representative examples, for forming L-tert-leucine can be varied with respect to which are intended to be illustrative, and not limiting. amounts of L-leucine dehydrogenase, 3.3-dimethyl-2-ox obutanoic acid, cofactor, ketoreductase, and alcohol for the 6. EXAMPLES efficient conversion of the substrate to the product (S)-2- amino-3,3-dimethylbutanoic acid. In some embodiments, at a Example 1 substrate loading of at about 75 g/L to 125 g/L, the process is capable of converting at least 80%, at least 85%, at least 90%, Biocatalytic Process for Preparation of at least 91%, at least 92%, at least 93%, at least 94%, at least Enantiomerically Pure L-Tert-Leucine from 95%, at least 96%, at least 97%, at least 98%, at least 99% or Trimethylpyruvic Acid (TMP) about 100% of the substrate to the corresponding product. 0157. This Example illustrates a biocatalytic process for 0151. In some embodiments, the present disclosure pro production of enantiomerically pure L-tert-leucine from tri vides a process for preparing an N-protected chiral amino methylpyruvic acid at a 20g scale using leucine dehydroge acid compound, where the process comprises: (i) a biocata nase (LeuDH) in the presence of a ketoreductase (KRED) lytic step for converting a compound of formula Ito a chiral recycling system. amino acid compound of formula IIa, and (ii) a chemical step 0158 Materials where the amino acid compound of formula IIa is converted to 0159. The materials used and their sources are provided in an N-protected chiral amino acid compound. Table 5. Specific vendors and grades are given for commer US 2013/0029385 A1 Jan. 31, 2013 20 cially available inputs, although it is expected that the Source volume to 200 mL. The leucine dehydrogenase (LeuDH) of of such reagents would not have an impact upon the reaction. SEQID NO: 26 (150 mg) and the engineered ketoreductase (KRED) of SEQ ID NO: 18 (150 mg) biocatalysts were TABLE 5 charged to the stirred mixture as powders. The reaction mix ture was heated to 40°C. (internal temperature) stirring at 600 Materials List rpm. The final temperature was reached within 30 min. The Material Amount reaction course was followed by taking samples periodically out of the reaction mixture and analyzing as described below. Trimethylpyruvic acid 20 g For the purpose of monitoring the reaction, t-O was the time (60% aqueous solution) (33.4 mL) Water 118.6 mL at which the KRED was added into the reaction mixture. After Ammonium hydroxide 19 mL in-process analyses indicated maximum conversion, the reac (28-30% aqueous solution) tion was cooled to room temperature (in this Example, the Ammonium chloride 9.0 g Isopropanol (min 99.7%) 20 mL. reaction was stirred for a total time of 24 h). As shown by the B-NAD" 100 mg reaction profile based on the in-process analyses (described Leucine dehydrogenase 150 mg further below) provided in Table 7 below, the biocatalytic (LeuDH) of SEQID NO: 26 reaction achieved nearly 88% conversion of TMP substrate to Ketoreductase (KRED) of 150 mg L-tert-leucine product in only 7.5 hours, and complete (i.e., SEQ ID NO: 18 99.9%) conversion of substrate to L-tert-leucine product in 24 hours. 0160 Summary of Reaction Conditions 0161. A summary of the reaction conditions used in the TABLE 7 biocatalytic process are shown below in Table 6. Reaction profile TABLE 6 Time (h) % Conversion Reactant parameters and conditions for biocatalytic process O O 1.5 53.7 Trimethylpyruvic acid (TMP) 100 g/L 3 64.O Ammonium chloride 1.1 equiv 4.5 75.6 pH 9.0 6 83.7 IPA 10% 7.5 87.9 B-NAD 0.50 g/L 22 99.5 LeuldH of SEQID NO: 26 0.75 g/L 24 99.9 KRED of SEQID NO: 18 0.75 g/L Temperature 40° C. 0164. White turbidity was observed as reaction progressed and the pH of the reaction mixture after complete conversion (0162 Reaction Protocol was 8.83 at room temperature. A sample (10 mL) of the 0163 A 250 mL 2-neck round bottom flask was equipped reaction mixture was taken and heated at 40°C. for an addi with pH probe and a magnetic stirrer. The pH probe also tional 24 h. No increase in impurities was observed by GC served as an internal thermometer. The flask was charged analysis. sequentially with 100 mL water and 20.0 g (33.4 mL) of trimethylpyruvic acid. The resulting aqueous Solution was pH (0165 L-Tert Leucine Product Isolation and Work-Up 1.1 (measured at 23°C.) and was stirred for 10 minutes at 400 0166 The reaction mixture was adjusted to pH 4.5 using rpm at room temperature to obtain homogeneity. The Solution 6.0 M HCl solution under controlled exothermic conditions. was cooled to 15° C. using ice water bath Ammonium Celite (3.0 g) was added at room temperature and stirred to hydroxide was added in portions to neutralize the acid and obtain homogeneity. The mixture was filtered through a sin attain a pH between 7 and 7.5, maintaining the temperature tered glass funnel and the filter cake washed with 10 mL of below 20° C. during addition. Approximately 13 mL of water. The combined aqueous solution was concentrated to ammonium hydroxide (28-30% aqueous solution) adjusted about one-third the original volume. Precipitation of a white the pH to 7.34 at 21°C. Ammonium chloride (1.1 equiv, 9.0 Solid was observed during concentration. The Suspension was g) was added and stirred until a clear Solution was obtained. cooled in an ice bath for a few hours and the solid was filtered, After addition of ammonium chloride, reaction mixture was washed with 20 mL of a 1:1 mixture of isopropanol: water allowed to reach room temperature and the pH at room tem followed by 20 mL of acetonitrile. A repeat washing with 1:1 perature approximately 7.0. The reaction mixture was read isopropanol: water (50 mL) and acetonitrile (50 mL) afforded justed to pH 9.0 by adding ammonium hydroxide in portions, 3.0 g (approx. yield=50%) of white solid. The solid was maintaining the temperature below 25° C. during addition. identified as the desired product, L-tert-leucine, with about Approximately 6 mL of ammonium hydroxide (28-30% 90% purity when compared with a standard sample of L-tert aqueous solution) adjusted the pH to 9.03 at 22°C. The pH leucine (peak area by HPLC ELSD). was found to have a significant impact on the rate of reaction (0167 Analytical Methods Reactions did not proceed to complete conversion in 24 h 0168 Analytical methods used in the biocatalytic process when NaOH was used to adjust pH to pH 9.5 or above. 20 mL and described below include: Method 1, a GC analysis isopropanol (IPA) was added to the reaction mixture. The pH method used to monitor % conversion of trimethylpyruvic dropped to 8.95 upon addition of IPA, but no further re acid substrate to L-tert-leucine product; Method 2, an HPLC adjustment to pH 9.0 was carried out. NAD (100 mg) was UV-ELSD method to monitor trimethylpyruvic acid substrate charged to the stirred mixture as a powder and 18.6 mL of to and L-tert-leucine product in the reaction; and Method 3. water as added to the reaction mixture to adjust the final an HPLC method to determine the enantiomeric excess of US 2013/0029385 A1 Jan. 31, 2013

