US0090 12189B2

(12) United States Patent (10) Patent No.: US 9,012,189 B2 Bastian et al. (45) Date of Patent: Apr. 21, 2015

(54) MODIFIED ALCOHOL DEHYDROGENASES “Request for Inter Partes Reexamination of U.S. Patent No. FOR THE PRODUCTION OF FUELS AND 8,158.404 Under 35 U.S.C. S 311 and 37 C.F.R.S 1913.”2583 pages, CHEMICALS U.S. Reexamination Control No. 95/002, 177 (filed Sep. 11, 2012). “Request for Inter Partes Reexamination of U.S. Patent No. (75) Inventors: Sabine Bastian, Pasadena, CA (US); 8,153,415 Under 35 U.S.C. S 311 and 37 C.F.R.S 1913.”3262 pages, Frances Arnold, La Cañada, CA (US); U.S. Reexamination Control No. 95/002,174 (filed Sep. 10, 2012). Peter Meinhold, Denver, CO (US) Overkamp et al., “Metabolic Engineering of Glycerol Production in Saccaromyces cerevisiae," AEM 68(6):2814-2821, American Soci (73) Assignees: Gevo, Inc., Englewood, CO (US); The ety for Microbiology, United States (2002). California Institute of Technology, Van Maris et al., “Directed Evolution of Pyruvate Decarboxylase Pasadena, CA (US) Negative Saccharomyces cerevisiae, Yielding a C-Independent, Glucose-Tolerant, and P-yruvate-Hyperpproducing Yeast. Appl. (*) Notice: Subject to any disclaimer, the term of this Environ. Microbiol. 70(1): 159-166, American Society for Microbiol patent is extended or adjusted under 35 ogy, United States (2004). U.S.C. 154(b) by 597 days. Würdig et al., “Vorkommen, Nachweis und Bestimmung von 2-und 3-methyl-2,3-dihydroxybuttersä und 2-hydroxyglutarav im Wein.” (21) Appl. No.: 13/025,805 Vitis 8:216-230, Bundesanstalt fur Zuechtungsforschung an (22) Filed: Feb. 11, 2011 Kulturpflanzen, Germany (1969). “Order Granting Inter Partes Reexamination of U.S. Patent No. (65) Prior Publication Data 8,133,715.” 35 pages, U.S. Reexamination Control No. 95/002,159 (mailed Dec. 3, 2012). US 2011 FO2O1072 A1 Aug. 18, 2011 “Order Granting Inter Partes Reexamination of U.S. Patent No. Related U.S. Application Data 8,158.404” 20 pages, U.S. Reexamination Control No. 95/002,177 (mailed Dec. 10, 2012). (60) Provisional application No. 61/304,069, filed on Feb. “Order Granting Inter Partes Reexamination of U.S. Patent No. 12, 2010, provisional application No. 61/308,568, 8,153,415.” 38 pages, U.S. Reexamination Control No. 95/002,174 filed on Feb. 26, 2010, provisional application No. (mailed Dec. 7, 2012). 61/282,641, filed on Mar. 10, 2010, provisional Written Opinion for International Application No. PCT/US2011/ application No. 61/352,133, filed on Jun. 7, 2010, 024482, 10 pages (mailed Apr. 9, 2012). provisional application No. 61/411.885, filed on Nov. Scannell et al., “Independent sorting-out of thousands of duplicated 9, 2010, provisional application No. 61/430,801, filed gene pairs in two yeast species descended from a whole-genome on Jan. 7, 2011. duplication.” PNAS 2007, Vo. 104(20): 8397-8402. “Office Action in Inter Partes Reexamination of U.S. Patent No. (51) Int. Cl. 8,133,715.” 46 pages, U.S. Reexamination Control No. 95/002,159 CI2P 7/16 (2006.01) (mailed Jun. 14, 2013). CI2N 15/8 (2006.01) “Office Action in Inter Partes Reexamination of U.S. Patent No. CI2N 9/04 (2006.01) 8,158.404” 28 pages, U.S. Reexamination Control No. 95/002,177 (52) U.S. Cl. (mailed Jun. 14, 2013). CPC ...... CI2N 15/81 (2013.01); C12N 9/0006 “Office Action in Inter Partes Reexamination of U.S. Patent No. (2013.01); CI2P 7/16 (2013.01); C12Y 8,153,415.” 47 pages, U.S. Reexamination Control No. 95/002,174 102/01005 (2013.01); C12Y 101/01 (2013.01); (mailed Jun. 14, 2013). Y02E 50/10 (2013.01) (58) Field of Classification Search * cited by examiner None Primary Examiner — Hope Robinson See application file for complete search history. (74) Attorney, Agent, or Firm — Cooley LLP (56) References Cited (57) ABSTRACT U.S. PATENT DOCUMENTS The present invention relates to recombinant microorganisms 6,753,314 B1* 6/2004 Giot et al...... 424,169 comprising biosynthetic pathways and methods of using said 8,017,376 B2 9, 2011 Dundon et al. recombinant microorganisms to produce various beneficial

8,133,715 B2 * 3/2012 Buelter et al...... 435.254.11 8,153,415 B2 * 4/2012 Buelter et al...... 435,254.11 metabolites. In various aspects of the invention, the recombi 8, 158.404 B2 * 4/2012 Lies et al...... 435/254.2 nant microorganisms may further comprise one or more 2007/00929.57 A1 4/2007 Donaldson et al. modifications resulting in the reduction or elimination of 3 2011/0076733 A1 3/2011 Urano et al. keto-acid (e.g., acetolactate and 2-aceto-2-hydroxybutyrate) and/or aldehyde-derived by-products. In various embodi FOREIGN PATENT DOCUMENTS ments described herein, the recombinant microorganisms may be microorganisms of the Saccharomyces clade, Crab WO WO 2009/086423 A2 T 2009 tree-negative yeast microorganisms, Crabtree-positive yeast WO WO 2012/O27642 A1 3, 2012 microorganisms, post-WGD (whole genome duplication) OTHER PUBLICATIONS yeast microorganisms, pre-WGD (whole genome duplica Bastian et al. tion) yeast microorganisms, and non-fermenting yeast micro “Request for Inter Partes Reexamination of U.S. Patent No. organisms. 8,133,715 Under 35 U.S.C.S 311 and 37 C.F.R.S 1913.”3131 pages, U.S. Reexamination Control No. 95/002,159 (filed Sep. 7, 2012). 10 Claims, 22 Drawing Sheets U.S. Patent Apr. 21, 2015 Sheet 1 of 22 US 9,012,189 B2

Glucose

2 NAD' -- s 2 NADH Glycolysis - 2 ATP 2?e-se pyruvate co,--- Als Hs H 2-aceto-lactate HcoH NAD(P)HNAD(P)' - d. KAR Hac 2,3-dihydroxy H3C O -isowaterate DHAD s o 2-keto-sowaterate co, KVD c H3 isobutyraldehyde NAD(P)HNAD(P) . . O ADH c s isobutara of

FIGURE 1 U.S. Patent Apr. 21, 2015 Sheet 2 of 22 US 9,012,189 B2

O O NAD(P)H NAD(P) O -- OH 3-ketoacid -reductase - -> OH OH 2,3-dihydrox-2-methyl 2-acetolactate butanoate (DH2MB) (AL)

O O NAD(P)H NAD(P)" O HO OH — — . OH 3-ketoacid reductase OH O 2-ethyl-2,3-dihydroxy-butyrate 2-aceto-2-hydroxy-butyrate

O O O. O. NAD(P)H NAD(P)" R OH R R R. R 3-ketoacid-k reductase 3-hydroxyacid

3-ketoacid

FIGURE 2 U.S. Patent Apr. 21, 2015 Sheet 3 of 22 US 9,012,189 B2

Ark orra ^i:SS: & -'. r . . . is -3-hycisas seiss oxy & is . . a. a .

S-8ietysaicrate-se: iaisiehysie

-- Y - & a S S ;, x Nii 3-8: Yi (3-3-3X3:333s

:seizy S (2,3s-2-retiyiis s: iris'xylisatioaie

FIGURE 3 U.S. Patent Apr. 21, 2015 Sheet 4 of 22 US 9,012,189 B2

s NADP" 5 NAP 3-hydroxypropionate alonate serialdehyde

FIGURE 3 (CONT.) U.S. Patent Apr. 21, 2015 Sheet 5 of 22 US 9,012,189 B2

NAD(P)" NAD(P)H 1-propanal aldehyde -N-4' dehydrogenase (ALDH) propionat e

Chis NAD(P)' NAD(P)H Cis isobutyraldehyde Nu aldehyde -- dehydrogenase (ALDH) r isobutyrat e NAD(P)" NAD(P)H 1-butana N aldehyde dehydrogenase (ALDH) - butyrate NAD(P)' NAD(P)H 2-methyl-1-butanal Nu aldehyde dehydrogenase (ALDH) s 2-methyl 1-butyrate NAD(P)' NAD(P)H 3-methyl-1-butanal N aldehyde dehydrogenase (ALDH) rC 3-methyl 1-butyrate

FIGURE 4 U.S. Patent Apr. 21, 2015 Sheet 6 of 22 US 9,012,189 B2

pyruvate

- a Cetolactate co, - synthase (ALS)

- - cases 3i 2,3-dihydrox-2-methyl

butanoate (DH2MB)

ii dihydroxy-acid dehydratase (DHAD)

2-ketoisowalerate (KIV) ^ 3.

ketoisowalerate

o, - decarboxylase (KVD) his 33. isobutyraldehyde -- x- -- i (P)H all A dehydrogenase (ADH) isobutyrate NAD(P)- isitar (BuOH) ---

FIGURE 5 U.S. Patent Apr. 21, 2015 Sheet 7 of 22 US 9,012,189 B2

2,3-dihydro-2-methy butariate (DH2MB)

3-methyl-1-butarial r *...*

FIGURE 6 U.S. Patent Apr. 21, 2015 Sheet 8 of 22 US 9,012,189 B2

" acetohydroxybutanoate

. NADPy" 2-aceto-2-hydroxy-butyrate s-X"* us six's: r / r y NPN-: keto-acid reducto 2-ethyl-2,3-dihydroxy-butyrate A isomerase (KARI) NAD(P) 2,3-dihydroxy-3-methyvalerate -eH : dihydroxy-acid dehydratase (DAD) 2-keto-3-methylwaterate ---s

A kitselaerate

- decarboxylase (KIMD)

2-methy--butara -- as - NAD(P)H- s NAD(P)- de ydrogenase (ADH)A. 2-methyl-1-butyrate 2-methyl-1-butano sus

FIGURE 7 U.S. Patent Apr. 21, 2015 Sheet 9 of 22 US 9,012,189 B2

33 83 U.S. Patent Apr. 21, 2015 Sheet 10 of 22 US 9,012,189 B2

xxxxx s

it ig 338ARides

Seiics

88,38 sis.

f

f

s st i

C

RS is . . . . & wer is \ was is is try S. is ... x X -- six is v a x\\\ a rr a U.S. Patent Apr. 21, 2015 Sheet 11 of 22 US 9,012,189 B2

{udd T. st L. c. ur C. L. c. ge is "

U.S. Patent Apr. 21, 2015 Sheet 13 of 22 US 9,012,189 B2

&ssssssssssss

U.S. Patent Apr. 21, 2015 Sheet 16 of 22 US 9,012,189 B2

i. U.S. Patent Apr. 21, 2015 Sheet 17 of 22 US 9,012,189 B2

r) |e?OLHOng!-e-

(1/3) uoeulueouoo U.S. Patent Apr. 21, 2015 Sheet 18 of 22 US 9,012,189 B2

U.S. Patent Apr. 21, 2015 Sheet 19 of 22 US 9,012,189 B2

U.S. Patent Apr. 21, 2015 Sheet 20 of 22 US 9,012,189 B2

| ~~~~)

w |

} eurone|

||| se U.S. Patent Apr. 21, 2015 Sheet 21 of 22 US 9,012,189 B2

| „ºs:(~~~~ | ,

| Ho,~~~~

/

| |-?oo | „~~~~

{} ( () | (~~~~•I }}~~~~ U.S. Patent Apr. 21, 2015 Sheet 22 of 22 US 9,012,189 B2

pyruvate st l CH

CA1 N acetyl-CoA -N

N COA . NAD(P)H- NAD(P)" acetyl-CoA : g t O coA1 -

reduced electron acceptors -ne' oxidized electron acceptor l NAD(P)H-NAD(P) - CNAD(P) C:A 1N --- NAD(P)H- -propano 1-butano NAD(P)'-N, ca