L-tert-leucine formed during the course of the reaction. (0176 The % conversion is calculated as follows: Results obtained by the Method 1 were verified using Method 2 (0169 Method 1: GC Analysis Method for Monitoring % % Conversion = Conversion (Area of Peak 1) 100 (0170 Sample Preparation: (Area of Peak 1) + (Area of Peak 2x Response factor) X 0171 In a 1.5 mL glass vial, samples from the reaction mixture (5uI) were diluted in 1.0 mL of acetonitrile contain ing 10 uL of pyridine. Methyl chloroformate (10 uI) was where the Response Factor is calculated as follows: added and the mixture was agitated for a few seconds. The exothermic reaction accompanied by release of CO was left standing until the precipitate settled. A sample (300 uL) from Response Factor = the Supernatant of the derivatized reaction sample was taken and analyzed by gas chromatography (GC). A Summary of the (Area of Peak 1 at time = x) analytical parameters and conditions are provided in Table 8 (Area of Peak 2 at time = 0) - (Area of Peak 2 at time = x) below. (0177 Method 3: HPLC Analysis of Enantiomeric Purity TABLE 8 (0178 HPLC Sample Preparation: Instrument Agilent 6890N series 0179 10 uL from the reaction mixture was taken and Column: HP-5 (30.0 m x 320 m x 0.25 m added 990 u, of mobile phase in a 2 mL HPLC glass vial. nominal) Injection is neat into the HPLC equilibrated and set up Temperature profile: 90°C. (hold for 2.8 min) then ramp to 210°C. according to the analytical parameters and conditions are (a) 20° C./min Pressure: 12 psi (constant) provided in Table 10 below. Inlet: 250° C. Split ratio 20:1 TABLE 10 Detector FID, 250° C. Run time 8.8 min Instrument Agilent HPLC 1200 series Retention times Trimethylpyruvic acid methyl ester (Peak Column Phenomenex Chirex 4.6 x 150 mm (5 m) 1): 2.50 min Mobile Phase 85% 3 mM CuSO 15% MeOH Derivatized L-tert-leucine (Peak 2): 6.6 min Flow Rate 1.00 mL/min Detection Wavelength 254.0 nm. Detector Temperature 25°C. 0172. The % conversion based on consumption of trim Injection Volume 10 IL ethylpyruvic acid was calculated as follows: Run time 20.0 min Retention Times: Peak 1 (L-tert-leucine): 11.319 min Peak 2 (D-tert-leucine): 19.560 (Area of Peak 1), 0 - (Area of Peak 1), % Conversion = (Area of Peak 1), o x 100 Example 2 (0173 Method 2: HPLC-UV-ELSD Analysis Method for 800 gram-scale Biocatalytic Process for Preparation Monitoring % Conversion of Enantiomerically Pure L-Tert-Leucine from (0174 HPLC Sample Preparation: Trimethylpyruvic Acid (TMP) 0175 20LL of the reaction mixture was taken and added to 980 uL of mobile phase A. Injection is neat into the HPLC 0180. This Example illustrates a scaled up version of the equilibrated and set up according to the analytical parameters biocatalytic process for production of enantiomerically pure L-tert-leucine from trimethylpyruvic acid as described in and conditions are provided in Table 9 below. Example 1 to near-kilogram Scale. Generally, the reaction protocol followed the same general scheme described in TABLE 9 Example 1, only at a larger Scale. Instrument Agilent HPLC 1200 series 0181. The materials used and their sources are provided in Column Mightysil RP18 GP 250 x 4.6 mm, 5 Im Table 11. (1x Aqua R18 guard column before & after analytical column). Mobile Phase A: 76% 20 mM NHOAc, pH 4.87 + 0.3 mM TABLE 11 dodecyltriethylammonium phosphate Regis Cat No. 404021 Material Quantity B: 24% MeCN Flow rate 1.0 mL/min Trimethyl pyruvate 800 g Run Time 15 min (TMP) ELSD detector 35o C. Ammonium formate 128 g temperature Ammonia 7SO L Column temperature 29° C. (25% w/w) Injection volume 10 IL Isopropanol (IPA) 800 mL. Gain 3 NAD 4 g UV Wavelength 230 mm Leu)H 12 g Retention Times Peak 1 (L-tert-leucine): 2.725 min (SEQID NO: 26) Peak 2 (TMP): 13.587 min KRED 12 g (SEQID NO: 18) US 2013/0029385 A1 Jan. 31, 2013 22