N-sa C

NAD(P)- Nu-ra

1-butano

FIGURE 20 US 9,012,189 B2 1. 2 MODIFIED ALCOHOL DEHYDROGENASES One of the primary reasons for the sub-optimal perfor FOR THE PRODUCTION OF FUELS AND mance observed in many existing microorganisms is the CHEMICALS undesirable conversion of pathway intermediates to unwanted by-products. The present inventors have identified CROSS REFERENCE TO RELATED various by-products, including 2,3-dihydroxy-2-methylbu APPLICATIONS tanoic acid (DH2MB) (CAS #14868-24-7), 2-ethyl-2,3-dihy droxybutyrate, 2,3-dihydroxy-2-methyl-butanonate, isobu This application claims priority to U.S. Provisional Appli tyrate, 3-methyl-1-butyrate, 2-methyl-1-butyrate, and cation Ser. No. 61/304,069, filed Feb. 12, 2010; U.S. Provi propionate, which are derived from various intermediates of sional Application Ser. No. 61/308,568, filed Feb. 26, 2010; 10 biosynthetic pathways used to produce fuels, chemicals, and amino acids. The accumulation of these by-products nega U.S. Provisional Application Ser. No. 61/282,641, filed Mar. tively impacts the synthesis and yield of desirable metabolites 10, 2010; U.S. Provisional Application Ser. No. 61/352,133, in a variety of fermentation reactions. Until now, the enzy filed Jun. 7, 2010; U.S. Provisional Application Ser. No. matic activities responsible for the production of these 61/411,885, filed Nov. 9, 2010; and U.S. Provisional Appli 15 unwanted by-products had not been characterized. More par cation Ser. No. 61/430,801, filed Jan. 7, 2011, each of which ticularly, the present application shows that the activities of a is herein incorporated by reference in its entirety for all pur 3-ketoacid reductase (3-KAR) and an aldehyde dehydroge poses. nase (ALDH) allow for the formation of these by-products from important biosynthetic pathway intermediates. ACKNOWLEDGMENT OF GOVERNMENTAL The present invention results from the study of these enzy SUPPORT matic activities and shows that the suppression of the 3-KAR and/or ALDH enzymes considerably reduces or eliminates This invention was made with government Support under the formation of unwanted by-products, and concomitantly Contract No. 2009-10006-05919, awarded by the United improves the yields and titers of beneficial metabolites. The States Department of Agriculture, and under Contract No. 25 present application shows moreover, that enhancement of the W911 NF-09-2-0022, awarded by the United States Army 3-KAR and/or ALDH enzymatic activities can be used to Research Laboratory. The government has certain rights in increase the production of various by-products, such 2,3- the invention. dihydroxy-2-methylbutanoic acid (DH2MB), 2-ethyl-2,3-di hydroxybutyrate, 2,3-dihydroxy-2-methyl-butanonate, TECHNICAL FIELD 30 isobutyrate, 3-methyl-1-butyrate, 2-methyl-1-butyrate, and propionate. Recombinant microorganisms and methods of producing Such organisms are provided. Also provided are methods of SUMMARY OF THE INVENTION producing beneficial metabolites including fuels, chemicals, and amino acids by contacting a suitable Substrate with 35 The present inventors have discovered that unwanted by recombinant microorganisms and enzymatic preparations products can accumulate during various fermentation pro therefrom. cesses, including fermentation of the biofuel candidate, isobutanol. The accumulation of these unwanted by-products DESCRIPTION OF THE TEXT FILE SUBMITTED results from the undesirable conversion of pathway interme ELECTRONICALLY 40 diates including the 3-keto acids, acetolactate and 2-aceto-2- hydroxybutyrate, and/or aldehydes, such as isobutyralde The contents of the text file submitted electronically here hyde, 1-butanal, 1-propanal, 2-methyl-1-butanal, and with are incorporated herein by reference in their entirety: A 3-methyl-1-butanal. The conversion of these intermediates to computer readable format copy of the Sequence Listing (file unwanted by-products can hinder the optimal productivity aC. GEVO 045 04US SeqList ST25.txt, date 45 and yield of a 3-keto acid- and/or aldehyde-derived products. recorded: Feb. 9, 2011, file size: 305 kilobytes). Therefore, the present inventors have developed methods for reducing the conversion of 3-keto acid and/or aldehyde inter BACKGROUND mediates to various fermentation by-products during pro cesses where a 3-keto acid and/or an aldehyde acts as a The ability of microorganisms to convert pyruvate to ben 50 pathway intermediate. eficial metabolites including fuels, chemicals, and amino In a first aspect, the present invention relates to a recombi acids has been widely described in the literature in recent nant microorganism comprising a biosynthetic pathway of years. See, e.g., Alper et al., 2009, Nature Microbiol. Rev. 7: which a 3-keto acid and/oran aldehyde is/are intermediate(s), 715-723. Recombinant engineering techniques have enabled wherein said recombinant microorganism is (a) Substantially the creation of microorganisms that express biosynthetic 55 free of an enzyme catalyzing the conversion of a 3-keto acid pathways capable of producing a number of useful products, to a 3-hydroxyacid; (b) Substantially free of an enzyme cata Such as valine, isoleucine, leucine, and panthothenic acid lyzing the conversion of an aldehyde to an acid by-product; (vitamin B5). In addition, fuels such as isobutanol have been (c) engineered to reduce or eliminate the expression or activ produced recombinantly in microorganisms expressing a het ity of an enzyme catalyzing the conversion of a 3-keto acid to erologous metabolic pathway (See, e.g., WO/2007/050671 to 60 a 3-hydroxyacid; and/or (d) engineered to reduce or eliminate Donaldson et al., and WO/2008/098227 to Liao, et al.). the expression or activity of an enzyme catalyzing the con Although engineered microorganisms represent potentially version of an aldehyde to acid by-product. In one embodi useful tools for the renewable production of fuels, chemicals, ment, the 3-keto acid is acetolactate. In another embodiment, and amino acids, many of these microorganisms have fallen the 3-keto acid is 2-aceto-2-hydroxybutyrate. short of commercial relevance due to their low performance 65 In one embodiment, the invention is directed to a recom characteristics, including low productivity, low titers, and low binant microorganism comprising a biosynthetic pathway yields. which uses the 3-keto acid, acetolactate, as an intermediate, US 9,012,189 B2 3 4 wherein said recombinant microorganism is engineered to about 60%, by at least about 65%, by at least about 70%, by reduce or eliminate the expression or activity of an enzyme at least about 75%, by at least about 80%, by at least about catalyzing the conversion of acetolactate to the corresponding 85%, by at least about 90%, by at least about 95%, or by at 3-hydroxyacid, DH2MB. In some embodiments, the enzyme least about 99% as compared to a recombinant microorgan catalyzing the conversion of acetolactate to DH2MB is a ism not comprising a reduction or deletion of the activity or 3-ketoacid reductase (3-KAR). expression of one or more endogenous proteins involved in In one embodiment, the invention is directed to a recom catalyzing the conversion of a 3-keto acid intermediate to a binant microorganism comprising a biosynthetic pathway 3-hydroxyacid by-product. In one embodiment, the 3-keto which uses the 3-keto acid, 2-aceto-2-hydroxybutyrate, as an acid intermediate is acetolactate and the 3-hydroxyacid by intermediate, wherein said recombinant microorganism is 10 product is DH2MB. In another embodiment, the 3-keto acid engineered to reduce or eliminate the expression or activity of intermediate is 2-aceto-2-hydroxybutyrate and the 3-hy an enzyme catalyzing the conversion of acetolactate to the droxyacid by-product is 2-ethyl-2,3-dihydroxybutanoate. corresponding 3-hydroxyacid, 2-ethyl-2,3-dihydroxybu In various embodiments described herein, the protein tanoate. In some embodiments, the enzyme catalyzing the involved in catalyzing the conversion of a 3-keto acid inter conversion of 2-aceto-2-hydroxybutyrate to 2-ethyl-2,3-di 15 mediate to a 3-hydroxyacid by-product is a ketoreductase. In hydroxybutanoate is a 3-ketoacid reductase (3-KAR). an exemplary embodiment, the ketoreductase is a 3-ketoacid In one embodiment, the invention is directed to a recom reductase (3-KAR). In another embodiment, the protein is a binant microorganism comprising a biosynthetic pathway short chain alcohol dehydrogenase. In yet another embodi which uses an aldehyde as an intermediate, wherein said ment, the protein is a medium chain alcohol dehydrogenase. recombinant microorganism is engineered to reduce or elimi In yet another embodiment, the protein is an aldose reductase. nate the expression or activity of an enzyme catalyzing the In yet another embodiment, the protein is a D-hydroxyacid conversion of the aldehyde to an acid by-product. In some dehydrogenase. In yet another embodiment, the protein is a embodiments, the enzyme catalyzing the conversion of the lactate dehydrogenase. In yet another embodiment, the pro aldehyde to an acid by-product is an aldehyde dehydrogenase tein is selected from the group consisting of YAL060W. (ALDH). 25 YJR159W, YGL157W, YBL114W, YOR120W, YKL055C, In one embodiment, the invention is directed to a recom YBR159W, YBR149W, YDL168W, YDR368W, YLR426W, binant microorganism comprising a biosynthetic pathway YCR107W, YIL124W, YML054C, YOL151W, YMR318C, which uses both a 3-keto acid and an aldehyde as intermedi YMR226C, YBR046C, YHR104W, YIR036C, YDL174C, ates, wherein said recombinant microorganism is (a) engi YDR541C, YBR 145W, YGL039W, YCR105W, YDL124W, neered to reduce or eliminate the expression or activity of an 30 YIR035C, YFLO56C, YNL274C, YLR255C, YGL185C, enzyme catalyzing the conversion of a 3-keto acid interme YGL256W, YJR096W, YMR226C, YJR155W, YPL275W, diate to a 3-hydroxyacid by-product; and (b) engineered to YOR388C, YLR070C, YMR083W, YER081 W, YJR139C, reduce or eliminate the expression or activity of an enzyme YDL243C, YPL113C, YOL165C, YML086C, YMR303C, catalyzing the conversion of an aldehyde intermediate to an YDL246C, YLR070C, YHR063C, YNL331C, YFLO57C, acid by-product. In one embodiment, the 3-keto acid is aceto 35 YIL155C, YOLO86C, YAL061W, YDR127W, YPR127W, lactate and the 3-hydroxyacid by-product is DH2MB. In YCI018W, YIL074C, YIL124W, and YEL071W genes of S. another embodiment, the 3-keto acid is 2-aceto-2-hydroxy cerevisiae and homologs thereof. butyrate and the 3-hydroxyacid by-product is 2-ethyl-2,3- In one embodiment, the endogenous protein is a 3-ketoacid dihydroxybutanoate. In some embodiments, the enzyme cata reductase (3-KAR). In an exemplary embodiment, the 3-ke lyzing the conversion of acetolactate to DH2MB is a 40 toacid reductase is the S. cerevisiae YMR226C (SEQID NO: 3-ketoacid reductase (3-KAR). In some other embodiments, 1) protein, used interchangeably herein with “TMA29. In the enzyme catalyzing the conversion of 2-aceto-2-hydroxy Some embodiments, the endogenous protein may be the S. butyrate to 2-ethyl-2,3-dihydroxybutanoate is a 3-ketoacid cerevisiae YMR226C (SEQID NO: 1) protein or a homolog reductase (3-KAR). In some other embodiments, the enzyme or variant thereof. In one embodiment, the homolog may be catalyzing the conversion of the aldehyde to an acid by 45 selected from the group consisting of Vanderwaltomzyma product is an aldehyde dehydrogenase (ALDH). In yet some polyspora (SEQID NO: 2), Saccharomyces castellii (SEQID other embodiments, the enzyme catalyzing the conversion of NO:3), Candida glabrata (SEQ ID NO: 4), Saccharomyces acetolactate to DH2MB is a 3-ketoacid reductase (3-KAR) bayanus (SEQ ID NO. 5), Zygosaccharomyces rouxi (SEQ and the enzyme catalyzing the conversion of the aldehyde to ID NO: 6), Kluyveromyces lactis (SEQ ID NO: 7), Ashbya an acid by-product is an aldehyde dehydrogenase (ALDH). In 50 gossypii (SEQID NO: 8), Saccharomyces kluyveri (SEQID yet some other embodiments, the enzyme catalyzing the con NO: 9), Kluyveromyces thermotolerans (SEQ ID NO: 10), version of 2-aceto-2-hydroxybutyrate to 2-ethyl-2,3-dihy Kluyveromyces waltii (SEQID NO: 11), Pichia stipitis (SEQ droxybutanoate is a 3-ketoacid reductase (3-KAR) and the ID NO: 12), Debaromyces hansenii (SEQID NO: 13), Pichia enzyme catalyzing the conversion of the aldehyde to an acid pastoris (SEQ ID NO: 14), Candida dubliniensis (SEQ ID by-product is an aldehyde dehydrogenase (ALDH). 55 NO: 15), Candida albicans (SEQ ID NO: 16), Yarrowia In various embodiments described herein, the recombinant lipolytica (SEQID NO: 17), Issatchenkia orientalis (SEQID microorganisms of the invention may comprise a reduction or NO: 18), Aspergillus nidulans (SEQID NO: 19), Aspergillus deletion of the activity or expression of one or more endog niger (SEQ ID NO: 20), Neurospora crassa (SEQ ID NO: enous proteins involved in catalyzing the conversion of a 21), Schizosaccharomyces pombe (SEQ ID NO: 22), and 3-keto acid intermediate to a 3-hydroxyacid by-product. In 60 Kluyveromyces marxianus (SEQID NO. 23). one embodiment, the activity or expression of one or more In one embodiment, the recombinant microorganism endogenous proteins involved in catalyzing the conversion of includes a mutation in at least one gene encoding for a 3-ke a 3-keto acid intermediate to a 3-hydroxyacid by-product is toacid reductase resulting in a reduction of 3-ketoacid reduc reduced by at least about 50%. In another embodiment, the tase activity of a polypeptide encoded by said gene. In another activity or expression of one or more endogenous proteins 65 embodiment, the recombinant microorganism includes a par involved in catalyzing the conversion of a 3-keto acid inter tial deletion of gene encoding for a 3-ketoacid reductase mediate to a 3-hydroxyacid by-product is reduced by at least resulting in a reduction of 3-ketoacid reductase activity of a US 9,012,189 B2 5 6 polypeptide encoded by the gene. In another embodiment, the 38), Kluyveromyces marxianus (SEQID NO:39), Schizosac recombinant microorganism comprises a complete deletion charomyces pombe (SEQ ID NO: 40), and Schizosaccharo of a gene encoding for a 3-ketoacid reductase resulting in a myces pombe (SEQID NO: 41). reduction of 3-ketoacid reductase activity of a polypeptide In one embodiment, the recombinant microorganism encoded by the gene. In yet another embodiment, the recom- 5 includes a mutation in at least one gene encoding for an binant microorganism includes a modification of the regula aldehyde dehydrogenase resulting in a reduction of aldehyde tory region associated with the gene encoding for a 3-ketoacid dehydrogenase activity of a polypeptide encoded by said reductase resulting in a reduction of expression of a polypep gene. In another embodiment, the recombinant microorgan tide encoded by said gene. In yet another embodiment, the recombinant microorganism comprises a modification of the 10 ism includes a partial deletion of gene encoding for an alde transcriptional regulator resulting in a reduction of transcrip hyde dehydrogenase resulting in a reduction of aldehyde tion of a gene encoding for a 3-ketoacid reductase. In yet dehydrogenase activity of a polypeptide encoded by the gene. another embodiment, the recombinant microorganism com In another embodiment, the recombinant microorganism prises mutations in all genes encoding for a 3-ketoacid reduc comprises a complete deletion of a gene encoding for an tase resulting in a reduction of activity of a polypeptide 15 aldehyde dehydrogenase resulting in a reduction of aldehyde encoded by the gene(s). In one embodiment, the 3-ketoacid dehydrogenase activity of a polypeptide encoded by the gene. reductase activity or expression is reduced by at least about In yet another embodiment, the recombinant microorganism 50%. In another embodiment, the 3-ketoacid reductase activ includes a modification of the regulatory region associated ity or expression is reduced by at least about 60%, by at least with the gene encoding for an aldehyde dehydrogenase about 65%, by at least about 70%, by at least about 75%, by 20 resulting in a reduction of expression of a polypeptide at least about 80%, by at least about 85%, by at least about encoded by said gene. In yet another embodiment, the recom 90%, by at least about 95%, or by at least about 99% as binant microorganism comprises a modification of the tran compared to a recombinant microorganism not comprising a Scriptional regulator resulting in a reduction of transcription reduction of the 3-ketoacid reductase activity or expression. of a gene encoding for an aldehyde dehydrogenase. In yet In one embodiment, said 3-ketoacid reductase is encoded by 25 another embodiment, the recombinant microorganism com the S. cerevisiae TMA29 (YMR226C) gene or a homolog prises mutations in all genes encoding for an aldehyde dehy thereof. drogenase resulting in a reduction of activity of a polypeptide In various embodiments described herein, the recombinant encoded by the gene(s). In one embodiment, the aldehyde microorganisms of the invention may comprise a reduction or dehydrogenase activity or expression is reduced by at least deletion of the activity or expression of one or more endog- 30 about 50%. In another embodiment, the aldehyde dehydro enous proteins involved in catalyzing the conversion of an genase activity or expression is reduced by at least about 60%, aldehyde to an acid by-product. In one embodiment, the activ by at least about 65%, by at least about 70%, by at least about ity or expression of one or more endogenous proteins 75%, by at least about 80%, by at least about 85%, by at least involved in catalyzing the conversion of an aldehyde to an about 90%, by at least about 95%, or by at least about 99% as acid by-product is reduced by at least about 50%. In another 35 compared to a recombinant microorganism not comprising a embodiment, the activity or expression of one or more endog reduction of the aldehyde dehydrogenase activity or expres enous proteins involved in catalyzing the conversion of an Sion. In one embodiment, said aldehyde dehydrogenase is aldehyde to an acid by-product is reduced by at least about encoded by the S. cerevisiae ALD6 gene or a homolog 60%, by at least about 65%, by at least about 70%, by at least thereof. about 75%, by at least about 80%, by at least about 85%, by 40 In various embodiments described herein, the recombinant at least about 90%, by at least about 95%, or by at least about microorganism may comprise a biosynthetic pathway which 99% as compared to a recombinant microorganism not com uses a 3-keto acid as an intermediate. In one embodiment, the prising a reduction or deletion of the activity or expression of 3-keto acid intermediate is acetolactate. The biosynthetic one or more endogenous proteins involved in catalyzing the pathway which uses acetolactate as an intermediate may be conversion of an aldehyde to an acid by-product. 45 selected from a pathway for the biosynthesis of isobutanol, In various embodiments described herein, the endogenous 2-butanol, 1-butanol. 2-butanone, 2,3-butanediol, acetoin, protein involved in catalyzing the conversion of an aldehyde diacetyl, Valine, leucine, pantothenic acid, isobutylene, 3-me to an acid by-product is an aldehyde dehydrogenase (ALDH). thyl-1-butanol, 4-methyl-1-pentanol, and coenzyme A. In In one embodiment, the aldehyde dehydrogenase is encoded another embodiment, the 3-keto acid intermediate is 2-aceto by a gene selected from the group consisting of ALD2, ALD3, 50 2-hydroxybutyrate. The biosynthetic pathway which uses ALD4, ALD5, ALD6, and HFD1, and homologs and variants 2-aceto-2-hydroxybutyrate as an intermediate may be thereof. In an exemplary embodiment, the aldehyde dehydro selected from a pathway for the biosynthesis of 2-methyl-1- genase is the S. cerevisiae ALD6 (SEQID NO:25) protein. In butanol, isoleucine, 3-methyl-1-pentanol, 4-methyl-1-hex Some embodiments, the aldehyde dehydrogenase is the S. anol, and 5-methyl-1-heptanol. cerevisiae ALD6 (SEQID NO: 25) protein or a homolog or 55 In various embodiments described herein, the recombinant variant thereof. In one embodiment, the homolog is selected microorganism may comprise a biosynthetic pathway which from the group consisting of Saccharomyces castelli (SEQID uses an aldehyde as an intermediate. The biosynthetic path NO: 26), Candida glabrata (SEQID NO: 27), Saccharomy way which uses an aldehyde as an intermediate may be ces bayanus (SEQ ID NO: 28), Kluyveromyces lactis (SEQ selected from a pathway for the biosynthesis of isobutanol, ID NO: 29), Kluyveromyces thermotolerans (SEQ ID NO: 60 1-butanol, 2-methyl-1-butanol, 3-methyl-1-butanol, 1-pro 30), Kluyveromyces waltii (SEQID NO:31), Saccharomyces panol. 1-pentanol, 1-hexanol, 3-methyl-1-pentanol, 4-me cerevisiae YJ789 (SEQ ID NO:32), Saccharomyces cerevi thyl-1-pentanol, 4-methyl-1-hexanol, and 5-methyl-1-hep siae JAY291 (SEQ ID NO: 33), Saccharomyces cerevisiae tanol. In various embodiments described herein, the aldehyde EC1118 (SEQ ID NO. 34), Saccharomyces cerevisiae intermediate may be selected from isobutyraldehyde, 1-buta DBY939 (SEQ ID NO: 35), Saccharomyces cerevisiae 65 nal, 2-methyl-1-butanal, 3-methyl-1-butanal, 1-propanal, AWRI1631 (SEQ ID NO: 36), Saccharomyces cerevisiae 1-pentanal, 1-hexanal, 3-methyl-1-pentanal, 4-methyl-1- RM11-1a (SEQ ID NO: 37), Pichia pastoris (SEQ ID NO: pentanal, 4-methyl-1-hexanal, and 5-methyl-1-heptanal. US 9,012,189 B2 7 8 In various embodiments described herein, the recombinant comprises at five exogenous genes that catalyze steps in the microorganism may comprise a biosynthetic pathway which conversion of pyruvate to isobutanol. uses a 3-keto acid and an aldehyde as intermediates. In one In one embodiment, one or more of the isobutanol pathway embodiment, the 3-keto acid intermediate is acetolactate. The genes encodes an enzyme that is localized to the cytosol. In biosynthetic pathway which uses acetolactate and an alde one embodiment, the recombinant microorganisms comprise hyde as intermediates may be selected from a pathway for the an isobutanol producing metabolic pathway with at least one biosynthesis of isobutanol, 1-butanol, and 3-methyl-1-bu isobutanol pathway enzyme localized in the cytosol. In tanol. In another embodiment, the 3-keto acid intermediate is another embodiment, the recombinant microorganisms com 2-aceto-2-hydroxybutyrate. The biosynthetic pathway which prise an isobutanol producing metabolic pathway with at least 10 two isobutanol pathway enzymes localized in the cytosol. In uses 2-aceto-2-hydroxybutyrate and an aldehyde as interme yet another embodiment, the recombinant microorganisms diates may be selected from a pathway for the biosynthesis of comprise an isobutanol producing metabolic pathway with at 2-methyl-1-butanol, 3-methyl-1-pentanol, 4-methyl-1-hex least three isobutanol pathway enzymes localized in the cyto anol, and 5-methyl-1-heptanol. Sol. In yet another embodiment, the recombinant microorgan In one embodiment, the invention is directed to a recom 15 isms comprise an isobutanol producing metabolic pathway binant microorganism for producing isobutanol, wherein said with at least four isobutanol pathway enzymes localized in the recombinant microorganism comprises an isobutanol pro cytosol. In an exemplary embodiment, the recombinant ducing metabolic pathway and wherein said microorganism microorganisms comprise an isobutanol producing metabolic is engineered to reduce or eliminate the expression or activity pathway with five isobutanol pathway enzymes localized in of an enzyme catalyzing the conversion of acetolactate to the cytosol. DH2MB. In some embodiments, the enzyme catalyzing the In various embodiments described herein, the isobutanol conversion of acetolactate to DH2MB is a 3-ketoacid reduc pathway genes encodes enzyme(s) selected from the group tase (3-KAR). In a specific embodiment, the 3-ketoacid consisting of acetolactate synthase (ALS), ketol-acid reduc reductase is encoded by the S. cerevisiae TMA29 toisomerase (KARI), dihydroxyacid dehydratase (DHAD), (YMR226C) gene or a homolog thereof. 25 2-keto-acid decarboxylase (KIVD), and alcohol dehydroge In another embodiment, the invention is directed to a nase (ADH). recombinant microorganism for producing isobutanol, In another aspect, the recombinant microorganism may be wherein said recombinant microorganism comprises an engineered to reduce the conversion of isobutanol to isobu isobutanol producing metabolic pathway and wherein said tyraldehyde by reducing and/or eliminating the expression of microorganism is engineered to reduce or eliminate the 30 one or more alcohol dehydrogenases. In a specific embodi expression or activity of an enzyme catalyzing the conversion ment, the alcohol dehydrogenase is encoded by a gene of isobutyraldehyde to isobutyrate. In some embodiments, the selected from the group consisting of ADH1, ADH2, ADH3, enzyme catalyzing the conversion of isobutyraldehyde to ADH4, ADH5, ADH6, and ADH7, and homologs and vari isobutyrate is an aldehyde dehydrogenase. In a specific ants thereof. embodiment, the aldehyde dehydrogenase is encoded by the 35 In another aspect, the present invention relates to modified S. cerevisiae ALD6 gene or a homolog thereof. alcohol dehydrogenase (ADH) enzymes that exhibit an In yet another embodiment, the invention is directed to a enhanced ability to convertisobutyraldehyde to isobutanol. In recombinant microorganism for producing isobutanol, general, cells expressing these improved ADH enzymes will wherein said recombinant microorganism comprises an produce increased levels of isobutanol during fermentation isobutanol producing metabolic pathway and wherein said 40 reactions. While the modified ADH enzymes of the present microorganism is (i) engineered to reduce or eliminate the invention have utility in isobutanol-producing fermentation expression or activity of an enzyme catalyzing the conversion reactions, it will be understood by those skilled in the art of acetolactate to DH2MB and (ii) engineered to reduce or equipped with this disclosure that the modified ADH eliminate the expression or activity of an enzyme catalyzing enzymes also have usefulness in fermentation reactions pro the conversion of isobutyraldehyde to isobutyrate. In some 45 ducing other alcohols such as 1-propanol, 2-propanol. 1-bu embodiments, the enzyme catalyzing the conversion of aceto tanol, 2-butanol. 1-pentanol, 2-methyl-1-butanol, and 3-me lactate to DH2MB is a 3-ketoacid reductase (3-KAR). In a thyl-1-butanol. specific embodiment, the 3-ketoacid reductase is encoded by In certain aspects, the invention is directed to alcoholdehy the S. cerevisiae TMA29 (YMR226C) gene or a homolog drogenases (ADHs), which have been modified to enhance thereof. In some embodiments, the enzyme catalyzing the 50 the enzyme’s ability to convert isobutyraldehyde to isobu conversion of isobutyraldehyde to isobutyrate is an aldehyde tanol. Examples of Such ADHS include enzymes having one dehydrogenase. In a specific embodiment, the aldehyde dehy or more mutations at positions corresponding to amino acids drogenase is encoded by the S. cerevisiae ALD6 gene or a selected from: (a) tyrosine 50 of the L. lactis Adha (SEQID homolog thereof. NO: 185); (b) glutamine 77 of the L. lactis Adha (SEQ ID In one embodiment, the isobutanol producing metabolic 55 NO:185); (c) valine 108 of the L. lactis Adh A (SEQID NO: pathway comprises at least one exogenous gene that catalyzes 185); (d) tyrosine 113 of the L. lactis Adha (SEQ ID NO: a step in the conversion of pyruvate to isobutanol. In another 185); (e) isoleucine 212 of the L. lactis Adh A (SEQ ID NO: embodiment, the isobutanol producing metabolic pathway 185); and (f) leucine 264 of the L. lactis Adha (SEQID NO: comprises at least two exogenous genes that catalyze steps in 185), wherein Adha (SEQID NO:185) is encoded by the L. the conversion of pyruvate to isobutanol. In yet another 60 lactis alcohol dehydrogenase (ADH) gene adhA (SEQ ID embodiment, the isobutanol producing metabolic pathway NO:184) or a codon-optimized version thereof (SEQID NO: comprises at least three exogenous genes that catalyze steps 206). in the conversion of pyruvate to isobutanol. In yet another In one embodiment, the modified ADH enzyme contains a embodiment, the isobutanol producing metabolic pathway mutation at the amino acid corresponding to position 50 of the comprises at least four exogenous genes that catalyze steps in 65 L. lactis Adh A (SEQID NO:185). In another embodiment, the conversion of pyruvate to isobutanol. In yet another the modified ADH enzyme contains a mutation at the amino embodiment, the isobutanol producing metabolic pathway acid corresponding to position 77 of the L. lactis Adha (SEQ US 9,012,189 B2 9 10 ID NO:185). In yet another embodiment, the modified ADH than the L. lactis Adha (SEQ ID NO: 185), which contain enzyme contains a mutation at the amino acid corresponding alterations corresponding to those set out above. Such ADH to position 108 of the L. lactis Adha (SEQID NO: 185). In enzymes may include, but are not limited to, the ADH yet another embodiment, the modified ADH enzyme contains enzymes listed in Table 97. a mutation at the amino acid corresponding to position 113 of 5 In some embodiments, the ADH enzymes to be modified the L. lactis AdhA (SEQID NO:185). In yet another embodi are NADH-dependent ADH enzymes. Examples of such ment, the modified ADH enzyme contains a mutation at the NADH-dependent ADH enzymes are described in commonly amino acid corresponding to position 212 of the L. lactis owned and co-pending U.S. Patent Publication No. 2010/ Adh A (SEQ ID NO: 185). In yet another embodiment, the 0.143997, which is herein incorporated by reference in its modified ADH enzyme contains a mutation at the amino acid 10 entirety for all purposes. In some embodiments, genes origi corresponding to position 264 of the L. lactis Adha (SEQID nally encoding NADPH-utilizing ADH enzymes are modi NO:185). fied to switch the co-factor preference of the enzyme to In one embodiment, the ADH enzyme contains two or NADH. more mutations at the amino acids corresponding to the posi As described herein, the modified ADHs will generally tions described above. In another embodiment, the ADH 15 exhibit an enhanced ability to convert isobutyraldehyde to enzyme contains three or more mutations at the amino acids isobutanol as compared to the wild-type or parental ADH. corresponding to the positions described above. In yet Preferably, the catalytic efficiency (k/K) of the modified another embodiment, the ADH enzyme contains four or more ADH enzyme is enhanced by at least about 5% as compared mutations at the amino acids corresponding to the positions to the wild-type or parental ADH. More preferably, the cata described above. In yet another embodiment, the ADH lytic efficiency of the modified ADH enzyme is enhanced by enzyme contains five or more mutations at the amino acids at least about 15% as compared to the wild-type or parental corresponding to the positions described above. In yet ADH. More preferably, the catalytic efficiency of the modi another embodiment, the ADH enzyme contains six muta fied ADH enzyme is enhanced by at least about 25% as tions at the amino acids corresponding to the positions compared to the wild-type or parental ADH. More preferably, described above. 25 the catalytic efficiency of the modified ADH enzyme is In one specific embodiment, the invention is directed to enhanced by at least about 50% as compared to the wild-type ADH enzymes wherein the tyrosine at position 50 is replaced or parental ADH. More preferably, the catalytic efficiency of with a phenylalanine or tryptophan residue. In another spe the modified ADH enzyme is enhanced by at least about 75% cific embodiment, the invention is directed to ADH enzymes as compared to the wild-type or parental ADH. More prefer wherein the glutamine at position 77 is replaced with an 30 ably, the catalytic efficiency of the modified ADH enzyme is arginine or serine residue. In another specific embodiment, enhanced by at least about 100% as compared to the wild-type the invention is directed to ADH enzymes wherein the valine or parental ADH. More preferably, the catalytic efficiency of at position 108 is replaced with a serine or alanine residue. In the modified ADH enzyme is enhanced by at least about another specific embodiment, the invention is directed to 200% as compared to the wild-type or parental ADH. More ADH enzymes wherein the tyrosine at position 113 is 35 preferably, the catalytic efficiency of the modified ADH replaced with a phenylalanine or glycine residue. In another enzyme is enhanced by at least about 500% as compared to specific embodiment, the invention is directed to ADH the wild-type or parental ADH. More preferably, the catalytic enzymes wherein the isoleucine at position 212 is replaced efficiency of the modified ADH enzyme is enhanced by at with a threonine or valine residue. In yet another specific least about 1000% as compared to the wild-type or parental embodiment, the invention is directed to ADH enzymes 40 ADH. More preferably, the catalytic efficiency of the modi wherein the leucine at position 264 is replaced with a valine fied ADH enzyme is enhanced by at least about 2000% as residue. In one embodiment, the ADH enzyme contains two compared to the wild-type or parental ADH. More preferably, or more mutations at the amino acids corresponding to the the catalytic efficiency of the modified ADH enzyme is positions described in these specific embodiments. In another enhanced by at least about 3000% as compared to the wild embodiment, the ADH enzyme contains three or more muta 45 type or parental ADH. Most preferably, the catalytic effi tions at the amino acids corresponding to the positions ciency of the modified ADH enzyme is enhanced by at least described in these specific embodiments. In yet another about 3500% as compared to the wild-type or parental ADH. embodiment, the ADH enzyme contains four or more muta In additional aspects, the invention is directed to modified tions at the amino acids corresponding to the positions ADH enzymes that have been codon optimized for expression described in these specific embodiments. In yet another 50 in certain desirable host organisms, such as yeast and E. coli. embodiment, the ADH enzyme contains five or more muta In other aspects, the present invention is directed to recom tions at the amino acids corresponding to the positions binant host cells comprising a modified ADH enzyme of the described in these specific embodiments. In yet another invention. According to this aspect, the present invention is embodiment, the ADH enzyme contains six mutations at the also directed to methods of using the modified ADH enzymes amino acids corresponding to the positions described in these 55 in any fermentation process, where the conversion of isobu specific embodiments. tyraldehyde to isobutanol is desired. In one embodiment In certain exemplary embodiments, the ADH enzyme com according to this aspect, the modified ADH enzymes may be prises a sequence selected SEQ ID NO: 189, SEQ ID NO: suitable for enhancing a host cells ability to produce isobu 191, SEQID NO: 193, SEQ ID NO: 195, SEQID NO: 197, tanol. In another embodiment according to this aspect, the SEQID NO: 199, SEQID NO: 201, SEQID NO: 203, SEQ 60 modified ADH enzymes may be suitable for enhancing a host IDNO: 205, SEQIDNO:208, SEQID NO:210, SEQID NO: cells ability to produce 1-propanol. 2-propanol, 1-butanol, 212, SEQID NO: 214, SEQID NO: 216, SEQID NO: 218, 2-butanol. 1-pentanol, 2-methyl-1-butanol, and 3-methyl-1- SEQID NO: 220, SEQID NO: 222, SEQID NO: 224, and butanol. homologs or variants thereof comprising corresponding In various embodiments described herein, the recombinant mutations as compared to the wild-type or parental enzyme. 65 microorganisms comprising a modified ADH may be further As alluded to in the preceding paragraph, further included engineered to express an isobutanol producing metabolic within the scope of the invention are ADH enzymes, other pathway. In one embodiment, the recombinant microorgan US 9,012,189 B2 11 12 ism may be engineered to express an isobutanol producing In some embodiments, the recombinant microorganisms metabolic pathway comprising at least one exogenous gene. may be Crabtree-positive recombinant yeast microorgan In one embodiment, the recombinant microorganism may be isms. In one embodiment, the Crabtree-positive yeast micro engineered to express an isobutanol producing metabolic organism is classified into a genera selected from the group pathway comprising at least two exogenous genes. In another consisting of Saccharomyces, Kluyveromyces, Zygosaccha embodiment, the recombinant microorganism may be engi romyces, Debaryomyces, Candida, Pichia and Schizosaccha neered to express an isobutanol producing metabolic pathway romyces. In additional embodiments, the Crabtree-positive comprising at least three exogenous genes. In another yeast microorganism is selected from the group consisting of embodiment, the recombinant microorganism may be engi Saccharomyces cerevisiae, Saccharomyces uvarum, Saccha neered to express an isobutanol producing metabolic pathway 10 romyces bayanus, Saccharomyces paradoxus, Saccharomy comprising at least four exogenous genes. In another embodi ces castelli, Kluyveromyces thermotolerans, Candida gla ment, the recombinant microorganism may be engineered to brata, Z. bailli, Z. rouxii, Debaryomyces hansenii, Pichia express an isobutanol producing metabolic pathway compris pastorius, Schizosaccharomyces pombe, and Saccharomyces ing five exogenous genes. Thus, the present invention further iOFiFi. provides recombinant microorganisms that comprise an 15 In some embodiments, the recombinant microorganisms isobutanol producing metabolic pathway and methods of may be post-WGD (whole genome duplication) yeast recom using said recombinant microorganisms to produce isobu binant microorganisms. In one embodiment, the post-WGD tanol. yeast recombinant microorganism is classified into a genera In various embodiments described herein, the isobutanol selected from the group consisting of Saccharomyces or Can pathway enzyme(s) is/are selected from acetolactate synthase dida. In additional embodiments, the post-WGD yeast is (ALS), ketol-acid reductoisomerase (KARI), dihydroxyacid selected from the group consisting of Saccharomyces cerevi dehydratase (DHAD), 2-keto-acid decarboxylase (KIVD). siae, Saccharomyces uvarum, Saccharomyces bavanus, Sac and alcohol dehydrogenase (ADH). charomyces paradoxus, Saccharomyces castelli, and Can In various embodiments described herein, the isobutanol dida glabrata. pathway enzymes may be derived from a prokaryotic organ 25 In some embodiments, the recombinant microorganisms ism. In alternative embodiments described herein, the isobu may be pre-WGD (whole genome duplication) yeast recom tanol pathway enzymes may be derived from a eukaryotic binant microorganisms. In one embodiment, the pre-WGD organism. An exemplary metabolic pathway that converts yeast recombinant microorganism is classified into a genera pyruvate to isobutanol may be comprised of a acetohydroxy selected from the group consisting of Saccharomyces, acid synthase (ALS) enzyme encoded by, for example, alsS 30 Kluyveromyces, Candida, Pichia, Issatchenkia, Debaryomy from B. subtilis, a ketol-acid reductoisomerase (KARI) ces, Hansenula, Pachysolen, Yarrowia and Schizosaccharo encoded by, for example ilvC from E. coli, a dihydroxy-acid myces. In additional embodiments, the pre-WGD yeast is dehydratase (DHAD), encoded by, for example, ilvD from L. selected from the group consisting of Saccharomyces lactis, a 2-keto-acid decarboxylase (KIVD) encoded by, for kluyveri, Kluyveromyces thermotolerans, Kluyveromyces example kiv) from L. lactis, and an alcohol dehydrogenase 35 marxianus, Kluyveromyces waltii, Kluyveromyces lactis, (ADH) (e.g. a modified ADH described herein), encoded by, Candida tropicalis, Pichia pastoris, Pichia anomala, Pichia for example, adhA from L. lactis with one or more mutations stipitis, Issatchenkia Orientalis, Issatchenkia Occidentalis, at positions Y50, Q77, V108, Y113, I212, and L264 as Debaryomyces hansenii, Hansenula anomala, Pachysolen described herein. tannophilis, Yarrowia lipolytica, and Schizosaccharomyces In various embodiments described herein, the recombinant 40 pombe. microorganisms may be microorganisms of the Saccharomy In some embodiments, the recombinant microorganisms ces clade, Saccharomyces sensu stricto microorganisms, may be microorganisms that are non-fermenting yeast micro Crabtree-negative yeast microorganisms, Crabtree-positive organisms, including, but not limited to those, classified into yeast microorganisms, post-WGD (whole genome duplica a genera selected from the group consisting of Tricosporon, tion) yeast microorganisms, pre-WGD (whole genome dupli 45 Rhodotorula, Myxozyma, or Candida. In a specific embodi cation) yeast microorganisms, and non-fermenting yeast ment, the non-fermenting yeast is C. xestobii. microorganisms. In another aspect, the present invention provides methods In some embodiments, the recombinant microorganisms of producing beneficial metabolites including fuels, chemi may be yeast recombinant microorganisms of the Saccharo cals, and amino acids using a recombinant microorganism as myces clade. 50 described herein. In one embodiment, the method includes In some embodiments, the recombinant microorganisms cultivating the recombinant microorganism in a culture may be Saccharomyces sensu stricto microorganisms. In one medium containing a feedstock providing the carbon source embodiment, the Saccharomyces sensu stricto is selected until a recoverable quantity of the metabolite is produced and from the group consisting of S. cerevisiae, S. kudriavZevi, S. optionally, recovering the metabolite. In one embodiment, the mikatae, S. bayanus, S. uvarum. S. carocanis and hybrids 55 microorganism produces the metabolite from a carbon Source thereof. at a yield of at least about 5 percent theoretical. In another In some embodiments, the recombinant microorganisms embodiment, the microorganism produces the metabolite at a may be Crabtree-negative recombinant yeast microorgan yield of at least about 10 percent, at least about 15 percent, isms. In one embodiment, the Crabtree-negative yeast micro about least about 20 percent, at least about 25 percent, at least organism is classified into a genera selected from the group 60 about 30 percent, at least about 35 percent, at least about 40 consisting of Saccharomyces, Kluyveromyces, Pichia, percent, at least about 45 percent, at least about 50 percent, at Issatchenkia, Hansenula, or Candida. In additional embodi least about 55 percent, at least about 60 percent, at least about ments, the Crabtree-negative yeast microorganism is selected 65 percent, at least about 70 percent, at least about 75 percent, from Saccharomyces kluyveri, Kluyveromyces lactis, at least about 80 percent, at least about 85 percent, at least Kluyveromyces marxianus, Pichia anomala, Pichia stipitis, 65 about 90 percent, at least about 95 percent, or at least about Hansenula anomala, Candida utilis and Kluyveromyces 97.5 percent theoretical. In one embodiment, the metabolite waltii. may be derived from a biosynthetic pathway which uses a US 9,012,189 B2 13 14 3-ketoacid as an intermediate. In one embodiment, the 3-keto FIG. 10 illustrates a 1H-COSY spectrum of the peak iso acid intermediate is acetolactate. Accordingly, the metabolite lated from LC1. The spectrum indicates that DH2MB methyl may be derived from a biosynthetic pathway which uses protons (doublet) at 0.95 ppm are coupled to methine proton acetolactate as an intermediate, including, but not limited to, (quartet) at 3.7 ppm. isobutanol, 2-butanol, 1-butanol, 2-butanone, 2,3-butanediol. 5 FIG. 11 illustrates a 1H-NMR spectrum of the peak iso acetoin, diacetyl, Valine, leucine, pantothenic acid, isobuty lated from LC1. The spectrum indicates the presence of lene, 3-methyl-1-butanol, 4-methyl-1-pentanol, and coen DH2MB: a singlet of methyl protons (a) at 1.2 ppm with Zyme A. In another embodiment, the 3-keto acid intermediate integral value 3, a doublet of methyl protons (b) at 0.95 ppm is 2-aceto-2-hydroxybutyrate. Accordingly, the metabolite with integral value 3 and a quartet of methine proton (c) at 3.7 may be derived from a biosynthetic pathway which uses 10 ppm with integral value of 1.84. Integral value of methine 2-aceto-2-hydroxybutyrate as an intermediate, including, but proton (c) is greater than 1 due to overlap with glucose reso not limited to, 2-methyl-1-butanol, isoleucine, 3-methyl-1- nance in the same region. pentanol, 4-methyl-1-hexanol, and 5-methyl-1-heptanol. In FIG. 12 illustrates a LC-MS analysis of the peak isolated another embodiment, the metabolite may be derived from a 15 from LC1. Several molecular ions were identified in the biosynthetic pathway which uses an aldehyde as an interme sample as indicated at the top portion of the figure. Further diate, including, but not limited to, isobutanol, 1-butanol, fragmentation (MS2) of 134 molecular ion indicated that 2-methyl-1-butanol, 3-methyl-1-butanol. 1-propanol. 1-pen isolated LC1 fraction contains hydroxyl carboxylic acid by tanol, 1-hexanol, 3-methyl-1-pentanol, 4-methyl-1-pentanol, characteristic loss of CO (*) and H2O+CO (**). 4-methyl-1-hexanol, and 5-methyl-1-heptanol. In yet another FIG. 13 illustrates the diastereomeric and enantiomeric embodiment, the metabolite may be derived from a biosyn structures of 2,3-dihydroxy-2-methylbutanoic acid (2R.3S)- thetic pathway which uses acetolactate and an aldehyde as 1a, (2S,3R)-1b, (2R,3R)-2a, (2S,3S)-2b. intermediates, including, but not limited to, isobutanol, 1-bu FIG. 14 illustrates the 1H spectrum of crystallized DH2MB tanol, and 3-methyl-1-butanol biosynthetic pathways. In yet in D.O.'HNMR (TSP) 1.1 (d. 6.5 Hz,3H), 1.3 (s.3H), 3.9 (q, another embodiment, the metabolite may be derived from a 25 6.5 Hz, 3H) biosynthetic pathway which uses 2-aceto-2-hydroxybutyrate FIG. 15 illustrates the 13C spectrum of crystallized and an aldehyde as intermediates, including, but not limited DH2MB in D.O.The spectrum indicates five different carbon to, 2-methyl-1-butanol, 3-methyl-1-pentanol, 4-methyl-1- resonances one of them being characteristic carboxylic acid hexanol, and 5-methyl-1-heptanol biosynthetic pathways. resonance at 181 ppm. In one embodiment, the recombinant microorganism is FIG. 16 illustrates the fermentation profile of isobutanol grown under aerobic conditions. In another embodiment, the and by-products from a single fermentation with GEVO3160. recombinant microorganism is grown under microaerobic Production aeration was reduced from an OTR of 0.8 mM/h to 0.3 mM/h at 93 h post inoculation. Open diamond-iBuOH, conditions. In yet another embodiment, the recombinant square-unknown quantified as DH2MB, asterisk=cell dry microorganism is grown under anaerobic conditions. 35 weight (cdw), and closed triangle total byproducts. FIG. 17 illustrates a structural alignment of the L. lactis BRIEF DESCRIPTION OF DRAWINGS Adha amino acid sequence with the structure of G. Stearo thermophilus (Pymol). Active site mutations are shown Illustrative embodiments of the invention are illustrated in (Y50F and L264V). Mutations in the co-factor binding the drawings, in which: 40 domain are also shown (I212T and N219Y). FIG. 1 illustrates an exemplary embodiment of an isobu FIG. 18 illustrates biosynthetic pathways utilizing aceto tanol pathway. lactate as an intermediate. Biosynthetic pathways for the pro FIG. 2 illustrates exemplary reactions capable of being duction of 1-butanol, isobutanol, 3-methyl-1-butanol, and catalyzed by 3-ketoacid reductases. 4-methyl-1-pentanol use both acetolactate and an aldehyde as FIG. 3 illustrates a non-limiting list of exemplary 3-ke 45 an intermediate. toacid reductases and their corresponding enzyme classifica FIG. 19 illustrates biosynthetic pathways utilizing 2-aceto tion numbers. 2-hydroxybutyrate as an intermediate. Biosynthetic path FIG. 4 illustrates exemplary reactions capable of being ways for the production of 2-methyl-1-butanol, 3-methyl-1- catalyzed by aldehyde dehydrogenases. pentanol, 4-methyl-1-hexanol, and 5-methyl-1-heptanol use FIG. 5 illustrates a strategy for reducing the production of 50 both 2-aceto-2-hydroxybutyrate and an aldehyde as an inter DH2MB and isobutyrate in isobutanol-producing recombi mediate. nant microorganisms. FIG. 20 illustrates additional biosynthetic pathways utiliz FIG. 6 illustrates a strategy for reducing the production of ing an aldehyde as an intermediate. DH2MB and 3-methyl-1-butyrate in 3-methyl-1-butanol producing recombinant microorganisms. 55 DETAILED DESCRIPTION FIG. 7 illustrates a strategy for reducing the production of 2-ethyl-2,3-dihydroxybutyrate and 2-methyl-1-butyrate in As used herein and in the appended claims, the singular 2-methyl-1-butanol producing recombinant microorganisms. forms “a,” “an and “the include plural referents unless the FIG. 8 illustrates a stacked overlay of LC4 chromatograms context clearly dictates otherwise. Thus, for example, refer showing a sample containing DH2MB and acetate (top) and a 60 ence to “a polynucleotide' includes a plurality of such poly sample containing acetate and DHIV (bottom). Elution order: nucleotides and reference to “the microorganism' includes DH2MB followed by acetate (top); lactate, acetate, DHIV. reference to one or more microorganisms, and so forth. isobutyrate, pyruvate (bottom). Unless defined otherwise, all technical and scientific terms FIG. 9 illustrates a chromatogram for sample fraction col used herein have the same meaning as commonly understood lected at retention time corresponding to DHIV collected on 65 to one of ordinary skill in the art to which this disclosure LC1 and analyzed by LC4 on an AS-11 Column with Con belongs. Although methods and materials similar or equiva ductivity Detection. lent to those described herein can be used in the practice of the US 9,012,189 B2 15 16 disclosed methods and compositions, the exemplary meth identical to the parent cell, but are still included within the ods, devices and materials are described herein. Scope of the term as used herein. Any publications discussed above and throughout the text The term “overexpression” refers to an elevated level (e.g., are provided solely for their disclosure prior to the filing date aberrant level) of mRNAS encoding for a protein(s) (e.g., an of the present application. Nothing herein is to be construed 5 TMA29 protein or homolog thereof), and/or to elevated levels as an admission that the inventors are not entitled to antedate of protein(s) (e.g., TMA29) in cells as compared to similar such disclosure by virtue of prior disclosure. corresponding unmodified cells expressing basal levels of The term “microorganism’ includes prokaryotic and mRNAS (e.g., those encoding Aft proteins) or having basal eukaryotic microbial species from the Domains Archaea, levels of proteins. In particular embodiments, TMA29, or Bacteria and Eucarya, the latter including yeast and filamen 10 homologs thereof, may be overexpressed by at least 2-fold, tous fungi, protozoa, algae, or higher Protista. The terms 3-fold, 4-fold, 5-fold, 6-fold, 8-fold, 10-fold, 12-fold, 15-fold “microbial cells' and “microbes’ are used interchangeably or more in microorganisms engineered to exhibit increased with the term microorganism. TMA29 mRNA, protein, and/or activity. The term "genus is defined as a taxonomic group of As used herein and as would be understood by one of related species according to the Taxonomic Outline of Bac 15 ordinary skill in the art, “reduced activity and/or expression' teria and Archaea (Garrity, G.M.Lilburn, T.G., Cole, JR. of a protein Such as an enzyme can mean either a reduced Harrison, S.H., Euzehy, J., and Tindall, B.J. (2007) The Taxo specific catalytic activity of the protein (e.g. reduced activity) nomic Outline of Bacteria and Archaca. TOBA Release 7.7, and/or decreased concentrations of the protein in the cell (e.g. Mar. 2007. Michigan State University Board of Trustees. reduced expression), while “deleted activity and/or expres The term “species” is defined as a collection of closely sion” or “eliminated activity and/or expression of a protein related organisms with greater than 97% 16S ribosomal RNA Such as an enzyme can mean either no or negligible specific sequence homology and greater than 70% genomic hybrid catalytic activity of the enzyme (e.g. deleted activity) and/or ization and sufficiently different from all other organisms so no or negligible concentrations of the enzyme in the cell (e.g. as to be recognized as a distinct unit. deleted expression). The terms “recombinant microorganism.” “modified 25 The term “wild-type microorganism’ describes a cell that microorganism,” and “recombinant host cell are used inter occurs in nature, i.e. a cell that has not been genetically changeably herein and refer to microorganisms that have modified. A wild-type microorganism can be genetically been genetically modified to express or overexpress endog modified to express or overexpress a first target enzyme. This enous polynucleotides, to express heterologous polynucle microorganism can act as a parental microorganism in the otides, such as those included in a vector, in an integration 30 generation of a microorganism modified to express or over construct, or which have an alteration in expression of an express a second target enzyme. In turn, the microorganism endogenous gene. By “alteration’ it is meant that the expres modified to express or overexpress a first and a second target sion of the gene, or level of a RNA molecule or equivalent enzyme can be modified to express or overexpress a third RNA molecules encoding one or more polypeptides or target enzyme. polypeptide subunits, or activity of one or more polypeptides 35 Accordingly, a "parental microorganism' functions as a or polypeptide Subunits is up regulated or down regulated, reference cell for Successive genetic modification events. Such that expression, level, or activity is greater than or less Each modification event can be accomplished by introducing than that observed in the absence of the alteration. For a nucleic acid molecule into the reference cell. The introduc example, the term “alter can mean “inhibit, but the use of tion facilitates the expression or overexpression of a target the word “alter is not limited to this definition. 40 enzyme. It is understood that the term “facilitates' encom The term "expression' with respect to a gene sequence passes the activation of endogenous polynucleotides encod refers to transcription of the gene and, as appropriate, trans ing a target enzyme through genetic modification of e.g., a lation of the resulting mRNA transcript to a protein. Thus, as promoter sequence in a parental microorganism. It is further will be clear from the context, expression of a protein results understood that the term “facilitates' encompasses the intro from transcription and translation of the open reading frame 45 duction of heterologous polynucleotides encoding a target sequence. The level of expression of a desired product in a enzyme in to a parental microorganism host cell may be determined on the basis of either the amount The term “engineer” refers to any manipulation of a micro of corresponding mRNA that is present in the cell, or the organism that results in a detectable change in the microor amount of the desired product encoded by the selected ganism, wherein the manipulation includes but is not limited sequence. For example, mRNA transcribed from a selected 50 to inserting a polynucleotide and/or polypeptide heterolo sequence can be quantitated by qRT-PCR or by Northern gous to the microorganism and mutating a polynucleotide hybridization (see Sambrook et al., Molecular Cloning: A and/or polypeptide native to the microorganism. Laboratory Manual, Cold Spring Harbor Laboratory Press The term “mutation” as used herein indicates any modifi (1989)). Protein encoded by a selected sequence can be quan cation of a nucleic acid and/or polypeptide which results in an titated by various methods, e.g., by ELISA, by assaying for 55 altered nucleic acid or polypeptide. Mutations include, for the biological activity of the protein, or by employing assays example, point mutations, deletions, or insertions of single or that are independent of Such activity. Such as western blotting multiple residues in a polynucleotide, which includes alter or radioimmunoassay, using antibodies that recognize and ations arising within a protein-encoding region of a gene as bind the protein. See Sambrook et al., 1989, supra. The poly well as alterations in regions outside of a protein-encoding nucleotide generally encodes a target enzyme involved in a 60 sequence. Such as, but not limited to, regulatory or promoter metabolic pathway for producing a desired metabolite. It is sequences. A genetic alteration may be a mutation of any understood that the terms "recombinant microorganism' and type. For instance, the mutation may constitute a point muta “recombinant host cell refer not only to the particular recom tion, a frame-shift mutation, a nonsense mutation, an inser binant microorganism but to the progeny or potential progeny tion, or a deletion of part or all of a gene. In addition, in some of such a microorganism. Because certain modifications may 65 embodiments of the modified microorganism, a portion of the occur in Succeeding generations due to either mutation or microorganism genome has been replaced with a heterolo environmental influences, such progeny may not, in fact, be gous polynucleotide. In some embodiments, the mutations US 9,012,189 B2 17 18 are naturally-occurring. In other embodiments, the mutations The term “specific productivity” or “specific production are identified and/or enriched through artificial selection rate' is defined as the amount of product formed per volume pressure. In still other embodiments, the mutations in the of medium per unit of time per amount of cells. Specific microorganism genome are the result of genetic engineering. productivity is reported ingram or milligram per literper hour The term “biosynthetic pathway', also referred to as per OD (g/L/h/OD). “metabolic pathway', refers to a set of anabolic or catabolic The term "yield' is defined as the amount of product biochemical reactions for converting one chemical species obtained per unit weight of raw material and may be into another. Gene products belong to the same “metabolic expressed as g product per g substrate (g/g). Yield may be pathway’ if they, in parallel or in series, act on the same expressed as a percentage of the theoretical yield. “Theoreti Substrate, produce the same product, or act on or produce a 10 cal yield' is defined as the maximum amount of product that metabolic intermediate (i.e., metabolite) between the same can be generated per a given amount of Substrate as dictated substrate and metabolite end product. by the stoichiometry of the metabolic pathway used to make As used herein, the term "isobutanol producing metabolic the product. For example, the theoretical yield for one typical pathway refers to an enzyme pathway which produces 15 conversion of glucose to isobutanol is 0.41 g/g. As such, a isobutanol from pyruvate. yield of isobutanol from glucose of 0.39 g/g would be The term "heterologous' as used herein with reference to expressed as 95% of theoretical or 95% theoretical yield. molecules and in particular enzymes and polynucleotides, The term “titer' is defined as the strength of a solution or indicates molecules that are expressed in an organism other the concentration of a Substance in Solution. For example, the than the organism from which they originated or are found in titer of a biofuel in a fermentation broth is described as g of nature, independently of the level of expression that can be biofuel in solution per liter of fermentation broth (g/L). lower, equal or higher than the level of expression of the “Aerobic conditions' are defined as conditions under molecule in the native microorganism. which the oxygen concentration in the fermentation medium On the other hand, the term “native' or “endogenous” as is sufficiently high for an aerobic or facultative anaerobic used herein with reference to molecules, and in particular 25 microorganism to use as a terminal electron acceptor. enzymes and polynucleotides, indicates molecules that are In contrast, "anaerobic conditions' are defined as condi expressed in the organism in which they originated or are tions under which the oxygen concentration in the fermenta found in nature, independently of the level of expression that tion medium is too low for the microorganism to use as a can be lower equal or higher than the level of expression of the terminal electron acceptor. Anaerobic conditions may be molecule in the native microorganism. It is understood that 30 expression of native enzymes or polynucleotides may be achieved by sparging a fermentation medium with an inert gas modified in recombinant microorganisms. such as nitrogen until oxygen is no longer available to the The term “feedstock' is defined as a raw material or mix microorganism as a terminal electron acceptor. Alternatively, ture of raw materials Supplied to a microorganism or fermen anaerobic conditions may be achieved by the microorganism tation process from which other products can be made. For 35 consuming the available oxygen of the fermentation until example, a carbon Source. Such as biomass or the carbon oxygen is unavailable to the microorganism as a terminal compounds derived from biomass area feedstock for a micro electron acceptor. Methods for the production of isobutanol organism that produces a biofuel in a fermentation process. under anaerobic conditions are described in commonly However, a feedstock may contain nutrients other than a owned and co-pending publication, US 2010/0143997, the carbon Source. 40 disclosures of which are herein incorporated by reference in The term “substrate' or "suitable substrate” refers to any its entirety for all purposes. Substance or compound that is converted or meant to be “Aerobic metabolism” refers to a biochemical process in converted into another compound by the action of an enzyme. which oxygen is used as a terminal electron acceptor to make The term includes not only a single compound, but also com energy, typically in the form of ATP from carbohydrates. binations of compounds, such as Solutions, mixtures and 45 Aerobic metabolism occurs e.g. via glycolysis and the TCA other materials which contain at least one substrate, orderiva cycle, wherein a single glucose molecule is metabolized com tives thereof. Further, the term “substrate' encompasses not pletely into carbon dioxide in the presence of oxygen. only compounds that provide a carbon Source Suitable for use In contrast, "anaerobic metabolism' refers to a biochemi as a starting material. Such as any biomass derived Sugar, but cal process in which oxygen is not the final acceptor of elec also intermediate and end product metabolites used in a path 50 trons contained in NADH. Anaerobic metabolism can be way associated with a recombinant microorganism as divided into anaerobic respiration, in which compounds other described herein. than oxygen serve as the terminal electron acceptor, and The term "C2-compound as used as a carbon source for substrate level phosphorylation, in which the electrons from engineered yeast microorganisms with mutations in all pyru NADH are utilized to generate a reduced product via a “fer vate decarboxylase (PDC) genes resulting in a reduction of 55 mentative pathway.” pyruvate decarboxylase activity of said genes refers to In “fermentative pathways”, NAD(P)H donates its elec organic compounds comprised of two carbon atoms, includ trons to a molecule produced by the same metabolic pathway ing but not limited to ethanol and acetate. that produced the electrons carried in NAD(P)H. For The term “fermentation' or “fermentation process” is example, in one of the fermentative pathways of certain yeast defined as a process in which a microorganism is cultivated in 60 strains, NAD(P)H generated through glycolysis transfers its a culture medium containing raw materials, such as feedstock electrons to pyruvate, yielding ethanol. Fermentative path and nutrients, wherein the microorganism converts raw mate ways are usually active under anaerobic conditions but may rials, such as a feedstock, into products. also occur under aerobic conditions, under conditions where The term “volumetric productivity” or “production rate” is NADH is not fully oxidized via the respiratory chain. For defined as the amount of product formed per volume of 65 example, above certain glucose concentrations, Crabtree medium per unit of time. Volumetric productivity is reported positive yeasts produce large amounts of ethanol under aero in gram per liter per hour (g/L/h). bic conditions. US 9,012,189 B2 19 20 The term “byproduct' or “by-product” means an undesired decreased, or eliminated) by modifying the common pro product related to the production of an amino acid, amino acid moter. Alternatively, any gene or combination of genes in an precursor, chemical, chemical precursor, biofuel, or biofuel operon can be modified to alter the function or activity of the precursor. encoded polypeptide. The modification can result in an The term “substantially free” when used in reference to the increase in the activity of the encoded polypeptide. Further, presence or absence of enzymatic activities (3-KAR, ALDH, the modification can impart new activities on the encoded PDC, GPD, etc.) in carbon pathways that compete with the polypeptide. Exemplary new activities include the use of desired metabolic pathway (e.g., an isobutanol-producing alternative substrates and/or the ability to function in alterna tive environmental conditions. metabolic pathway) means the level of the enzyme is Substan A “vector” is any means by which a nucleic acid can be tially less than that of the same enzyme in the wild-type host, 10 propagated and/or transferred between organisms, cells, or wherein less than about 50% of the wild-type level is pre cellular components. Vectors include viruses, bacteriophage, ferred and less than about 30% is more preferred. The activity pro-viruses, plasmids, phagemids, transposons, and artificial may be less than about 20%, less than about 10%, less than chromosomes such as YACs (yeast artificial chromosomes), about 5%, or less than about 1% of wild-type activity. Micro BACs (bacterial artificial chromosomes), and PLACs (plant organisms which are “substantially free” of a particular enzy 15 artificial chromosomes), and the like, that are “episomes.” matic activity (3-KAR, ALDH, PDC, GPD, etc.) may be that is, that replicate autonomously or can integrate into a created through recombinant means or identified in nature. chromosome of a host cell. A vector can also be a naked RNA The term “non-fermenting yeast’ is a yeast species that polynucleotide, a naked DNA polynucleotide, a polynucle fails to demonstrate an anaerobic metabolism in which the otide composed of both DNA and RNA within the same electrons from NADH are utilized to generate a reduced prod Strand, a poly-lysine-conjugated DNA or RNA, a peptide uct via a fermentative pathway such as the production of conjugated DNA or RNA, a liposome-conjugated DNA, or ethanol and CO2 from glucose. Non-fermentative yeast can the like, that are not episomal in nature, or it can be an be identified by the “Durham Tube Test” (J. A. Barnett, R. W. organism which comprises one or more of the above poly Payne, and D. Yarrow. 2000. Yeasts Characteristics and Iden nucleotide constructs such as an agrobacterium or a bacte tification. 3" edition. p. 28-29. Cambridge University Press, 25 rium. Cambridge, UK) or by monitoring the production of fermen “Transformation” refers to the process by which a vector is tation productions such as ethanol and CO. introduced into a host cell. Transformation (or transduction, The term “polynucleotide' is used herein interchangeably or transfection), can be achieved by any one of a number of means including chemical transformation (e.g. lithium with the term “nucleic acid and refers to an organic polymer acetate transformation), electroporation, microinjection, composed of two or more monomers including nucleotides, 30 biolistics (or particle bombardment-mediated delivery), or nucleosides or analogs thereof, including but not limited to agrobacterium mediated transformation. single stranded or double stranded, sense or antisense deox The term “enzyme” as used herein refers to any substance yribonucleic acid (DNA) of any length and, where appropri that catalyzes or promotes one or more chemical or biochemi ate, single Stranded or double stranded, sense or antisense cal reactions, which usually includes enzymes totally or par ribonucleic acid (RNA) of any length, including siRNA. The 35 tially composed of a polypeptide, but can include enzymes term “nucleotide' refers to any of several compounds that composed of a different molecule including polynucleotides. consist of a ribose or deoxyribose Sugar joined to a purine or The term “protein,” “peptide.” or “polypeptide' as used a pyrimidine base and to a phosphate group, and that are the herein indicates an organic polymer composed of two or more basic structural units of nucleic acids. The term “nucleoside' amino acidic monomers and/or analogs thereof. As used refers to a compound (as guanosine or adenosine) that con 40 herein, the term "amino acid' or "amino acidic monomer sists of a purine or pyrimidine base combined with deoxyri refers to any natural and/or synthetic amino acids including glycine and both D or L optical isomers. The term “amino bose or ribose and is found especially in nucleic acids. The acid analog refers to an amino acid in which one or more term “nucleotide analog or “nucleoside analog refers, individual atoms have been replaced, either with a different respectively, to a nucleotide or nucleoside in which one or atom, or with a different functional group. Accordingly, the more individual atoms have been replaced with a different 45 term polypeptide includes amino acidic polymer of any atom or with a different functional group. Accordingly, the length including full length proteins, and peptides as well as term polynucleotide includes nucleic acids of any length, analogs and fragments thereof. A polypeptide of three or DNA, RNA, analogs and fragments thereof. A polynucleotide more amino acids is also called a protein oligomer or oli of three or more nucleotides is also called nucleotidic oligo gopeptide mer or oligonucleotide. 50 The term “homolog used with respect to an original enzyme or gene of a first family or species, refers to distinct It is understood that the polynucleotides described herein enzymes or genes of a second family or species which are include “genes' and that the nucleic acid molecules described determined by functional, structural or genomic analyses to herein include “vectors' or “plasmids.” Accordingly, the term be an enzyme or gene of the second family or species which “gene’’, also called a “structural gene' refers to a polynucle corresponds to the original enzyme or gene of the first family otide that codes for a particular sequence of amino acids, 55 or species. Most often, homologs will have functional, struc which comprise all or part of one or more proteins or tural or genomic similarities. Techniques are known by which enzymes, and may include regulatory (non-transcribed) DNA homologs of an enzyme or gene can readily be cloned using sequences. Such as promoter sequences, which determine for genetic probes and PCR. Identity of cloned sequences as example the conditions under which the gene is expressed. homolog can be confirmed using functional assays and/or by The transcribed region of the gene may include untranslated 60 genomic mapping of the genes. regions, including introns, 5'-untranslated region (UTR), and A protein has "homology' or is "homologous' to a second 3'-UTR, as well as the coding sequence. protein if the amino acid sequence encoded by a gene has a The term “operon” refers to two or more genes which are similar amino acid sequence to that of the second gene. Alter transcribed as a single transcriptional unit from a common natively, a protein has homology to a second protein if the two promoter. In some embodiments, the genes comprising the 65 proteins have “similar amino acid sequences. (Thus, the operon are contiguous genes. It is understood that transcrip term “homologous proteins’ is defined to mean that the two tion of an entire operon can be modified (i.e., increased, proteins have similar amino acid sequences). US 9,012,189 B2 21 22 The term “analog or “analogous' refers to nucleic acid or Each of the biosynthetic pathways listed in Table 1 shares protein sequences or protein structures that are related to one the common 3-keto acid intermediate, acetolactate. There another in function only and are not from common descent or fore, the product yield from these biosynthetic pathways will do not share a common ancestral sequence. Analogs may in part depend upon the amount of acetolactate that is avail differ in sequence but may share a similar structure, due to able to downstream enzymes of said biosynthetic pathways. convergent evolution. For example, two enzymes are analogs Another example of a 3-keto acid which is common to or analogous if the enzymes catalyze the same reaction of many biosynthetic pathways is 2-aceto-2-hydroxybutyrate, which is formed from pyruvate and 2-ketobutyrate by the conversion of a substrate to a product, are unrelated in action of the enzyme acetolactate synthase (also known as sequence, and irrespective of whether the two enzymes are acetohydroxy acid synthase). Amongst the biosynthetic path related in structure. 10 ways using 2-aceto-2-hydroxybutyrate as intermediate Recombinant Microorganisms with Reduced By-Product include pathways for the production of 2-methyl-1-butanol, Accumulation isoleucine, 3-methyl-1-pentanol, 4-methyl-1-hexanol, and Yeast cells convert Sugars to produce pyruvate, which is 5-methyl-1-heptanol. Engineered biosynthetic pathways for then utilized in a number of pathways of cellular metabolism. the synthesis of these beneficial 2-aceto-2-hydroxybutyrate In recent years, yeast cells have been engineered to produce a 15 derived metabolites are found in Table 2 and FIG. 19. number of desirable products via pyruvate-driven biosyn thetic pathways. In many of these biosynthetic pathways, the TABLE 2 initial pathway step is the conversion of endogenous pyruvate to a 3-keto acid. Biosynthetic Pathways Utilizing 2-Aceto-2-Hydroxybutyrate as an As used herein, a "3-keto acid refers to an organic com Intermediate pound which contains a carboxylic acid moiety on the C1 Biosynthetic Pathway Reference carbon and a ketone moiety on the C3 carbon. For example, acetolactate and 2-hydroxy-2-methyl-3-oxobutanoic acid are 2-Methyl-1-Butanol WO/2008/098227 (Liao et al.), WO/2009/076480 (Picataggio et al.), 3-keto acids with a ketone group at the C3 carbon (See, e.g., and Atsumi et al., 2008, Nature 451: 86-89 FIG. 2). 25 Isoleucine McCourt et al., 2006, Amino Acids 31: 173-210 An example of a 3-keto acid which is common to many 3-Methyl-1-Pentanol WO/2010/045629 (Liao et al.), Zhang et al., biosynthetic pathways is acetolactate, which is formed from 2008, Proc Nati AcadSci USA 105: 2O653-20658 pyruvate by the action of the enzyme acetolactate synthase 4-Methyl-1-Hexanol W WO/2010/045629 (Liao et al.), Zhang et al., (also known as acetohydroxy acid synthase). Amongst the 2008, Proc Nati AcadSci USA 105: biosynthetic pathways using acetolactate as intermediate 30 2O653-20658 include pathways for the production of isobutanol, 2-butanol, 5-Methyl-1-Heptanol WO/2010/045629 (Liao et al.), Zhang et al., 1-butanol, 2-butanone, 2,3-butanediol, acetoin, diacetyl, 2008, Proc Natl AcadSci USA 105; Valine, leucine, pantothenic acid, isobutylene, 3-methyl-1- 2O653-20658 butanol, 4-methyl-1-pentanol, and coenzyme A. Engineered The contents of each of the references in this table are herein incorporated by reference in their entireties for all purposes. biosynthetic pathways for the synthesis of these beneficial 35 acetolactate-derived metabolites are found in Table 1 and Each of the biosynthetic pathways listed in Table 2 shares FIG. 18. the common 3-keto acid intermediate, 2-aceto-2-hydroxybu TABLE 1. Biosynthetic Pathways Utilizing Acetolactate as an Intermediate Biosynthetic Pathway Reference' Isobutanol US 2009/0226991 (Feldman et al.), US 2011/0020889 (Feldman et al.), and US 2010/0143997 (Buelter et al.) 1-Butanol WO/2010/017230 (Lynch), WO 2010/031772 (Wu et al.), and KR2011 002130 (Lee et al.) 2-Butanol WO/2007/130518 (Donaldson et al.), WO/2007/130521 (Donaldson et al.), and WO 2009/134276 (Donaldson et al.) 2-Butanone WO/2007/130518 (Donaldson et al.), WO/2007/130521 (Donaldson et al.), and WO 2009/134276 (Donaldson et al.) 2-3-Butanediol WO/2007/130518 (Donaldson et al.), WO/2007/130521 (Donaldson et al.), and WO 2009/134276 (Donaldson et al.) Acetoin WO/2007/130518 (Donaldson et al.), WO/2007/130521 (Donaldson et al.), and WO 2009/134276 (Donaldson et al.) Diacetyl Gonzalez et al., 2000, J. Biol. Chem 275:35876-85 and Ehsani et al., 2009, App. Environ. Micro. 75:3196-205 Valine WO/2001/021772 (Yocum et al.) and McCourt et al., 2006, Amino Acids 31: 173-210 Leucine WO/2001/021772 (Yocum et al.) and McCourt et al., 2006, Amino Acids 31: 173-210 Pantothenic Acid WO/2001/021772 (Yocum et al.) 3-Methyl-1-Butanol WO/2008/098227 (Liao et al.), Atsumi et al., 2008, Nature 451:86-89, and Connor et al., 2008, Appi. Environ. Microbiol. 74:5769-5775 4-Methyl-1-Pentanol WO/2010/045629 (Liao et al.), Zhang et al., 2008, Proc Natl Acad Sci USA 1 OS: 206S3-20658 Coenzyme A WO/2001/021772 (Yocum et al.) The contents of each of the references in this table are herein incorporated by reference in their entireties for all purposes, US 9,012,189 B2 23 24 tyrate. Therefore, the product yield from these biosynthetic including, but not limited to, those listed in Tables 1-3. The pathways will in part depend upon the amount of acetolactate present inventors have found that Suppressing these newly that is available to downstream enzymes of said biosynthetic characterized enzymatic activities considerably reduces or pathways. eliminates the formation of unwanted by-products, and con Likewise, yeast cells can be engineered to produce a num comitantly improves the yields and titers of beneficial ber of desirable products via biosynthetic pathways that uti metabolites. lize an aldehyde as a pathway intermediate. Engineered bio synthetic pathways comprising an aldehyde intermediate Reduced Accumulation of 3-Hydroxyacids from 3-Keto include biosynthetic pathways for the production of isobu Acids tanol, 1-butanol, 2-methyl-1-butanol, 3-methyl-1-butanol, 10 As described herein, the present inventors have discovered 1-propanol. 1-pentanol, 1-hexanol, 3-methyl-1-pentanol, that unwanted by-products, 3-hydroxyacids, can accumulate 4-methyl-1-pentanol, 4-methyl-1-hexanol, and 5-methyl-1- during fermentation reactions with microorganisms compris heptanol (See Table 3 and FIGS. 18, 19, and 20). ing a pathway involving a 3-keto acid intermediate. TABLE 3 Biosynthetic Pathways Utilizing an Aldehyde as an Intermediate Biosynthetic Aldehyde Pathway Intermediate Reference Isobutanol Isobutyraldehyde US 2009/0226991 (Feldman et al.), US 2011/0020889 (Feldman et al.), and US 2010/0143997 (Buelter et al.) 1-Butanol 1-Butanal WO/2010/017230 (Lynch), WO 2010/031772 (Wu et al.), WO/2010/045629 (Liao et al.), WO/2007/041269 (Donaldson et al.), WO 2008/052991 (Raamsdonket al.), WO/2008/143704 (Buelter et al.), and WO/2008/08O124 (Gunawardena et al.) 2-Methyl-1- 2-Methyl-1- WO/2008/098227 (Liao et al.), WO/2009/076480 (Picataggio Butanol Butanal et al.), and Atsumi et al., 2008, Nature 451:86-89 3-Methyl-1- 3-Methyl-1- WO/2008/098227 (Liao et al.), Atsumi et al., 2008, Nature Butanol Butanal 451:86-89, and Connor et al., 2008, Appi. Environ. Microbii. 74: 5769-5775 -Propano -Propanal WO/2008/098227 (Liao et al.) -Pentano -Pentanal WO/2010/045629 (Liao et al.), Zhanget al., 2008, Proc Nati AcadSci USA 105: 20653-20658 -Hexanol -Hexanal WO/2010/045629 (Liao et al.), Zhanget al., 2008, Proc Nati AcadSci USA 105: 20653-20658 3-Methyl-1- 3-Methyl-1- WO/2010/045629 (Liao et al.), Zhanget al., 2008, Proc Nati Pentanol Pentanal AcadSci USA 105: 20653-20658 4-Methyl-1- 4-Methyl-1- WO/2010/045629 (Liao et al.), Zhanget al., 2008, Proc Nati Pentanol Pentanal AcadSci USA 105: 20653-20658 4-Methyl-1- 4-Methyl-1- WO/2010/045629 (Liao et al.), Zhanget al., 2008, Proc Nati Hexanol Hexanal AcadSci USA 105: 20653-20658 5-Methyl-1- 5-Methyl-1- WO/2010/045629 (Liao et al.), Zhanget al., 2008, Proc Nati Heptanol Heptanal AcadSci USA 105: 20653-20658 The contents of each of the references in this table are herein incorporated by reference in their entireties for all purposes,