TABLE 11-continued temperature and a calibrated pH probe and meter used at certain stages. For isolation the vessel was fitted with a short Material Quantity path distillation head. Deionized water 5.5 L 0190. The reaction vessel was charged with 10.0 g crude (DIW) L-tert-Leucine (e.g., made as in Example 2) and 70 mL water, Acetone 11 L and the stirrer set at 150 rpm. 21.4 g of NaOH added to the Celite 545 250 g addition funnel and fed dropwise into the reaction over 5 min (temperature remains <35 C) resulting in a straw-colored solution. The addition funnel was rinsed with 10 mL water 0182 Reaction Protocol and then charged with 14.4 g MOC-C1, 14.4 g. The MOC-Cl 0183. A 10 L two neck round bottom flask in a water bath was fed into the reaction dropwise over 30-40 min (tempera was charged 800 g TMP substrate in 5 L water. Overhead stirrer (equipped with flat two-blade paddle; 8 cm diameter) ture does not exceed 50° C.). HPLC was used to monitor the was started and water bath heated to 40.0°C. The ammonia reaction for completion of conversion. was added in portions until all TMP was dissolved (approx. (0191 Workup Procedure temperature increase of 4°C.). The flask was then charged 0.192 Upon complete conversion a pH probe was inserted ammonium formate and IPA. The pH was adjusted with ~170 and the flaskallowed to cool <25 C. Approximately 12 mL of mL ammonia to pH 9.0+0.1 at 40.0°C.:-0.5° C. The solution 12 NHCl was fed drop wise using the addition funnel until the was charged with 12g of KRED of SEQID NO: 18, 12g of pH was 2 while keeping the temperature <30°C. The reaction LeulDH of SEQID NO: 26, and 4 g of NAD in 500 mL water. was charged with 60 mL ethyl acetatate which was mixed for The reaction volume was adjusted to 8 L with water as nec 2 min and then layers allowed to separate (emulsions were essary. After 18 h the reaction was complete (as determined filtered through an “M” sintered frit). The upper organic layer by Method 1 analysis of Example 1). was removed and stored. The pH was re-adjusted to ~pH 3 and the extraction was repeated two more time with 40 mL 0184 Workup Procedure then 20 mL of ethyl acetate. The extracted upper organic 0185. Charged 250 g Celite 545 and stir for about 15 min layers were combined and charged the 10 mL saturated NaCl then filtered off Celite with a 5 LP4 fritted funnel Distilled off with mixing for 2 min followed by layer separation. The ammonia, acetone, IPA and water at 50° C. under reduced organic layers (~130 mL) were charged to a reaction vessel pressure up to 60 mbar, until a vessel volume of -2.5 L is fitted with a distillation kit. Heat was applied to distill solvent achieved. Charged residue in a 15 L bucket with 10 L cold (with 200 rpm stirring) until a vessel volume of 80 mL is acetone (4 vols) and mixed for 5 min. Closed bucket was achieved. Continue distillation with heptane fed dropwise to stored overnight at 4° C. Precipitate was filtered off and maintain -80 mL vessel volume. Continue distillation and washed two times with 500 mL cold acetone. Filter cake was feed until the pot temperature is -98.2 C. Cool to 80° C. and dried at 50° C. under reduced pressure yielding 624 g L-tert feed iPrCAc, 8 mL. Cool to 65° C. and charge product seed Leucine as a beige powder having >95% e.e. The crude beige crystals, 0.1 g. Slowly cool to ambient and then to 5 C. Collect powder from this workup procedure could be immediately product by filtration on an “M” sintered glass funnel; wash used in the reaction with the N-protecting groups MOC or through with recycled liquors until all solid is collected. Wash BOC as described in Example 3, to produce the N-protected solid with heptane, 15 mL, and suction dry. Vacuum dry to version of L-tert-leucine. constant weight at 40°C. Example 3 (0193 Analytical Methods 0194 HPLC was utilized for in-process analysis of % conversion and final product quality assay. H-NMR was ulti Preparation of MOC-Protected L-Tert-Leucine lized only for final product quality assay for general purity 0186 This Example illustrates a process for production of and as a check for residual solvents. a MOC-protected L-tert-leucine compound made using (0195 Sample Preparation: L-tert-leucine prepared biocatalytically using Leul)H and an 0196. During the reaction, with mixer on, a 200 uLaliquot engineered KRED recycling system as described in is withdrawn via pipette and diluted into 3000 uL of acetoni Examples 1 or 2. trile/water (50:50). This sample is shaken by hand to ensure 0187 Materials used in the process are shown in Table 12. homogeneity. The sample is further diluted 1:2 prior to injec tion. TABLE 12 0197) The analytical instrumentation and relevant param eters are shown in Table 13. Material Quantity L-tert-Leucine 10.0 g TABLE 13 Deionized water (DIW) 80 mL. Methyl chloroformate (MOC-CI) 14.4 g Instrument Agilent 1100 NaOH, 50% wt wt aqueous 21.4 g Column TOSOH Bioscience, #19533, TSKGel Amide30 Hydrochloric acid, 12N 12 mL (4.6 mm x 10 cm, 5um) Isopropyl acetate (iProAc) 8.0 mL. Column temp 30° C. Heptane (Hept) 125 mL. Detection 205 nm. Ethyl acetate (EtOAc) 120 mL. Flow rate 1.2 mL/min NaCl, saturated aqueous 10 mL. Mobile phase C = water; D = acetonitrile Time (min) 9.6 C 96 D 0188 Equipment and Reaction Protocol Timetable: O O.2 99.8 (0189 A 300 mL round-bottomed flask fitted with a 2.OO 1 99.0 mechanical stirrer, thermocouple, and graduated addition 1O.OO 60.9 40 funnel was used for the reaction, extractive workup. Tempera 12.00 O.2 99.8 ture was controlled with a waterfice bath for below ambient US 2013/0029385 A1 Jan. 31, 2013

TABLE 13-continued suspension. Charged TEA, 7.71 g, over 5 min so that T remains <25°C., which resulted in a straw-colored solution. Injection volume 10 IL Sample Concentration ~5 mg/mL Charged reaction vessel with NaOH, 4.57 g, over 2 min, Retention Times tert-leucine: 7.8 min resulting in an opaque solution. Prepared a solution of MOC-tert-leucine: 1.7 min BOCO, 20.8 g., in THF, 30 mL, and fed dropwise to reaction vessel over 30-40 min, without allowing temperature to exceed 32°C. The reaction was monitored for complete con version by HPLC as described in Example 3, except that Example 4 retention time for BOC-tent-leucine was 1.5 min. Preparation of BOC-Protected L-Tert-Leucine (0206 Workup Procedure 0207 Checked pH and adjusted with 50% NaOH to pH>9. 0198 This Example illustrates a process for production of 2. Charged vessel with heptanes (30 mL) with mixing. Upper a BOC-protected L-tert-leucine compound made using organic layer was removed and a pH probe and flask cooled to L-tert-leucine prepared biocatalytically using a ketoreduc <25° C. before charging with -32 mL 3 NHCl dropwise (pH tase (KRED) recycling system as described in Examples 1 or -3.5) and keeping temperature <30° C. Remove cooling and 2 charged with ethyl acetate (50 mL) followed by mixing for 2 0199. Materials used in the process are shown in Table 14. min and removal of upper organic layer. pH was readjusted to -3.5, if necessary, and the vessel charged with ethyl acetate TABLE 1.4 (70 mL) and heptane (30 mL). Upper organic layer removed and stored. Charged again with ethyl acetate (25 mL) and Material Quantity heptane (25 mL). Upper organic layer removed and stored. L-tert-Leucine 10 g All organic phases were combined and filtered through an Deionized water (DIW) 70 mL. “M” sintered frit if an emulsion were present. Triethylamine (TEA) 7.71 g di-tert-butyl dicarbonate (BOCO) 20.8 g. 0208. The combined organic phases were charged with NaOH, 50% wt wt aqueous 9.15 g sat’d NaCl, 10 mL, mixed for 2 min and allowed to separate. Tetrahydrofuran (THF) 30 mL. Lower layer was removed (~235 mL) and charged to a crys Heptane 285 mL. tallizer fitted with a distillation kit. Heat was applied to distill Ethyl acetate 145 mL. solvent until a vessel volume of 100 mL is achieved. Continue NaCl, saturated aqueous 10 mL. distillation and begin feeding heptanes dropwise so as to maintain ~100 mL vessel volume. Continue distillation and 0200 Equipment and Reaction Protocol feed until the pottemperature 98°C. Cool to 5°C. and collect 0201 Reaction Vessel: product by filtration on an “M” sintered glass funnel Wash 0202. A 300 mL round-bottomed flask fitted with a through with recycled liquors until all solid is collected. Wash mechanical stirrer, thermocouple, and graduated addition solid with pre-cooled (5° C.) heptane, 15 mL. Vacuum dry to funnel was used for the reaction and extractive workup. Tem constant weight at 40°C. perature was controlled with a waterfice bath for below ambi ent temperature. A calibrated pH probe and meter was used at 0209 All publications, patents, patent applications and certain stages. other documents cited in this application are hereby incorpo 0203 Isolation Vessel: rated by reference in their entireties for all purposes to the 0204. A 500 mL round-bottomed flask fitted with a same extent as if each individual publication, patent, patent mechanical stirrer, thermocouple with temperature controller application or other document were individually indicated to and heating mantle, graduated addition funnel, and distilla be incorporated by reference for all purposes. tion kit was used for the solvent exchange and crystallization. 0210 While various specific embodiments have been 0205 Reaction vessel was charged 10.0 g crude L-tert illustrated and described, it will be appreciated that various Leucine product (as made in Example 2), and 70 mL to the changes can be made without departing from the spirit and reaction vessel and set stirrer at 150 rpm. Solution was a Scope of the invention(s).

SEQUENCE LISTING

<16 Os NUMBER OF SEO ID NOS: 26

<21 Os SEQ ID NO 1 &211s LENGTH: 792 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Variant of ketoreductase from Novosphingobium aromaticivorans

<4 OOs SEQUENCE: 1

atgcc.gcttgaaatgacgat tictotcaac aatgtggtcg ccgt.cgtcac cqgcgcggcg 60

ggaggcatcg gcc.gc.gaact ggtcaaggcg atgaaggcc.g. ccaacgc.cat cqt catcgc.c 12O

accga catgg cqc.cct cq9C catgtcgaa gg.cgcggacc attatct coa gCacgacgtg 18O US 2013/0029385 A1 Jan. 31, 2013 24

- Continued acgagcgagg cc.ggctggala gC9gtC9C9 gC9ctggCCC aggaaaagta C9ggcgc.gtC 24 O gatgcgctgg to acaacgc gggcatctic atcgt cacga agttcgaaga cactic.cgctg 3OO tccgatttico accogcdtgaa cacggtcaac gtcgatt coat catcatcgg tacgcaggit c 360 Ctgctg.ccgc tigctcaagga aggcggcaag gcgc.gc.gcag ggggcgc.ctic ggtggtcaac 42O ttct coagcg tcgggggtct gcgcggcgcg gcgttcaatig cigcc tattg Caccagcaa.g 48O gcggcggtga agatgctic to gaagtgcctic ggcgcggaat tccggcgct cqgct acaac 54 O atcc.gcgt.ca act cogtgca t cc.gggcggc atcgataccc catgct cqg Ctcgatcatg 6OO gacaagtacg tcgaacticgg cqctg.ccc.cc ticgc.gcgagg togcc caggc cgcgatggaa 660 atgcgccacc catcggit cq catgggtc.gc cct gcc.gaaa tiggcggcgg C9tggtctat 72 O Ctctgctc.cg acgcagcaag Ctt.cgtcacc tic acggaat tctgatgga C9gcggctt C 78O agcc aggt ct ga 792

<210s, SEQ ID NO 2 &211s LENGTH: 263 212. TYPE: PRT <213> ORGANISM: Novosphingobium aromaticivorans

<4 OOs, SEQUENCE: 2

Met Pro Lieu. Glu Met Thir Ile Ala Lieu. Asn. Asn. Wal Wall Ala Wal Wall 1. 5 1O 15 Thr Gly Ala Ala Gly Gly Ile Gly Arg Glu Lieu Val Lys Ala Met Lys 2O 25 3O Ala Ala Asn Ala Ile Val Ile Ala Thr Asp Met Ala Pro Ser Ala Asp 35 4 O 45 Val Glu Gly Ala Asp His Tyr Lieu Gln His Asp Val Thir Ser Glu Ala SO 55 6 O Gly Trp Lys Ala Val Ala Ala Lieu Ala Glin Glu Lys Tyr Gly Arg Val 65 70 7s 8O Asp Ala Lieu Val His Asn Ala Gly Ile Ser Ile Val Thr Llys Phe Glu 85 90 95 Asp Thr Pro Lieu. Ser Asp Phe His Arg Val Asn Thr Val Asn Val Asp 1OO 105 11 O Ser Ile Ile Ile Gly Thr Glin Val Lieu. Lieu Pro Lieu Lleu Lys Glu Gly 115 12 O 125 Gly Lys Ala Arg Ala Gly Gly Ala Ser Val Val Asn. Phe Ser Ser Val 13 O 135 14 O Gly Gly Lieu. Arg Gly Ala Ala Phe Asn Ala Ala Tyr Cys Thir Ser Lys 145 150 155 160 Ala Ala Wall Lys Met Lieu. Ser Lys Cys Lieu. Gly Ala Glu Phe Ala Ala 1.65 17O 17s Lieu. Gly Tyr Asn. Ile Arg Val Asn. Ser Val His Pro Gly Gly Ile Asp 18O 185 19 O Thr Pro Met Leu Gly Ser Ile Met Asp Llys Tyr Val Glu Lieu. Gly Ala 195 2OO 2O5 Ala Pro Ser Arg Glu Val Ala Glin Ala Ala Met Glu Met Arg His Pro 21 O 215 22O Ile Gly Arg Met Gly Arg Pro Ala Glu Met Gly Gly Gly Val Val Tyr 225 23 O 235 24 O

Lieu. Cys Ser Asp Ala Ala Ser Phe Val Thr Cys Thr Glu Phe Val Met

US 2013/0029385 A1 Jan. 31, 2013 27

- Continued <4 OOs, SEQUENCE: 6

Met Pro Lieu. Glu Met Thir Ile Ala Lieu. Asn. Asn. Wal Wall Ala Wal Wall 1. 5 1O 15 Thr Gly Ala Ala Gly Gly Ile Gly Arg Glu Lieu Val Lys Ala Met Lys 2O 25 3O Ala Ala Asn Ala Ile Val Ile Ala Thr Asp Met Ala Pro Ser Ala Asp 35 4 O 45 Val Glu Gly Ala Asp His Tyr Lieu Gln His Asp Val Thir Ser Glu Ala SO 55 6 O Gly Trp Lys Ala Val Ala Ala Lieu Ala Glin Glu Lys Tyr Gly Arg Val 65 70 7s 8O Asp Ala Lieu Val His Asn Ala Gly Ile Ser Ile Val Thr Llys Phe Glu 85 90 95 Asp Thr Pro Lieu. Ser Asp Phe His Arg Val Asn Thr Val Asn Val Asp 1OO 105 11 O Ser Ile Ile Ile Gly Thr Glin Val Lieu. Lieu Pro Lieu Lleu Lys Glu Gly 115 12 O 125 Gly Lys Ala Arg Ala Gly Gly Ala Ser Val Val Asn. Phe Ser Ser Val 13 O 135 14 O Gly Gly Lieu. Arg Gly Ala Ala Phe Asn Ala Ala Tyr Cys Thir Ser Lys 145 150 155 160 Ala Ala Wall Lys Met Lieu. Ser Lys Cys Lieu. Gly Ala Glu Phe Ala Ala 1.65 17O 17s Lieu. Gly Tyr Asn. Ile Arg Val Asn. Ser Val His Pro Gly Gly Ile Asp 18O 185 19 O Thr Pro Met Leu Gly Ser Gly Met Asp Llys Tyr Val Glu Lieu. Gly Ala 195 2OO 2O5 Ala Pro Ser Arg Glu Val Ala Glin Ala Ala Met Glu Met Arg His Pro 21 O 215 22O Ile Gly Arg Met Gly Arg Pro Ala Glu Met Gly Gly Gly Val Val Tyr 225 23 O 235 24 O Lieu. Cys Ser Asp Ala Ala Ser Phe Val Thr Cys Thr Glu Phe Val Met 245 250 255 Asp Gly Gly Phe Ser Glin Val 26 O

<210s, SEQ ID NO 7 &211s LENGTH: 792 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Variant of ketoreductase from Novosphingobium aromaticivorans