Each of the biosynthetic pathways listed in Table 3 have an As used herein, a "3-hydroxyacid' is an organic compound aldehyde intermediate. For example, the aldehyde intermedi 45 which contains a carboxylic acid moiety on the C1 carbon and ate in the isobutanol producing metabolic pathway is isobu an alcohol moiety on the C3 carbon. 3-hydroxyacids can be tyraldehyde (See FIG. 1), while pathways for the production obtained from 3-keto acids by chemical reduction of the of 1-butanol, 2-methyl-1-butanol, 3-methyl-1-butanol, 1-pro 3-keto acid ketone moiety to an alcohol moiety. For example, panol. 1-pentanol, 1-hexanol, 3-methyl-1-pentanol, 4-me reduction of the ketone moiety in acetolactate or 2-hydroxy thyl-1-pentanol, 4-methyl-1-hexanol, and 5-methyl-1-hep 50 2-methyl-3-oxobutanoic acid results in the formation of 3-hy tanol utilize 1-butanal, 2-methyl-1-butanal, 3-methyl-1- droxyacid 2,3-dihydroxy-2-methylbutanoic acid (DH2MB) butanal, 1-propanal, 1-pentanal, 1-hexanal, 3-methyl-1- (See, e.g., FIG. 2). pentanal, 4-methyl-1-pentanal, 4-methyl-1-hexanal, and The present inventors have discovered that the 3-hydroxy 5-methyl-1-heptanol as aldehyde intermediates, respectively. acid by-product, 2,3-dihydroxy-2-methylbutanoic acid (CAS Therefore, the product yield in biosynthetic pathways that 55 it 14868-24-7) (DH2MB), accumulates during fermentation utilize these aldehyde intermediates will in part depend upon reactions with microorganisms comprising biosynthetic path the amount of the aldehyde intermediate that is available to downstream enzymes of said biosynthetic pathways. ways involving the 3-keto acid intermediate, acetolactate. As described herein, the present inventors have discovered The accumulation of this by-product was found to hinder the enzymatic activities responsible for the accumulation of 60 optimal productivity and yield of the biosynthetic pathways unwanted by-products derived from 3-keto acid and/or alde target metabolite. The present inventors found that the pro hyde intermediates. Specifically, they have determined that a duction of DH2MB is caused by the reduction of acetolactate. 3-ketoacid reductase and an aldehyde dehydrogenase are To reduce or eliminate the activity responsible for the pro responsible for the conversion of 3-keto acids and aldehydes, duction of DH2MB, the corresponding enzymatic activity respectively, to unwanted by-products. The activities of these 65 catalyzing this reaction had to be identified and reduced or enzymes are shown to hinder the optimal productivity and eliminated. The inventors have found in S. cerevisiae that one yield of 3-keto acid- and/or aldehyde-derived products, Such enzyme catalyzing the conversion of acetolactate to US 9,012,189 B2 25 26 DH2MB is YMR226C (also known as TMA29). This the first In various embodiments described herein, the protein report of a protein in yeast that converts acetolactate to involved in catalyzing the conversion of the 3-keto acid inter DH2MB. mediate to the 3-hydroxyacid by-product is a ketoreductase. The present inventors have also discovered that the 3-hy In an exemplary embodiment, the ketoreductase is a 3-ke droxyacid by-product, 2-ethyl-2,3-dihydroxybutanoate, toacid reductase (3-KAR). As used herein, the term “3-ke accumulates during fermentation reactions with microorgan toacid reductase' refers to a ketoreductase (i.e. ketone reduc isms comprising biosynthetic pathways involving the 3-keto tase) active towards the 3-oxo group of a 3-keto acid. An acid intermediate, 2-aceto-2-hydroxybutyrate. The accumu illustration of exemplary reactions capable of being catalyzed lation of this by-product was found to hinder optimal produc by 3-ketoacid reductases is shown in FIG. 2. Suitable 3-ke tivity and yield of the biosynthetic pathway's target metabo 10 toacid reductases are generally found in the enzyme classifi lite. The present inventors found that the production of cation subgroup 1.1.1.X., the final digit X being dependent 2-ethyl-2,3-dihydroxybutanoate is caused by the reduction of upon the Substrate. A non-limiting list of exemplary 3-ke 2-aceto-2-hydroxybutyrate. To reduce or eliminate the activ toacid reductases and their corresponding enzyme classifica ity responsible for the production of 2-ethyl-2,3-dihydrox 15 tion number is shown in FIG. 3. ybutanoate, the corresponding enzymatic activity catalyzing In an exemplary embodiment, the 3-ketoacid reductase is this reaction had to be identified and reduced or eliminated. the S. cerevisiae YMR226C (SEQ ID NO: 1) protein, used The inventors have found in S. cerevisiae, the enzyme interchangeably herein with “TMA29. In some embodi YMR226C (also known as TMA29) which catalyzes the con ments, the 3-ketoacid reductase is the S. cerevisiae version of acetolactate to DH2MB also catalyzes the conver YMR226C (SEQID NO: 1) protein or a homolog or variant sion of 2-aceto-2-hydroxybutyrate to 2-ethyl-2,3-dihydrox thereof. In one embodiment, the homolog may be selected ybutanoate. This the first report of a protein in yeast that from the group consisting of Vanderwaltonzyma polyspora converts 2-aceto-2-hydroxybutyrate to 2-ethyl-2,3-dihydrox (SEQ ID NO: 2), Saccharomyces castellii (SEQ ID NO:3), ybutanoate. Candida glabrata (SEQID NO: 4), Saccharomyces bayanus The present inventors describe herein multiple strategies 25 (SEQIDNO:5), Zgosaccharomyces rouxii (SEQID NO:6), for reducing the conversion of the 3-keto acid intermediate to K. lactis (SEQID NO: 7), Ashbya gossypii (SEQID NO: 8), the corresponding 3-hydroxyacid by-product, a process Saccharomyces kluyveri (SEQ ID NO: 9), Kluyveromyces which is accompanied by an increase in the yield of desirable thermotolerans (SEQ ID NO: 10), Kluyveromyces waltii metabolites. In one embodiment, the 3-keto acid intermediate (SEQID NO: 11), Pichia stipitis (SEQID NO: 12), Debaro is acetolactate and the corresponding 3-hydroxyacid is 30 myces hansenii (SEQ ID NO: 13), Pichia pastoris (SEQ ID DH2MB. As described herein, reducing the conversion of NO: 14), Candida dubliniensis (SEQ ID NO: 15), Candida acetolactate to DH2MB enables the increased production of albicans (SEQIDNO:16), Yarrowia lipolytica (SEQID NO: beneficial metabolites such as isobutanol, 2-butanol, 1-bu 17), Issatchenkia Orientalis (SEQ ID NO: 18), Aspergillus tanol, 2-butanone, 2,3-butanediol, acetoin, diacetyl, valine, nidulans (SEQID NO: 19), Aspergillus niger (SEQID NO: leucine, pantothenic acid, isobutylene, 3-methyl-1-butanol, 35 20), Neurospora crassa (SEQ ID NO: 21), Schizosaccharo 4-methyl-1-pentanol, and coenzyme A which are derived myces pombe (SEQID NO: 22), and Kluyveromyces marx from biosynthetic pathways which use acetolactate as an ianus (SEQID NO:23). intermediate. In another embodiment, the 3-keto acid inter In one embodiment, the recombinant microorganism of the mediate is 2-aceto-2-hydroxybutyrate and the corresponding invention includes a mutation in at least one gene encoding 3-hydroxyacid is 2-ethyl-2,3-dihydroxybutanoate. As 40 for a 3-ketoacid reductase resulting in a reduction of 3-ke described herein, reducing the conversion of 2-aceto-2-hy toacid reductase activity of a polypeptide encoded by said droxybutyrate to 2-ethyl-2,3-dihydroxybutanoate enables the gene. In another embodiment, the recombinant microorgan increased production of beneficial metabolites such as 2-me ism includes a partial deletion of a gene encoding for a 3-ke thyl-1-butanol, isoleucine, 3-methyl-1-pentanol. 4-methyl-1- toacid reductase gene resulting in a reduction of 3-ketoacid hexanol, and 5-methyl-1-heptanol. 45 reductase activity of a polypeptide encoded by the gene. In Accordingly, one aspect of the invention is directed to a another embodiment, the recombinant microorganism com recombinant microorganism comprising a biosynthetic path prises a complete deletion of a gene encoding for a 3-ketoacid way which uses a 3-keto acid as an intermediate, wherein said reductase resulting in a reduction of 3-ketoacid reductase recombinant microorganism is substantially free of an activity of a polypeptide encoded by the gene. In yet another enzyme that catalyzes the conversion of the 3-keto acid inter 50 embodiment, the recombinant microorganism includes a mediate to a 3-hydroxyacid by-product. In one embodiment, modification of the regulatory region associated with the gene the 3-keto acid intermediate is acetolactate and the 3-hy encoding for a 3-ketoacid reductase resulting in a reduction of droxyacid by-product is DH2MB. In another embodiment, expression of a polypeptide encoded by said gene. In yet the 3-keto acid intermediate is 2-aceto-2-hydroxybutyrate another embodiment, the recombinant microorganism com and the 3-hydroxyacid by-product is 2-ethyl-2,3-dihydrox 55 prises a modification of the transcriptional regulator resulting ybutanoate. in a reduction of transcription of gene encoding for a 3-ke In another aspect, the invention is directed to a recombinant toacid reductase. In yet another embodiment, the recombi microorganism comprising a biosynthetic pathway which nant microorganism comprises mutations in all genes encod uses a 3-keto acid as an intermediate, wherein said recombi ing for a 3-ketoacid reductase resulting in a reduction of nant microorganism is engineered to reduce or eliminate the 60 activity of a polypeptide encoded by the gene(s). In one expression or activity of an enzyme catalyzing the conversion embodiment, said 3-ketoacid reductase gene is the S. cerevi of the 3-keto acid intermediate to a 3-hydroxyacid by-prod siae TMA29 (YMR226C) gene or a homolog thereof. As uct. In one embodiment, the 3-keto acid intermediate is aceto would be understood in the art, naturally occurring homologs lactate and the 3-hydroxyacid by-product is DH2MB. In of TMA29 in yeast other than S. cerevisiae can similarly be another embodiment, the 3-keto acid intermediate is 2-aceto 65 inactivated using the methods of the present invention. 2-hydroxybutyrate and the 3-hydroxyacid by-product is TMA29 homologs and methods of identifying such TMA29 2-ethyl-2,3-dihydroxybutanoate. homologs are described herein. US 9,012,189 B2 27 28 As is understood by those skilled in the art, there are several mediate to the 3-hydroxyacid by-product. In one embodi additional mechanisms available for reducing or disrupting ment, the 3-hydroxyacid by-product is DH2MB, derived the activity of a protein Such as 3-ketoacid reductase, includ from the 3-keto acid, acetolactate. Accordingly, in one ing, but not limited to, the use of a regulated promoter, use of embodiment, the desirable fermentation product is derived a weak constitutive promoter, disruption of one of the two 5 from any biosynthetic pathway in which acetolactate acts as copies of the gene in a diploidyeast, disruption of both copies an intermediate, including, but not limited to, isobutanol, of the gene in a diploid yeast, expression of an anti-sense 2-butanol, 1-butanol. 2-butanone, 2,3-butanediol, acetoin, nucleic acid, expression of an siRNA, over expression of a diacetyl, Valine, leucine, pantothenic acid, isobutylene, 3-me negative regulator of the endogenous promoter, alteration of thyl-1-butanol, 4-methyl-1-pentanol, and coenzyme A. In the activity of an endogenous or heterologous gene, use of a 10 another embodiment, the 3-hydroxyacid by-product is heterologous gene with lower specific activity, the like or 2-ethyl-2,3-dihydroxybutanoate, derived from the 3-keto combinations thereof. acid, 2-aceto-2-hydroxybutyrate. Accordingly, in another As described herein, the recombinant microorganisms of embodiment, the desirable fermentation product is derived the present invention are engineered to produce less of the from any biosynthetic pathway in which 2-aceto-2-hydroxy 3-hydroxyacid by-product than an unmodified parental 15 butyrate acts as an intermediate, including, but not limited to, microorganism. In one embodiment, the recombinant micro 2-methyl-1-butanol, isoleucine, 3-methyl-1-pentanol, 4-me organism produces the 3-hydroxyacid by-product from a car thyl-1-hexanol, and 5-methyl-1-heptanol. bon source at a carbon yield of less than about 20 percent. In In further embodiments, additional enzymes potentially another embodiment, the microorganism is produces the catalyzing the conversion of a 3-ketoacid intermediate to a 3-hydroxyacid by-product from a carbon Source at a carbon 3-hydroxyacid by-product are deleted from the genome of a yield of less than about 10, less than about 5, less than about recombinant microorganism comprising a biosynthetic path 2, less than about 1, less than about 0.5, less than about 0.1, or way which uses a 3-ketoacid as an intermediate. Endogenous less than about 0.01 percent. In one embodiment, the 3-hy yeast genes with the potential to convert of a 3-ketoacid droxyacid by-product is DH2MB, derived from the 3-keto intermediate to a 3-hydroxyacid by-product include ketore acid, acetolactate. In another embodiment, the 3-hydroxyacid 25 ductases, short chain alcohol dehydrogenases, medium chain by-product is 2-ethyl-2,3-dihydroxybutanoate, derived from alcohol dehydrogenases, members of the aldose reductase the 3-keto acid, 2-aceto-2-hydroxybutyrate. family, members of the D-hydroxyacid dehydrogenase fam In one embodiment, the 3-hydroxyacid by-product carbon ily, alcohol dehydrogenases, and lactate dehydrogenases. In yield derived from the 3-ketoacid is reduced by at least about one embodiment, the 3-hydroxyacid by-product is DH2MB, 50% in a recombinant microorganism as compared to a paren 30 derived from the3-ketoacid, acetolactate. In another embodi tal microorganism that does not comprise a reduction or dele ment, the 3-hydroxyacid by-product is 2-ethyl-2,3-dihydrox tion of the activity or expression of one or more endogenous ybutanoate, derived from the 3-keto acid, 2-aceto-2-hydroxy proteins involved in catalyzing the conversion of the 3-ke butyrate. toacid intermediate to the 3-hydroxyacid by-product. In Methods for identifying additional enzymes catalyzing the another embodiment, the 3-hydroxyacid by-product derived 35 conversion of a 3-ketoacid intermediate to a 3-hydroxyacid from the 3-ketoacid is reduced by at least about 60%, by at by-product are outlined as follows: endogenous yeast genes least about 65%, by at least about 70%, by at least about 75%, coding for ketoreductases, short chain alcohol dehydrogena by at least about 80%, by at least about 85%, by at least about ses, medium chain alcohol dehydrogenases, members of the 90%, by at least about 95%, by at least about 99%, by at least aldose reductase family, members of the D-hydroxyacid about 99.9%, or by at least about 100% as compared to a 40 dehydrogenase family, alcohol dehydrogenases, and lactate parental microorganism that does not comprise a reduction or dehydrogenases are deleted from the genome of a yeast strain deletion of the activity or expression of one or more endog comprising a biosynthetic pathway in which a 3-ketoacid enous proteins involved in catalyzing the conversion of the (e.g., acetolactate or 2-aceto-2-hydroxybutyrate) is an inter 3-ketoacid to the 3-hydroxyacid by-product. In one embodi mediate. These deletion strains are compared to the parent ment, the 3-hydroxyacid by-product is DH2MB, derived 45 strain by fermentation and analysis of the fermentation broth from the 3-keto acid, acetolactate. In another embodiment, for the presence and concentration of the corresponding 3-hy the 3-hydroxyacid by-product is 2-ethyl-2,3-dihydroxybu droxyacid by-product (e.g., DH2MB or 2-ethyl-2,3-dihy tanoate, derived from the 3-keto acid, 2-aceto-2-hydroxybu droxybutanoate, derived from acetolactate and 2-aceto-2-hy tyrate. droxybutyrate, respectively). In S. cerevisiae, deletions that In an additional embodiment, the yield of a desirable fer 50 reduce the production of the 3-hydroxyacid by-product are mentation product is increased in the recombinant microor combined by construction of strains carrying multiple dele ganisms comprising a reduction or elimination of the activity tions. Candidate genes can include, but are not limited to, or expression of one or more endogenous proteins involved in YAL060W, YJR159W, YGL157W, YBL114W, YOR120W, catalyzing the conversion of the 3-ketoacid intermediate to YKL055C, YBR159W, YBR149W, YDL168W, YDR368W, the 3-hydroxyacid by-product. In one embodiment, the yield 55 YLR426W, YCR107W, YILL24W, YML054C, YOL151W, of a desirable fermentation product is increased by at least YMR318C, YBR046C, YHR104W, YIR036C, YDL174C, about 1% as compared to a parental microorganism that does YDR541C, YBR 145W, YGL039W, YCR105W, YDL124W, not comprise a reduction or elimination of the activity or YIR035C, YFLO56C, YNL274C, YLR255C, YGL185C, expression of one or more endogenous proteins involved in YGL256W, YJR096W, YJR155W, YPL275W, YOR388C, catalyzing the conversion of the 3-ketoacid intermediate to 60 YLR070C, YMR083W, YER081W, YJR139C, YDL243C, the 3-hydroxyacid by-product. In another embodiment, the YPL113C, YOL165C, YML086C, YMR303C, YDL246C, yield of a desirable fermentation product is increased by at YLR070C, YHR063C, YNL33 1C, YFLO57C, YIL155C, least about 5%, by at least about 10%, by at least about 25%, YOLO86C, YAL061W, YDR127W, YPR127W, YCL018W, or by at least about 50% as compared to a parental microor YIL074C,YIL124W, and YEL071 W. Many of these deletion ganism that does not comprise a reduction or elimination of 65 strains are available commercially (for example Open Bio the activity or expression of one or more endogenous proteins systems YSC1054). These deletion strains are transformed involved in catalyzing the conversion of the 3-ketoacid inter with a plasmid pGV2435 from which the ALS gene (e.g., the US 9,012,189 B2 29 30 B. subtilis alsS) is expressed under the control of the CUP1 ism which exhibits inherently low or undetectable promoter. The transformants are cultivated in YPD medium endogenous enzyme activity responsible for the production containing 150 g/L glucose in shake flasks at 30°C., 75 rpm of the 3-hydroxyacid by-product (e.g., DH2MB or 2-ethyl-2, in a shaking incubator for 48 hours. After 48 h samples from 3-dihydroxybutanoate) is to analyze yeast strains by incubat the shake flasks are analyzed by HPLC for the concentration ing the yeast cells with a 3-keto acid (e.g., acetolactate or of the 3-hydroxyacid by-product (e.g., DH2MB and 2-ethyl 2-aceto-2-hydroxybutyrate) and analyze the broth for the pro 2,3-dihydroxybutanoate, derived from acetolactate and 2-ac duction of the corresponding 3-hydroxyacid by-product (e.g., eto-2-hydroxybutyrate, respectively). As would be under DH2MB or 2-ethyl-2,3-dihydroxybutanoate, derived from stood in the art, naturally occurring homologs of 3-ketoacid acetolactate and 2-aceto-2-hydroxybutyrate, respectively). reductase genes (e.g., TMA29) in yeast other than S. cerevi 10 siae can similarly be inactivated. 3-ketoacid reductase gene The recombinant microorganisms described herein which (e.g., TMA29) homologs and methods of identifying Such produce a beneficial metabolite derived from a biosynthetic 3-ketoacid reductase gene homologs are described herein. pathway which uses a 3-keto acid as an intermediate may be Another way to screen the deletion library is to incubate further engineered to reduce or eliminate enzymatic activity yeast cells with the 3-ketoacid intermediate (e.g., acetolactate 15 for the conversion of pyruvate to products other than the or 2-aceto-2-hydroxybutyrate) and analyze the broth for the 3-keto acid (e.g., acetolactate and/or 2-aceto-2-hydroxybu production of the corresponding 3-hydroxyacid by-product tyrate). In one embodiment, the enzymatic activity of pyru (e.g., DH2MB or 2-ethyl-2,3-dihydroxybutanoate, derived vate decarboxylase (PDC), lactate dehydrogenase (LDH), from acetolactate and 2-aceto-2-hydroxybutyrate, respec pyruvate oxidase, pyruvate dehydrogenase, and/or glycerol tively). 3-phosphate dehydrogenase (GPD) is reduced or eliminated. Some of the listed genes are the result of tandem duplica In a specific embodiment, the beneficial metabolite is pro tion or whole genome duplication events and are expected to duced in a recombinant PDC-minus GPD-minus yeast micro have similar substrate specificities. Examples are YAL061W organism that overexpresses an acetolactate synthase (ALS) (BDH1), and YAL060W (BDH2), YDR368W (YPR1) and gene. In another specific embodiment, the ALS is encoded by YOR120W (GCY1). Deletion of just one of the duplicated 25 the B. subtilis als.S. genes is likely not to result in a phenotype. These gene pairs Reduced Accumulation of Acid By-Products from Aldehyde have to be analyzed in Strains carrying deletions in both Intermediates genes. As described further in the Examples, the present inventors An alternative approach to find additional endogenous have also discovered that unwanted acid by-products (e.g., activity responsible for the production of the 3-hydroxyacid 30 isobutyrate in the case of isobutanol), can accumulate during by-product (e.g., DH2MB or 2-ethyl-2,3-dihydroxybu fermentation reactions with microorganisms comprising a tanoate, derived from acetolactate and 2-aceto-2-hydroxybu pathway involving an aldehyde intermediate (e.g., isobutyral tyrate, respectively) is to analyze yeast strains that overex dehyde in the case of isobutanol). press the genes Suspected of encoding the enzyme As used herein, an "acid by-product” refers to an organic responsible for production of the 3-hydroxyacid by-product. 35 compound which contains a carboxylic acid moiety. An acid Such strains are commercially available for many of the can by-product can be obtained by the oxidation of an aldehyde. didate genes listed above (for example Open Biosystems For example, the oxidation of isobutyraldehyde results in the YSC3870). The ORF overexpressing strains are processed in formation of isobutyric acid (See, e.g., FIG. 4). the same way as the deletion strains. They are transformed The present inventors have found that accumulation of with a plasmid for ALS expression and screened for 3-hy 40 these acid by-products hinders the optimal productivity and droxyacid by-product (e.g., DH2MB or 2-ethyl-2,3-dihy yield of the biosynthetic pathway which utilize aldehyde droxybutanoate) production levels. To narrow the list of pos intermediates. The present inventors found that the produc sible genes causing the production of the 3-hydroxyacid tion of these acid by-products is caused by dehydrogenation by-product (e.g., DH2MB or 2-ethyl-2,3-dihydroxybu of the corresponding aldehyde. To reduce or eliminate the tanoate), their expression can be analyzed in fermentation 45 activity responsible for the production of the acid by-product, samples. Genes that are not expressed during a fermentation the corresponding enzymatic activity catalyzing this reaction that produced the 3-hydroxyacid by-product (e.g., DH2MB had to be identified and reduced or eliminated. The inventors or 2-ethyl-2,3-dihydroxybutanoate) can be excluded from the have found in S. cerevisiae that one such enzyme catalyzing list of possible targets. This analysis can be done by extraction the conversion of aldehydes to acid by-products is aldehyde of RNA from fermenter samples and submitting these 50 dehydrogenase. samples to whole genome expression analysis, for example, The present inventors describe herein multiple strategies by Roche NimbleGen. for reducing acid by-product formation, a process which is As described herein, strains that naturally produce low accompanied by an increase in the yield of desirable metabo levels of one or more 3-hydroxyacid by-products can also lites such as isobutanol, 1-butanol, 2-methyl-1-butanol, have applicability for producing increased levels of desirable 55 3-methyl-1-butanol, 1-propanol. 1-pentanol, 1-hexanol, fermentation products that are derived from biosynthetic 3-methyl-1-pentanol, 4-methyl-1-pentanol, 4-methyl-1-hex pathways comprising a 3-ketoacid intermediate. As would be anol, and 5-methyl-1-heptanol. understood by one skilled in the art equipped with the instant Accordingly, one aspect of the invention is directed to a disclosure, Strains that naturally produce low levels of one or recombinant microorganism comprising a biosynthetic path more 3-hydroxyacid by-products may inherently exhibit low 60 way which uses an aldehyde as an intermediate, wherein said or undetectable levels of endogenous enzyme activity, result recombinant microorganism is substantially free of an ing in the reduced conversion of 3-ketoacids to 3-hydroxyac enzyme that catalyzes the conversion of an aldehyde to an ids, a trait favorable for the production of a desirable fermen acid by-product. tation product such as isobutanol. Described herein are In another aspect, the invention is directed to a recombinant several approaches for identifying a native host microorgan 65 microorganism comprising a biosynthetic pathway which ism which is substantially free of 3-ketoacid reductase activ uses an aldehyde as an intermediate, wherein said recombi ity. For example, one approach to finding a host microorgan nant microorganism is engineered to reduce or eliminate the US 9,012,189 B2 31 32 expression or activity of one or more enzymes catalyzing the In one embodiment, the recombinant microorganism conversion of the aldehyde to an acid by-product. includes a mutation in at least one gene encoding for an In one embodiment, the aldehyde intermediate is isobu aldehyde dehydrogenase resulting in a reduction of aldehyde tyraldehyde and the acid by-product is isobutyrate. In another dehydrogenase activity of a polypeptide encoded by said embodiment, the aldehyde intermediate is 1-butanal and the gene. In another embodiment, the recombinant microorgan acid by-product is butyrate. In yet another embodiment, the ism includes a partial deletion of gene encoding for an alde aldehyde intermediate is 2-methyl-1-butanal and the acid by hyde dehydrogenase resulting in a reduction of aldehyde product is 2-methyl-1-butyrate. In yet another embodiment, dehydrogenase activity of a polypeptide encoded by the gene. the aldehyde intermediate is 3-methyl-1-butanal and the acid In another embodiment, the recombinant microorganism 10 comprises a complete deletion of a gene encoding for an by-product is 3-methyl-1-butyrate. In yet another embodi aldehyde dehydrogenase resulting in a reduction of aldehyde ment, the aldehyde intermediate is 1-propanal and the acid dehydrogenase activity of a polypeptide encoded by the gene. by-product is propionate. In yet another embodiment, the In yet another embodiment, the recombinant microorganism aldehyde intermediate is 1-pentanal and the acid by-product includes a modification of the regulatory region associated is pentanoate. In yet another embodiment, the aldehyde inter 15 with the gene encoding for an aldehyde dehydrogenase mediate is 1-hexanal and the acid by-product is hexanoate. In resulting in a reduction of expression of a polypeptide yet another embodiment, the aldehyde intermediate is 3-me encoded by said gene. In yet another embodiment, the recom thyl-1-pentanal and the acid by-product is 3-methyl-1-pen binant microorganism comprises a modification of the tran tanoate. In yet another embodiment, the aldehyde intermedi Scriptional regulator resulting in a reduction of transcription ate is 4-methyl-1-pentanal and the acid by-product is of a gene encoding for an aldehyde dehydrogenase. In yet 4-methyl-1-pentanoate. In yet another embodiment, the alde another embodiment, the recombinant microorganism com hyde intermediate is 4-methyl-1-hexanal and the acid by prises mutations in all genes encoding for an aldehyde dehy product is 4-methyl-1-hexanoate. In yet another embodiment, drogenase resulting in a reduction of activity of a polypeptide the aldehyde intermediate is 5-methyl-1-heptanal and the encoded by the gene(s). In one embodiment, said aldehyde acid by-product is 5-methyl-1-heptanoate. 25 dehydrogenase is encoded by a gene selected from the group In various embodiments described herein, the protein consisting of ALD2, ALD3, ALD4, ALD5, ALD6, and involved in catalyzing the conversion of an aldehyde to acid HFD1, and homologs and variants thereof. As would be by-product is an aldehyde dehydrogenase (ALDH). understood in the art, naturally occurring homologs of alde As used herein, the term "aldehyde dehydrogenase' refers hyde dehydrogenase in yeast other than S. cerevisiae can to an enzyme catalyzing the reaction: 30 similarly be inactivated using the methods of the present invention. Aldehyde dehydrogenase homologs and methods an aldehyde--oxidized cofactor+HO-an acid-reduced of identifying such aldehyde dehydrogenase homologs are cofactor-i-H' described herein. An illustration of exemplary reactions capable of being As is understood by those skilled in the art, there are several catalyzed by aldehyde dehydrogenases is shown in FIG. 4. 35 additional mechanisms available for reducing or disrupting Suitable aldehyde dehydrogenases are generally found in the the activity of a protein such as aldehyde dehydrogenase, enzyme classification subgroup EC 1.2.1.X., wherein the final including, but not limited to, the use of a regulated promoter, digit X is dependent upon the substrate or the cofactor. For use of a weak constitutive promoter, disruption of one of the example, EC 1.2.1.3 catalyzes the following reaction: an two copies of the gene in a diploid yeast, disruption of both aldehyde--NAD"+HO—an acid-i-NADH+H"); EC 1.2.1.4 40 copies of the gene in a diploid yeast, expression of an anti catalyzes the following reaction: an aldehyde--NADP+ sense nucleic acid, expression of an siRNA, over expression HO—an acid+NADPH-i-H); and EC1.2.1.5 catalyzes the fol of a negative regulator of the endogenous promoter, alteration lowing reaction: an aldehyde--NAD(P)+HO-an acid--NAD of the activity of an endogenous or heterologous gene, use of (P)H-i-H". a heterologous gene with lower specific activity, the like or As described herein, the protein involved in catalyzing the 45 combinations thereof. conversion of an aldehyde to an acid by-product is an alde As would be understood by one skilled in the art, the hyde dehydrogenase (ALDH). In one embodiment, the alde activity or expression of more than one aldehyde dehydroge hyde dehydrogenase is encoded by a gene selected from the nase can be reduced or eliminated. In one specific embodi group consisting of ALD2, ALD3, ALD4, ALD5, ALD6, and ment, the activity or expression of ALD4 and ALD6 or HFD1, and homologs and variants thereof. In an exemplary 50 homologs or variants thereof is reduced or eliminated. In embodiment, the aldehyde dehydrogenase is the S. cerevisiae another specific embodiment, the activity or expression of aldehyde dehydrogenase ALD6 (SEQ ID NO: 25) or a ALD5 and ALD6 or homologs or variants thereof is reduced homolog or variant thereof. In one embodiment, the homolog or eliminated. In yet another specific embodiment, the activ may be selected from the group consisting of Saccharomyces ity or expression of ALD4, ALD5, and ALD6 or homologs or castelli (SEQ ID NO: 26), Candida glabrata (SEQ ID NO: 55 variants thereof is reduced or eliminated. In yet another spe 27), Saccharomyces bayanus (SEQID NO: 28), Kluyveromy cific embodiment, the activity or expression of the cytosoli ces lactis (SEQID NO: 29), Kluyveromyces thermotolerans cally localized aldehyde dehydrogenases ALD2, ALD3, and (SEQ ID NO:30), Kluyveromyces waltii (SEQ ID NO: 31), ALD6 or homologs or variants thereof is reduced or elimi Saccharomyces cerevisiae YJ789 (SEQID NO:32), Saccha nated. In yet another specific embodiment, the activity or romyces cerevisiae JAY291 (SEQID NO:33), Saccharomy 60 expression of the mitochondrially localized aldehyde dehy ces cerevisiae EC 1118 (SEQ ID NO. 34), Saccharomyces drogenases, ALD4 and ALD5 or homologs or variants cerevisiae DBY939 (SEQ ID NO:35), Saccharomyces cer thereof, is reduced or eliminated. evisiae AWRI1631 (SEQID NO:36), Saccharomyces cerevi As described herein, the recombinant microorganisms of siae RM11-1a (SEQ ID NO: 37), Pichia pastoris (SEQ ID the present invention are engineered to produce less of the NO: 38), Kluyveromyces marxianus (SEQ ID NO: 39), 65 acid by-product than an unmodified parental microorganism. Schizosaccharomyces pombe (SEQ ID NO: 40), and In one embodiment, the recombinant microorganism pro Schizosaccharomyces pombe (SEQID NO: 41). duces the acid by-product from a carbon Source at a carbon US 9,012,189 B2 33 34 yield of less than about 50 percent as compared to a parental 1-pentanol biosynthetic pathway. In yet another embodiment, microorganism. In another embodiment, the microorganism the acid by-product is hexanoate, derived from 1-hexanal, an is produces the acid by-product from a carbon Source from a intermediate of the 1-hexanol biosynthetic pathway. In yet carbon source at a carbonyield of less than about 25, less than another embodiment, the acid by-product is 3-methyl-1-pen about 10, less than about 5, less than about 1, less than about tanoate, derived from 3-methyl-1-pentanal, an intermediate 0.5, less than about 0.1, or less than about 0.01 percent as of the 3-methyl-1-pentanol biosynthetic pathway. In yet compared to a parental microorganism. In one embodiment, another embodiment, the acid by-product is 4-methyl-1-pen the acid by-product is isobutyrate, derived from isobutyral tanoate, derived from 4-methyl-1-pentanal, an intermediate dehyde, an intermediate of the isobutanol biosynthetic path of the 4-methyl-1-pentanol biosynthetic pathway. In yet way. In another embodiment, the acid by-product is butyrate, 10 another embodiment, the acid by-product is 4-methyl-1-hex derived from 1-butanal, an intermediate of the 1-butanol bio anoate, derived from 4-methyl-1-hexanal, an intermediate of synthetic pathway. In yet another embodiment, the acid by the 4-methyl-1-hexanol biosynthetic pathway. In yet another product is 2-methyl-1-butyrate, derived from 2-methyl-1-bu embodiment, the acid by-product is 5-methyl-1-heptanoate, tanal, an intermediate of the 2-methyl-1-butanol biosynthetic derived from 5-methyl-1-heptanal, an intermediate of the pathway. In yet another embodiment, the acid by-product is 15 5-methyl-1-heptanol biosynthetic pathway. 3-methyl-1-butyrate, derived from 3-methyl-1-butanal, an In an additional embodiment, the yield of a desirable fer intermediate of the 3-methyl-1-butanol biosynthetic path mentation product is increased in the recombinant microor way. In yet another embodiment, the acid by-product is pro ganisms comprising a reduction or elimination of the activity pionate, derived from 1-propanal, an intermediate of the or expression of one or more proteins involved in catalyzing 1-propanol biosynthetic pathway. In yet another embodi the conversion of an aldehyde to acid by-product. In one ment, the acid by-product is pentanoate, derived from 1-pen embodiment, the yield of a desirable fermentation product is tanal, an intermediate of the 1-pentanol biosynthetic pathway. increased by at least about 1% as compared to a parental In yet another embodiment, the acid by-product is hexanoate, microorganism that does not comprise a reduction or elimi derived from 1-hexanal, an intermediate of the 1-hexanol nation of the activity or expression of one or more endog biosynthetic pathway. In yet another embodiment, the acid 25 enous proteins involved in catalyzing the conversion of an by-product is 3-methyl-1-pentanoate, derived from 3-methyl aldehyde to acid by-product. In another embodiment, the 1-pentanal, an intermediate of the 3-methyl-1-pentanol bio yield of a desirable fermentation product is increased by at synthetic pathway. In yet another embodiment, the acid by least about 5%, by at least about 10%, by at least about 25%, product is 4-methyl-1-pentanoate, derived from 4-methyl-1- or by at least about 50% as compared to a parental microor pentanal, an intermediate of the 4-methyl-1-pentanol 30 ganism that does not comprise a reduction or elimination of biosynthetic pathway. In yet another embodiment, the acid the activity or expression of one or more endogenous proteins by-product is 4-methyl-1-hexanoate, derived from 4-methyl involved in catalyzing the conversion of an aldehyde to acid 1-hexanal, an intermediate of the 4-methyl-1-hexanol biosyn by-product. As described herein, the desirable fermentation thetic pathway. In yet another embodiment, the acid by-prod product may be derived from any biosynthetic pathway in uct is 5-methyl-1-heptanoate, derived from 5-methyl-1- 35 which an aldehyde acts as an intermediate, including, but not heptanal, an intermediate of the 5-methyl-1-heptanol limited to, isobutanol, 1-butanol, 2-methyl-1-butanol, 3-me biosynthetic pathway. thyl-1-butanol. 1-propanol. 1-pentanol, 1-hexanol, 3-methyl In one embodiment, the acid by-product carbon yield from 1-pentanol, 4-methyl-1-pentanol, 4-methyl-1-hexanol, and the corresponding aldehyde is reduced by at least about 50% 5-methyl-1-heptanol biosynthetic pathways. in a recombinant microorganism as compared to a parental 40 Methods for identifying additional enzymes catalyzing the microorganism that does not comprise a reduction or deletion conversion of an aldehyde to acid by-product are outlined as of the activity or expression of one or more proteins involved follows: endogenous yeast genes coding for putative alde in catalyzing the conversion of an aldehyde to an acid by hyde and alcohol dehydrogenases are deleted from the product. In another embodiment, the acid by-product carbon genome of a yeast strain. These deletion strains are compared yield from acetolactate is reduced by at least about 60%, by at 45 to the parent strain by enzymatic assay. Many of these dele least about 65%, by at least about 70%, by at least about 75%, tion strains are available commercially (for example Open by at least about 80%, by at least about 85%, by at least about Biosystems YSC1054). 90%, by at least about 95%, by at least about 99%, by at least Another way to screen the deletion library is to incubate about 99.9%, or by at least about 100% as compared to a yeast cells with an aldehyde (e.g., isobutyraldehyde or 1-bu parental microorganism that does not comprise a reduction or 50 tanal) and analyze the broth for the production of the corre deletion of the activity or expression of one or more proteins sponding acid by-product (e.g., isobutyrate or butyrate, involved in catalyzing the conversion of an aldehyde to an derived from isobutyraldehyde or 1-butanal, respectively). acid by-product. In one embodiment, the acid by-product is An alternative approach to find additional endogenous isobutyrate, derived from isobutyraldehyde, an intermediate activity responsible for the production of the acid by-product of the isobutanol biosynthetic pathway. In another embodi 55 (e.g., isobutyrate or butyrate, derived from isobutyraldehyde ment, the acid by-product is butyrate, derived from 1-butanal, or 1-butanal, respectively) is to analyze yeast strains that an intermediate of the 1-butanol biosynthetic pathway. In yet overexpress the genes Suspected of encoding the enzyme another embodiment, the acid by-product is 2-methyl-1-bu responsible for production of the acid by-product. Such tyrate, derived from 2-methyl-1-butanal, an intermediate of strains are commercially available for many of the candidate the 2-methyl-1-butanol biosynthetic pathway. In yet another 60 genes listed above (for example Open Biosystems YSC3870). embodiment, the acid by-product is 3-methyl-1-butyrate, The ORF overexpressing strains are screened for increased derived from 3-methyl-1-butanal, an intermediate of the acid by-product production levels. Alternatively, the cell 3-methyl-1-butanol biosynthetic pathway. In yet another lysates of the ORF overexpressing strains are assayed for embodiment, the acid by-product is propionate, derived from increased aldehyde oxidation activity. To narrow the list of 1-propanal, an intermediate of the 1-propanol biosynthetic 65 possible genes causing the production of acid by-products, pathway. In yet another embodiment, the acid by-product is their expression can be analyzed in fermentation samples. pentanoate, derived from 1-pentanal, an intermediate of the Genes that are not expressed during a fermentation that pro US 9,012,189 B2 35 36 duces an acid by-product can be excluded from the list of acids occurs via a 2-oxoacid dehydrogenase in S. cerevisiae. possible targets. This analysis can be done by extraction of J. Gen. Microbiol. 138: 2029-2033). Reducing the expression RNA from fermenter samples and submitting these samples of or deleting one or more ketoacid dehydrogenases and to whole genome expression analysis, for example, by Roche homologs or variants thereof, will generally lead to a reduced NimbleGen. production of isobutyrate and a concomitant increase in As described herein, strains that naturally produce low isobutanol yield. levels of one or more acid by-products can also have appli The reduction in expression of or deletion of genes in S. cability for producing increased levels of desirable fermen cerevisiae and other yeast can be achieved by methods known tation products that are derived from biosynthetic pathways to those of skill in the art, such as allelic replacement or comprising an aldehyde intermediate. As would be under 10 exchange, as well as gene disruption by the insertion of stood by one skilled in the art equipped with the instant another gene or marker cassette. disclosure, Strains that naturally produce low levels of one or Another strategy described herein for reducing the produc more acid by-products may inherently exhibit low or unde tion of the by-product isobutyrate is to increase the activity tectable levels of endogenous enzyme activity, resulting in the and/or expression of an alcohol dehydrogenase (ADH) reduced conversion of aldehydes to acid by-products, a trait 15 responsible for the conversion of isobutyraldehyde to isobu favorable for the production of a desirable fermentation prod tanol. This strategy prevents competition by endogenous uct such as isobutanol. Described herein are several enzymes for the isobutanol pathway intermediate, isobutyral approaches for identifying a native host microorganism dehyde. An increase in the activity and/or expression of the which is substantially free of aldehyde dehydrogenase activ alcohol dehydrogenase may be achieved by various means. ity. For example, one approach to finding a host microorgan For example, alcohol dehydrogenase activity can be ism which exhibits inherently low or undetectable endog increased by utilizing a promoter with increased promoter enous enzyme activity responsible for the production of the strength, by increasing the copy number of the alcohol dehy acid by-product (e.g., isobutyrate or butyrate) is to analyze drogenase gene, or by utilizing an alternative or modified yeast strains by incubating the yeast cells with an aldehyde alcohol dehydrogenase with increased specific activity. (e.g., isobutyraldehyde or 1-butanal) and analyze the broth 25 An alternative strategy described herein for reducing the for the production of the corresponding acid by-product (e.g., production of the by-product isobutyrate is to utilize an alco isobutyrate or butyrate, derived from isobutyraldehyde or holdehydrogenase (ADH) in the isobutanol pathway respon 1-butanal, respectively). sible for the conversion of isobutyraldehyde to isobutanol As described above, one strategy reducing the production which exhibits a decrease in Michaelis-Menten constant of the acid by-product, isobutyrate, is to reduce or eliminate 30 (K). This strategy also prevents competition by endogenous the activity or expression of one or more endogenous alde enzymes for the isobutanol pathway intermediate, isobutyral hyde dehydrogenase proteins present in yeast that may be dehyde. converting isobutyraldehyde to isobutyrate. Another strategy described herein for reducing the produc Another strategy for reducing the production of isobutyrate tion of the by-product isobutyrate is to utilize an alcohol is the reduction or elimination of activity or expression of one 35 dehydrogenase (ADH) in the isobutanol pathway responsible more endogenous yeast alcohol dehydrogenases. Reducing for the conversion of isobutyraldehyde to isobutanol which the expression of or deleting one or more alcohol dehydroge exhibits increased activity and a decrease in Michaelis nases including, but not limited to, S. cerevisiae ADH1, Menten constant (K). This strategy also prevents competi ADH2, ADH3, ADH4, ADH5, ADH6, ADH7, and SFA1, and tion by endogenous enzymes for the isobutanol pathway homologs or variants thereof, will generally lead to a reduced 40 intermediate, isobutyraldehyde. production of isobutyrate and a concomitant increase in Further, by utilizing a modified ADH enzyme, the present isobutanol yield. The reduction and/or deletion of additional inventors may establish a situation in which the forward reac dehydrogenases are envisioned herein and are considered tion (i.e. the isobutyraldehyde conversion to isobutanol) is the within the scope of the present invention. These dehydroge favored reaction over the reverse reaction (i.e. the conversion nases include additional alcohol dehydrogenases such as S. 45 of isobutanol to isobutyraldehyde). cerevisiae BDH1, BDH2, SOR1, SOR2, and XYL1, and The strategies described above generally lead to a decrease homologs or variants thereof, as well as aryl alcohol dehy in isobutyrate yield, which is accompanied by an increase in drogenases such as AAD3, AAD4, AAD6, AAD10, AAD14, isobutanol yield. Hence, the above strategies are useful for AAD15, AAD16, and YPL088W, and homologs or variants decreasing the isobutyrate yield and/ortiterand for increasing thereof. 50 the ratio of isobutanol yield over isobutyrate yield. In another embodiment, the invention provides recombi In one embodiment, the isobutyrate yield (mol isobutyrate nant microorganisms engineered to reduce and/or deletion per mol glucose) is less than about 5%. In another embodi one or more additional genes encoding carbonyl/aldehyde ment, the isobutyrate yield (mol isobutyrate per mol glucose) reductases. These carbonyl/aldehyde reductases include S. is less than about 1%. In yet another embodiment, the isobu cerevisiae ARI1,YPR1, TMA29,YGL039W, and UGA2, and 55 tyrate yield (mol isobutyrate per mol glucose) is less than homologs or variants thereof. about 0.5%, less than about 0.1%, less than about 0.05%, or An additional strategy described herein for reducing the less than about 0.01%. production of the by-product isobutyrate is to reduce or elimi In one embodiment, the isobutanol to isobutyrate yield nate the activity or expression of endogenous proteins present ratio is at least about 2. In another embodiment, the isobutanol in yeast that may be producing isobutyrate from the isobu 60 to isobutyrate yield is at least about 5. In yet another embodi tanol pathway intermediate 2-ketoisovalerate. Such enzymes ment, the isobutanol to isobutyrate yield ratio at least about are generally referred to asketoacid dehydrogenases (KDH). 20, at least about 100, at least about 500, or at least about Elimination or reduction of the activity or expression of these 1OOO. endogenous proteins can reduce or eliminate the production The recombinant microorganisms described herein which of the unwanted byproduct, isobutyrate. KDH enzyme activ 65 produce a beneficial metabolite derived from a biosynthetic ity has been identified in S. cerevisiae (Dickinson, J. R., and pathway which uses an aldehyde as an intermediate may be I. W. Dawes, 1992, The catabolism of branched-chain amino further engineered to reduce or eliminate enzymatic activity US 9,012,189 B2 37 38 for the conversion of pyruvate to products other than a 3-keto one embodiment, the homolog may be selected from the acid (e.g., acetolactate and/or 2-aceto-2-hydroxybutyrate). In group consisting of Vanderwaltomzyma polyspora (SEQ ID one embodiment, the enzymatic activity of pyruvate decar NO: 2), Saccharomyces castellii (SEQID NO:3), Candida boxylase (PDC), lactate dehydrogenase (LDH), pyruvate oxi glabrata (SEQID NO: 4), Saccharomyces bayanus (SEQID dase, pyruvate dehydrogenase, and/or glycerol-3-phosphate NO: 5), Zgosaccharomyces rouxii (SEQID NO: 6), K. lactis dehydrogenase (GPD) is reduced or eliminated. (SEQ ID NO: 7), Ashbya gossypii (SEQID NO: 8), Saccha In a specific embodiment, the beneficial metabolite is pro romyces kluyveri (SEQ ID NO: 9), Kluyveromyces thermo duced in a recombinant PDC-minus GPD-minus yeast micro tolerans (SEQ ID NO: 10), Kluyveromyces waltii (SEQ ID organism that overexpresses an acetolactate synthase (ALS) NO: 11), Pichia stipitis (SEQ ID NO: 12), Debaromyces gene. In another specific embodiment, the ALS is encoded by 10 hansenii (SEQID NO: 13), Pichiapastoris (SEQID NO:14), the B. subtilis als.S. Candida dubliniensis (SEQ ID NO: 15), Candida albicans Reduced Accumulation of 3-Hydroxyacid By-Products and (SEQ ID NO: 16), Yarrowia lipolytica (SEQ ID NO: 17), Acid By-Products Issatchenkia Orientalis (SEQID NO: 18), Aspergillus nidu The present inventors describe herein multiple strategies lans (SEQID NO: 19), Aspergillus niger (SEO ID NO. 20), for reducing the conversion of a 3-keto acid intermediate to a 15 Neurospora crassa (SEQID NO: 21), Schizosaccharomyces corresponding 3-hydroxyacid by-product, a process which is pombe (SEQ ID NO: 22), and Kluyveromyces marxianus accompanied by an increase in the yield of desirable metabo (SEQID NO. 23). lites. The present inventors also describe herein multiple In various embodiments described herein, the protein strategies for reducing the conversion of an aldehyde inter involved in catalyzing the conversion of an aldehyde to an mediate to a corresponding acid by-product, a process which acid by-product is an aldehyde dehydrogenase (ALDH). In is accompanied by a further increase in the yield of desirable one embodiment, the aldehyde dehydrogenase is encoded by metabolites. a gene selected from the group consisting of ALD2, ALD3. Accordingly, in one aspect, the invention is directed to a ALD4, ALD5, ALD6, and HFD1, and homologs and variants recombinant microorganism comprising a biosynthetic path thereof. In an exemplary embodiment, the aldehyde dehydro way which uses a 3-keto acid as an intermediate and an 25 genase is the S. cerevisiae aldehyde dehydrogenase ALD6 aldehyde as an intermediate, wherein said recombinant (SEQ ID NO: 25) or homolog or variant thereof. In one microorganism is (i) Substantially free of an enzyme that embodiment, the homolog may be selected from the group catalyzes the conversion of the 3-keto acid intermediate to a consisting of Saccharomyces castelli (SEQID NO: 26), Can 3-hydroxyacid by-product and (ii) substantially free of an dida glabrata (SEQ ID NO: 27), Saccharomyces bayanus enzyme that catalyzes the conversion of an aldehyde to an 30 (SEQ ID NO: 28), Kluyveromyces lactis (SEQ ID NO: 29), acid by-product. In one embodiment, the 3-keto acid inter Kluyveromyces thermotolerans (SEQID NO: 30), Kluyvero mediate is acetolactate. The biosynthetic pathway which uses myces waltii (SEQ ID NO: 31), Saccharomyces cerevisiae acetolactate and an aldehyde as intermediates may be YJ789 (SEQID NO:32), Saccharomyces cerevisiae JAY291 selected from a pathway for the biosynthesis of isobutanol, (SEQ ID NO: 33), Saccharomyces cerevisiae EC1118 (SEQ 1-butanol, and 3-methyl-1-butanol. In another embodiment, 35 ID NO. 34), Saccharomyces cerevisiae DBY939 (SEQ ID the 3-keto acid intermediate is 2-aceto-2-hydroxybutyrate. NO: 35), Saccharomyces cerevisiae AWRI1631 (SEQ ID The biosynthetic pathway which uses 2-aceto-2-hydroxybu NO:36), Saccharomyces cerevisiae RM11-1a (SEQID NO: tyrate and an aldehyde as intermediates may be selected from 37), Pichia pastoris (SEQID NO:38), Kluyveromyces marx a pathway for the biosynthesis of 2-methyl-1-butanol, 3-me ianus (SEQID NO:39), Schizosaccharomyces pombe (SEQ thyl-1-pentanol, 4-methyl-1-hexanol, and 5-methyl-1-hep 40 ID NO: 40), and Schizosaccharomyces pombe (SEQID NO: tanol. 41). In another aspect, the invention is directed to a recombinant The recombinant microorganisms described herein which microorganism comprising a biosynthetic pathway which produce a beneficial metabolite derived from a biosynthetic uses a 3-keto acid as an intermediate and an aldehyde as an pathway which uses a 3-keto acid and an aldehyde as an intermediate, wherein said recombinant microorganism is (i) 45 intermediate may be further engineered to reduce or eliminate engineered to reduce or eliminate the expression or activity of enzymatic activity for the conversion of pyruvate to products an enzyme catalyzing the conversion of the 3-keto acid inter other than a 3-keto acid (e.g., acetolactate and/or 2-aceto-2- mediate to a 3-hydroxyacid by-product and (ii) engineered to hydroxybutyrate). In one embodiment, the enzymatic activity reduce or eliminate the expression or activity of one or more of pyruvate decarboxylase (PDC), lactate dehydrogenase enzymes catalyzing the conversion of the aldehyde to an acid 50 (LDH), pyruvate oxidase, pyruvate dehydrogenase, and/or by-product. In one embodiment, the 3-keto acid intermediate glycerol-3-phosphate dehydrogenase (GPD) is reduced or is acetolactate. The biosynthetic pathway which uses aceto eliminated. lactate and an aldehyde as intermediates may be selected from In a specific embodiment, the beneficial metabolite is pro a pathway for the biosynthesis of isobutanol, 1-butanol, and duced in a recombinant PDC-minus GPD-minus yeast micro 3-methyl-1-butanol. In another embodiment, the 3-keto acid 55 organism that overexpresses an acetolactate synthase (ALS) intermediate is 2-aceto-2-hydroxybutyrate. The biosynthetic gene. In another specific embodiment, the ALS is encoded by pathway which uses 2-aceto-2-hydroxybutyrate and an alde the B. subtilis als.S. hyde as intermediates may be selected from a pathway for the Illustrative Embodiments of Strategies for Reducing Accu biosynthesis of 2-methyl-1-butanol, 3-methyl-1-pentanol, mulation of 3-Hydroxyacid by-Products and/or Acid 4-methyl-1-hexanol, and 5-methyl-1-heptanol. 60 by-Products In various embodiments described herein, the protein In a specific illustrative embodiment, the recombinant involved in catalyzing the conversion of the 3-keto acid inter microorganism comprises an isobutanol producing metabolic mediate to the 3-hydroxyacid by-product is a ketoreductase. pathway of which acetolactate and isobutyraldehyde are In an exemplary embodiment, the ketoreductase is a 3-ke intermediates, wherein said recombinant microorganism is toacid reductase (3-KAR). In a further exemplary embodi 65 Substantially free of enzymes catalyzing the conversion of the ment, the 3-ketoacid reductase is the S. cerevisiae YMR226C acetolactate intermediate to DH2MB and of the isobutyral (SEQ ID NO: 1) protein or a homolog or variant thereof. In dehyde intermediate to isobutyrate. In another specific US 9,012,189 B2 39 40 embodiment, the recombinant microorganism comprises an Overexpression of Enzymes Converting DH2MB into Isobu isobutanol producing metabolic pathway of which acetolac tanol Pathway Intermediates tate and isobutyraldehyde are intermediates, wherein said A different approach to reduce or eliminate the production recombinant microorganism is (i) engineered to reduce or of 2,3-dihydroxy-2-methylbutanoic acid (CASH 14868-24-7) eliminate the expression or activity of one or more enzymes in isobutanol producing yeast is to overexpress an enzyme catalyzing the conversion of acetolactate to DH2MB and (ii) that converts DH2MB into an isobutanol pathway intermedi engineered to reduce or eliminate the expression or activity of ate. One way to accomplish this is through the use of an one or more enzymes catalyzing the conversion of isobutyral enzyme that catalyzes the interconversion of DH2MB and dehyde to isobutyrate. In one embodiment, the enzyme cata acetolactate, but favors the oxidation of DH2MB. Therefore, 10 in one embodiment, the present invention provides a recom lyzing the conversion of acetolactate to DH2MB is a 3-ke binant microorganism for producing isobutanol, wherein said toacid reductase (3-KAR). In another embodiment, the recombinant microorganism overexpresses an endogenous or enzyme catalyzing the conversion of isobutyraldehyde to heterologous protein capable of converting DH2MB into isobutyrate is an aldehyde dehydrogenase (ALDH). A non acetolactate. limiting example of Such a pathway in which a 3-ketoacid 15 In one embodiment, the endogenous or heterologous pro reductase (3-KAR) and an aldehyde dehydrogenase (ALDH) tein kinetically favors the oxidative reaction. In another are eliminated is depicted in FIG. 5. embodiment, the endogenous or heterologous protein has a In a further specific illustrative embodiment, the recombi low K for DH2MB and a high K for acetolactate. In yet nant microorganism comprises a 3-methyl-1-butanol produc another embodiment, the endogenous or heterologous protein ing metabolic pathway of which acetolactate and 3-methyl has a low K for the oxidized form of its cofactor and a high 1-butanal are intermediates, wherein said recombinant K for the corresponding reduced form of its cofactor. In yet microorganism is Substantially free of enzymes catalyzing another embodiment, the endogenous or heterologous protein the conversion of the acetolactate intermediate to DH2MB has a higherk for the oxidative reaction than for the reduc and of the 3-methyl-1-butanal intermediate to 3-methyl-1- tive direction. This endogenous or heterologous protein butyrate. In another specific embodiment, the recombinant 25 should preferably have the ability to use aredox cofactor with microorganism comprises a 3-methyl-1-butanol producing a high concentration of its oxidized form versus its reduced metabolic pathway of which acetolactate and 3-methyl-1- form. butanal are intermediates, wherein said recombinant micro In one embodiment, the endogenous or heterologous pro organism is (i) engineered to reduce or eliminate the expres tein is encoded by a gene selected from the group consisting sion or activity of one or more enzymes catalyzing the 30 ofYAL060W.YJR159W,YGL157W,YBL114W,YOR120W, conversion of acetolactate to DH2MB and (ii) engineered to YKL055C, YBR159W, YBR149W, YDL168W, YDR368W, reduce or eliminate the expression or activity of one or more YLR426W, YCR107W, YILL24W, YML054C, YOL151W, enzymes catalyzing the conversion of 3-methyl-1-butanal to YMR318C, YBR046C, YHR104W, YIR036C, YDL174C, 3-methyl-1-butyrate. In one embodiment, the enzyme cata YDR541C, YBR 145W, YGL039W, YCR105W, YDL124W, lyzing the conversion of acetolactate to DH2MB is a 3-ke 35 YIR035C, YFLO56C, YNL274C, YLR255C, YGL185C, toacid reductase (3-KAR). In another embodiment, the YGL256W, YJR096W, YJR155W, YPL275W, YOR388C, enzyme catalyzing the conversion of 3-methyl-1-butanal to YLR070C, YMR083W, YER081W, YJR139C, YDL243C, 3-methyl-1-butyrate is an aldehyde dehydrogenase (ALDH). YPL113C, YOL165C, YML086C, YMR303C, YDL246C, A non-limiting example of Such a pathway in which a 3-ke YLR070C, YHR063C, YNL33 1C, YFLO57C, YIL155C, toacid reductase (3-KAR) and an aldehyde dehydrogenase 40 YOLO86C, YAL061W, YDR127W, YPR127W, YCL018W, (ALDH) are eliminated is depicted in FIG. 6. YIL074C, YIL124W, and YEL071 W. In addition, heterolo In a further specific illustrative embodiment, the recombi gous genes can be overexpressed in isobutanol producing nant microorganism comprises a 2-methyl-1-butanol produc yeast. For examples beta-hydroxy acid dehydrogenases ing metabolic pathway of which acetolactate and 2-methyl (EC1.1.1.45 and EC1.1.1.60) would be candidates for over 1-butanal are intermediates, wherein said recombinant 45 expression. microorganism is Substantially free of enzymes catalyzing In another embodiment, the endogenous or heterologous the conversion of the 2-aceto-2-hydroxybutyrate intermedi protein kinetically that favors the reductive reaction is engi ate to 2-ethyl-2,3-dihydroxybutyrate and of the 2-methyl-1- neered to favor the oxidative reaction. In another embodi butanal intermediate to 2-methyl-1-butyrate. In another spe ment, the protein is engineered to have a low K for DH2MB cific embodiment, the recombinant microorganism 50 and a high K for acetolactate. In yet another embodiment, comprises a 2-methyl-1-butanol producing metabolic path the protein is engineered to have a low K for the oxidized way of which 2-aceto-2-hydroxybutyrate and 2-methyl-1- form of its cofactor and a high K for the corresponding butanal are intermediates, wherein said recombinant micro reduced form of its cofactor. In yet another embodiment, the organism is (i) engineered to reduce or eliminate the protein is engineered to have a higher k, for the oxidative expression or activity of one or more enzymes catalyzing the 55 reaction than for the reductive direction. This engineered conversion of 2-aceto-2-hydroxybutyrate to 2-ethyl-2,3-di protein should preferably have the ability to use a redox hydroxybutyrate and (ii) engineered to reduce or eliminate cofactor with a high concentration of its oxidized form versus the expression or activity of one or more enzymes catalyzing its reduced form. the conversion of 2-methyl-1-butanal to 2-methyl-1-butyrate. Alternatively, an enzyme could be overexpressed that In one embodiment, the enzyme catalyzing the conversion of 60 isomerizes DH2MB into DHIV. This approach represents a 2-aceto-2-hydroxybutyrate to 2-ethyl-2,3-dihydroxybutyrate novel pathway for the production of isobutanol from pyru is a 3-ketoacid reductase (3-KAR). In another embodiment, Vate. Thus, in one embodiment, the present invention pro the enzyme catalyzing the conversion of 2-methyl-1-butanal vides a recombinant microorganism for producing isobu to 2-methyl-1-butyrate is an aldehyde dehydrogenase tanol, wherein said recombinant microorganism (ALDH). A non-limiting example of such a pathway in which 65 overexpresses an endogenous or heterologous protein a 3-ketoacid reductase (3-KAR) and an aldehyde dehydroge capable of converting DH2MB into 2,3-dihydroxyisovaler nase (ALDH) are eliminated is depicted in FIG. 7. ate. US 9,012,189 B2 41 42 Overexpression of Enzymes Converting 2-Ethyl-2,3-Dihy Use of Overexpressed Ketol-Acid Reductoisomerase (KARI) droxybutanoate into Biosynthetic Pathway Intermediates and/or Modified Ketol-Acid Reductoisomerase (KARI) to A different approach to reduce or eliminate the production Reduce the Production of DH2MB of 2-ethyl-2,3-dihydroxybutanoate in yeast is to overexpress As described herein, the conversion of acetolactate to an enzyme that converts 2-ethyl-2,3-dihydroxybutanoate into DH2MB competes with the isobutanol pathway for the inter a biosynthetic pathway intermediate. This approach is useful mediate acetolactate. In the current yeast isobutanol produc for any biosynthetic pathway which uses 2-aceto-2-hydroxy tion strains, ketol-acid reductoisomerase (KARI) catalyzes butyrate as an intermediate, including, but not limited to, the conversion of acetolactate to DHIV. 2-methyl-1-butanol, isoleucine, 3-methyl-1-pentanol, 4-me In one embodiment, the present invention provides recom thyl-1-hexanol, and 5-methyl-1-heptanol. One way to accom 10 binant microorganisms having an overexpressed ketol-acid reductoisomerase (KARI), which catalyzes the conversion of plish this is through the use of an enzyme that catalyzes the acetolactate to 2,3-dihydroxyisovalerate (DHIV). The over interconversion of 2-ethyl-2,3-dihydroxybutanoate and 2-ac expression of KARI has the effect of reducing DH2MB pro eto-2-hydroxybutyrate, but favors the oxidation of 2-ethyl-2, duction. In one embodiment, the KARI has at least 0.01 U/mg 3-dihydroxybutanoate. Therefore, in one embodiment, the 15 ofactivity in the lysate. In another embodiment, the KARI has present invention provides a recombinant microorganism for at least 0.03 U/mg of activity in the lysate. In yet another producing a product selected from 2-methyl-1-butanol, iso embodiment, the KARI has at least 0.05, 0.1, 0.5, 1,2,5, or 10 leucine, 3-methyl-1-pentanol, 4-methyl-1-hexanol, and U/mg of activity in the lysate. 5-methyl-1-heptanol wherein said recombinant microorgan In a preferred embodiment, the overexpressed KARI is ism overexpresses an endogenous or heterologous protein engineered to exhibit a reduced K for acetolactate as com capable of converting 2-ethyl-2,3-dihydroxybutanoate into pared to a wild-type or parental KARI. The use of the modi 2-aceto-2-hydroxybutyrate. fied KARI with lower K for acetolactate is expected to In one embodiment, the endogenous or heterologous pro reduce the production of the by-product DH2MB. A KARI tein kinetically favors the oxidative reaction. In another with lower Substrate K is identified by Screening homologs. embodiment, the endogenous or heterologous protein has a 25 In the alternative, the KARI can be engineered to exhibit low K for 2-ethyl-2,3-dihydroxybutanoate and a high K. reduced K by directed evolution using techniques known in for 2-aceto-2-hydroxybutyrate. In yet another embodiment, the art. the endogenous or heterologous protein has a low K for the In each of these embodiments, the KARI may be a variant oxidized form of its cofactor and a high K for the corre enzyme that utilizes NADH (rather than NADPH) as a co 30 factor. Such enzymes are described in the commonly owned sponding reduced form of its cofactor. In yet another embodi and co-pending publication, US 2010/0143997, which is ment, the endogenous or heterologous protein has a higher herein incorporated by reference in its entirety for all pur kfor the oxidative reaction than for the reductive direction. poses. This endogenous or heterologous protein should preferably Use of Overexpressed Dihydroxy Acid Dehydratase (DHAD) have the ability to use a redox cofactor with a high concen 35 to Reduce the Production of DH2MB tration of its oxidized form versus its reduced form. As described herein, the present inventors have found that In one embodiment, the endogenous or heterologous pro overexpression of the isobutanol pathway enzyme, dihy tein is encoded by a gene selected from the group consisting droxyacid dehydratase (DHAD), reduces the production of ofYAL060W.YJR159W,YGL157W,YBL114W,YOR120W, the by-product, DH2MB. YKL055C, YBR159W, YBR149W, YDL168W, YDR368W, 40 Accordingly, in one embodiment, the present invention YLR426W, YCR107W, YILL24W, YML054C, YOL151W, provides recombinant microorganisms having an dihydroxy YMR318C, YBR046C, YHR104W, YIR036C, YDL174C, acid dehydratase (DHAD), which catalyzes the conversion of YDR541C, YBR 145W, YGL039W, YCR105W, YDL124W, 2,3-dihydroxyisovalerate (DHIV) to 2-ketoisovalerate YIR035C, YFLO56C, YNL274C, YLR255C, YGL185C, (KIV). The overexpression of DHAD has the effect of reduc YGL256W, YJR096W, YJR 155W, YPL275W, YOR388C, 45 ing DH2MB production. In one embodiment, the DHAD has YLR070C, YMR083W, YER081W, YJR139C, YDL243C, at least 0.01 U/mg of activity in the lysate. In another embodi YPL113C, YOL165C, YML086C, YMR303C, YDL246C, ment, the DHAD has at least 0.03 U/mg of activity in the YLR070C, YHR063C, YNL33 1C, YFLO57C, YIL155C, lysate. In yet another embodiment, the DHAD has at least YOLO86C, YAL061W, YDR127W, YPR127W, YCL018W, 0.05, 0.1, 0.5, 1, 2, 5, or 10 U/mg of activity in the lysate. YIL074C, YIL124W, and YEL071 W. In addition, heterolo 50 Recombinant Microorganisms for the Production of 3-Hy gous genes can be overexpressed in isoleucine producing droxyacids yeast. For examples beta-hydroxy acid dehydrogenases The present invention provides in additional aspects (EC1.1.1.45 and EC1.1.1.60) would be candidates for over recombinant microorganisms for the production of 3-hy expression. droxyacids as a product or a metabolic intermediate. In one Alternatively an enzyme could be overexpressed that 55 embodiment, these 3-hydroxyacid-producing recombinant isomerizes 2-ethyl-2,3-dihydroxybutanoate into 2,3-dihy microorganisms express acetolactate synthase (ALS) and a droxy-3-methylvalerate. This approach represents a novel 3-ketoacid reductase catalyzing the reduction of 2-acetolac pathway for the production of 2-methyl-1-butanol, isoleu tate to DH2MB. In another embodiment, these 3-hydroxy cine, 3-methyl-1-pentanol. 4-methyl-1-hexanol, and 5-me acid-producing recombinant microorganisms express aceto thyl-1-heptanol from pyruvate. Thus, in one embodiment, the 60 lactate synthase (ALS) and a 3-ketoacid reductase catalyzing present invention provides a recombinant microorganism for the reduction of 2-aceto-2-hydroxybutyrate into 2-ethyl-2,3- producing a product selected from 2-methyl-1-butanol, iso dihydroxybutyrate. leucine, 3-methyl-1-pentanol, 4-methyl-1-hexanol, and These 3-hydroxyacid-producing recombinant microorgan 5-methyl-1-heptanol, wherein said recombinant microorgan isms may be further engineered to reduce or eliminate enzy ism overexpresses an endogenous or heterologous protein 65 matic activity for the conversion of pyruvate to products other capable of converting 2-ethyl-2,3-dihydroxybutanoate into than acetolactate. In one embodiment, the enzymatic activity C.B-dihydroxy-3-methylvalerate. of pyruvate decarboxylase (PDC), lactate dehydrogenase US 9,012,189 B2 43 44 (LDH), pyruvate oxidase, pyruvate dehydrogenase, and/or In accordance with this additional aspect, the present glycerol-3-phosphate dehydrogenase (GPD) is reduced or invention also provides a method of producing an acid prod eliminated. uct, comprising: (a) providing an acid product-producing In a specific embodiment, DH2MB is produced in a recom recombinant microorganism that expresses an aldehyde binant PDC-minus GPD-minus yeast microorganism that dehydrogenase catalyzing the conversion of an aldehyde to overexpresses an ALS gene and expresses a 3-ketoacid reduc acid product, and (b) cultivating said recombinant microor tase. In one embodiment, the 3-ketoacid reductase is natively ganism in a culture medium containing a feedstock providing expressed. In another embodiment, the 3-ketoacid reductase the carbon source, until a recoverable quantity of the desired is heterologously expressed. In yet another embodiment, the acid product is produced. 10 The Microorganism in General 3-ketoacid reductase is overexpressed. In a specific embodi The recombinant microorganisms provided herein can ment, the 3-ketoacid reductase is encoded by the S. cerevisiae express a plurality of heterologous and/or native enzymes TMA29 gene or a homolog thereof. In another specific involved in pathways for the production of beneficial metabo embodiment, the ALS is encoded by the B. subtilis AlsS. lites such as isobutanol, 2-butanol, 1-butanol, 2-butanone, In another specific embodiment, 2-ethyl-2,3-dihydroxybu 15 2,3-butanediol, acetoin, diacetyl, Valine, leucine, pantothenic tyrate is produced in a recombinant PDC-minus GPD-minus acid, isobutylene, 3-methyl-1-butanol, coenzyme A, 2-me yeast microorganism that overexpresses an ALS gene and thyl-1-butanol, isoleucine, 1-pentanol, 1-hexanol, 3-methyl expresses a 3-ketoacid reductase. In one embodiment, the 1-pentanol, 4-methyl-1-pentanol, 4-methyl-1-hexanol, 5-me 3-ketoacid reductase is natively expressed. In another thyl-1-heptanol, and 1-propanol from a Suitable carbon embodiment, the 3-ketoacid reductase is heterologously Source. A non-limiting list of beneficial metabolites produced expressed. In yet another embodiment, the 3-ketoacid reduc in engineered biosynthetic pathways is found herein at Tables tase is overexpressed. In a specific embodiment, the 3-ke 1-3. toacid reductase is encoded by the S. cerevisiae TMA29 gene As described herein, “engineered’ or “modified microor or a homolog thereof. In another specific embodiment, the ganisms are produced via the introduction of genetic material ALS is encoded by the B. subtilis AlsS. 25 into a host or parental microorganism of choice and/or by In accordance with these additional aspects, the present modification of the expression of native genes, thereby modi invention also provides a method of producing 2,3-dihy fying or altering the cellular physiology and biochemistry of droxy-2-methylbutanoic acid (DH2MB), comprising: (a) the microorganism. Through the introduction of genetic providing a DH2MB-producing recombinant microorganism material and/or the modification of the expression of native that expresses acetolactate synthase (ALS) and a 3-ketoacid 30 genes the parental microorganism acquires new properties, reductase catalyzing the reduction of 2-acetolactate to e.g., the ability to produce a new, or greater quantities of an DH2MB, and (b) cultivating said recombinant microorgan intracellular and/or extracellular metabolite. As described ism in a culture medium containing a feedstock providing the herein, the introduction of genetic material into and/or the carbon source, until a recoverable quantity of DH2MB is modification of the expression of native genes in a parental produced. 35 microorganism results in a new or modified ability to produce In accordance with these additional aspects, the present beneficial metabolites such as isobutanol, 2-butanol, 1-bu invention also provides a method of producing 2-ethyl-2,3- tanol, 2-butanone, 2,3-butanediol, acetoin, diacetyl, valine, dihydroxybutyrate, comprising: (a) providing a 2-ethyl-2,3- leucine, pantothenic acid, isobutylene, 3-methyl-1-butanol, dihydroxybutyrate-producing recombinant microorganism coenzyme A, 2-methyl-1-butanol, isoleucine, 1-pentanol, that expresses acetolactate synthase (ALS) and a 3-ketoacid 40 1-hexanol, 3-methyl-1-pentanol, 4-methyl-1-pentanol, 4-me reductase catalyzing the reduction of 2-aceto-2-hydroxybu thyl-1-hexanol, 5-methyl-1-heptanol, and 1-propanol from a tyrate to 2-ethyl-2,3-dihydroxybutyrate, and (b) cultivating Suitable carbon source. The genetic material introduced into said recombinant microorganism in a culture medium con and/or the genes modified for expression in the parental taining a feedstock providing the carbon Source, until a recov microorganism contains gene(s), or parts of genes, coding for erable quantity of 2-ethyl-2,3-dihydroxybutyrate is pro 45 one or more of the enzymes involved in a biosynthetic path duced. way for the production of one or more metabolites selected Recombinant Microorganisms for the Production of Acid from isobutanol, 2-butanol, 1-butanol, 2-butanone, 2,3-bu Products tanediol, acetoin, diacetyl, Valine, leucine, pantothenic acid, The present invention provides in additional aspects isobutylene, 3-methyl-1-butanol, coenzyme A, 2-methyl-1- recombinant microorganisms for the production of acid prod 50 butanol, isoleucine, 1-pentanol, 1-hexanol, 3-methyl-1-pen ucts derived from aldehydes. In one embodiment, these acid tanol, 4-methyl-1-pentanol, 4-methyl-1-hexanol, 5-methyl product producing recombinant microorganisms express an 1-heptanol, and 1-propanol and may also include additional aldehyde dehydrogenase catalyzing the conversion of an elements for the expression and/or regulation of expression of aldehyde to a corresponding acid product. These acid product these genes, e.g., promoter sequences. producing recombinant microorganisms may be further engi 55 In addition to the introduction of a genetic material into a neered to reduce or eliminate competing enzymatic activity host or parental microorganism, an engineered or modified for the undesirable conversion of metabolites upstream of the microorganism can also include alteration, disruption, dele desired acid product. tion or knocking-out of a gene or polynucleotide to alter the In a specific embodiment, the acid product is produced in a cellular physiology and biochemistry of the microorganism. recombinant yeast microorganism that overexpresses an alde 60 Through the alteration, disruption, deletion or knocking-out hyde dehydrogenase. In one embodiment, the aldehyde dehy of a gene or polynucleotide the microorganism acquires new drogenase is natively expressed. In another embodiment, the or improved properties (e.g., the ability to produce a new aldehyde dehydrogenase is heterologously expressed. In yet metabolite or greater quantities of an intracellular metabolite, another embodiment, the aldehyde dehydrogenase is overex to improve the flux of a metabolite down a desired pathway, pressed. In a specific embodiment, the aldehyde dehydroge 65 and/or to reduce the production of by-products). nase is encoded by the S. cerevisiae ALD6 gene or a homolog Recombinant microorganisms provided herein may also thereof. produce metabolites in quantities not available in the parental US 9,012,189 B2 45 46 microorganism. A "metabolite' refers to any Substance pro the enzymatic anabolic or catabolic activity of the reference duced by metabolism or a Substance necessary for or taking polypeptide. Furthermore, the amino acid sequences encoded part in a particular metabolic process. A metabolite can be an by the DNA sequences shown herein merely illustrate organic compound that is a starting material (e.g., glucose or embodiments of the disclosure. pyruvate), an intermediate (e.g., 2-ketoisovalerate), or an end In addition, homologs of enzymes useful for generating product (e.g., isobutanol) of metabolism. Metabolites can be metabolites are encompassed by the microorganisms and used to construct more complex molecules, or they can be methods provided herein. broken down into simpler ones. Intermediate metabolites As used herein, two proteins (or a region of the proteins) may be synthesized from other metabolites, perhaps used to are Substantially homologous when the amino acid sequences make more complex Substances, or broken down into simpler 10 have at least about 30%, 40%, 50% 60%, 65%, 70%, 75%, compounds, often with the release of chemical energy. 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, The disclosure identifies specific genes useful in the meth 98%, or 99% identity. To determine the percent identity of ods, compositions and organisms of the disclosure; however two amino acid sequences, or of two nucleic acid sequences, it will be recognized that absolute identity to Such genes is not the sequences are aligned for optimal comparison purposes necessary. For example, changes in a particular gene or poly 15 (e.g., gaps can be introduced in one or both of a first and a nucleotide comprising a sequence encoding a polypeptide or second amino acid or nucleic acid sequence for optimal align enzyme can be performed and screened for activity. Typically ment and non-homologous sequences can be disregarded for Such changes comprise conservative mutations and silent comparison purposes). In one embodiment, the length of a mutations. Such modified or mutated polynucleotides and reference sequence aligned for comparison purposes is at polypeptides can be screened for expression of a functional least 30%, typically at least 40%, more typically at least 50%, enzyme using methods known in the art. even more typically at least 60%, and even more typically at Due to the inherent degeneracy of the genetic code, other least 70%, 80%, 90%, 100% of the length of the reference polynucleotides which encode Substantially the same or func sequence. The amino acid residues or nucleotides at corre tionally equivalent polypeptides can also be used to clone and sponding amino acid positions or nucleotide positions are express the polynucleotides encoding Such enzymes. 25 then compared. When a position in the first sequence is occu As will be understood by those of skill in the art, it can be pied by the same amino acid residue or nucleotide as the advantageous to modify a coding sequence to enhance its corresponding position in the second sequence, then the mol expression in a particular host. The genetic code is redundant ecules are identical at that position (as used hereinamino acid with 64 possible codons, but most organisms typically use a or nucleic acid “identity” is equivalent to amino acid or subset of these codons. The codons that are utilized most 30 nucleic acid “homology’). The percent identity between the often in a species are called optimal codons, and those not two sequences is a function of the number of identical posi utilized very often are classified as rare or low-usage codons. tions shared by the sequences, taking into account the number Codons can be substituted to reflect the preferred codon usage of gaps, and the length of each gap, which need to be intro of the host, in a process sometimes called "codon optimiza duced for optimal alignment of the two sequences. tion” or “controlling for species codon bias.” 35 When “homologous is used in reference to proteins or Optimized coding sequences containing codons preferred peptides, it is recognized that residue positions that are not by a particular prokaryotic or eukaryotic host (Murray et al., identical often differ by conservative amino acid substitu 1989, Nucl Acids Res. 17: 477-508) can be prepared, for tions. A "conservative amino acid substitution' is one in example, to increase the rate of translation or to produce which an amino acid residue is substituted by another amino recombinant RNA transcripts having desirable properties, 40 acid residuehaving a side chain (Rgroup) with similar chemi Such as a longer half-life, as compared with transcripts pro cal properties (e.g., charge or hydrophobicity). In general, a duced from a non-optimized sequence. Translation stop conservative amino acid substitution will not substantially codons can also be modified to reflect host preference. For change the functional properties of a protein. In cases where example, typical stop codons for S. cerevisiae and mammals two or more amino acid sequences differ from each other by are UAA and UGA, respectively. The typical stop codon for 45 conservative substitutions, the percent sequence identity or monocotyledonous plants is UGA, whereas insects and E. degree of homology may be adjusted upwards to correct for coli commonly use UAA as the stop codon (Dalphin et al., the conservative nature of the substitution. Means for making 1996, NuclAcids Res. 24:216-8). Methodology for optimiz this adjustment are well known to those of skill in the art (See, ing a nucleotide sequence for expression in a plant is pro e.g., Pearson W. R., 1994, Methods in Mol Biol 25: 365-89). vided, for example, in U.S. Pat. No. 6,015,891, and the ref 50 The following six groups each contain amino acids that are erences cited therein. conservative substitutions for one another: 1) Serine (S), Those of skill in the art will recognize that, due to the Threonine (T); 2) Aspartic Acid (D), Glutamic Acid (E); 3) degenerate nature of the genetic code, a variety of DNA Asparagine (N). Glutamine (Q); 4) Arginine (R), Lysine (K); compounds differing in their nucleotide sequences can be 5) Isoleucine (I), Leucine (L), Alanine (A), Valine (V), and 6) used to encode a given enzyme of the disclosure. The native 55 Phenylalanine (F), Tyrosine (Y), Tryptophan (W). DNA sequence encoding the biosynthetic enzymes described Sequence homology for polypeptides, which is also above are referenced herein merely to illustrate an embodi referred to as percent sequence identity, is typically measured ment of the disclosure, and the disclosure includes DNA using sequence analysis software. See commonly owned and compounds of any sequence that encode the amino acid co-pending application US 2009/0226991. A typical algo sequences of the polypeptides and proteins of the enzymes 60 rithm used comparing a molecule sequence to a database utilized in the methods of the disclosure. In similar fashion, a containing a large number of sequences from different organ polypeptide can typically tolerate one or more amino acid isms is the computer program BLAST. When searching a Substitutions, deletions, and insertions in its amino acid database containing sequences from a large number of differ sequence without loss or significant loss of a desired activity. ent organisms, it is typical to compare amino acid sequences. The disclosure includes such polypeptides with different 65 Database searching using amino acid sequences can be mea amino acid sequences than the specific proteins described Sured by algorithms described in commonly owned and co herein so long as the modified or variant polypeptides have pending application US 2009/0226991. US 9,012,189 B2 47 48 It is understood that a range of microorganisms can be are functional in the yeast cell. In one embodiment, the Xylu modified to include a recombinant metabolic pathway Suit lokinase (XK) gene is overexpressed. able for the production of beneficial metabolites from aceto In one embodiment, the microorganism has reduced or no lactate- and/or aldehyde intermediate-requiring biosynthetic pyruvate decarboxylase (PDC) activity. PDC catalyzes the pathways. In various embodiments, microorganisms may be 5 decarboxylation of pyruvate to acetaldehyde, which is then selected from yeast microorganisms. Yeast microorganisms reduced to ethanol by ADH via an oxidation of NADH to for the production of a metabolite such as isobutanol, 2-bu NAD+. Ethanol production is the main pathway to oxidize the tanol, 1-butanol, 2-butanone, 2,3-butanediol, acetoin, NADH from glycolysis. Deletion of this pathway increases diacetyl, Valine, leucine, pantothenic acid, isobutylene, 3-me the pyruvate and the reducing equivalents (NADH) available thyl-1-butanol, coenzyme A, 2-methyl-1-butanol, isoleucine, 10 1-pentanol, 1-hexanol, 3-methyl-1-pentanol, 4-methyl-1- for the biosynthetic pathway. Accordingly, deletion of PDC pentanol, 4-methyl-1-hexanol, 5-methyl-1-heptanol, and genes can further increase the yield of desired metabolites. 1-propanol may be selected based on certain characteristics: In another embodiment, the microorganism has reduced or One characteristic may include the property that the micro no glycerol-3-phosphate dehydrogenase (GPD) activity. organism is selected to convert various carbon sources into 15 GPD catalyzes the reduction of dihydroxyacetone phosphate beneficial metabolites such as isobutanol, 2-butanol, 1-bu (DHAP) to glycerol-3-phosphate (G3P) via the oxidation of tanol, 2-butanone, 2,3-butanediol, acetoin, diacetyl, valine, NADH to NAD+. Glycerol is then produced from G3P by leucine, pantothenic acid, isobutylene, 3-methyl-1-butanol, Glycerol-3-phosphatase (GPP). Glycerol production is a sec coenzyme A, 2-methyl-1-butanol, isoleucine, 1-pentanol, ondary pathway to oxidize excess NADH from glycolysis. 1-hexanol, 3-methyl-1-pentanol, 4-methyl-1-pentanol, 4-me Reduction or elimination of this pathway would increase the thyl-1-hexanol, 5-methyl-1-heptanol, and 1-propanol. The pyruvate and reducing equivalents (NADH) available for the term “carbon source’ generally refers to a substance suitable biosynthetic pathway. Thus, deletion of GPD genes can fur to be used as a source of carbon for prokaryotic or eukaryotic ther increase the yield of desired metabolites. cell growth. Examples of Suitable carbon sources are In yet another embodiment, the microorganism has described in commonly owned and co-pending application 25 reduced or no PDC activity and reduced or no GPD activity. US 2009/0226991. Accordingly, in one embodiment, the PDC-minus/GPD-minus yeast production strains are recombinant microorganism herein disclosed can convert a described in commonly owned and co-pending publications, variety of carbon Sources to products, including but not lim US 2009/0226991 and US 2011/0020889, both of which are ited to glucose, galactose, mannose, Xylose, arabinose, lac herein incorporated by reference in their entireties for all tose, Sucrose, and mixtures thereof. 30 purposes. The recombinant microorganism may thus further include In one embodiment, the yeast microorganisms may be a pathway for the production of isobutanol, 2-butanol, 1-bu selected from the "Saccharomyces Yeast Clade', as described tanol, 2-butanone, 2,3-butanediol, acetoin, diacetyl, valine, in commonly owned and co-pending application US 2009/ leucine, pantothenic acid, isobutylene, 3-methyl-1-butanol, O226991. coenzyme A, 2-methyl-1-butanol, isoleucine, 1-pentanol, 35 The term "Saccharomyces sensu stricto' taxonomy group 1-hexanol, 3-methyl-1-pentanol, 4-methyl-1-pentanol, 4-me is a cluster of yeast species that are highly related to S. thyl-1-hexanol, 5-methyl-1-heptanol, and 1-propanol from cerevisiae (Rainieri et al., 2003, J. Biosci Bioengin 96: 1-9). five-carbon (pentose) Sugars including Xylose. Most yeast Saccharomyces sensu stricto yeast species include but are not species metabolize Xylose via a complex route, in which limited to S. cerevisiae, S. kudriavZevi, S. mikatae, S. baya xylose is first reduced to xylitol via a xylose reductase (XR) 40 nus, S. uvarum, S. Carocanis and hybrids derived from these enzyme. The xylitol is then oxidized to xylulose via a xylitol species (Masneufetal., 1998, Yeast 7:61-72). dehydrogenase (XDH) enzyme. The xylulose is then phos An ancient whole genome duplication (WGD) event phorylated via a xylulokinase (XK) enzyme. This pathway occurred during the evolution of the hemiascomycete yeast operates inefficiently in yeast species because it introduces a and was discovered using comparative genomic tools (Kellis redox imbalance in the cell. The xylose-to-xylitol step uses 45 et al., 2004, Nature 428: 617-24; Dujon et al., 2004, Nature NADH as a cofactor, whereas the xylitol-to-xylulose step 430:35-44; Langkjaeret al., 2003, Nature 428: 848-52; Wolfe uses NADPH as a cofactor. Other processes must operate to et al., 1997, Nature 387: 708-13). Using this major evolution restore the redox imbalance within the cell. This often means ary event, yeast can be divided into species that diverged from that the organism cannot grow anaerobically on Xylose or a common ancestor following the WGD event (termed “post other pentose Sugars. Accordingly, a yeast species that can 50 WGD yeast herein) and species that diverged from the yeast efficiently ferment Xylose and other pentose Sugars into a lineage prior to the WGD event (termed “pre-WGD yeast” desired fermentation product is therefore very desirable. herein). Thus, in one aspect, the recombinant microorganism is Accordingly, in one embodiment, the yeast microorganism engineered to express a functional exogenous Xylose may be selected from a post-WGDyeast genus, including but isomerase. Exogenous Xylose isomerases functional in yeast 55 not limited to Saccharomyces and Candida. The favored post are known in the art. See, e.g., Rajgarhia et al., US2006/ WGDyeast species include: S. cerevisiae, S. uvarum, S. baya 0234364, which is herein incorporated by reference in its nus, S. paradoxus, S. Castelli, and C. glabrata. entirety. In an embodiment according to this aspect, the exog In another embodiment, the yeast microorganism may be enous Xylose isomerase gene is operatively linked to pro selected from a pre-whole genome duplication (pre-WGD) moter and terminator sequences that are functional in the 60 yeast genus including but not limited to Saccharomyces, yeast cell. In a preferred embodiment, the recombinant Kluyveromyces, Candida, Pichia, Issatchenkia, Debaryomy microorganism further has a deletion or disruption of a native ces, Hansenula, Yarrowia and, Schizosaccharomyces. Repre gene that encodes for an enzyme (e.g., XRand/or XDH) that sentative pre-WGDyeast species include: S. kluyveri, K. ther catalyzes the conversion of xylose to xylitol. In a further motolerans, K. marxianus, K. Waltii, K. lactis, C. tropicalis, P preferred embodiment, the recombinant microorganism also 65 pastoris, Panomala, P stipitis, I. Orientalis, I. Occidentalis, I. contains a functional, exogenous Xylulokinase (XK) gene Scutulata, D. hansenii, H. anomala, Y lipolytica, and S. operatively linked to promoter and terminator sequences that pombe. US 9,012,189 B2 49 50 A yeast microorganism may be either Crabtree-negative or (KARI), 3) Dihydroxy-acid dehydratase (DHAD), 4) Keto Crabtree-positive as described in described in commonly isovalerate decarboxylase (KIVD), and 5) an Alcohol dehy owned and co-pending application US 2009/0226991. In one drogenase (ADH) (FIG. 1). In another embodiment, the yeast embodiment the yeast microorganism may be selected from microorganism is engineered to overexpress these enzymes. yeast with a Crabtree-negative phenotype including but not For example, these enzymes can be encoded by native genes. limited to the following genera: Saccharomyces, Kluyvero Alternatively, these enzymes can be encoded by heterologous myces, Pichia, Issatchenkia, Hansenula, and Candida. Crab genes. For example, ALS can be encoded by the alss gene of tree-negative species include but are not limited to: S. B. subtilis, alss of L. lactis, or the ilvK gene of K. pneumonia. kluyveri, K. lactis, K. marxianus, P anomala, P stipitis, I. For example, KARI can be encoded by the ilvC genes of E. Orientalis, I. Occidentalis, I. Scutulata, H. anomala, and C. 10 coli, C. glutamicum, M. maripaludis, or Piromyces sp E2. For utilis. In another embodiment, the yeast microorganism may example, DHAD can be encoded by the ilvD genes of E. coli, be selected from yeast with a Crabtree-positive phenotype, C. glutamicum, or L. lactis. For example, KIVD can be including but not limited to Saccharomyces, Kluyveromyces, encoded by the kiv D gene of L. lactis. ADH can be encoded Zygosaccharomyces, Debaryomyces, Pichia and Schizosac by ADH2, ADH6, or ADH7 of S. cerevisiae or adh A of L. charomyces. Crabtree-positive yeast species include but are 15 lactis. not limited to: S. cerevisiae, S. uvarum, S. bayanus, S. para In one embodiment, pathway steps 2 and 5 may be carried doxus, S. castelli, K. thermotolerans, C. glabrata, Z. bailli, Z. out by KARI and ADH enzymes that utilize NADH (rather rouxii, D. hansenii, P. pastorius, and S. pombe. than NADPH) as a co-factor. Such enzymes are described in Another characteristic may include the property that the the commonly owned and co-pending publication, US 2010/ microorganism is that it is non-fermenting. In other words, it 0.143997, which is herein incorporated by reference in its cannot metabolize a carbon Source anaerobically while the entirety for all purposes. The present inventors have found yeast is able to metabolize a carbon Source in the presence of that utilization of NADH-dependent KARI and ADH oxygen. Nonfermenting yeast refers to both naturally occur enzymes to catalyze pathway steps 2 and 5, respectively, ring yeasts as well as genetically modified yeast. During Surprisingly enables production of isobutanol under anaero anaerobic fermentation with fermentative yeast, the main 25 bic conditions. Thus, in one embodiment, the recombinant pathway to oxidize the NADH from glycolysis is through the microorganisms of the present invention may use an NADH production of ethanol. Ethanol is produced by alcohol dehy dependent KARI to catalyze the conversion of acetolactate drogenase (ADH) via the reduction of acetaldehyde, which is (+NADH) to produce 2,3-dihydroxyisovalerate. In another generated from pyruvate by pyruvate decarboxylase (PDC). embodiment, the recombinant microorganisms of the present In one embodiment, a fermentative yeast can be engineered to 30 invention may use an NADH-dependent ADH to catalyze the be non-fermentative by the reduction or elimination of the conversion of isobutyraldehyde (+NADH) to produce isobu native PDC activity. Thus, most of the pyruvate produced by tanol. In yet another embodiment, the recombinant microor glycolysis is not consumed by PDC and is available for the ganisms of the present invention may use both an NADH isobutanol pathway. Deletion of this pathway increases the dependent KARI to catalyze the conversion of acetolactate pyruvate and the reducing equivalents available for the bio 35 (+NADH) to produce 2,3-dihydroxyisovalerate, and an synthetic pathway. Fermentative pathways contribute to low NADH-dependent ADH to catalyze the conversion of isobu yield and low productivity of desired metabolites such as tyraldehyde (+NADH) to produce isobutanol. isobutanol. Accordingly, deletion of PDC genes may increase In another embodiment, the yeast microorganism may be yield and productivity of desired metabolites such as isobu engineered to have increased ability to convert pyruvate to tanol. 40 isobutanol. In one embodiment, the yeast microorganism may In some embodiments, the recombinant microorganisms be engineered to have increased ability to convert pyruvate to may be microorganisms that are non-fermenting yeast micro isobutyraldehyde. In another embodiment, the yeast micro organisms, including, but not limited to those, classified into organism may be engineered to have increased ability to a genera selected from the group consisting of Tricosporon, convert pyruvate to keto-isovalerate. In another embodiment, Rhodotorula, Myxozyma, or Candida. In a specific embodi 45 the yeast microorganism may be engineered to have increased ment, the non-fermenting yeast is C. xestobii. ability to convert pyruvate to 2,3-dihydroxyisovalerate. In Isobutanol-Producing Yeast Microorganisms another embodiment, the yeast microorganism may be engi As described herein, in one embodiment, a yeast microor neered to have increased ability to convert pyruvate to aceto ganism is engineered to convert a carbon source. Such as lactate. glucose, to pyruvate by glycolysis and the pyruvate is con 50 Furthermore, any of the genes encoding the foregoing Verted to isobutanol via an isobutanol producing metabolic enzymes (or any others mentioned herein (or any of the regu pathway (See, e.g., WO/2007/050671, WO/2008/098227, latory elements that control or modulate expression thereof)) and Atsumi et al., 2008, Nature 45: 86-9). Alternative path may be optimized by genetic/protein engineering techniques, ways for the production of isobutanol have been described in Such as directed evolution or rational mutagenesis, which are WO/2007/050671 and in Dickinson et al., 1998, J Biol Chem 55 known to those of ordinary skill in the art. Such action allows 273:25751-6. those of ordinary skill in the art to optimize the enzymes for Accordingly, in one embodiment, the isobutanol producing expression and activity in yeast. metabolic pathway to convert pyruvate to isobutanol can be In addition, genes encoding these enzymes can be identi comprised of the following reactions: fied from other fungal and bacterial species and can be 1. 2 pyruvate->acetolactate--CO 60 expressed for the modulation of this pathway. A variety of 2. acetolactate--NAD(P)H->2,3-dihydroxyisovalerate-- organisms could serve as sources for these enzymes, includ NAD(P)" ing, but not limited to, Saccharomyces spp., including S. 3. 2.3-dihydroxyisovalerate->alpha-ketoisovalerate cerevisiae and S. uvarum, Kluyveromyces spp., including K. 4. alpha-ketoisovalerate->isobutyraldehyde--CO thermotolerans, K. lactis, and K. marxianus, Pichia spp., 5. isobutyraldehyde--NAD(P)H->isobutanol--NAD(P)+ 65 Hansenula spp., including H. polymorpha, Candida spp., These reactions are carried out by the enzymes 1) Aceto TrichospOron spp., Yamadazyma spp., including Y. spp. Stipi lactate Synthase (ALS), 2) Ketol-acid Reducto-Isomerase tis, Torulaspora pretoriensis, Issatchenkia Orientalis, US 9,012,189 B2 51 52 Schizosaccharomyces spp., including S. pombe, Cryptococ microorganism is (i) engineered to reduce or eliminate the cus spp., Aspergillus spp., Neurospora spp., or Ustilago spp. expression or activity of an enzyme catalyzing the conversion Sources of genes from anaerobic fungi include, but not lim of acetolactate to DH2MB and (ii) engineered to reduce or ited to, Piromyces spp., Orpinomyces spp., or Neocallinastix eliminate the expression or activity of an enzyme catalyzing spp. Sources of prokaryotic enzymes that are useful include, the conversion of isobutyraldehyde to isobutyrate. In some but not limited to, Escherichia. coli, Zymomonas mobilis, embodiments, the enzyme catalyzing the conversion of aceto Staphylococcus aureus, Bacillus spp., Clostridium spp., lactate to DH2MB is a 3-ketoacid reductase (3-KAR). In a Corynebacterium spp., Pseudomonas spp., Lactococcus spp., specific embodiment, the 3-ketoacid reductase is encoded by Enterobacter spp., and Salmonella spp. the S. cerevisiae TMA29 (YMR226C) gene or a homolog or In one embodiment, the invention is directed to a recom 10 variant thereof. In one embodiment, the homolog is selected binant microorganism for producing isobutanol, wherein said from the group consisting of Vanderwaltonzyma polyspora recombinant microorganism comprises an isobutanol pro (SEQ ID NO: 2), Saccharomyces castellii (SEQ ID NO:3), ducing metabolic pathway and wherein said microorganism Candida glabrata (SEQID NO: 4), Saccharomyces bayanus is engineered to reduce or eliminate the expression or activity (SEQIDNO:5), Zgosaccharomyces rouxii (SEQID NO:6), of an enzyme catalyzing the conversion of acetolactate to 15 Kluyveromyces lactis (SEQID NO:7), Ashbya gossypii (SEQ DH2MB. In some embodiments, the enzyme catalyzing the ID NO: 8), Saccharomyces kluyveri (SEQ ID NO: 9), conversion of acetolactate to DH2MB is a 3-ketoacid reduc Kluyveromyces thermotolerans (SEQID NO: 10), Kluyvero tase (3-KAR). In a specific embodiment, the 3-ketoacid myces waltii (SEQID NO: 11), Pichia stipitis (SEQID NO: reductase is encoded by the S. cerevisiae TMA29 12), Debaromyces hansenii (SEQ ID NO: 13), Pichia pas (YMR226C) gene or a homolog thereof. In one embodiment, toris (SEQID NO: 14), Candida dubliniensis (SEQ ID NO: the homolog may be selected from the group consisting of 15), Candida albicans (SEO ID NO 16), Yarrowia lipolytica Vanderwaltomzyma polyspora (SEQ ID NO: 2), Saccharo (SEQID NO: 17), Issatchenkia orientalis (SEQID NO: 18), myces castellii (SEQID NO:3), Candida glabrata (SEQID Aspergillus nidulans (SEQ ID NO: 19), Aspergillus niger NO: 4), Saccharomyces bayanus (SEQID NO. 5), Zygosac (SEQ ID NO: 20), Neurospora crassa (SEQ ID NO: 21), charomyces rouxii (SEQ ID NO: 6), Kluyveromyces lactis 25 Schizosaccharomyces pombe (SEQ ID NO: 22), and (SEQ ID NO: 7), Ashbya gossypii (SEQID NO: 8), Saccha Kluyveromyces marxianus (SEQ ID NO. 23). In some romyces kluyveri (SEQ ID NO: 9), Kluyveromyces thermo embodiments, the enzyme catalyzing the conversion of isobu tolerans (SEQ ID NO: 10), Kluyveromyces waltii (SEQ ID tyraldehyde to isobutyrate is an aldehyde dehydrogenase. In a NO: 11), Pichia stipitis (SEQ ID NO: 12), Debaromyces specific embodiment, the aldehyde dehydrogenase is the S. hansenii (SEQID NO: 13), Pichiapastoris (SEQID NO: 14), 30 cerevisiae aldehyde dehydrogenase ALD6 (SEQID NO:25) Candida dubliniensis (SEQ ID NO: 15), Candida albicans or a homolog or variant thereof. In one embodiment, the (SEQ ID NO: 16), Yarrowia lipolytica (SEQ ID NO: 17), homolog is selected from the group consisting of Saccharo Issatchenkia Orientalis (SEQID NO: 18), Aspergillus nidu myces castelli (SEQID NO: 26), Candida glabrata (SEQID lans (SEQID NO: 19), Aspergillus niger (SEQ ID NO: 20), NO: 27), Saccharomyces bayanus (SEQ ID NO: 28), Neurospora crassa (SEQID NO: 21), Schizosaccharomyces 35 Kluyveromyces lactis (SEQID NO: 29), Kluyveromyces ther pombe (SEQ ID NO: 22), and Kluyveromyces marxianus motolerans (SEQID NO:30), Kluyveromyces waltii (SEQID (SEQ ID NO. 23). NO:31), Saccharomyces cerevisiae YJ789 (SEQID NO:32), In another embodiment, the invention is directed to a Saccharomyces cerevisiae JAY291 (SEQ ID NO: 33), Sac recombinant microorganism for producing isobutanol, charomyces cerevisiae EC 1118 (SEQID NO:34), Saccharo wherein said recombinant microorganism comprises an 40 myces cerevisiae DBY939 (SEQID NO:35), Saccharomyces isobutanol producing metabolic pathway and wherein said cerevisiae AWRI1631 (SEQ ID NO: 36), Saccharomyces microorganism is engineered to reduce or eliminate the cerevisiae RM11-1a (SEQID NO:37), Pichia pastoris (SEQ expression or activity of an enzyme catalyzing the conversion ID NO: 38), Kluyveromyces marxianus (SEQ ID NO:39), of isobutyraldehyde to isobutyrate. In some embodiments, the Schizosaccharomyces pombe (SEQ ID NO: 40), and enzyme catalyzing the conversion of isobutyraldehyde to 45 Schizosaccharomyces pombe (SEQID NO: 41). isobutyrate is an aldehyde dehydrogenase. In an exemplary In one embodiment, the isobutanol producing metabolic embodiment, the aldehyde dehydrogenase is the S. cerevisiae pathway comprises at least one exogenous gene that catalyzes aldehyde dehydrogenase ALD6 (SEQ ID NO: 25) or a a step in the conversion of pyruvate to isobutanol. In another homolog or variant thereof. In one embodiment, the homolog embodiment, the isobutanol producing metabolic pathway is selected from the group consisting of Saccharomyces cas 50 comprises at least two exogenous genes that catalyze steps in telli (SEQID NO: 26), Candida glabrata (SEQID NO: 27), the conversion of pyruvate to isobutanol. In yet another Saccharomyces bayanus (SEQ ID NO: 28), Kluyveromyces embodiment, the isobutanol producing metabolic pathway lactis (SEQ ID NO: 29), Kluyveromyces thermotolerans comprises at least three exogenous genes that catalyze steps (SEQ ID NO:30), Kluyveromyces waltii (SEQ ID NO: 31), in the conversion of pyruvate to isobutanol. In yet another Saccharomyces cerevisiae YJ789 (SEQID NO:32), Saccha 55 embodiment, the isobutanol producing metabolic pathway romyces cerevisiae JAY291 (SEQID NO:33), Saccharomy comprises at least four exogenous genes that catalyze steps in ces cerevisiae EC 1118 (SEQ ID NO. 34), Saccharomyces the conversion of pyruvate to isobutanol. In yet another cerevisiae DBY939 (SEQ ID NO:35), Saccharomyces cer embodiment, the isobutanol producing metabolic pathway evisiae AWRI1631 (SEQID NO:36), Saccharomyces cerevi comprises at five exogenous genes that catalyze steps in the siae RM11-1a (SEQ ID NO: 37), Pichia pastoris (SEQ ID 60 conversion of pyruvate to isobutanol. NO: 38), Kluyveromyces marxianus (SEQ ID NO: 39), In one embodiment, one or more of the isobutanol pathway Schizosaccharomyces pombe (SEQ ID NO: 40), and genes encodes an enzyme that is localized to the cytosol. In Schizosaccharomyces pombe (SEQID NO: 41). one embodiment, the recombinant microorganisms comprise In yet another embodiment, the invention is directed to a an isobutanol producing metabolic pathway with at least one recombinant microorganism for producing isobutanol, 65 isobutanol pathway enzyme localized in the cytosol. In wherein said recombinant microorganism comprises an another embodiment, the recombinant microorganisms com isobutanol producing metabolic pathway and wherein said prise an isobutanol producing metabolic pathway with at least US 9,012,189 B2 53 54 two isobutanol pathway enzymes localized in the cytosol. In embodiment, the recombinant microorganism is further engi yet another embodiment, the recombinant microorganisms neered to reduce or eliminate the expression or activity of an comprise an isobutanol producing metabolic pathway with at enzyme catalyzing the conversion of acetolactate to DH2MB least three isobutanol pathway enzymes localized in the cyto as described herein. In another embodiment, the recombinant Sol. In yet another embodiment, the recombinant microorgan microorganism is further engineered to reduce or eliminate isms comprise an isobutanol producing metabolic pathway the expression or activity of an enzyme catalyzing the con with at least four isobutanol pathway enzymes localized in the version of isobutyraldehyde to isobutyrate as described cytosol. In an exemplary embodiment, the recombinant herein. microorganisms comprise an isobutanol producing metabolic In addition to the isobutanol biosynthetic pathway, other pathway with five isobutanol pathway enzymes localized in 10 the cytosol. Isobutanol producing metabolic pathways in biosynthetic pathways utilize ADH enzymes for the conver which one or more genes are localized to the cytosol are sion of an aldehyde to an alcohol. For example, ADH described in commonly owned and co-pending U.S. applica enzymes convert various aldehydes to alcohols as part of tion Ser. No. 12/855,276, which is herein incorporated by biosynthetic pathways for the production of 1-propanol, reference in its entirety for all purposes. 15 2-propanol, 1-butanol, 2-butanol. 1-pentanol, 2-methyl-1-bu Expression of Modified Alcohol Dehydrogenases in the Pro tanol, 3- and methyl-1-butanol. duction of Isobutanol As used herein, the terms “ADH or “ADH enzyme” or Another strategy described herein for reducing the produc "alcohol dehydrogenase' are used interchangeably herein to tion of the by-product isobutyrate is to increase the activity refer to an enzyme that catalyzes the conversion of isobutyral and/or expression of an alcohol dehydrogenase (ADH) dehyde to isobutanol. ADH sequences are available from a responsible for the conversion of isobutyraldehyde to isobu vastarray of microorganisms, including, but not limited to, L. tanol. This strategy prevents competition by endogenous lactis (SEQ ID NO: 175), Streptococcus pneumoniae, Sta enzymes for the isobutanol pathway intermediate, isobutyral phylococcus aureus, and Bacillus cereus. ADH enzymes dehyde. An increase in the activity and/or expression of ADH modifiable by the methods of the present invention include, may be achieved by various means. For example, ADH activ 25 but are not limited to those, disclosed in commonly owned ity can be increased by utilizing a promoter with increased and co-pending U.S. Patent Publication No. 2010/0143997. A promoter strength or by increasing the copy number of the representative list of ADH enzymes modifiable by the meth alcohol dehydrogenase gene. ods described herein can be found in Table 97. In alternative embodiments, the production of the by-prod Modified ADH Enzymes uct isobutyrate may be reduced by utilizing an ADH with 30 In accordance with the invention, any number of mutations increased specific activity for isobutyraldehyde. Such ADH can be made to the ADH enzymes, and in one embodiment, enzymes with increased specific activity for isobutyralde multiple mutations can be made to result in an increased hyde may be identified in nature, or may result from modifi ability to convertisobutyraldehyde to isobutanol. Such muta cations to the ADH enzyme. Such as the modifications tions include point mutations, frame shift mutations, dele described herein. In some embodiments, these modifications 35 tions, and insertions, with one or more (e.g., one, two, three, will produce a decrease in the Michaelis-Menten constant four, five, or six, etc.) point mutations preferred. In an exem (K) for isobutyraldehyde. Through the use of such modified plary embodiment, the modified ADHenzyme comprises one ADH enzymes, competition by endogenous enzymes for or more mutations at positions corresponding to amino acids isobutyraldehyde is further limited. In one embodiment, the selected from: (a) tyrosine 50 of the L. lactis Adha (SEQID isobutyrate yield (mol isobutyrate per mol glucose) in a 40 NO: 185); (b) glutamine 77 of the L. lactis Adha (SEQ ID recombinant microorganism comprising a modified ADH as NO:185); (c) valine 108 of the L. lactis Adh A (SEQID NO: described herein is less than about 5%. In another embodi 185); (d) tyrosine 113 of the L. lactis Adha (SEQ ID NO: ment, the isobutyrate yield (mol isobutyrate per mol glucose) 185); (e) isoleucine 212 of the L. lactis Adh A (SEQ ID NO: in a recombinant microorganism comprising a modified ADH 185); and (f) leucine 264 of the L. lactis Adha (SEQID NO: as described herein is less than about 1%. In yet another 45 185), wherein Adha (SEQID NO:185) is encoded by the L. embodiment, the isobutyrate yield (mol isobutyrate per mol lactis alcohol dehydrogenase (ADH) gene adhA (SEQ ID glucose) in a recombinant microorganism comprising a NO:184) or a codon-optimized version thereof (SEQID NO: modified ADH as described herein is less than about 0.5%, 206). less thanabout 0.1%, less than about 0.05%, or less thanabout Mutations may be introduced into the ADHenzymes of the O.O1%. 50 present invention using any methodology known to those Further, by utilizing a modified ADH enzyme, the present skilled in the art. Mutations may be introduced randomly by, inventors may establish a situation in which the forward reac for example, conducting a PCR reaction in the presence of tion (i.e. the isobutyraldehyde conversion to isobutanol) is the manganese as a divalent metal ion cofactor. Alternatively, favored reaction over the reverse reaction (i.e. the conversion oligonucleotide directed mutagenesis may be used to create of isobutanol to isobutyraldehyde). 55 the modified ADH enzymes which allows for all possible The strategies described above generally lead to a decrease classes of base pair changes at any determined site along the in isobutyrate yield, which is accompanied by an increase in encoding DNA molecule. In general, this technique involves isobutanol yield. Hence, the above strategies are useful for annealing an oligonucleotide complementary (except for one decreasing the isobutyrate yield and/ortiterand for increasing or more mismatches) to a single Stranded nucleotide sequence the ratio of isobutanol yield over isobutyrate yield. 60 coding for the ADH enzyme of interest. The mismatched Accordingly, in one aspect, the present application oligonucleotide is then extended by DNA polymerase, gen describes the generation of modified ADHs with enhanced erating a double-stranded DNA molecule which contains the activity that can facilitate improved isobutanol production desired change in sequence in one Strand. The changes in when co-expressed with the remaining four isobutanol path sequence can, for example, result in the deletion, Substitution, way enzymes. In one embodiment according to this aspect, 65 orinsertion of anamino acid. The double-stranded polynucle the present application is directed to recombinant microor otide can then be inserted into an appropriate expression ganisms comprising one or more modified ADHS. In one vector, and a mutant or modified polypeptide can thus be US 9,012,189 B2 55 56 produced. The above-described oligonucleotide directed more contiguous groups within the reference sequence. As a mutagenesis can, for example, be carried out via PCR. practical matter, whether a given amino acid sequence is, for Enzymes for use in the compositions and methods of the example, at least 50% identical to the amino acid sequence of invention include any enzyme having the ability to convert a reference protein can be determined conventionally using isobutyraldehyde to isobutanol. Such enzymes include, but known computer programs such as those described above for are not limited to, the L. lactis Adh A, the S. pneumoniae nucleic acid sequence identity determinations, or using the Adh A, the S. aureus Adha, and the Bacillus cereus Adh A, CLUSTALW program (Thompson, J. D., et al., Nucleic Acids amongst others. Additional ADH enzymes modifiable by the Res. 22:4673.4680 (1994)). methods of the present invention include, but are not limited In one aspect, amino acid Substitutions are made at one or to those, disclosed in commonly owned and co-pending U.S. 10 Patent Publication No. 2010/0143997. A representative list of more of the above identified positions (i.e., amino acid posi ADH enzymes modifiable by the methods described herein tions equivalent or corresponding to Y50, Q77, V108, Y113, can be found in Table 16. As will be understood by one of I212, or L264 of L. lactis Adha (SEQID NO:185)). Thus, the ordinary skill in the art, modified ADH enzymes may be amino acids at these positions may be substituted with any obtained by recombinant or genetic engineering techniques 15 otheramino acid including Ala, ASn, Arg, Asp, Cys, Gln, Glu, that are routine and well-known in the art. Modified ADH Gly. His, Ile, Leu, Lys, Met, Phe, Pro, Ser. Thr, Trp, Tyr, and enzymes can, for example, be obtained by mutating the gene Val. A specific example of a ADH enzyme which exhibits an or genes encoding the ADH enzyme of interest by site-di increased ability to convert isobutyraldehyde to isobutanol is rected or random mutagenesis. Such mutations may include an ADH in which (1) the tyrosine at position 50 has been point mutations, deletion mutations, and insertional muta replaced with a phenylalanine or tryptophan residue, (2) the tions. For example, one or more point mutations (e.g., Sub glutamine at position 77 has been replaced with an arginine or stitution of one or more amino acids with one or more differ serine residue, (3) the valine at position 108 has been replaced ent amino acids) may be used to construct modified ADH with a serine or alanine residue, (4) the tyrosine at position enzymes of the invention. 113 has been replaced with a phenylalanine or glycine resi The invention further includes homologous ADH enzymes 25 due, (5), the isoleucine at position 212 has been replaced with which are 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 75%, a threonine or valine residue, and/or (6) the leucine at position 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% identical at 264 is replaced with a valine residue. the amino acid level to a wild-type ADH enzyme (e.g., L. Polypeptides having the ability to convert isobutyralde lactis Adha or E. coli Adh A) and exhibit an increased ability hyde to isobutanol for use in the invention may be isolated to convert isobutyraldehyde to isobutanol. Also included 30 from their natural prokaryotic or eukaryotic sources accord within the invention are ADH enzymes, which are 50%, 60%, ing to standard procedures for isolating and purifying natural 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% proteins that are well-known to one of ordinary skill in the art identical at the amino acid level to an ADH enzyme compris (see, e.g., Houts, G. E., et al., J. Virol. 29:517 (1979)). In ing the amino acid sequence set out in SEQID NO: 185 and addition, polypeptides having the ability to convert isobu exhibit an increased ability to convert isobutyraldehyde to 35 tyraldehyde to isobutanol may be prepared by recombinant isobutanol as compared to the unmodified wild-type enzyme. DNA techniques that are familiar to one of ordinary skill in The invention also includes nucleic acid molecules, which the art (see, e.g., Kotewicz, M. L., et al., Nucl. Acids Res. encode the above-described ADH enzymes. 16:265 (1988); Soltis, D. A., and Skalka, A. M., Proc. Natl. The invention also includes fragments of ADH enzymes Acad. Sci. USA 85:33723376 (1988)). which comprise at least 50, 100, 150, 200,250,300,350, 400, 40 In one aspect of the invention, modified ADH enzymes are 450, 500, 550, or 600 amino acid residues and retain one or made by recombinant techniques. To clone a gene or other more activities associated with ADH enzymes. Such frag nucleic acid molecule encoding an ADH enzyme which will ments may be obtained by deletion mutation, by recombinant be modified in accordance with the invention, isolated DNA techniques that are routine and well-known in the art, or by which contains the ADH enzyme gene or open reading frame enzymatic digestion of the ADH enzyme(s) of interest using 45 may be used to construct a recombinant DNA library. Any any of a number of well-known proteolytic enzymes. The vector, well known in the art, can be used to clone the ADH invention further includes nucleic acid molecules, which enzyme of interest. However, the vector used must be com encode the above described modified ADH enzymes and patible with the host in which the recombinant vector will be ADH enzyme fragments. transformed. By a protein or protein fragment having an amino acid 50 Prokaryotic vectors for constructing the plasmid library sequence at least, for example, 50% “identical to a reference include plasmids such as those capable of replication in E. amino acid sequence, it is intended that the amino acid coli such as, for example, pBR322, ColE1, pSC101, puC sequence of the protein is identical to the reference sequence vectors (pUC18, puC19, etc.: In: Molecular Cloning. A except that the protein sequence may include up to 50 amino Laboratory Manual, Cold Spring Harbor Laboratory Press, acid alterations per each 100 amino acids of the amino acid 55 Cold Spring Harbor, N.Y. (1982); and Sambrook et al., In: sequence of the reference protein. In other words, to obtain a Molecular Cloning A Laboratory Manual (2d ed.) Cold protein having an amino acid sequence at least 50% identical Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. to a reference amino acid sequence, up to 50% of the amino (1989)). Bacillus plasmids include pC194, puB110, pE194, acid residues in the reference sequence may be deleted or pC221 pC217, etc. Such plasmids are disclosed by Glyczan, Substituted with another amino acid, or a number of amino 60 T. In: The Molecular Biology Bacilli, Academic Press, York acids up to 50% of the total amino acid residues in the refer (1982), 307 329. Suitable Streptomyces plasmids include ence sequence may be inserted into the reference sequence. pIJ101 (Kendall et al., J. Bacteriol. 169:41.774183 (1987)). These alterations of the reference sequence may occur at the Pseudomonas plasmids are reviewed by John et al., (Rad. amino (N—) and/or carboxy (C ) terminal positions of the Insec. Dis. 8:693 704 (1986)), and Igaki, (Jpn. J. Bacteriol. reference amino acid sequence and/or anywhere between 65 33:729 742 (1978)). Broad-host range plasmids or cosmids, those terminal positions, interspersed either individually such as pCP13 (Darzins and Chakrabarty, J. Bacteriol. 159:9 among residues in the reference sequence and/or in one or 18 (1984)) can also be used for the present invention. US 9,012,189 B2 57 58 Suitable hosts for cloning the ADH nucleic acid molecules L. Coleohominis, Pediococcus sp., including P acidilactici, of interest are prokaryotic hosts. One example of a prokary Bacillus sp., including B. cereus, B. thuringiensis, B. coagul otic host is E. coli. However, the desired ADH nucleic acid lans, B. anthracis, B. Weihenstephanensis, B. mycoides, and molecules of the present invention may be cloned in other B. amyloliquefaciens, Leptotrichia sp., including L. goodfel prokaryotic hosts including, but not limited to, hosts in the lowii, L. buccalis, and L. hofstadii, Actinobacillus sp., includ genera Escherichia, Bacillus, Streptomyces, Pseudomonas, ing A. pleuropneumoniae, Streptococcus sp., including S. Salmonella, Serratia, and Proteus. Sanguinis, S. parasanguinis, S. gordonii, S. pneumoniae, and Eukaryotic hosts for cloning and expression of the ADH S. mitis, Streptobacillus sp., including S. moniliformis, Sta enzyme of interest include yeast and fungal cells. A particu phylococcus sp., including S. aureus, Eikenella sp., including larly preferred eukaryotic host is yeast. Expression of the 10 E. corrodens, Weissella sp., including W. paramesenteroides, desired ADH enzyme in such eukaryotic cells may require the Kingella sp., including K. Oralis, and Rothia sp., including R. use of eukaryotic regulatory regions which include eukary dentocariosa, and Exiguobacterium sp. otic promoters. Cloning and expressing the ADH nucleic acid The nucleotide sequences for several ADH enzymes are molecule in eukaryotic cells may be accomplished by well known. For instance, the sequences of ADH enzymes are known techniques using well known eukaryotic vector sys 15 available from a vast array of microorganisms, including, but temS. not limited to, L. lactis (SEQID NO:185), S. pneumoniae, S. In accordance with the invention, one or more mutations aureus, and Bacillus cereus. ADH enzymes modifiable by the may be made in any ADH enzyme of interest in order to methods of the present invention include, but are not limited increase the ability of the enzyme to convertisobutyraldehyde to those, disclosed in commonly owned and co-pending U.S. to isobutanol, or confer other properties described herein Patent Publication No. 2010/0143997. A representative list of upon the enzyme, in accordance with the invention. Such ADH enzymes modifiable by the methods described herein mutations include point mutations, frame shift mutations, can be found in Table 97. deletions, and insertions. Preferably, one or more point muta In addition, any method can be used to identify genes that tions, resulting in one or more amino acid Substitutions, are encode for ADH enzymes with a specific activity. Generally, used to produce ADH enzymes having an enhanced ability to 25 homologous or analogous genes with similar activity can be convert isobutyraldehyde to isobutanol. In a preferred aspect identified by functional, structural, and/or genetic analysis. In of the invention, one or more mutations at positions equiva most cases, homologous or analogous genes with similar lent or corresponding to positionY50 (e.g., Y50W or Y50F), activity will have functional, structural, or genetic similari Q77 (e.g., Q77S or Q77R), V108 (e.g. V108S or V108A), ties. Techniques known to those skilled in the art may be Y113 (e.g., Y113F or Y113G), I212 (e.g., I212T or I212V), 30 Suitable to identify homologous genes and homologous and/or L264 (e.g. L264V) of the L. lactis Adh A (SEQID NO: enzymes. Generally, analogous genes and/or analogous 185) enzyme may be made to produce the desired result in enzymes can be identified by functional analysis and will other ADH enzymes of interest. have functional similarities. Techniques known to those The corresponding positions of the ADH enzymes identi skilled in the art may be suitable to identify analogous genes fied herein (e.g. the L. lactis Adh A of SEQID NO:185) may 35 and analogous enzymes. For example, to identify homolo be readily identified for other ADH enzymes by one of skill in gous or analogous genes, proteins, or enzymes, techniques the art. Thus, given the defined region and the assays may include, but not limited to, cloning a gene by PCR using described in the present application, one with skill in the art primers based on a published sequence of a gene/enzyme or can make one or a number of modifications, which would by degenerate PCR using degenerate primers designed to result in an increased ability to convert isobutyraldehyde to 40 amplify a conserved region among a gene. Further, one isobutanol in any ADH enzyme of interest. skilled in the art can use techniques to identify homologous or In a preferred embodiment, the modified ADH enzymes analogous genes, proteins, or enzymes with functional have from 1 to 6 amino acid substitutions selected from homology or similarity. Techniques include examining a cell positions corresponding to Y50, Q77, V108, Y113, I212, or or cell culture for the catalytic efficiency or the specific activ L264 as compared to the wild-type ADH enzymes. In other 45 ity of an enzyme through in vitro enzyme assays for said embodiments, the modified ADH enzymes have additional activity, then isolating the enzyme with said activity through amino acid Substitutions at other positions as compared to the purification, determining the protein sequence of the enzyme respective wild-type ADH enzymes. Thus, modified ADH through techniques such as Edman degradation, design of enzymes may have at least about 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, PCR primers to the likely nucleic acid sequence, amplifica 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 50 tion of said DNA sequence through PCR, and cloning of said 29, 30, 31, 32,33, 34,35, 36, 37,38, 39, 40 different residues nucleic acid sequence. To identify homologous or analogous in other positions as compared to the respective wild-type genes with similar activity, techniques also include compari ADH enzymes. As will be appreciated by those of skill in the Son of data concerning a candidate gene or enzyme with art, the number of additional positions that may have amino databases such as BRENDA, KEGG, or MetaCYC. The can acid substitutions will depend on the wild-type ADH enzyme 55 didate gene or enzyme may be identified within the above used to generate the variants. Thus, in some instances, up to mentioned databases in accordance with the teachings herein. 50 different positions may have amino acid substitutions. Furthermore, enzymatic activity can be determined pheno It is understood that various microorganisms can act as typically. “sources' for genetic material encoding ADH enzymes Suit Methods of Making ADH Enzymes with Enhanced Catalytic able for use in a recombinant microorganism provided herein. 60 Efficiency For example. In addition, genes encoding these enzymes can The present invention further provides methods of engi be identified from other fungal and bacterial species and can neering ADH enzymes to enhance their catalytic efficiency. be expressed for the modulation of this pathway. A variety of One approach to increasing the catalytic efficiency of ADH organisms could serve as Sources for these enzymes, includ enzymes is by saturation mutagenesis with NNK libraries. ing, but not limited to, Lactococcus sp., including L. lactis, 65 These libraries may be screened for increases in catalytic Lactobacillus sp., including L. brevis, L. buchneri, L. hilgar efficiency in order to identify, which single mutations con dii, L. fermentum, L. reuteri, L. vaginalis, L. antri, L. Oris, and tribute to an increased ability to convert isobutyraldehyde to US 9,012,189 B2 59 60 isobutanol. Combinations of mutations at aforementioned chromosomal or other DNA in the microorganism or in any residues may be investigated by any method. For example, a cellular compartment, such as a replicating vector in the cyto combinatorial library of mutants may be designed based on plasm. An expression vector also comprises a promoter that the results of the Saturation mutagenesis studies. drives expression of an RNA, which typically is translated Another approach is to use random oligonucleotide into a polypeptide in the microorganism or cell extract. For mutagenesis to generate diversity by incorporating random efficient translation of RNA into protein, the expression vec mutations, encoded on a synthetic oligonucleotide, into the tor also typically contains a ribosome-binding site sequence enzyme. The number of mutations in individual enzymes positioned upstream of the start codon of the coding sequence within the population may be controlled by varying the length of the gene to be expressed. Other elements, such as enhanc of the target sequence and the degree of randomization during 10 ers, secretion signal sequences, transcription termination synthesis of the oligonucleotides. The advantages of this sequences, and one or more marker genes by which host more defined approach are that all possible amino acid muta microorganisms containing the vector can be identified and/ tions and also coupled mutations can be found. or selected, may also be present in an expression vector. If the best variants from the experiments described above Selectable markers, i.e., genes that confer antibiotic resis do not display sufficient activity, directed evolution via error 15 tance or sensitivity, are used and confer a selectable pheno prone PCR may be used to obtain further improvements. type on transformed cells when the cells are grown in an Error-prone PCR mutagenesis of the ADH enzyme may be appropriate selective medium. performed followed by screening for ADH activity. The various components of an expression vector can vary Enhanced ADH Catalytic Efficiency widely, depending on the intended use of the vector and the In one aspect, the catalytic efficiency of the modified ADH host cell(s) in which the vector is intended to replicate or drive enzyme is enhanced. As used herein, the phrase “catalytic expression. Expression vector components Suitable for the efficiency” refers to the property of the ADH enzyme that expression of genes and maintenance of vectors in E. coli, allows it to convert isobutyraldehyde to isobutanol. yeast, Streptomyces, and other commonly used cells are In one embodiment, the catalytic efficiency of the modified widely known and commercially available. For example, Suit ADH is enhanced as compared to the wild-type or parental 25 able promoters for inclusion in the expression vectors of the ADH. Preferably, the catalytic efficiency of the modified disclosure include those that function in eukaryotic or ADH enzyme is enhanced by at least about 5% as compared prokaryotic host microorganisms. Promoters can comprise to the wild-type or parental ADH. More preferably, the cata regulatory sequences that allow for regulation of expression lytic efficiency of the modified ADH enzyme is enhanced by relative to the growth of the host microorganism or that cause at least about 15% as compared to the wild-type or parental 30 the expression of a gene to be turned on or offin response to ADH. More preferably, the catalytic efficiency of the modi a chemical or physical stimulus. For E. coli and certain other fied ADH enzyme is enhanced by at least about 25% as bacterial host cells, promoters derived from genes for biosyn compared to the wild-type or parental ADH. More preferably, thetic enzymes, antibiotic-resistance conferring enzymes, the catalytic efficiency of the modified ADH enzyme is and phage proteins can be used and include, for example, the enhanced by at least about 50% as compared to the wild-type 35 galactose, lactose (lac), maltose, tryptophan (trp), beta-lacta or parental ADH. More preferably, the catalytic efficiency of mase (bla), bacteriophage lambda PL, and T5 promoters. In the modified ADH enzyme is enhanced by at least about 75% addition, synthetic promoters, such as the tac promoter (U.S. as compared to the wild-type or parental ADH. More prefer Pat. No. 4,551,433), can also be used. For E. coli expression ably, the catalytic efficiency of the modified ADH enzyme is vectors, it is useful to include an E. coli origin of replication, enhanced by at least about 100% as compared to the wild-type 40 such as from puC, p1 P. p1, and p3R. or parental ADH. More preferably, the catalytic efficiency of Thus, recombinant expression vectors contain at least one the modified ADH enzyme is enhanced by at least about expression system, which, in turn, is composed of at least a 200% as compared to the wild-type or parental ADH. More portion of a biosynthetic gene coding sequences operably preferably, the catalytic efficiency of the modified ADH linked to a promoter and optionally termination sequences enzyme is enhanced by at least about 500% as compared to 45 that operate to effect expression of the coding sequence in the wild-type or parental ADH. More preferably, the catalytic compatible host cells. The host cells are modified by trans efficiency of the modified ADH enzyme is enhanced by at formation with the recombinant DNA expression vectors of least about 1000% as compared to the wild-type or parental the disclosure to contain the expression system sequences ADH. More preferably, the catalytic efficiency of the modi either as extrachromosomal elements or integrated into the fied ADH enzyme is enhanced by at least about 2000% as 50 chromosome. compared to the wild-type or parental ADH. More preferably, Moreover, methods for expressing a polypeptide from a the catalytic efficiency of the modified ADH enzyme is nucleic acid molecule that are specific to a particular micro enhanced by at least about 3000% as compared to the wild organism (i.e. a yeast microorganism) are well known. For type or parental ADH. Most preferably, the catalytic effi example, nucleic acid constructs that are used for the expres ciency of the modified ADH enzyme is enhanced by at least 55 sion of heterologous polypeptides within Kluyveromyces and about 3500% as compared to the wild-type or parental ADH. Saccharomyces are well known (see, e.g., U.S. Pat. Nos. Gene Expression of Modified ADH Enzymes 4,859,596 and 4,943,529, each of which is incorporated by Provided herein are methods for the expression of one or reference herein in its entirety for Kluyveromyces and, e.g., more of the modified ADH enzyme genes involved the pro Gellissen et al., Gene 190(1):87–97 (1997) for Saccharomy duction of beneficial metabolites and recombinant DNA 60 ces. Yeast plasmids have a selectable marker and an origin of expression vectors useful in the method. Thus, included replication, also known as Autonomously Replicating within the scope of the disclosure are recombinant expression Sequences (ARS). In addition certain plasmids may also con vectors that include Such nucleic acids. The term expression tain a centromeric sequence. These centromeric plasmids are vector refers to a nucleic acid that can be introduced into a generally a single or low copy plasmid. Plasmids without a host microorganism or cell-free transcription and translation 65 centromeric sequence and utilizing either a 2 micron (S. cer system. An expression vector can be maintained permanently evisiae) or 1.6 micron (K. lactis) replication origin are high or transiently in a microorganism, whether as part of the copy plasmids. The selectable marker can be either pro US 9,012,189 B2 61 62 totrophic, such as HIS3, TRP1, LEU2, URA3 or ADE2, or The S. cerevisiae gene TMA29 is also known as antibiotic resistance. Such as, bar, ble, hph, or kan. YMR226C. The open reading frame (ORF) YMR226C is A nucleic acid of the disclosure can be amplified using found on the S. cerevisiae Chromosome XIII at positions cDNA, mRNA synthetic DNA, or alternatively, genomic 722395 . . . 721592. The chromosomal location of YMR226C DNA, as a template and appropriate oligonucleotide primers is a region that is highly syntenic to chromosomes in many according to standard PCR amplification techniques and related yeast Byrne, K. P. and K. H. Wolfe (2005) “The Yeast those procedures described in the Examples section below. Gene Order Browser: combining curated homology and Syn The nucleic acid so amplified can be cloned into an appropri tenic context reveals gene fate in polyploid species. Genome ate vector and characterized by DNA sequence analysis. Fur Res. 15(10): 1456-61. Scannell, D. R. K. P. Byrne, J. L. thermore, oligonucleotides corresponding to nucleotide 10 sequences can be prepared by standard synthetic techniques, Gordon, S. Wong, and K. H. Wolfe (2006) “Multiple rounds e.g., using an automated DNA synthesizer. of speciation associated with reciprocal gene loss in polyp It is also understood that an isolated nucleic acid molecule loidy yeasts.” Nature 440: 341-5. Scannell, D. R. A. C. Frank, encoding a polypeptide homologous to the enzymes G. C. Conant, K. P. Byrne, M. Woolfit, and K. H. Wolfe (2007) described herein can be created by introducing one or more 15 “Independent sorting-out of thousands of duplicated gene nucleotide Substitutions, additions ordeletions into the nucle pairs in two yeast species descended from a whole-genome otide sequence encoding the particular polypeptide. Such that duplication.” Proc Natl AcadSci USA 104: 8397-402. one or more amino acid Substitutions, additions or deletions For example, locations of the Syntenic versions of are introduced into the encoded protein. Mutations can be YMR226C from other yeast species can be found on Chro introduced into the polynucleotide by Standard techniques, mosome 13 in Candida glabrata, Chromosome 1 in Z'gosac Such as site-directed mutagenesis and PCR-mediated charomyces rouxi, Chromosome 2 in K. lactis, Chromosome mutagenesis. In contrast to those positions where it may be 6 in Ashbya gossypii, Chromosome 8 in S. kluyveri, Chromo desirable to make a non-conservative amino acid substitu some 4 in K. thermotolerance and Chromosome 8 from the tions (see above), in some positions it is preferable to make inferred ancestral yeast species Gordon, J. L. K. P. Byrne, conservative amino acid substitutions. A "conservative amino 25 and K. H. Wolfe (2009) Additions, losses, and rearrange acid substitution' is one in which the amino acid residue is ments on the evolutionary route from a reconstructed ancestor replaced with an amino acid residue having a similar side to the modern Saccharomyces cerevisiae genome. PLOS chain. Families of amino acid residues having similar side Genet. 5: e1000485. chains have been defined in the art. These families include Using this syntenic relationship, species-specific versions amino acids with basic side chains (e.g., lysine, arginine, 30 histidine), acidic side chains (e.g., aspartic acid, glutamic of this gene are readily identified and examples can be found acid), uncharged polar side chains (e.g., glycine, asparagine, in Table 4. glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., alanine, Valine, leucine, isoleucine, proline, TABLE 4 phenylalanine, methionine, tryptophan), beta-branched side 35 YMR226C and honologs thereof. chains (e.g., threonine, Valine, isoleucine) and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine). Species Gene Name SEQ ID NO: Although the effect of anamino acid change varies depend S. cerevisiae YMR226C 1 ing upon factors such as phosphorylation, glycosylation, K. polyspora Kpol 1043p53 2 intra-chain linkages, tertiary structure, and the role of the 40 S. casteli Scas 594.12d 3 C. glabrata CAGLOM11242g 4 amino acid in the active site or a possible allosteric site, it is S. bayantis Sbay 651.2 5 generally preferred that the substituted amino acid is from the Z. rotaxi ZYROOA05742p 6 same group as the amino acid being replaced. To some extent K. iactis KLLAOBO8371g 7 the following groups contain amino acids, which are inter A. goSSpi AFR561Wp 8 45 S. kluyveri SAKLOHO4730g 9 changeable: the basic amino acids lysine, arginine, and histi K. thermotoierans KLTHOD13002p 10 dine; the acidic amino acids aspartic and glutamic acids; the K. waiti Kwal 26.9160 11 neutral polar amino acids serine, threonine, cysteine, glutamine, asparagine and, to a lesser extent, methionine; the nonpolar aliphatic amino acids glycine, alanine, Valine, iso In addition to synteny, fungal homologs to the S. cerevisiae leucine, and leucine (however, because of size, glycine and 50 TMA29 gene may be identified by one skilled in the art alanine are more closely related and valine, isoleucine and through tools such as BLAST and sequence alignment. These leucine are more closely related); and the aromatic amino other homologs may be deleted in a similar manner from the acids phenylalanine, tryptophan, and tyrosine. In addition, respective yeast species to eliminate the accumulation of the although classified in different categories, alanine, glycine, 3-hydroxyacid by-product. Examples of homologous pro and serine seem to be interchangeable to Some extent, and 55 teins can be found in Vanderwaltomzyma polyspora (SEQID cysteine additionally fits into this group, or may be classified NO: 2), Saccharomyces castellii (SEQID NO:3), Candida with the polar neutral amino acids. glabrata (SEQID NO: 4), Saccharomyces bayanus (SEQID Methods in General NO: 5), Zgosaccharomyces rouxii (SEQID NO: 6), K. lactis Identification of 3-Ketoacid Reductase Homologs (SEQ ID NO: 7), Ashbya gossypii (SEQID NO: 8), Saccha Any method can be used to identify genes that encode for 60 romyces kluyveri (SEQ ID NO: 9), Kluyveromyces thermo enzymes with 3-ketoacid reductase activity, including, but tolerans (SEQ ID NO: 10), Kluyveromyces waltii (SEQ ID not limited to S. cerevisiae TMA29. Generally, genes that are NO: 11), Pichia stipitis (SEQ ID NO: 12), Debaromyces homologous or similar to 3-ketoacid reductases such as hansenii (SEQID NO: 13), Pichiapastoris (SEQID NO:14), TMA29 can be identified by functional, structural, and/or Candida dubliniensis (SEQ ID NO: 15), Candida albicans genetic analysis. In most cases, homologous or similar genes 65 (SEQ ID NO: 16), Yarrowia lipolytica (SEQ ID NO: 17), and/or homologous or similar enzymes will have functional, Issatchenkia Orientalis (SEQID NO: 18), Aspergillus nidu structural, or genetic similarities. lans (SEQID NO: 19), Aspergillus niger (SEQ ID NO: 20), US 9,012,189 B2 63 64 Neurospora crassa (SEQID NO: 21), Schizosaccharomyces rangements on the evolutionary route from a reconstructed pombe (SEQ ID NO: 22), and Kluyveromyces marxianus ancestor to the modern Saccharomyces cerevisiae genome.” (SEQ ID NO. 23). PLoS Genet. 5: e1000485. Techniques known to those skilled in the art may be suit Using this syntenic relationship, species-specific versions able to identify additional homologous genes and homolo of this gene are readily identified and examples can be found gous enzymes. Generally, analogous genes and/or analogous in Table 5. enzymes can be identified by functional analysis and will have functional similarities. Techniques known to those TABLE 5 skilled in the art may be suitable to identify analogous genes ALD6 and honologs thereof. and analogous enzymes. For example, to identify homolo 10 gous or analogous genes, proteins, or enzymes, techniques Species Gene Name SEQ ID NO: may include, but not limited to, cloning a dehydratase gene by S. cerevisiae YPLO61W 25 PCR using primers based on a published sequence of a gene/ S. casteli Scas 664.24 26 enzyme or by degenerate PCR using degenerate primers C. glabrata CAGLOHO5137g 27 15 S. bayantis Sbay 623.4 28 designed to amplify a conserved region among dehydratase K. iactis KLLAOE23057 29 genes. Further, one skilled in the art can use techniques to K. thermotoierans KLTHOE12210g 30 identify homologous or analogous genes, proteins, or K. waiti Kwal 27.119760 31 enzymes with functional homology or similarity. Techniques include examining a cell or cell culture for the catalytic activ In addition to synteny, fungal homologs to the S. cerevisiae ity of an enzyme through in vitro enzyme assays for said ALD6 gene may be identified by one skilled in the art through activity (e.g. as described herein or in Kiritani, K. Branched tools such as BLAST and sequence alignment. These other Chain Amino Acids Methods Enzymology, 1970), then iso homologs may be deleted in a similar manner from the lating the enzyme with said activity through purification, respective yeast species to eliminate the accumulation of the determining the protein sequence of the enzyme through 25 aldehyde by-product. Examples of homologous proteins can techniques such as Edman degradation, design of PCR prim be found in Saccharomyces castelli (SEQID NO: 26), Can ers to the likely nucleic acid sequence, amplification of said dida glabrata (SEQ ID NO: 27), Saccharomyces bayanus DNA sequence through PCR, and cloning of said nucleic acid (SEQ ID NO: 28), Kluyveromyces lactis (SEQ ID NO: 29), sequence. To identify homologous or similar genes and/or Kluyveromyces thermotolerans (SEQID NO: 30), Kluyvero homologous or similar enzymes, analogous genes and/or 30 myces waltii (SEQ ID NO: 31), Saccharomyces cerevisiae analogous enzymes or proteins, techniques also include com YJ789 (SEQID NO:32), Saccharomyces cerevisiae JAY291 parison of data concerning a candidate gene or enzyme with (SEQID NO:33), Saccharomyces cerevisiae EC1118 (SEQ databases such as BRENDA, KEGG, or MetaCYC. The can ID NO. 34), Saccharomyces cerevisiae DBY939 (SEQ ID didate gene or enzyme may be identified within the above NO: 35), Saccharomyces cerevisiae AWRI1631 (SEQ ID mentioned databases in accordance with the teachings herein. 35 NO:36), Saccharomyces cerevisiae RM11-1a (SEQID NO: Identification of Aldehyde Dehydrogenase Homologs 37), Pichia pastoris (SEQID NO:38), Kluyveromyces marx Any method can be used to identify genes that encode for ianus (SEQID NO:39), Schizosaccharomyces pombe (SEQ enzymes with aldehyde dehydrogenase activity, including, ID NO: 40), and Schizosaccharomyces pombe (SEQID NO: but not limited, to the S. cerevisiae ALD6. Generally, genes 41). that are homologous or similar to aldehyde dehydrogenases 40 Identification of an ADH or KDH in a Microorganism such as ALD6 can be identified by functional, structural, Any method can be used to identify genes that encode for and/or genetic analysis. In most cases, homologous or similar enzymes with alcohol dehydrogenase (ADH) or ketoacid genes and/or homologous or similar enzymes will have func dehydrogenase (KDH) activity. Alcohol dehydrogenase tional, structural, or genetic similarities. (ADH) can catalyze the reversible conversion of isobutanol to The S. cerevisiae gene ALD6 is also known by its system 45 isobutyraldehyde. Ketoacid dehydrogenases (KDH) can cata atic name YPL061W. The open reading frame (ORF) lyze the conversion of 2-ketoisovalerate to isobutyryl-CoA, YPLO61 W is found on the S. cerevisiae Chromosome XVI at which can be converted further to isobutyrate by the action of positions 432585 . . . 434087. The chromosomal location of transacetylase and carboxylic acid kinase enzymes. Gener YPL061W is a region that is highly syntenic to chromosomes ally, genes that are homologous or similar to known alcohol in many related yeast Byrne, K. P. and K. H. Wolfe (2005) 50 dehydrogenases and ketoacid dehydrogenases can be identi “The Yeast Gene Order Browser: combining curated homol fied by functional, structural, and/or genetic analysis. In most ogy and Syntenic context reveals gene fate in polyploid spe cases, homologous or similar alcohol dehydrogenase genes cies. Genome Res. 15: 1456-61. Scannell, D. R. K. P. Byrne, and/or homologous or similar alcohol dehydrogenase J. L. Gordon, S. Wong, and K. H. Wolfe (2006) “Multiple enzymes will have functional, structural, or genetic similari rounds of speciation associated with reciprocal gene loss in 55 ties. Likewise, homologous or similar ketoacid dehydroge polyploidy yeasts.” Nature 440: 341-5. Scannell, D. R. A. C. nase genes and/or homologous or similar ketoacid dehydro Frank, G.C. Conant, K. P. Byrne, M. Woolfit, and K. H. Wolfe genase enzymes will have functional, structural, or genetic (2007) “Independent sorting-out of thousands of duplicated similarities. gene pairs in two yeast species descended from a whole Identification of PDC and GPD in a Yeast Microorganism genome duplication.” Proc Natl Acad Sci USA 104: 8397 60 Any method can be used to identify genes that encode for 402. enzymes with pyruvate decarboxylase (PDC) activity or glyc For example, locations of the Syntenic versions of erol-3-phosphate dehydrogenase (GPD) activity. Suitable YPL061W from other yeast species can be found on Chro methods for the identification of PDC and GPD are described mosome 8 in Candida glabrata, Chromosome 5 in K. lactis, in commonly owned and co-pending publications, US 2009/ Chromosome 5 in K. thermotolerans and Chromosome 8 65 0226991 and US 2011/0020889, both of which are herein from the inferred ancestral yeast species Gordon, J. L., K. P. incorporated by reference in their entireties for all pur Byrne, and K. H. Wolfe (2009) Additions, losses, and rear poses. US 9,012,189 B2 65 66 Genetic Insertions and Deletions also refers to the elimination of enzymatic activity as com Any method can be used to introduce a nucleic acid mol pared to a comparable yeast cell of the same species. Thus, ecule into yeast and many Such methods are well known. For yeast cells lacking 3-ketoacid reductase, PDC, ALDH or example, transformation and electroporation are common glycerol-3-phosphate dehydrogenase (GPD) activity are con methods for introducing nucleic acid into yeast cells. See, sidered to have reduced 3-ketoacid reductase, PDC, ALDH or e.g., Gietz et al., 1992, Nuc Acids Res. 27: 69-74; Ito et al., glycerol-3-phosphate dehydrogenase (GPD) activity since 1983, J. Bacteriol. 153: 163-8; and Becker et al., 1991, Meth most, if not all, comparable yeast strains have at least some ods in Enzymology 194: 182-7. 3-ketoacid reductase, PDC, ALDH, or glycerol-3-phosphate In an embodiment, the integration of a gene of interest into dehydrogenase (GPD) activity. Such reduced enzymatic a DNA fragment or target gene of a yeast microorganism 10 activities can be the result of lower enzyme concentration, occurs according to the principle of homologous recombina lower specific activity of an enzyme, or a combination tion. According to this embodiment, an integration cassette thereof. Many different methods can be used to make yeast containing a module comprising at least one yeast marker having reduced enzymatic activity. For example, a yeast cell gene and/or the gene to be integrated (internal module) is can be engineered to have a disrupted enzyme-encoding locus flanked on either side by DNA fragments homologous to 15 using common mutagenesis or knock-out technology. See, those of the ends of the targeted integration site (recombino e.g., Methods in Yeast Genetics (1997 edition), Adams, genic sequences). After transforming the yeast with the cas Gottschling, Kaiser, and Stems, Cold Spring Harbor Press sette by appropriate methods, a homologous recombination (1998). In addition, certain point-mutation(s) can be intro between the recombinogenic sequences may result in the duced which results in an enzyme with reduced activity. Also internal module replacing the chromosomal region in included within the scope of this invention are yeast strains between the two sites of the genome corresponding to the which when found in nature, are substantially free of one or recombinogenic sequences of the integration cassette. (Orr more activities selected from 3-ketoacid reductase, PDC, Weaver et al., 1981, PNAS USA 78: 6354-58). ALDH, or glycerol-3-phosphate dehydrogenase (GPD) In an embodiment, the integration cassette for integration activity. of a gene of interest into a yeast microorganism includes the 25 Alternatively, antisense technology can be used to reduce heterologous gene under the control of an appropriate pro enzymatic activity. For example, yeast can be engineered to moter and terminator together with the selectable marker contain a cDNA that encodes an antisense molecule that flanked by recombinogenic sequences for integration of a prevents an enzyme from being made. The term “antisense heterologous gene into the yeast chromosome. In an embodi molecule' as used herein encompasses any nucleic acid mol ment, the heterologous gene includes an appropriate native 30 ecule that contains sequences that correspond to the coding gene desired to increase the copy number of a native gene(s). Strand of an endogenous polypeptide. An antisense molecule The selectable marker gene can be any marker gene used in also can have flanking sequences (e.g., regulatory sequences). yeast, including but not limited to, HIS3, TRP1, LEU2, Thus antisense molecules can be ribozymes or antisense oli URA3, bar, ble, hph, and kan. The recombinogenic sequences gonucleotides. A ribozyme can have any general structure can be chosen at will, depending on the desired integration 35 including, without limitation, hairpin, hammerhead, or site suitable for the desired application. axhead structures, provided the molecule cleaves RNA. In another embodiment, integration of a gene into the chro Yeast having a reduced enzymatic activity can be identified mosome of the yeast microorganism may occur via random using many methods. For example, yeast having reduced integration (Kooistra et al., 2004, Yeast 21: 781-792). 3-ketoacid reductase, PDC, ALDH, or glycerol-3-phosphate Additionally, in an embodiment, certain introduced marker 40 dehydrogenase (GPD) activity can be easily identified using genes are removed from the genome using techniques well common methods, which may include, for example, measur known to those skilled in the art. For example, URA3 marker ing glycerol formation via liquid chromatography. loss can be obtained by plating URA3 containing cells in Overexpression of Heterologous Genes FOA (5-fluoro-orotic acid) containing medium and selecting Methods for overexpressing a polypeptide from a native or for FOA resistant colonies (Boeke et al., 1984, Mol. Gen. 45 heterologous nucleic acid molecule are well known. Such Genet. 197: 345-47). methods include, without limitation, constructing a nucleic The exogenous nucleic acid molecule contained within a acid sequence Such that a regulatory element promotes the yeast cell of the disclosure can be maintained within that cell expression of a nucleic acid sequence that encodes the desired in any form. For example, exogenous nucleic acid molecules polypeptide. Typically, regulatory elements are DNA can be integrated into the genome of the cell or maintained in 50 sequences that regulate the expression of other DNA an episomal state that can stably be passed on (“inherited') to sequences at the level of transcription. Thus, regulatory ele daughter cells. Such extra-chromosomal genetic elements ments include, without limitation, promoters, enhancers, and (such as plasmids, mitochondrial genome, etc.) can addition the like. For example, the exogenous genes can be under the ally contain selection markers that ensure the presence of control of an inducible promoter or a constitutive promoter. Such genetic elements in daughter cells. Moreover, the yeast 55 Moreover, methods for expressing a polypeptide from an cells can be stably or transiently transformed. In addition, the exogenous nucleic acid molecule in yeast are well known. For yeast cells described herein can contain a single copy, or example, nucleic acid constructs that are used for the expres multiple copies of a particular exogenous nucleic acid mol sion of exogenous polypeptides within Kluyveromyces and ecule as described above. Saccharomyces are well known (see, e.g., U.S. Pat. Nos. Reduction of Enzymatic Activity 60 4,859.596 and 4,943,529, for Kluyveromyces and, e.g., Gel Yeast microorganisms within the scope of the invention lissen et al., Gene 190(1):87–97 (1997) for Saccharomyces). may have reduced enzymatic activity Such as reduced 3-ke Yeast plasmids have a selectable marker and an origin of toacid reductase, PDC, ALDH, or glycerol-3-phosphate replication. In addition certain plasmids may also contain a dehydrogenase (GPD) activity. The term “reduced as used centromeric sequence. These centromeric plasmids are gen herein with respect to a particular enzymatic activity refers to 65 erally a single or low copy plasmid. Plasmids without a cen a lower level of enzymatic activity than that measured in a tromeric sequence and utilizing either a 2 micron (S. cerevi comparable yeast cell of the same species. The term reduced siae) or 1.6 micron (K. lactis) replication originare high copy US 9,012,189 B2 67 68 plasmids. The selectable marker can be either prototrophic, products have little or no value. These extra products also such as HIS3, TRP1, LEU2, URA3 or ADE2, or antibiotic require additional capital and operating costs to separate resistance, such as, bar, ble, hph, or kan. these products from the desired metabolite. In another embodiment, heterologous control elements can In one aspect, the present invention provides a method of be used to activate or repress expression of endogenous producing a beneficial metabolite derived from a recombinant genes. Additionally, when expression is to be repressed or microorganism comprising a biosynthetic pathway. eliminated, the gene for the relevant enzyme, protein or RNA In one embodiment, the method includes cultivating a can be eliminated by known deletion techniques. recombinant microorganism comprising a biosynthetic path As described herein, any yeast within the scope of the way which uses a 3-ketoacid as an intermediate in a culture disclosure can be identified by selection techniques specific 10 to the particular enzyme being expressed, over-expressed or medium containing a feedstock providing the carbon source repressed. Methods of identifying the strains with the desired until a recoverable quantity of the beneficial metabolite is phenotype are well known to those skilled in the art. Such produced and optionally, recovering the metabolite. In one methods include, without limitation, PCR, RT-PCR, and embodiment, the 3-ketoacid intermediate is acetolactate. In nucleic acid hybridization techniques such as Northern and 15 an exemplary embodiment, said recombinant microorganism Southern analysis, altered growth capabilities on a particular is engineered to reduce or eliminate the expression or activity Substrate or in the presence of a particular Substrate, a chemi of an enzyme catalyzing the conversion of acetolactate to cal compound, a selection agent and the like. In some cases, DH2MB. The beneficial metabolite may be derived from any immunohistochemistry and biochemical techniques can be biosynthetic pathway which uses acetolactate as intermedi used to determine ifa cell contains aparticular nucleic acid by ate, including, but not limited to, biosynthetic pathways for detecting the expression of the encoded polypeptide. For the production of isobutanol, 2-butanol, 1-butanol, 2-bu example, an antibody having specificity for an encoded tanone, 2,3-butanediol, acetoin, diacetyl, Valine, leucine, pan enzyme can be used to determine whether or not a particular tothenic acid, isobutylene, 3-methyl-1-butanol, 4-methyl-1- yeast cell contains that encoded enzyme. Further, biochemi pentanol, and coenzyme A. In a specific embodiment, the cal techniques can be used to determine if a cell contains a 25 beneficial metabolite is isobutanol. In another embodiment, particular nucleic acid molecule encoding an enzymatic the 3-ketoacid intermediate is 2-aceto-2-hydroxybutyrate. In polypeptide by detecting a product produced as a result of the an exemplary embodiment, said recombinant microorganism expression of the enzymatic polypeptide. For example, trans is engineered to reduce or eliminate the expression or activity forming a cell with a vector encoding acetolactate synthase of an enzyme catalyzing the conversion of 2-aceto-2-hy and detecting increased acetolactate concentrations com 30 droxybutyrate to 2-ethyl-2,3-dihydroxybutyrate. The benefi pared to a cell without the vector indicates that the vector is cial metabolite may be derived from any biosynthetic path both present and that the gene product is active. Methods for way which uses 2-aceto-2-hydroxybutyrate as intermediate, detecting specific enzymatic activities or the presence of par including, but not limited to, biosynthetic pathways for the ticular products are well known to those skilled in the art. For production of 2-methyl-1-butanol, isoleucine, 3-methyl-1- example, the presence of acetolactate can be determined as 35 pentanol, 4-methyl-1-hexanol, and 5-methyl-1-heptanol. described by Hugenholtz and Starrenburg, 1992, Appl. Micro. In another embodiment, the method includes cultivating a Biot. 38:17-22. recombinant microorganism comprising a biosynthetic path Increase of Enzymatic Activity way which uses an aldehyde as an intermediate in a culture Yeast microorganisms of the invention may be further medium containing a feedstock providing the carbon source engineered to have increased activity of enzymes (e.g., 40 until a recoverable quantity of the beneficial metabolite is increased activity of enzymes involved in an isobutanol pro produced and optionally, recovering the metabolite. In an ducing metabolic pathway). The term “increased as used exemplary embodiment, said recombinant microorganism is herein with respect to a particular enzymatic activity refers to engineered to reduce or eliminate the expression or activity of a higher level of enzymatic activity than that measured in a an enzyme catalyzing the conversion of an aldehyde to acid comparable yeast cell of the same species. For example, 45 by-product. The beneficial metabolite may be derived from overexpression of a specific enzyme can lead to an increased any biosynthetic pathway which uses an aldehyde as inter level of activity in the cells for that enzyme. Increased activi mediate, including, but not limited to, biosynthetic pathways ties for enzymes involved in glycolysis or the isobutanol for the production of isobutanol, 1-butanol, 2-methyl-1-bu pathway would result in increased productivity and yield of tanol, 3-methyl-1-butanol. 1-propanol, 1-pentanol, 1-hex isobutanol. 50 anol, 3-methyl-1-pentanol, 4-methyl-1-pentanol, 4-methyl Methods to increase enzymatic activity are known to those 1-hexanol, and 5-methyl-1-heptanol. In a specific skilled in the art. Such techniques may include increasing the embodiment, the beneficial metabolite is isobutanol. expression of the enzyme by increased copy number and/or In another embodiment, the method includes cultivating a use of a strong promoter, introduction of mutations to relieve recombinant microorganism comprising a biosynthetic path negative regulation of the enzyme, introduction of specific 55 way which uses acetolactate and an aldehyde as intermediates mutations to increase specific activity and/or decrease the Km in a culture medium containing a feedstock providing the for the substrate, or by directed evolution. See, e.g., Methods carbon source until a recoverable quantity of the beneficial in Molecular Biology (vol. 231), ed. Arnold and Georgiou, metabolite is produced and optionally, recovering the Humana Press (2003). metabolite. In an exemplary embodiment, said recombinant Methods of Using Recombinant Microorganisms for High 60 microorganism is engineered to (i) reduce or eliminate the Yield Fermentations expression or activity of an enzyme catalyzing the conversion For a biocatalyst to produce a beneficial metabolite most of acetolactate to DH2MB and (ii) reduce or eliminate the economically, it is desirable to produce said metabolite at a expression or activity of an enzyme catalyzing the conversion high yield. Preferably, the only product produced is the of an aldehyde to acid by-product. The beneficial metabolite desired metabolite, as extra products (i.e. by-products) lead to 65 may be derived from any biosynthetic pathway which uses a reduction in the yield of the desired metabolite and an acetolactate and an aldehyde as intermediate, including, but increase in capital and operating costs, particularly if the extra not limited to, biosynthetic pathways for the production of US 9,012,189 B2 69 70 isobutanol, 1-butanol, and 3-methyl-1-butanol. In a specific percent, at least about 35 percent, at least about 40 percent, at embodiment, the beneficial metabolite is isobutanol. least about 45 percent, at least about 50 percent, at least about In another embodiment, the method includes cultivating a 55 percent, at least about 60 percent, at least about 65 percent, recombinant microorganism comprising a biosynthetic path at least about 70 percent, at least about 75 percent, at least way which uses 2-aceto-2-hydroxybutyrate and an aldehyde about 80 percent, at least about 85 percent, at least about 90 as intermediates in a culture medium containing a feedstock percent, at least about 95 percent, or at least about 97.5% providing the carbon source until a recoverable quantity of theoretical. In a specific embodiment, the beneficial metabo the beneficial metabolite is produced and optionally, recov lite is isobutanol. ering the metabolite. In an exemplary embodiment, said This invention is further illustrated by the following recombinant microorganism is engineered to (i) reduce or 10 eliminate the expression or activity of an enzyme catalyzing examples that should not be construed as limiting. The con the conversion of 2-aceto-2-hydroxybutyrate to 2-ethyl-2,3- tents of all references, patents, and published patent applica dihydroxybutyrate and (ii) reduce or eliminate the expression tions cited throughout this application, as well as the Figures or activity of an enzyme catalyzing the conversion of an and the Sequence Listing, are incorporated herein by refer aldehyde to acid by-product. The beneficial metabolite may 15 ence for all purposes. be derived from any biosynthetic pathway which uses 2-ac eto-2-hydroxybutyrate and an aldehyde as intermediate, including, but not limited to, biosynthetic pathways for the production of 2-methyl-1-butanol, 3-methyl-1-pentanol, EXAMPLES 4-methyl-1-hexanol, and 5-methyl-1-heptanol. In another embodiment, the present invention provides a General Methods for Examples 1-26 method of producing a beneficial metabolite derived from an alcohol dehydrogenase (ADH)-requiring biosynthetic path Sequences: Amino acid and nucleotide sequences dis way. In one embodiment, the method includes cultivating a closed herein are shown in Table 6. TABLE 6 Amino Acid and Nucleotide Sequences of Enzymes and Genes Disclosed in Various Examples. Corresponding Protein Enz. Source Gene (SEQID NO) (SEQID NO) ALS B. subtiis Bs alsS1 coSc (SEQ ID NO: 42) Bs. AlsS1 (SEQID NO:43) KARI E. coi Ec ilvC coSce''' (SEQID NO:44) Ec IIvce (SEQID NO: 45) E. coi Ec ilvC coSc2P1-4 (SEQID NO:46) Ec ilvC coSc2P1-4 (SEQID NO:47) KIVD L. lactis Ll kiv D2 coEc (SEQID NO: 48) Ll Kivd2 (SEQID NO:49) DHAD L. lactis Ll ilvD coSc (SEQID NO: 50) Ll Ilv) (SEQID NO: 51) S. cerevisiae Sc ILV3AN (SEQID NO:52) Sc Ilv3AN (SEQID NO:53) ADH D. melanogaster Dm ADH (SEQID NO:54) Dm Adh (SEQID NO: 55) L. lactis Ll adh A (SEQID NO: 56) Ll Adh A (SEQID NO: 57) L. lactis Ll adhA coSci (SEQID NO:58) Ll Adha (SEQID NO:59) L. lactis Ll adhA coSc' (SEQID NO: 60) Ll AdhAF-lis (SEQID NO: 61) recombinant microorganism comprising a modified ADH Media: Medium used was standard yeast medium (see, for described herein in a culture medium containing a feedstock example Sambrook, J., Russel, D. W. Molecular Cloning. A providing the carbon source until a recoverable quantity of Laboratory Manual. 3rd ed. 2001, Cold Spring Harbor, N.Y.: the beneficial metabolite is produced and optionally, recov 45 Cold Spring Harbor Laboratory Press and Guthrie, C. and ering the metabolite. The beneficial metabolite may be Fink, G. R. eds. Methods in Enzymology Part B: Guide to derived from any ADH-requiring biosynthetic pathway, Yeast Genetics and Molecular and Cell Biology 350:3-623 including, but not limited to, biosynthetic pathways for the (2002)). YP medium contains 1% (w/v) yeast extract, 2% production of 1-propanol. 2-propanol, 1-butanol, 2-butanol, 50 (w/v) peptone. YPD is YP containing 2% glucose unless 1-pentanol, 2-methyl-1-butanol, and 3-methyl-1-butanol. In a specified otherwise.YPE is YP containing 25 mL/L ethanol. specific embodiment, the beneficial metabolite is isobutanol. SC medium is 6.7 g/L DifcoTMYeast Nitrogen Base, 14 g/L In a method to produce a beneficial metabolite from a SigmaTM Synthetic Dropout Media supplement (includes carbon Source, the yeast microorganism is cultured in an amino acids and nutrients excluding histidine, tryptophan, appropriate culture medium containing a carbon Source. In 55 uracil, and leucine), 0.076 g/L histidine, 0.076 g/L tryp certain embodiments, the method further includes isolating tophan, 0.380 g/L leucine, and 0.076 g/L uracil. SCD is the beneficial metabolite from the culture medium. For containing 2% (w/v) glucose unless otherwise noted. Drop example, isobutanol may be isolated from the culture medium out versions of SC and SCD media are made by omitting one by any method known to those skilled in the art, Such as or more of histidine (-H), tryptophan (-W), leucine (-L), or distillation, pervaporation, or liquid-liquid extraction. 60 uracil (-U). Solid versions of the above described media con In one embodiment, the recombinant microorganism may tain 2% (w/v) agar. produce the beneficial metabolite from a carbon source at a Cloning techniques: Standard molecular biology methods yield of at least 5 percent theoretical. for cloning and plasmid construction were generally used, In another embodiment, the microorganism may produce unless otherwise noted (Sambrook, J., Russel, D.W. Molecu the beneficial metabolite from a carbon source at a yield of at 65 lar Cloning, A Laboratory Manual. 3 ed. 2001, Cold Spring least about 10 percent, at least about 15 percent, about least Harbor, N.Y.: Cold Spring Harbor Laboratory Press). Cloning about 20 percent, at least about 25 percent, at least about 30 techniques included digestion with restriction enzymes, PCR US 9,012,189 B2 71 72 to generate DNA fragments (KOD Hot Start Polymerase, were spread over YPD plates containing 0.2 g/L G418 selec Cati 71086, Merck, Darmstadt, Germany), ligations of two tive plates. Transformants were then single colony purified DNA fragments using the DNA Ligation Kit (Mighty Mix onto G418 selective plates. Catil TAK 6023, Clontech Laboratories, Madison, Wis.), and K. marxianus transformations: K. marxianus Strains were bacterial transformations into competent E. coli cells 5 grown in 3 mL of an appropriate culture medium at 250 rpm (Xtreme Efficiency DH5a Competent Cells, Catil ABP-CE and 30° C. overnight. The following day, cultures were CC02096P. Allele Biotechnology, San Diego, Calif.). Plas diluted in 50 mL of the same medium and grown to an ODoo mid DNA was purified from E. coli cells using the Qiagen of between 1 and 4. The cells were collected in a sterile 50 mL QIAprep Spin Miniprep Kit (Catil 27106, Qiagen, Valencia, conical tube by centrifugation (1600xg, 5 min at room tem Calif.). DNA was purified from agarose gels using the 10 perature). The cells were resuspended in 10 mL of electropo ration buffer (10 mM Tris-C1, 270 mM sucrose, 1 mM Zymoclean Gel DNA Recovery Kit (Zymo Research, MgCl, pH 7.5), and collected at 1600xg for 5 min at room Orange, Calif.; Catalog #D4002) according to manufactur temperature. The cells were resuspended in 10 mLIB (YPE, er's protocols. 25 mM DTT, 20 mM HEPES, pH 8.0: prepared fresh by Colony PCR: Yeast colony PCR used the FailSafetM PCR 15 diluting 100 uL of 2.5 M DTT and 200 uL of 1 mM HEPES, System (EPICENTRE(R) Biotechnologies, Madison, Wis.; pH 8.0 into 10 mL ofYPD). The cells were incubated for 30 Catalog #FS99250) according to manufacturer's protocols. A min, 250 rpm, 30° C. (tube standing vertical). The cells were PCR cocktail containing 15uL of Master Mix E buffer, 10.5 collected at 1600xg for 5 min at room temperature and resus LL water, 2 LL of each primer at 10 uM concentration, 0.5 L pended in 10 mL of chilled electroporation buffer. The cells polymerase enzyme mix from the kit was added to a 0.2 mL 20 were pelleted at 1600xg for 5 min at 4°C. The cells were PCR tube for each sample (30 ul, each). For each candidate a resuspended in 1 mL of chilled electroporation buffer and Small amount of cells was added to the reaction tube using a transferred to a microfuge tube. The cells were collected by sterile pipette tip. Presence of the positive PCR product was centrifugation at >10,000xg for 20 sec at 4°C. The cells were assessed using agarose gel electrophoresis. resuspended in appropriate amount of chilled electroporation SOE PCR: The PCR reactions were incubated in a ther- 25 buffer for a final biomass concentration of 30-38 OD/mL. mocycler using the following PCR conditions: 1 cycle of 94° 400 uL of cells was added to a chilled electroporation cuvette C.x2 min, 35 cycles of 94°C.x30s, 53° C.x30s, 72° C.x2 min (0.4 cm gap), 50 uL of SOE PCR product (or water control) and 1 cycle of 72°C.x10 min. A master mix was made such was added and mixed by pipetting up and down, and the cuvette was incubated on ice for 30 min. The samples were that each reaction contained the following: 3 ul. MgSO (25 electroporated at 1.8 kV, 1000 Ohm, 25uF. The samples were mM), 5 L 10xKOD buffer, 5 L 50% DMSO, 5 L dNTP then transferred to a 50 mL tube with 1 mL of an appropriate mix (2 mM each), 1 uL KOD, 28 uI dHO, 1.5uL forward culture medium, and the samples were incubated for over primer (10 uM), 1.5 L reverse primer (10 uM), 0.5 L night at 250 rpm at 30° C. After incubation the cells were template (plasmid or genomic DNA). plated onto appropriate agar plates. Genomic DNA Isolation: The Zymo Research ZR Fungal/ 35 K. lactis transformations: K. lactis strains were grown in 3 Bacterial DNA Kit (Zymo Research Orange, Calif.; Catalog mLYPD at 250 rpm and 30°C. overnight. The following day, #D6005) was used for genomic DNA isolation according to cultures were diluted in 50 mLYPD and allowed to grow until manufacturer's protocols with the following modifications. they reached an ODoo of -0.8. Cells from 50 mLYPD cul Following resuspension of pellets, 200 uL was transferred to tures were collected by centrifugation (2700 rcf. 2 min, 25° 2 separate ZR BashingBeadTM Lysis Tubes (to maximize 40 C.). The cells were washed with 50 mL sterile water and yield). Following lysis by bead beating, 400 uL of supernatant collected by centrifugation at 2700 rcf for 2 min at RT. The from each of the ZR BashingBeadTM Lysis Tubes was trans cells were washed again with 25 mL sterile water and col ferred to 2 separate Zymo-SpinTM IV Spin Filters and centri lected by centrifugation at 2700 rcf for 2 min at RT. The cells fuged at 7,000 rpm for 1 min. Following the spin, 1.2 mL of were resuspended in 1 mL 100 mM lithium acetate and trans Fungal/Bacterial DNA Binding Buffer was added to each 45 ferred to a 1.5 mL Eppendorf tube. The cells were collected filtrate. In 800 ul aliquots, filtrate from both filters was trans by centrifugation for 10 sec at 18,000 rcfat RT. The cells were ferred to a single Zymo-SpinTM IIC Column in a collection resuspended in a volume of 100 mM lithium acetate that was tube and centrifuged at 10,000xg for 1 min. For the elution approximately 4x the volume of the cell pellet. A volume of step, instead of eluting in 100 uL of EB (elution buffer, 10-15uL of DNA, 72 uL 50% PEG (3350), 10uL 1 M lithium Qiagen), 50 uL of EB was added, incubated 1 min then the 50 acetate, 3 ul. denatured salmon sperm DNA, and sterile water columns were centrifuged for 1 min. This elution step was were combined to a final volume of 100 uL for each transfor repeated for a final elution volume of 100 uL. mation. In a 1.5 mL tube, 15 LL of the cell Suspension was S. cerevisiae Transformations. S. cerevisiae strains were added to the DNA mixture and the transformation suspension grown in YPD containing 1% ethanol. Transformation-com was vortexed with 5 short pulses. The transformation was petent cells were prepared by resuspension of S. cerevisiae 55 incubated for 30 min at 30°C., followed by incubation for 22 cells in 100 mM lithium acetate. Once the cells were pre min at 42°C. The cells were collected by centrifugation for 10 pared, a mixture of DNA (final volume of 15 uL with sterile sec at 18,000 rcfat RT. The cells were resuspended in 400 uL water), 72 uL 50% PEG, 10 uL 1M lithium acetate, and 3 ul. of an appropriate medium and spread over agar plates con of denatured salmon sperm DNA (10 mg/mL) was prepared taining an appropriate medium to select for transformed cells. for each transformation. In a 1.5 mL tube, 15 uL of the cell 60 Analytical Chemistry: suspension was added to the DNA mixture (100 uL), and the Gas Chromatography (method GC1). Analysis of volatile transformation Suspension was Vortexed for 5 short pulses. organic compounds, including ethanol and isobutanol was The transformation was incubated for 30 min at 30°C., fol performed on a Agilent 5890/6890/7890 gas chromatograph lowed by incubation for 22 min at 42°C. The cells were fitted with an Agilent 7673 Autosampler, a ZB-FFAP column collected by centrifugation (18,000xg, 10 seconds, 25°C.). 65 (J&W; 30 m length, 0.32 mm ID, 0.25 uM film thickness) or The cells were resuspended in 350 LLYPD and after an equivalent connected to a flame ionization detector (FID). overnight recovery shaking at 30° C. and 250 rpm, the cells The temperature program was as follows: 200° C. for the US 9,012,189 B2 73 74 injector, 300° C. for the detector, 100° C. oven for 1 minute, was made using a dilution series of a standard protein stock of 70° C./minute gradient to 23.0°C., and then hold for 2.5 min. 500 lug/mL BSA. An appropriate dilution of cell lysate was Analysis was performed using authentic standards (>99%, made in water to obtain ODsos measurements of each lysate obtained from Sigma-Aldrich, and a 5-point calibration curve that fell within linear range of the BioRad protein standard with 1-pentanol as the internal standard. curve. Ten uL of the lysate dilution was added to 500 uL of High Performance Liquid Chromatography (method diluted BioRad protein assay dye, samples were mixed by LC1): Analysis of organic acid metabolites including 2.3- Vortexing, and incubated at room temperature for 6 min. dihydroxyisovalerate (DHIV), 2,3-dihydroxy-2-methylbu Samples were transferred to cuvettes and read at 595 nm in a tanoic acid (DH2MB), isobutyrate and glucose was per spectrophotometer. The linear regression of the standards was formed on an Agilent 1200 or equivalent High Performance 10 Liquid Chromatography system equipped with a Bio-Rad used to calculate the protein concentration of each sample. Micro-guard Cation HCartridge and two Phenomenex Rezex Alcohol Dehydrogenase (ADH) Assay. Cells were thawed RFQ-Fast Fruit H+ (8%), 100x7.8-mm columns in series, or on ice and resuspended in lysis buffer (100 mM Tris-HCl pH equivalent. Organic acid metabolites were detected using an 7.5). 1000 uL of glass beads (0.5 mm diameter) were added to Agilent 1100 or equivalent UV detector (210 nm) and a 15 a 1.5 mL Eppendorf tube and 875 uL of cell suspension was refractive index detector. The column temperature was 60° C. added. Yeast cells were lysed using a Retsch MM301 mixer This method was isocratic with 0.0180 NHSO in Milli-Q mill (Retsch Inc. Newtown, Pa.), mixing 6x1 min each at full water as mobile phase. Flow was set to 1.1 mL/min. Injection speed with 1 min incubations on ice between each bead Volume was 20 L and run time was 16 min. Quantitation of beating step. The tubes were centrifuged for 10 min at organic acid metabolites was performed using a 5-point cali 23,500xg at 4°C. and the supernatant was removed for use. bration curve with authentic standards (>99% or highest These lysates were held on ice until assayed. Yeast lysate purity available), with the exception of DHIV (2,3-dihy protein concentrations were determined as described. droxy-3-methyl-butanoate, CAS 1756-18-9), which was syn Dilutions of the samples were made such that an activity thesized according to Ciofi et al. (Ciofi, E. et al. Anal Bio reading could be obtained. Generally the samples from Strains chem 1980, 104, pp. 485) and DH2MB which quantified 25 expected to have low ADH activity were diluted 1:5 in lysis based on the assumption that DHIV and DH2MB exhibit the buffer (100 mM Tris-HCl pH 7.5) and the samples from same response factor. In this method, DHIV and DH2MB strains with expected high ADH activity such as strains where co-elute, hence their concentrations are reported as the Sum of the ADH gene is expressed from a high copy number plasmid the two concentrations. were diluted 1:40 to 1:100. Reactions were performed in High Performance Liquid Chromatography (method 30 triplicate using 10 uI of appropriately diluted cell extract LC4): Analysis of oxoacids, including 2,3-dihydroxyisoval with 90 uL of reaction buffer (100 mM Tris-HCl, pH 7.5:150 erate (DHIV, CAS 1756-18-9), 2,3-dihydroxy-2-methylbu uMNADH: 11 mM isobutyraldehyde) in a 96-well plate in a tyrate acid (DH2MB), lactate, acetate, acetolactate, isobu SpectraMax(R340PC multi-plate reader (Molecular Devices, tyrate, and pyruvate) was performed on a Agilent-1 100 High Sunnyvale, Calif.). The reaction was followed at 340 nm for Performance Liquid Chromatography system equipped with 35 5 minutes, with absorbance readings every 10 seconds. The an IonPac AS11-HC Analytical column (Dionex: 9 um, 4.6x reactions were performed at 30°C. The reactions were per 250 mm) coupled with an IonPac AG 11-HC guard column formed in complete buffer and also in buffer with no sub (Dionex: 13 lum, 4.6x50 mm) and an IonPac ATC-3 Anion Strate. Trap column (Dionex: 9x24mm). Acetolactate was detected Isobutyraldehyde Oxidation Assay (ALD6 assay): Cell using a UV detector at 225 nm, while all other analytes were 40 pellets were thawed on ice and resuspended in lysis buffer (10 detected using a conductivity detector (ED50-suppressed mM sodium phosphate pH7.0, 1 mM dithiothreitol, 5% w/v. conductivity with ASRS 4 mm in AutoSuppression recycle glycerol). One mL of glass beads (0.5 mm diameter) was mode, 200 mA Suppressor current). The column temperature added to a 1.5 mL Eppendorf tube for each sample and 850LL was 35°C. Injection size was 10 uL. This method used the of cell Suspension were added. Yeast cells were lysed using a following elution profile:0.25 mMNaOH for 3 min, followed 45 Retsch MM301 mixer mill (Retsch Inc. Newtown, Pa.), mix by a linear gradient from 0.25 to 5 mM NaOH in 22 min and ing 6x1 min each at full speed with 1 min incubation on ice a second linear gradient from 5 mM to 38.25 mM in 0.1 min, between. The tubes were centrifuged for 10 min at 21,500xg followed by 38.25 mM NaOH for 4.9 min and a final linear at 4°C. and the supernatant was transferred to a fresh tube. gradient from 38.25 mM to 0.25 mM for 0.1 min before Extracts were held on ice until assayed. Yeast lysate protein re-equilibrating at 0.25 mM NaOH for 7 min. Flow was set at 50 concentrations were determined as described. 2 mL/min. Analysis was performed using a 4-point calibra The method used to measure enzyme activity of enzymes tion curve with authentic standards (>99%, or highest purity catalyzing the oxidation of isobutyraldehyde to isobutyrate in available), with the following exceptions: DHIV was synthe cell lysates was modified from Meaden et al. 1997, Yeast 13: sized according to Cioffi et al. (Ciofi, E. et al. Anal Biochem 1319-1327 and Postma et al. 1988, Appl. Environ. Microbiol. 1980, 104, pp. 485). DH2MB was synthesized as described in 55 55:468-477. Briefly, for each sample, 10u Lofundiluted cell Example 8 and quantified based on the assumption that DHIV lysate was added to 6 wells of a UV microtiter plate. Three and DH2MB exhibit the same response factor. Racemic wells received 90 uL assay buffer containing 50 mM HEPES acetolactate was made by hydrolysis of Ethyl-2-acetoxy-2- NaOH at pH 7.5, 0.4 mM NADP", 3.75 mM MgCl, and 0.1 methylacetoacetate (EAMMA) with NaOH (Krampitz, L.O. mM, 1 mM, or 10 mM isobutyraldehyde. The other 3 wells Methods in Enzymology 1957, 3, 277-283.). In this method, 60 received 90 uL of no substrate buffer (same as assay buffer but DHIV and DH2MB are separated (FIG. 8). without isobutyraldehyde). The buffers were mixed with the Enzyme Assays lysate in the wells by pipetting up and down. The reactions Determination of protein concentration: Protein concen were then monitored at 340 nm for 5 minutes, with absor tration: (of yeast lysate or of purified protein) was determined bance readings taken every 10 seconds in a SpectraMax(R) using the BioRad Bradford Protein Assay Reagent Kit (Catil 65 340PC plate reader (Molecular Devices, Sunnyvale, Calif.). 500-0006, BioRad Laboratories, Hercules, Calif.) and using The reactions were performed at 30° C. The V, for each BSA for the standard curve. A standard curve for the assay sample was determined by Subtracting the background read US 9,012,189 B2 75 76 ing of the no Substrate control. A no lysate control was also cells were lysed using a Retsch MM301 mixer mill (Retsch performed in triplicate for each Substrate concentration. Inc. Newtown, Pa.), mixing 6x1 min each at full speed with 1 ALS Assay: For ALS assays described in Examples 1-18, min incubation on ice between. The tubes were centrifuged cells were thawed on ice and resuspended in lysis buffer (50 for 10 min at 21.500xg at 4° C. and the supernatant was mM potassium phosphate buffer pH 6.0 and 1 mM MgSO). transferred to a fresh tube. Extracts were held on ice until 1000 uL of glass beads (0.5 mm diameter) were added to a 1.5 assayed. Yeast lysate protein concentration was determined mL Eppendorf tube and 875ul of cell suspension was added. using the BioRad Bradford Protein Assay Reagent Kit (Cath Yeast cells were lysed using a Retsch MM301 mixer mill 500-0006, BioRad Laboratories, Hercules, Calif.) and using (Retsch Inc. Newtown, Pa.), mixing 6x1 min each at full BSA for the standard curve as described. speed with 1 min incubations on ice between each bead 10 Enzymatic synthesis of (S)-2-acetolactate ((S)-AL) was beating step. The tubes were centrifuged for 10 min at performed in an anaerobic flask. The reaction was carried out 23,500xg at 4°C. and the supernatant was removed for use. in a total volume of 55 mL containing 20 mM potassium These lysates were held on ice until assayed. Protein content phosphate pH 7.0, 1 mM MgCl, 0.05 mM thiamine pyro of the lysates was measured as described. All ALS assays phosphate (TPP), and 200 mM sodium pyruvate. The synthe were performed in triplicate for each lysate, both with and 15 sis was initiated by the addition of 65 units of purified B. without Substrate. To assay each lysate, 15 uL of lysate was subtilis AlsS, and the reaction was incubated at 30° C. in a mixed with 135 uL of buffer (50 mM potassium phosphate static incubator for 7.5h. buffer pH 6.0, 1 mM MgSO4, 1 mMthiamin-pyrophosphate, Chemical synthesis of racemic 2-acetolactate ((R/S)-2- 110 mM pyruvate), and incubated for 15 minutes at 30° C. AL) was performed by mixing 50 L of ethyl-2-acetoxy-2- Buffers were prepared at room temperature. A no substrate methylacetoacetate (EAMMA) with 990 u, of water. 260 uL control (buffer without pyruvate) and a no lysate control (lysis of 2 N NaOH was then added in 10 uL increments with 15 buffer instead of lysate) were also included. After incubation seconds of Vortexing after each addition. The solution was 21.5uL of 35% HSO was added to each reaction and incu then mixed on an orbital shaker for 20 minutes. bated at 37° C. for 1 h. Chemical synthesis of racemic AHB ((R/S)-AHB) was For ALS assays described in Examples 19-25, cells were 25 performed by mixing 50 uL of ethyl-2-acetoxy-2-ethyl-3- thawed on ice and resuspended in lysis buffer (100 mM oxobutanoate with 990 uL of water. 2 N NaOH was then NaPO pH 7.0, 5 mM MgCl, and 1 mM DTT). One mL of added in 10 uI, increments with 15 seconds of vortexing after glass beads (0.5 mm diameter) were added to a 1.5 mL Eppen each addition. The NaOH was added until the pH of the dorf tube and 800 uL of the cell suspension was added to the solution was 12 (~180 uL of 2 N NaOH). The solution was tube containing glass beads. Yeast cells were lysed using a 30 then mixed on an orbital shaker for 20 minutes. Retsch MM301 mixer mill (Retsch Inc. Newtown, Pa.) and a For determination of (S)-AL, (R/S)-AL or (R/S)-AHB cooling block by mixing six times for 1 min each at 30 reduction activity, 10uLofundiluted cell lysate was added to cycles/second with 1 min icing in between mixing. The tubes 6 wells of a UV microtiter plate. Three wells received 90 uL were centrifuged for 10 min at 21,500xg at 4° C. and the assay buffer containing 100 mM. KPO at pH 7.0, 150 uM Supernatant was removed. Extracts were held on ice until 35 NADPH, and 5 mM (S)-AL or 10 mM (R/S)-AL or 10 mM assayed. Yeast lysate protein concentration was determined (R/S)-AHB as substrate. The other 3 wells received 90 uL of using the BioRad Bradford Protein Assay Reagent Kit (Catil assay buffer but without substrate. The buffers were mixed 500-0006, BioRad Laboratories, Hercules, Calif.) and using with the lysate in the wells by pipetting up and down. The BSA for the standard curve as described. All ALS assays were reactions were then monitored at 340 nm, with absorbance performed in triplicate for each lysate. All buffers, lysates and 40 readings taken every 10 seconds in a SpectraMax(R 340PC reaction tubes were pre-cooled on ice. To assay each lysate, plate reader (Molecular Devices, Sunnyvale, Calif.). The 15 uL of lysate (diluted with lysis buffer as needed) was reactions were performed at 30°C. The (S)-AL (R/S)-AL or mixed with 135uL of assay buffer (50mMKPi, pH 7.0, 1 mM (R/S)-AHB reduction activity for each sample was deter MgSO 1 mM thiamin-pyrophosphate, 110 mM pyruvate), mined by Subtracting the background reading of the no Sub and incubated for 15 min at 30° C. A no substrate control 45 strate control. A no lysate control was also performed in (buffer without pyruvate) and a no lysate control (lysis buffer triplicate. instead of lysate) were also included. After incubation each DHAD Enzyme Assay: Cell pellets were thawed on ice and reaction was mixed with 21.5uL of 35% HSO, incubated at resuspended in lysis buffer (50 mM Tris pH 8.0, 5 mM 37°C. for 1 h and centrifuged for 5 min at 5,000xg to remove MgSO and G Biosciences Yeast/Fungal ProteaseArrestTM any insoluble precipitants. 50 (St. Louis, Mo., USA, Catalog #788–333)). One mL of glass All assay samples were analyzed for the assay Substrate beads (0.5 mm diameter) was added to a 1.5 mL Eppendorf (pyruvate) and product (acetoin) via high performance liquid tube for each sample and 850 u, of cell suspension were chromatography an HP-1200 High Performance Liquid added. Yeast cells were lysed using a Retsch MM301 mixer Chromatography system equipped with two Restek RFQ mill (Retsch Inc. Newtown, Pa.), mixing 6x1 min each at full 150x4.6 mm columns in series. Organic acid metabolites 55 speed with 1 min incubation on ice between. The tubes were were detected using an HP-1100 UV detector (210 nm) and centrifuged for 10 min at 21.500xg at 4°C. and the superna refractive index. The column temperature was 60° C. This tant was transferred to a fresh tube. Extracts were held on ice method was isocratic with 0.0180 NHSO (in Milli-Q water) until assayed. Yeast lysate protein concentration was deter as mobile phase. Flow was set to 1.1 mL/min. Injection vol mined as described. Protein from each sample was diluted in ume was 20 uL and run time was 8 min. Analysis was per 60 DHAD assay buffer (50 mM Tris pH8, 5 mM MgSO) to a formed using authentic standards (>99%, obtained from final concentration of 0.5g/LL. Three samples of each lysate Sigma-Aldrich) and a 5-point calibration curve. were assayed, along with no lysate controls. 10 uL of each TMA29 enzyme assay: Cell pellets were thawed on ice and sample (or DHAD assay buffer) was added to 0.2 mL PCR resuspended in lysis buffer (10 mM sodium phosphate pH7.0, tubes. Using a multi-channel pipette, 90 uL of the substrate 1 mM dithiothreitol, 5% w/v glycerol). One mL of glass beads 65 was added to each tube (substrate mix was prepared by adding (0.5 mm diameter) was added to a 1.5 mL Eppendorf tube for 4 mL DHAD assay buffer to 0.5 mL 100 mM DHIV). each sample and 850 uL of cell suspension were added. Yeast Samples were put in a thermocycler (Eppendorf Mastercy US 9,012,189 B2 77 78 cler) at 35°C. for 30 min followed by a 5 min incubation at pGV2485 carrying the KARI, DHAD and ADH 95°C. Samples were cooled to 4°C. on the thermocycler, then (Ec ilvC Q11OV. Ll ilvD coSc, and L1 adhA, respec centrifuged at 3000xg for 5 minutes. Finally, 75uI of super tively) as described. natant was transferred to new PCR tubes and analyzed by HPLC as follows 100 uL DNPH reagent (12 mM 2,4-Dini To start fermentation cultures, Small overnight cultures of trophenyl Hydrazine 10 mM Citric Acid pH 3.0 80% Aceto the transformed strains were started inYPD medium contain nitrile 20% MilliOHO) was added to 100 uL of each sample. ing 1% ethanol and 0.2 g/L G418 and incubated overnight at Samples were incubated for 30 min at 70° C. in a thermo 30° C. and 250 rpm. Three biological replicates of each strain cycler (Eppendorf, Mastercycler). Analysis of keto-isovaler were tested. The next morning, the ODoo of these cultures ate and isobutyraldehyde was performed on an HP-1200 High 10 was determined and an appropriate amount used to inoculate Performance Liquid Chromatography system equipped with 50 mL of the same medium in a 50 mL baffled flask to an an Eclipse XDB C-18 reverse phase column (Agilent) and a ODoo of approximately 0.1. These precultures were incu bated at 30° C. and 250 rpm overnight. When the cultures had C-18 reverse phase column guard (Phenomenex). Ketoisov reached an ODoo of approximately 5-6 they were centrifuged alerate and isobutyraldehyde were detected using an 15 HP-1100 UV detector (210 nm). The column temperature at 2700 rpm for 5 min at 25°C. in 50 mL Falcon tubes. The was 50°C. This method was isocratic with 70% acetonitrile to cells from one 50 mL culture (one clone) were resuspended in water as mobile phase with 2.5% dilute phosphoric acid (4%). YPD containing 8% glucose, 0.2 g/L G418, 1% (v/v) ethanol Flow was set to 3 mL/min. Injection size was 10 uL and run (containing 3 g/L ergosterol and 132 g/L Tween-80), and time is 2 min. buffered at pH 6.5 with 200 mMMES. The cultures were then transferred into 250 mL unbaffled flasks and incubated at 30° C. and 75 rpm. At the 72 h timepoint, samples from each fermentation Example 1 flask were taken for determining ODoo, ADH activity, and 25 for analysis by GC1 and LC1. To prepare samples for GC1 Increased Isobutanol/Isobutyrate Ratio by Increasing and LC1 analysis, an appropriate Volume of cell culture was ADH Activity in S. cerevisiae spun in a microcentrifuge for 10 minutes at maximum speed and the supernatant was removed for GC1 and LC1 analysis. The purpose of this example is to demonstrate that Cell pellets were prepared for ADHassays by centrifuging 14 increased alcohol dehydrogenase activity results in an 30 mL of culture medium at 3000xg for 5 minutes at 4°C. The increased isobutanol yield, a decreased isobutyrate yield, and Supernatant was removed and the cells washed in 3 mL cold, an increase in the ratio of isobutanol yield to isobutyrate yield. sterile water. The tubes were then centrifuged as per above for Strains and plasmids disclosed in this example are shown in Tables 7 and 8, respectively. 2 minutes, the Supernatant removed, and the tubes reweighed 35 to determine total cell weight. The Falcon tubes were stored at TABLE 7 -80°C. ADH assays were performed as described. Table 9 shows the ODoo for each strain during the course Genotype of Strains Disclosed in Example 1. of the fermentation. During the 72 h of this fermentation, the GEVO Number Genotype ODoo of the strains were similar: they started at an ODoo of 40 around 7 and ended at an ODoofaround 9. The invitro ADH GEVO2843 S. cerevisiae, MATaura3 leu2 his3 trp1 pdc1A::PCP1:BS alsS1 coSc:Tcyc1: enzymatic activity of lysates from GEVO2843 transformed PPok1: Ll kiv D2: PENo2: Sp HIS5 with the two plasmids was measured for the 72 h timepoint. pdc5A::LEU2: bla: P: ILV3AN: Ps: Table 9 shows the ADH activity in the lysates as measured in Ec ilvC coSc910 vitro. The strain carrying the plasmid with no ADH pdc6A::URA3: bla: P: Ll kiv D2: 45 P: Dm ADH (p.GV2011) showed an activity of about 0.04U/mg. The strain {evolved for C2 supplement-independence, carrying the plasmid with the L1 adhA gene, (pGV2485), had glucose tolerance and faster growth approximately 7-fold more ADH activity.