<4 OO > SEQUENCE: 7 atgcc.gcttgaaatgacgat tict ct caac aatgtggtcg ccgt.cgt cac cqgcgcggcg 6 O ggaggcatcg gcc.gcgaact ggt caaggcg atgaaggc.cg C caacgc.cat cqt catcgc.c 12 O accgacatgg cqc cotcggc cgatgtcgaa ggcgcggacc attat ct coa gcacgacgtg 18O acgagcgagg cc.ggctggala gC9gtC9C9 gC9ctggCCC aggaaaagta C9ggcgc.gtC 24 O gatgcgctgg to acaacgc gggcatctic atcgt cacga agttcgaaga cactic.cgctg 3OO tccgatttico accogcdtgaa cacggtcaac gtcgatt coat catcatcgg tacgcaggit c 360

Ctgctg.ccgc tigctcaagga aggcggcaag gcgc.gc.gcag ggggcgc.ctic ggtggtcaac 42O US 2013/0029385 A1 Jan. 31, 2013 28

- Continued ttct coagcg tcgggggtct gcgcggctic gcgttcaatig cigcc tattg Caccagcaa.g 48O gcggcggtga agatgctic to gaagtgcctic ggcgcggaat tccggcgct cqgct acaac 54 O atcc.gcgt.ca act cogtgca t cc.gggcggc atcgataccc catgct cqg Ctcgatcatg 6OO gacaagtacg tcgaacticgg cqctg.ccc.cc ticgc.gcgagg togcc caggc cgcgatggaa 660 atgcgccacc catcggit cq catgggtc.gc cct gcc.gaaa tiggcggcgg C9tggtctat 72 O Ctctgctc.cg acgcagcaag Ctt.cgtcacc tic acggaat tctgatgga C9gcggctt C 78O agcc aggt ct ga 792

<210s, SEQ ID NO 8 &211s LENGTH: 263 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Variant of ketoreductase from Novosphingobium aromaticivorans

<4 OOs, SEQUENCE: 8

Met Pro Lieu. Glu Met Thir Ile Ala Lieu. Asn. Asn. Wal Wall Ala Wal Wall 1. 5 1O 15 Thr Gly Ala Ala Gly Gly Ile Gly Arg Glu Lieu Val Lys Ala Met Lys 2O 25 3O Ala Ala Asn Ala Ile Val Ile Ala Thr Asp Met Ala Pro Ser Ala Asp 35 4 O 45 Val Glu Gly Ala Asp His Tyr Lieu Gln His Asp Val Thir Ser Glu Ala SO 55 6 O Gly Trp Lys Ala Val Ala Ala Lieu Ala Glin Glu Lys Tyr Gly Arg Val 65 70 7s 8O Asp Ala Lieu Val His Asn Ala Gly Ile Ser Ile Val Thr Llys Phe Glu 85 90 95 Asp Thr Pro Lieu. Ser Asp Phe His Arg Val Asn Thr Val Asn Val Asp 1OO 105 11 O Ser Ile Ile Ile Gly Thr Glin Val Lieu. Lieu Pro Lieu Lleu Lys Glu Gly 115 12 O 125 Gly Lys Ala Arg Ala Gly Gly Ala Ser Val Val Asn. Phe Ser Ser Val 13 O 135 14 O Gly Gly Lieu. Arg Gly Ser Ala Phe Asn Ala Ala Tyr Cys Thir Ser Lys 145 150 155 160 Ala Ala Wall Lys Met Lieu. Ser Lys Cys Lieu. Gly Ala Glu Phe Ala Ala 1.65 17O 17s Lieu. Gly Tyr Asn. Ile Arg Val Asn. Ser Val His Pro Gly Gly Ile Asp 18O 185 19 O Thr Pro Met Leu Gly Ser Ile Met Asp Llys Tyr Val Glu Lieu. Gly Ala 195 2OO 2O5 Ala Pro Ser Arg Glu Val Ala Glin Ala Ala Met Glu Met Arg His Pro 21 O 215 22O Ile Gly Arg Met Gly Arg Pro Ala Glu Met Gly Gly Gly Val Val Tyr 225 23 O 235 24 O Lieu. Cys Ser Asp Ala Ala Ser Phe Val Thr Cys Thr Glu Phe Val Met 245 250 255 Asp Gly Gly Phe Ser Glin Val 26 O

US 2013/0029385 A1 Jan. 31, 2013 31

- Continued Thr Gly Ala Ala Gly Gly Ile Gly Arg Glu Lieu Val Lys Ala Met Lys 2O 25 3O Ala Ala Asn Ala Ile Val Ile Ala Thr Asp Met Ala Pro Ser Ala Asp 35 4 O 45 Val Glu Gly Ala Asp His Tyr Lieu Gln His Asp Val Thir Ser Glu Ala SO 55 6 O Gly Trp Lys Ala Val Ala Ala Lieu Ala Glin Glu Lys Tyr Gly Arg Val 65 70 7s 8O Asp Ala Lieu Val His Asn Ala Gly Ile Ser Ile Val Thr Llys Phe Glu 85 90 95 Asp Thr Pro Lieu. Ser Asp Phe His Arg Val Asn Thr Val Asn Val Asp 1OO 105 11 O Ser Ile Ile Ile Gly Thr Glin Val Lieu. Lieu Pro Lieu Lleu Lys Glu Gly 115 12 O 125 Gly Lys Ala Arg Ala Gly Gly Ala Ser Val Val Asn. Phe Ser Ser Val 13 O 135 14 O Gly Gly Lieu. Arg Gly Trp Ala Phe Asn Ala Ala Tyr Cys Thir Ser Lys 145 150 155 160 Ala Ala Wall Lys Met Lieu. Ser Lys Cys Lieu. Gly Ala Glu Phe Ala Ala 1.65 17O 17s Lieu. Gly Tyr Asn. Ile Arg Val Asn. Ser Val His Pro Gly Gly Ile Asp 18O 185 19 O Thr Pro Met Lieu. Gly Ser Ile Met Asp Llys Tyr Val Glu Lieu. Gly Ala 195 2OO 2O5 Ala Pro Ser Arg Glu Val Ala Glin Ala Ala Met Glu Met Arg His Pro 21 O 215 22O Ile Gly Arg Met Gly Arg Pro Ala Glu Met Gly Gly Gly Val Val Tyr 225 23 O 235 24 O Lieu. Cys Ser Asp Ala Ala Ser Phe Val Thr Cys Thr Glu Phe Val Met 245 250 255 Asp Gly Gly Phe Ser Glin Val 26 O

<210s, SEQ ID NO 13 &211s LENGTH: 792 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Variant of ketoreductase from Novosphingobium aromaticivorans

<4 OOs, SEQUENCE: 13 atgcc.gcttgaaatgacgat tict ct caac aatgtggtcg ccgt.cgt cac cqgcgcggcg 6 O ggaggcatcg gcc.gcgaact ggt caaggcg atgaaggc.cg C caacgc.cat cqt catcgc.c 12 O accgacatgg cqc cotcggc cgatgtcgaa ggcgcggacc attat ct coa gcacgacgtg 18O acgagcgagg cc.ggctggala gC9gtC9C9 gC9ctggCCC aggaaaagta C9ggcgc.gtC 24 O gatgcgctgg to acaacgc gggcatctic atcgt cacga agttcgaaga cactic.cgctg 3OO tccgatttico accogcdtgaa cacggtcaac gtcgatt coat catcatcgg tacgcaggit c 360 Ctgctg.ccgc tigctcaagga aggcggcaag gcgc.gc.gcag ggggcgc.ctic ggtggtcaac 42O ttct coagcg tcgggggtct gcgcggcgcg gcgttcaatig cigcc tattg Caccagcaa.g 48O gcggcggtga agatgctic to gaagtgcctic ggcgcggaat tccggcgct cqgct acaac 54 O US 2013/0029385 A1 Jan. 31, 2013 32