TABLE 9 50 TABLE 8 ODoo and Alcohol Dehydrogenase Activity of Strain GEVO2843 Plasmids Disclosed in Example 1. Transformed with Plasmids pCV2011 or pGV2485 After 72 h of Plasmid Name Relevant Genes/Usage Genotype Fermentation. 55 pGV2011 2L plasmid expressing Priors:Ec ilvC coSc', GEVO2843 ADH activity KARI, and DHAD PTEF1:Ll ilvD coSc, 2 ori, bla, G418R transformed with OD6oo Umg pGV2485 2L plasmid expressing Prous:Ec ilvC coSc', KARI, DHAD, P:Ll ilvD coSc, pGV2011 8.5 O.04 and ADH PENo2:Ll adhA, pGV2485 9.1 O.29 2 ori, bla, G418R 60

S. cerevisiae strain GEVO2843, which expresses a single Isobutanol and isobutyrate titers after 72 h of fermentation alcohol dehydrogenase (D. melanogaster ADH, Dm ADH) are shown in Table 10. The isobutanol titer in the strain with from its chromosomal DNA was transformed with 2L plas 65 low ADH activity of 0.04U/mg was significantly lower com mids pGV2011 carrying only the KARI and DHAD pared to the strain with high ADH activity of 0.29 U/mg. The (Ec ilvC Q11OV and Ll ilvD coSc, respectively) or isobutyrate titer in the strain with low ADH activity of 0.04 US 9,012,189 B2 79 80 U/mg was significantly higher compared to the Strain with mids pGV2543 carrying KARI, DHAD, KIVD and his high ADH activity of 0.29 U/mg. Table 6 also shows the yield tagged, codon-optimized wild-type ADH (Ec ilvC''', for isobutyrate and isobutanol after 72 h offermentation. The Ll ilvD coSc, and L1 adhA coSc", respectively) or isobutanol yield in the strain with low ADH activity of 0.04 pGV2545 carrying KARI, DHAD, KIVD and his-tagged, U/mg was significantly lower compared to the strain with 5 codon-optimized mutant ADH (Ec ilvC''', Ll ilvD high ADH activity of 0.29 U/mg. The isobutyrate yield in the coSc, and L1 adhA''' coSc', respectively). These strains strain with low ADH activity of 0.04U/mg was significantly were cultured and evaluated for ADH enzyme activity and the higher compared to the strain with high ADH activity of 0.29 production of extracellular metabolites by GC1 and LC1 as U/mg. described. TABLE 10 Titers and Yields for Isobutanol and Isobutyrate in Strain GEVO2843 Transformed with Plasmids pGV2011 or pGV2485 After 72 h of Fermentation. Isobutanol Isobutyrate titer titer isobutanol yield Isobutryate yield Yield ratio g/L) g/L) mol/mol glucose mol/mol glucose (isobutanol isobutyrate) pGV2011 3.2 3.8 O.22 O.22 1.O pGV2485 4.7 1.9 O.33 O.11 3.0

Example 2 The kinetic parameters of the gene products of Ll adhA coSc' L1 adhA''' coSc' (L1 adhA' and Further Increased Isobutanol/Isobutyrate Ratio by L1 adha AREl-hiso , respectively)tivel are shownhown in Table 13. 25 Use of Variant ADH Ll Adha' in S. cerevisiae TABLE 13 The purpose of this example is to demonstrate that expres- Comparison of Kinetic Parameters of Wild-Type Ll adh A" with sion of an alcohol dehydrogenase with increased k, and Modified Ll adh A RE Meatbuy aldehyde with NADH decreased Kresults in a further increase in isobutanol yield, 30 decrease in isobutyrate yield, and increase in the ratio of KAf kcal ka/KM isobutanol yield to isobutyrate yield. Variant mM isobutyraldehydel (s M-1 *s- Ll adh Aisé 11.7 51 4400 TABLE 11 Ll adh Arth 1.6 84 497OO Genotype of Strains Disclosed in Example 2. GEVO Table 14 shows the ODoo for each strain during the course Number Genotype of the fermentation. During the 72 h of this fermentation, the GEVO2843 S. cerevisiae, MATaura3 leu2 his3 trp1 40 ODoo of the strains were similar: they started at an ODoo of pdc1A::PCP:BS alss1 coSc:Tool: around 6 and ended at an ODoofaround 9. The in vitro ADH PPok1: Ll kiv D2: PENo2: Sp HIS5 enzymatic activity of lysates from GEVO2843 transformed pittieri ILV3AN: P : with the two plasmids was measured for the 72 h timepoint. Ec ilvC coSc pdc6A::URA3:bla: P: Ll kiv D2: P3: Dm ADH Table 14 shows the ADH activity in the lysates aS measured in {evolved for C2 Supplement-independence, glucose 45 vitro as described above. The strain carrying the plasmid with tolerance and faster growth L1 adhA coSc' (pGV2543) showed an activity of about 0.38 U/mg. The strain carrying the plasmid with the L1 adhA''' coSc" gene, (p.GV2545), had approximately TABLE 12 7-fold more ADH activity. 50 Plasmids Disclosed in Example 2. TABLE 1.4 Plasmid ODoo, and Alcohol Dehydrogenase Activity of Strain GEVO2843 Name Relevant Genes/Usage Genotype Transformed with Plasmids pCV2543 or pGV2545. After 72 h of Fermentation. pGV2543 21 plasmid expressing Priors:Ec ilvC coSc', 55 KARI, DHAD, KVD, PTEF1:Ll ilvD coSc, ADH activity and ADH (Ll Adha'i) EastE2. GEVO2843 transformed with OD6oo Umg Po2: Ll Adh A', 2 ori, bla, G418R pGV2543 8.5 O.38 pGV2545 21 plasmid expressing Ports:Ec ilvC coSce''', pGV2545 8.8 2.46 KARI, DHAD, KIVD, P:Ll ilvD coSc, 60 and ADH (Ll AdhA'") Peak:Ll kivD coEc, P:2 ori, bla,Ll G418RAdha'', Isobutanol and isobutyrate titers and yield after 72 h of fermentation are shown in Table 15. The isobutanol titer and yield in the strain carrying pGV 2543 was lower compared to S. cerevisiae strain GEVO2843, which expresses a single 65 the strain carrying pCV2545. The isobutyrate titer and yield alcohol dehydrogenase (D. melanogaster ADH, Dm ADH) in the Strain carrying pGV2543 was significantly higher com from its chromosomal DNA was transformed with 2L plas- pared to the strain carrying pCV2545. US 9,012,189 B2 81 82 TABLE 1.5 Titers and Yields for Isobutanol and Isobutyrate in Strain GEVO2843 Transformed with Plasmids pGV 2453 or pGV2485 After 72 h of Fermentation. GEVO2843 transformed Isobutanol Isobutyrate isobutanol yield Isobutryate yield Yield ratio with g/L) g/L) mol/mol glucose mol/mol glucose (Isobutanol isobutyrate) pGV2543 4.6 1.3 O.28 pGV2545 4.9 O.3 O.29