- Continued atcc.gcgt.ca act cogtgca t cc.gggc.ccg atcgataccc catgct cqg Ctcgatcatg 6OO gacaagtacg tcgaacticgg cqctg.ccc.cc ticgc.gcgagg togcc caggc cgcgatggaa 660 atgcgccacc catcggit cq catgggtc.gc cct gcc.gaaa tiggcggcgg C9tggtctat 72 O

Ctctgctc.cg acgcagcaag Ctt.cgtcacc tic acggaat tctgatgga C9gcggctt C 78O agcc aggt ct ga 792

<210s, SEQ ID NO 14 &211s LENGTH: 263 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Variant of ketoreductase from Novosphingobium aromaticivorans

<4 OOs, SEQUENCE: 14

Met Pro Lieu. Glu Met Thir Ile Ala Lieu. Asn. Asn. Wal Wall Ala Wal Wall 1. 5 1O 15 Thr Gly Ala Ala Gly Gly Ile Gly Arg Glu Lieu Val Lys Ala Met Lys 2O 25 3O Ala Ala Asn Ala Ile Val Ile Ala Thr Asp Met Ala Pro Ser Ala Asp 35 4 O 45 Val Glu Gly Ala Asp His Tyr Lieu Gln His Asp Val Thir Ser Glu Ala SO 55 6 O Gly Trp Lys Ala Val Ala Ala Lieu. Ala Glin Glu, Llys Tyr Gly Arg Val 65 70 7s 8O Asp Ala Lieu Val His Asn Ala Gly Ile Ser Ile Val Thr Llys Phe Glu 85 90 95 Asp Thr Pro Lieu. Ser Asp Phe His Arg Val Asn Thr Val Asn Val Asp 1OO 105 11 O Ser Ile Ile Ile Gly Thr Glin Val Lieu. Lieu Pro Lieu Lleu Lys Glu Gly 115 12 O 125 Gly Lys Ala Arg Ala Gly Gly Ala Ser Val Val Asn. Phe Ser Ser Val 13 O 135 14 O Gly Gly Lieu. Arg Gly Ala Ala Phe Asn Ala Ala Tyr Cys Thir Ser Lys 145 150 155 160 Ala Ala Wall Lys Met Lieu. Ser Lys Cys Lieu. Gly Ala Glu Phe Ala Ala 1.65 17O 17s Lieu. Gly Tyr Asn. Ile Arg Val Asn. Ser Val His Pro Gly Pro Ile Asp 18O 185 19 O Thr Pro Met Leu Gly Ser Ile Met Asp Llys Tyr Val Glu Lieu. Gly Ala 195 2OO 2O5 Ala Pro Ser Arg Glu Val Ala Glin Ala Ala Met Glu Met Arg His Pro 21 O 215 22O Ile Gly Arg Met Gly Arg Pro Ala Glu Met Gly Gly Gly Val Val Tyr 225 23 O 235 24 O Lieu. Cys Ser Asp Ala Ala Ser Phe Val Thr Cys Thr Glu Phe Val Met 245 250 255 Asp Gly Gly Phe Ser Glin Val 26 O

<210s, SEQ ID NO 15 &211s LENGTH: 792 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence

US 2013/0029385 A1 Jan. 31, 2013 35

- Continued

Val Glu Gly Ala Asp His Tyr Lieu Gln His Asp Val Thir Ser Glu Ala SO 55 6 O Gly Trp Lys Ala Val Ala Ala Lieu Ala Glin Glu Lys Tyr Gly Arg Val 65 70 7s 8O Asp Ala Lieu Val His Asn Ala Gly Ile Ser Ile Val Thr Llys Phe Glu 85 90 95 Asp Thr Pro Lieu. Ser Asp Phe His Arg Val Asn Thr Val Asn Val Asp 1OO 105 11 O Ser Ile Ile Ile Gly Thr Glin Val Lieu. Lieu Pro Lieu Lleu Lys Glu Gly 115 12 O 125 Gly Lys Ala Arg Ala Gly Gly Ala Ser Val Val Asn. Phe Ser Ser Val 13 O 135 14 O Gly Gly Lieu. Arg Gly Ala Ala Phe Asn Ala Ala Tyr Cys Thir Ser Lys 145 150 155 160 Ala Ala Wall Lys Met Lieu. Ser Lys Cys Lieu. Gly Ala Glu Phe Ala Ala 1.65 17O 17s Lieu. Gly Tyr Asn. Ile Arg Val Asn. Ser Val His Pro Gly Val Ile Asp 18O 185 19 O Thr Pro Met Leu Gly Ser Ile Met Asp Llys Tyr Val Glu Lieu. Gly Ala 195 2OO 2O5 Ala Pro Ser Arg Glu Val Ala Glin Ala Ala Met Glu Met Arg His Pro 21 O 215 22O Ile Gly Arg Met Gly Arg Pro Ala Glu Met Gly Gly Gly Val Val Tyr 225 23 O 235 24 O Lieu. Cys Ser Asp Ala Ala Ser Phe Val Thr Cys Thr Glu Phe Val Met 245 250 255 Asp Gly Gly Phe Ser Glin Val 26 O

<210s, SEQ ID NO 19 &211s LENGTH: 792 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Variant of ketoreductase from Novosphingobium aromaticivorans

<4 OOs, SEQUENCE: 19 atgcc.gcttgaaatgacgat tict ct caac aatgtggtcg ccgt.cgt cac cqgcgcggcg 6 O ggaggcatcg gcc.gcgaact ggt caaggcg atgaaggc.cg C caacgc.cat cqt catcgc.c 12 O accgacatgg cqc cotcggc cgatgtcgaa ggcgcggacc attat ct coa gcacgacgtg 18O acgagcgagg cc.ggctggala gC9gtC9C9 gC9ctggCCC aggaaaagta C9ggcgc.gtC 24 O gatgcgctgg to acaacgc gggcatctic atcgt cacga agttcgaaga cactic.cgctg 3OO tccgatttico accogcdtgaa cacggtcaac gtcgatt coat catcatcgg tacgcaggit c 360 Ctgctg.ccgc tigctcaagga aggcggcaag gcgc.gc.gcag ggggcgc.ctic ggtggtcaac 42O ttct coagcg tcgggggtct gcgcggcgcg gcgttcaatig cigcc tattg Caccagcaa.g 48O gcggcggtga agatgctic to gaagtgcctic ggcgcggaat tccggcgct cqgct acaac 54 O atcc.gcgt.ca act cogtgca t cc.gggcggc atcgataccc catgct cqg Ctcgatcatg 6OO gacaagtact ttgaacticgg cqctg.ccc.cc ticgc.gcgagg togcc caggc cgcgatggaa 660 atgcgccacc catcggit cq catgggtc.gc cct gcc.gaaa tiggcggcgg C9tggtctat 72 O US 2013/0029385 A1 Jan. 31, 2013 36