Example 3 TABLE 17-continued Further Increased Isobutanol/Isobutyrate Ratio in S. Plasmids Disclosed in Example 3. cerevisiae by Expression of RE1 15 Plasmid Relevant Genes? Name Usage Genotype The purpose of this example is to demonstrate that expres sion of an alcohol dehydrogenase with increased k and PPok1:Ll kiv D2 coEc decreased K results in an increase in isobutanol yield and a P:Ll adh A coSc''', decrease in isobutyrate yield in fermentations performed in 2 ori, bla, G418R fermenter vessels. A fermentation was performed to compare performance of S. cerevisiae strain GEVO3128 was transformed with S. cerevisiae Strains GEVO3519 and GEVO3523. Isobutanol either 2L plasmid pGV2524 or pGV2546, to generate strains GEVO3519 and GEVO3523, respectively as described. and isobutyrate titers and yields were measured during the 25 fermentation. GEVO3519 carries a 2L plasmid pGV2524 that Inoculum cultures of GEVO3519 and GEVO3523 were contains genes encoding the following enzymes: KARI, started by inoculating 500 mL baffled flasks containing 80 DHAD, KIVD and his-tagged, codon-optimized wild-type mL of YPD medium 0.2 g/L G418 antibiotic, 1% v/v ethanol, and 0.019 g/L tryptophan. The cultures were incubated for Lactococcus lactis ADH. GEVO3523 carries a 2L plasmid approximately 34 h. The orbital shaker was set at 250 rpm and pGV2524 that contains genes encoding the following 30 30° C. in both experiments. Similar cell mass was achieved enzymes: KARI, DHAD, KIVD and an improved variant of for GEVO3519 and GEVO3523 strains. The cell density the his-tagged, codon-optimized Lactococcus lactis ADH achieved after incubation was 8.0 ODoo. Batch fermenta having decreased K and increased k. These strains were tions were conducted in YPD medium containing 80 g/L evaluated for isobutanol, isobutyraldehyde, glucose con 35 glucose, 0.2 g/L G418, 1% V/v ethanol, and 0.019 g/L tryp Sumption by LC1 and GC1, as well as for ODoo during a tophan using 2 L top drive motor DasGip Vessels with a fermentation in DasGipfermenter vessels. working volume of 0.9 L per vessel. Vessels were sterilized, TABLE 16