- Continued

Ctctgctc.cg acgcagcaag Ctt.cgtcacc tic acggaat tctgatgga C9gcggctt C 78O agcc aggt ct ga 792

<210s, SEQ ID NO 2 O &211s LENGTH: 263 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Variant of ketoreductase from Novosphingobium aromaticivorans

<4 OOs, SEQUENCE: 2O

Met Pro Lieu. Glu Met Thir Ile Ala Lieu. Asn. Asn. Wal Wall Ala Wal Wall 1. 5 1O 15 Thr Gly Ala Ala Gly Gly Ile Gly Arg Glu Lieu Val Lys Ala Met Lys 2O 25 3O Ala Ala Asn Ala Ile Val Ile Ala Thr Asp Met Ala Pro Ser Ala Asp 35 4 O 45 Val Glu Gly Ala Asp His Tyr Lieu Gln His Asp Val Thir Ser Glu Ala SO 55 6 O Gly Trp Lys Ala Val Ala Ala Lieu Ala Glin Glu Lys Tyr Gly Arg Val 65 70 7s 8O Asp Ala Lieu Val His Asn Ala Gly Ile Ser Ile Val Thr Llys Phe Glu 85 90 95 Asp Thr Pro Lieu. Ser Asp Phe His Arg Val Asn Thr Val Asn Val Asp 1OO 105 11 O Ser Ile Ile Ile Gly Thr Glin Val Lieu. Lieu Pro Lieu Lleu Lys Glu Gly 115 12 O 125 Gly Lys Ala Arg Ala Gly Gly Ala Ser Val Val Asn. Phe Ser Ser Val 13 O 135 14 O Gly Gly Lieu. Arg Gly Ala Ala Phe Asn Ala Ala Tyr Cys Thir Ser Lys 145 150 155 160 Ala Ala Wall Lys Met Lieu. Ser Lys Cys Lieu. Gly Ala Glu Phe Ala Ala 1.65 17O 17s Lieu. Gly Tyr Asn. Ile Arg Val Asn. Ser Val His Pro Gly Gly Ile Asp 18O 185 19 O Thr Pro Met Leu Gly Ser Ile Met Asp Llys Tyr Phe Glu Lieu. Gly Ala 195 2OO 2O5 Ala Pro Ser Arg Glu Val Ala Glin Ala Ala Met Glu Met Arg His Pro 21 O 215 22O Ile Gly Arg Met Gly Arg Pro Ala Glu Met Gly Gly Gly Val Val Tyr 225 23 O 235 24 O Lieu. Cys Ser Asp Ala Ala Ser Phe Val Thr Cys Thr Glu Phe Val Met 245 250 255 Asp Gly Gly Phe Ser Glin Val 26 O

<210s, SEQ ID NO 21 &211s LENGTH: 792 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Variant of ketoreductase from Novosphingobium aromaticivorans

<4 OOs, SEQUENCE: 21

US 2013/0029385 A1 Jan. 31, 2013 40

- Continued tittgcc at C C Caagcgcga caa.cattcca acg tatgtgg cc.gc.cgaccg gatggcggaa 1 O2O gaacggattgaaacgatgcg caaag.cgc.gc agt caattitt to aaaatgg to accatatt 108 O ttaa.gc.cgcc gtc.gc.gc.ccg ctaa 1104

<210s, SEQ ID NO 26 &211s LENGTH: 367 212. TYPE: PRT <213> ORGANISM: Geobacillus stearothermophilus

<4 OOs, SEQUENCE: 26 Met Glu Lieu Phe Lys Tyr Met Glu Thr Tyr Asp Tyr Glu Glin Val Lieu. 1. 5 1O 15 Phe Cys Glin Asp Llys Glu Ser Gly Lieu Lys Ala Ile Ile Ala Ile His 2O 25 3O Asp Thir Thr Lieu. Gly Pro Ala Leu Gly Gly Thr Arg Met Trp Met Tyr 35 4 O 45 Asn Ser Glu Glu Glu Ala Lieu. Glu Asp Ala Lieu. Arg Lieu Ala Arg Gly SO 55 6 O Met Thr Tyr Lys Asn Ala Ala Ala Gly Lieu. Asn Lieu. Gly Gly Gly Lys 65 70 7s 8O Thr Val Ile Ile Gly Asp Pro Arg Lys Asp Lys Asn. Glu Ala Met Phe 85 90 95 Arg Ala Phe Gly Arg Phe Ile Glin Gly Lieu. ASn Gly Arg Tyr Ile Thr 1OO 105 11 O Ala Glu Asp Val Gly. Thir Thr Val Ala Asp Met Asp Ile Ile Tyr Glin 115 12 O 125 Glu Thir Asp Tyr Val Thr Gly Ile Ser Pro Glu Phe Gly Ser Ser Gly 13 O 135 14 O Asn Pro Ser Pro Ala Thr Ala Tyr Gly Val Tyr Arg Gly Met Lys Ala 145 150 155 160 Ala Ala Lys Glu Ala Phe Gly Ser Asp Ser Lieu. Glu Gly Llys Val Val 1.65 17O 17s Ala Val Glin Gly Val Gly Asn Val Ala Tyr His Lieu. Cys Arg His Lieu 18O 185 19 O His Glu Glu Gly Ala Lys Lieu. Ile Val Thir Asp Ile Asn Lys Glu Val 195 2OO 2O5 Val Ala Arg Ala Val Glu Glu Phe Gly Ala Lys Ala Val Asp Pro Asn 21 O 215 22O Asp Ile Tyr Gly Val Glu. Cys Asp Ile Phe Ala Pro Cys Ala Lieu. Gly 225 23 O 235 24 O Gly Ile Ile Asin Asp Glin Thir Ile Pro Glin Lieu Lys Ala Lys Val Ile 245 250 255 Ala Gly Ser Ala Asn. Asn Glin Lieu Lys Glu Pro Arg His Gly Asp Ile 26 O 265 27 O Ile His Glu Met Gly Ile Val Tyr Ala Pro Asp Tyr Val Ile Asn Ala 27s 28O 285 Gly Gly Val Ile Asn. Wall Ala Asp Glu Lieu. Tyr Gly Tyr Asn Arg Glu 29 O 295 3 OO Arg Ala Met Lys Lys Ile Glu Glin Ile Tyr Asp Asn. Ile Glu Lys Val 3. OS 310 315 32O

Phe Ala Ile Ala Lys Arg Asp Asn. Ile Pro Thr Tyr Val Ala Ala Asp 3.25 330 335 US 2013/0029385 A1 Jan. 31, 2013 41

- Continued

Arg Met Ala Glu Glu Arg Ile Glu Thir Met Arg Lys Ala Arg Ser Glin 34 O 345 35. O Phe Lieu. Glin Asn Gly His His Ile Lieu. Ser Arg Arg Arg Ala Arg 355 360 365

1. A process for converting a compound of formula I which 18. (canceled) is a Substrate for an amino acid dehydrogenase to a chiral 19. (canceled) amino acid of formula IIa: 20. (canceled) 21. The process of claim 16, wherein the engineered ketoreductase comprises an amino acid sequence selected R CO2H R is CO2H from the group consisting of SEQID NO: 2, 4, 6, 8, 10, 12, 14, r N. 16, 18, 20, 22, and 24. O 22. (canceled) 23. (canceled) I IIa 24. (canceled) 25. (canceled) comprising contacting the compound of formula I with a 26. The process of claim 1 wherein the lower secondary reaction medium comprising an amino acid dehydroge alcohol is isopropanol and the ketone is acetone. nase, an ammonium ion donor selected from ammonia 27. (canceled) or an ammonium salt, a cofactor selected from NAD"/ 28. The process of claim 1, wherein the reaction medium is NADH or NADP/NADPH, and a cofactor regenerating at a pH of about 8.5 to about 10. system comprising an engineered ketoreductase which 29. (canceled) has less than 5% of the activity of the amino acid dehy 30. (canceled) drogenase with compound of formula I and a lower 31. The process of claim 1, wherein the reaction medium is secondary alcohol, under conditions where the com at a temperature of about 35°C. to about 40°C. pound of formula I is converted to the chiral amino acid 32. The process of claim 1, wherein the secondary alcohol of formula IIa and the lower secondary alcohol is con is present in at least 1.5 fold stoichiometric excess of sub verted to a ketone. Strate. 2. The process of claim 1, wherein R is a substituted or 33. The process of claim 1, wherein the ketoreductase is unsubstituted —(C-C)alkyl, —(C-C)alkenyl, —(C-C) capable of recycling cofactor by converting isopropanol alkynyl. —(C-C)cycloalkyl, heterocycloalkyl, aryl, or het (IPA) to acetone in a reaction medium of 3 to 20% IPA at a pH eroaryl. of about 9.0 to 10.5 with an activity at least 2.0-fold greater 3. (canceled) than a reference ketoreductase of SEQID NO: 2. 4. (canceled) 34. The process of claim 1, wherein the activity of the 5. (canceled) ketoreductase with compound of formula I is less than 2%, 6. The process of claim 1, wherein the amino acid dehy 1%, 0.5%, 0.1%, 0.05%, or 0.01% of the activity of the amino drogenase is an L-amino acid dehydrogenase is selected from acid dehydrogenase with the compound of formula I, or L-alanine dehydrogenase, L-aspartate dehydrogenase, optionally no activity with the compound of formula I. L-erythro-3,5-diaminohexanoate dehydrogenase, L-leucine 35. (canceled) dehydrogenase, L-glutamate dehydrogenase, lysine dehydro 36. The process of claim 1, wherein the compound of genase, L-phenylalanine dehydrogenase, L-serine dehydro formula IIa is 3.3-dimethyl-2-oxobutanoic acid and the prod genase. L-valine dehydrogenase, L-2,4-diaminopentanoate uct of formula I is (S)-2-amino-3,3-dimethylbutanoic acid dehydrogenase, L-glutamate synthase, L-diaminopimelate and the amino acid dehydrogenase comprises a L-leucine dehydrogenase, L-N-methylalanine dehydrogenase, L-lysine dehydrogenase. 6-dehydrogenase, and L-tryptophan dehydrogenase. 37. (canceled) 7. (canceled) 38. The process of claim 36, wherein the leucine dehydro 8. (canceled) genase is engineered leucine dehydrogenase or is derived 9. (canceled) from a wild-type leucine dehydrogenase from Bacillus, 10. The process of claim 1, wherein the amino acid dehy Clostridium, Corynebacterium, Geobacillus, Natronobacte drogenase is a D-amino acid dehydrogenase is selected from rium, Thermoactinomyces, Thermos, Thermomicrobium, or a D-alanine dehydrogenase, D-threonine dehydrogenase, and Carderia. D-proline dehydrogenase. 39. (canceled) 11. (canceled) 40. The process of claim 38, wherein the leucine dehydro 12. (canceled) genase comprises the amino acid sequence of SEQ ID NO: 13. (canceled) 26. 14. (canceled) 41. A process for producing (S)-2-amino-3,3-dimethylbu 15. (canceled) tanoic acid, comprising: 16. The process of claim 1, wherein the engineered ketore contacting 3.3-dimethyl-2-oxobutanoic acid with a leucine ductase is derived from a wild-type ketoreductase from Lac dehydrogenase in a reaction medium comprising an tobacillus kefir, Lactobacillus brevis, Lactobacillus minor, or ammonium ion donor, cofactor NAD"/NADH or Novosphingobium aromaticivorans. NADP/NADPH, and a cofactor recycling system com 17. (canceled) prising a ketoreductase and a lower secondary alcohol, US 2013/0029385 A1 Jan. 31, 2013 42

under conditions where the 3.3-dimethyl-2-oxobutanoic 51. (canceled) acid is converted to (S)-2-amino-3,3-dimethylbutanoic 52. (canceled) acid, 53. (canceled) wherein the 3.3-dimethyl-2-oxobutanoic acid is at about 75 54. (canceled) g/L to 125 g/L, the secondary alcohol is about 1.5 fold 55. (canceled) stoichiometric excess of the 3.3-dimethyl-2-oxobu 56. (canceled) tanoic acid, the cofactor is at about 0.30g/L to 0.70 g/L, 57. (canceled) and the leucine dehydrogenase and ketoreductase are 58. (canceled) each independently at about 0.5 to about 1.0 g/L. 59. (canceled) 42. (canceled) 60. (canceled) 43. The process of claim 41, wherein the secondary alcohol 61. (canceled) is isopropanol, wherein the isopropanol is at about 7% to 12% 62. (canceled) volume of the reaction medium by (weight/volume). 63. (canceled) 44. A process for converting a compound mixture of for 64. (canceled) mula IId which comprises a Substrate for an amino acid dehy 65. (canceled) drogenase to a composition of formula I and a chiral amino 66. (canceled) acid of formula IIa: 67. (canceled) 68. (canceled) 69. (canceled) R CO2H 70. (canceled) 71. (canceled) N 72. (canceled) 73. (canceled) IId I 74. (canceled) R CO2H 75. The process of claim 1, wherein the process further comprises: N a chemical step comprising contacting the chiral amino acid compound of formula IIa with a compound com IIa prising an N-protecting group under conditions where the N-protecting group reacts with the chiral amino acid compound of formula IIa to form an N-protected chiral comprising contacting the compound mixture of formula amino acid compound. IId with an enantioselective amino acid dehydrogenase in a reaction medium comprising NAD"/NADH or 76. (canceled) NADP/NADPH and a cofactor recycling system com 77. (canceled) prising a ketoreductase and a lower alkyl ketone, under 78. (canceled) conditions where the compound mixture of formula IId 79. (canceled) is converted to the composition of formula I and a chiral 80. The process of claim 75, wherein the N-protecting amino acid of formula IIa, and the lower alkyl ketone is group is selected from Cbz, FMOC, BOC and MOC. converted to a lower secondary alcohol. 81. (canceled) 45. (canceled) 82. (canceled) 46. (canceled) 83. (canceled) 47. (canceled) 84. (canceled) 48. (canceled) 85. (canceled) 49. (canceled) 50. (canceled)