Genotype of Strains Disclosed in Example 3. GEVO Number Genotype GEVO3128 S. cerevisiae, MATaura3 leu2 his3 trp1

pdc1A::PCP1:BS alss1 coSc:Tcyc1:PPok1:Ll kiv DkivD2:PENo2:Sp HIS5 pdc5A::LEU2:bla:P: ILV3AN:P:Ec ilvC coSce'' pdc6A::Pter:Ll ilvD:Priors:Ec ilvC coSc''':Pevo2:Ll adhA:Pe:Sc TRP1 {evolved for C2 Supplement-independence, glucose tolerance and faster growth GEVO3519 GEVO3128 transformed with plasmid pCV2524 GEVO3523 GEVO3128 transformed with plasmid pCV2546

TABLE 17 55 along with the appropriate dissolved oxygen and pH probes, for 60 minutes at 121° C. Dissolved oxygen probes were Plasmids Disclosed in Example 3. calibrated post sterilization in order to allow for polarization, however, pH probes were calibrated prior to sterilization. The id Result Genes Genotype pH was controlled at pH 6.0 using 6N KOH and 2N HSO. 60 During the growth phase of the culture the oxygen transfer pGV2524, 21 plasmid P:Ec ilvC coSc'P'-', rate (OTR) was 10 mM/h and during the production phase of Prer:Ll ilvD coSc. the culture the OTR was 0.2 mM/h. o: SE, Table 18 shows the isobutanol titer and yield (as % theo 2f bia, G418R s retical) as calculated for the production phase of the culture. pGV2546 21 plasmid P:Ec ilvC coSc'P'', 65 Both isobutanol titer and yield are increased in strain PTEF1:Ll ilvD coSc, GEVO3523 carrying the alcohol dehydrogenase with decreased K and increased k.egg Table 18 also shows the US 9,012,189 B2 83 84 isobutyrate titer, reported as maximum titer reached, and the 3' end of the DNA fragment with the 5' end of the yield as carbon yield in 96. Both isobutyrate titer and yield are Ps promoter region from pCV 1954, using primers decreased in strain GEVO3523 carrying the alcoholdehydro oGV2834 and oGV2835. Another PCR reaction amplified a genase with decreased K and increased k. DNA fragment (D) comprising the downstream flanking region of ALD6 and a region of overlap at the 5' end of the TABLE 18 DNA fragment with the 3' end of the hph hygromycin resis tance ORF from pGV2074, using primers oGV2836 and Isobutanol and Isobutyrate Titers and Yields. oGV2837. Another PCR reaction amplified a DNA fragment Isobutanol (B) comprising the Ps. cc promoter region from Isobutanol yield Isobutyrate Isobutyrate yield pGV1954 with a region of overlap at the 5' end of the DNA Strain titer g/L (96 theor. titer g/L % C-yield fragment with the 3' end of the upstream flanking region of GEVO3519 3.9 0.4 SOS 2.1 O.82 0.04 4.O.O.O ALD6 (fragment A) and a region at the 3' end of the DNA GEVO3523 S.O. O.3 59.52.1 O40 OO1 2.O. O.O fragment with the 5' end of the hph hygromycin resistance ORF from pGV2074, using primers oGV2631 and oGV2632. 15 Another PCR reaction amplified a DNA fragment (C) com Example 4 prising the hph hygromycin resistance ORF from pGV2074 with a region of overlap at the 5' end of the DNA fragment Decreased Isobutyrate and Acetate Production in with the 3' end of the Ps, a promoter region from Fermentations with Deletion of ALD6 Gene in S. pGV 1954 (fragment B) and a region of overlap at the 3' end of cerevisiae the DNA fragment with the 5' end of the downstream flanking region of ALD6 (fragment D), using primers oGV2633 and The following example illustrates that deletion of the oGV2634. DNA fragments A and B were combined by PCR ALD6 gene leads to a decrease in isobutyrate and acetate using primers oGV2834 and oGV2632 to generate DNA frag production in fermentations. ment AB and DNA fragments C and D were combined by Construction of ALD6 Deletion Strains: PCR was used to PCR using primers oGV2633 and oGV2837 to generate DNA generate a DNA fragment that contained a deletion allele of fragment CD. DNA fragments AB and CD were combined by ALD6 for deletion of ALD6 from S. cerevisiae. One PCR PCR using primers oGV2834 and oGV2837 to generate the reaction amplified a DNA fragment (A) comprising the final DNA fragment ABCD that contained the deletion allele upstream flanking region of ALD6 and a region of overlap at of ALD6. TABL E 19 Primer Sequences Disclosed in Example 4. oGV No. Sequence

oGW968 ACTCGCCGATAGTGGAAACCGACG (SEQ ID NO: 62)

oGW1965 CAAACTGTGATGGACGACACC (SEQ ID NO: 63)

oGW2631. CAATACGTTATGCCGTAATGAAG (SEQ ID NO: 64)

oGW2632 GCTTTTTACCCATTATTGATATAGTGTTTAAGCGAATG (SEO ID NO: 65)

oGW2633 CACTATATCAATAATGGGTAAAAAGCCTGAACT CAC (SEQ ID NO: 66)

oGW2634 TTATTCCTTTGCCCTCGGACG (SEO ID NO: 67)

oGV268 O TGCACTGCTGTCTTCACTTC (SEQ ID NO: 68)

oGW2796 TGTCAGCGCTTCAGACTC (SEQ ID NO: 69)

oGW2797 AAGTATTTTTAAGGATTCGCTC (SEO ID NO: 7O)

oGW2798 CTTCATTACGGCATAACGTATTGAAGTATTTTTAAGGATTCGCTC (SEO ID NO : 71.)

oGW28 OO CGTCCGAGGGCAAAGGAATAAGATAGTTATCATTATGTAAGTGCG (SEO ID NO: 72)

oGW28O1 GGGAGTTTAGCAATCAGC (SEO ID NO: 73)

oGW28O2. TGGTTGACCCGCAAACTTC (SEO ID NO: 74)

US 9,012,189 B2 87 88 TABLE 19 - continued Primer Sequences Disclosed in Example 4. oGV No. Sequence dGW2833 TAAAGCGCTGGGTGGACAACCG (SEQ ID NO: 1.OO) oGW2.834 GCACCGAGACGTCATTGTTG (SEQ ID NO: 101) oGW2835 CTTCATTACGGCATAACGTATTGTAAACACGCCAGGCTTGACC (SEQ ID NO: 102) oGW2836 CGTCCGAGGGCAAAGGAATAATCCATTCGGTGGTGTTAAGC (SEQ ID NO: 103) oGW2837 ATGGCGAAATGGCAGTACTC (SEQ ID NO: 104) dGW2838 ACCAACGACCCAAGAATC (SEQ ID NO: 105) oGW2839 CTTTGCGACAGTGACAAC (SEQ ID NO: 106) oGW2.84 O CCTCACGTAAGGGCATGATAG (SEO ID NO : 107) oGW2841 GCATTGCAGCGGTATTGTCAGG (SEQ ID NO: 108) oGW2842 CAGCAGCCACATAGTATACC (SEQ ID NO: 109) oGV2843. CTTCATTACGGCATAACGTATTGAGCCGTCGTTTGACATGTTG (SEQ ID NO: O) dGW2844 CGTCCGAGGGCAAAGGAATAACGCTCCATTTGGAGGGATCG (SEQ ID NO: 1) oGW2845 GAATGCGCTTGCTGCTAGGG (SEQ ID NO: oGW2846. CAGCTCTTGCTGCAGGTAACAC (SEQ ID NO: 3) oGW2847 GGCACAATCTTGGAGCCGTTAG (SEQ ID NO: 4) oGW2848 ACCAAGCCATCAAGGTTGTC (SEQ ID NO: 5) oGW2849 TGGGTGATGGTTTGGCGAATGC (SEQ ID NO: 6) dGW2896 GAAATGATGACATGTGGAAATATAACAG (SEQ ID NO: 7)

50 Strains to demonstrate decreased isobutyrate and acetate KARI, DHAD, KIVD and ADH (Ec ilvC coSc'P'', production by deletion of ALD6 were constructed by trans Ll ilvD coSc, Ll kiv D2 coEc, and L1 adhA, respectively). formation of GEVO3198 with the ABCD DNA fragment that Transformants were selected for resistance to 0.2 g/L G418 contained the deletion allele of ALD6. Transformants were and 0.1 g/L hygromycin and purified by re-streaking onto selected for resistance to 0.1 g/L hygromycin and transfor- 55 media containing 0.1 g/L hygromycin and 0.2 g/L G418, mant colonies were screened by colony PCR for the correct generating strains GEVO3714, GEVO3715 and GEVO3716. integration of the ABCD DNA fragment using primer pairs An ALD6 control strain containing an isobutanol production oGV284.0/oGV2680, oGV968/oGV2841, and oGV2838/ pathway, GEVO3466, was generated by transformation of oGV2839. Strains GEVO3711, GEVO3712 and GEVO3713 GEVO3198 with plasmid pGV2247. Transformants were were identified by this colony PCR as having ALD6 deleted 60 selected for resistance to 0.2 g/L G418 and purified by re by correct integration of the ABCD DNA fragment. streaking onto media containing 0.2 g/L G418. Strains containing an isobutanol production pathway to Construction of ald2A, ald3A, ald4A, ald5A and hfd1A demonstrate decreased isobutyrate and acetate production by Deletion Strains: PCR was used to generate separate DNA deletion of ALD6 were constructed by transformation of 6s fragments that contained individual deletionalleles of ALD2, GEVO3711, GEVO3712 and GEVO3713 with a 2 origin of ALD3, ALD4, ALD5 and HFD1 for deletion of ALD2, replication plasmid, pGV2247, carrying genes expressing ALD3, ALD4, ALD5 and HFD1 individually from S. cerevi US 9,012,189 B2 89 90 siae in separate strains. Additionally, PCR was used to gen resistance to 0.1 g/L hygromycin and transformant colonies erate a DNA fragment that contained a deletion allele cover were screened by colony PCR for the correct integration of ing both ALD2 and ALD3, which are adjacent genes in the S. the four-component DNA fragment using the primer pairs cerevisiae genome, for deletion of ALD2 and ALD3 together listed in Table 21. Strain GEVO3567 was identified by this (ald2A ald3A) from S. cerevisiae in an individual strain. Four colony PCR as having ALD2 correctly deleted; strain component fragments containing the upstream flanking GEVO3568 was identified by this colony PCR as having region, the Ps. promoter region from pCV1954, the ALD3 correctly deleted; strain GEVO3569 was identified by hph hygromycin resistance ORF from pCV2074 and the this colony PCR as having ALD2 and ALD3 together cor downstream flanking region for each individual gene were rectly deleted; strain GEVO3579 was identified by this generated by PCR as for the generation of the ABCD frag 10 colony PCR as having ALD4 correctly deleted; strains ment for deletion of ALD6 except using the primer pairs listed GEVO3705, GEVO3706 and GEVO3707 were identified by in Table 20. The four-component fragment for deletion of this colony PCR as having ALD5 correctly deleted; and ALD2 and ALD3 together contained the upstream flanking strains GEVO3720, GEVO3721 and GEVO3722 were iden region from ALD2 and the downstream flanking region from tified by this colony PCR as having HFD1 correctly deleted. ALD3 and was similarly constructed by PCR using the primer 15 Strains containing an isobutanol production pathway and pairs listed in Table 20. The Ps, 2 promoter region from with deletion of ALD2, ALD3 and ALD5 individually or with pGV 1954 was always amplified with primer pair oGV263.1/ deletion of ALD2 and ALD3 together were constructed by oGV2632 and the hph hygromycin resistance ORF from transformation of strains GEVO3567, GEVO3568, pGV2074 was always amplified with primer pair oGV2633/ GEVO3569, GEVO3705, GEVO3706 and GEVO3707 with plasmid pGV2247. Transformants were selected for resis OGV2634. tance to 0.2 g/L G418 and 0.1 g/L hygromycin and purified by re-streaking onto media containing 0.1 g/L hygromycin and TABLE 20 0.2 g/L G418, generating strains GEVO3586, GEVO3587. Primers Used to Amplify Upstream and Downstream Regions for Gene GEVO3588, GEVO3590, GEVO3591, GEVO3592, Deletions. 25 GEVO3593, GEVO3594, GEVO3595, GEVO3708, Gene Primer Pairs for GEVO3709 and GEVO3710. Strains GEVO3579, Deletion Primer Pairs for Upstream Region Downstream Region GEVO3720, GEVO3721 and GEVO3722 were generated from GEVO3466 and therefore contained plasmid pGV2247. ald2A oGV2796/oCV2797, oGV2796/ OGV28OO.oGV28O1 OGV2798 ald3A OGV2806, CGV28O8 OGV281 OjoGV2811 30 TABLE 21 ald2A ald3A OGV2796, CGV2798 OGV281 OjoGV2811 ald4A OGV2816 OGV2818 OGV282OoGV2821 Primers Used to Screen Colonies for Verification of Gene Deletions. ald5A OGV2826, OCGV2827 OGV2828. oGV2829 ald6A OGV2834foGV2835 OGV2836, CGV2837 Gene hf1A OGV2842 foGV2843 OGV2844foGV2845 Deletion Primer Pairs 35 ald2A oGV2802/oGV2632, oGV968/oGV2803, oGV2804/oGV2805 ald3A oGV2812/oGV2632, oGV968/oGV2813, oGV2814/oGV2815 Strains with deletion of ALD2, ALD3, ALD4, ALD5 and ald2A oGV2802/oGV2632, oGV968/oGV2813, oGV2804/oGV2805, HFD1 individually and with deletion of ALD2 and ALD3 ald3A OGV2814foGV2815 together were constructed by transformation of GEVO3198 aldAA oGV2822foGV2632, oGV968/oGV2896, oGV2824/oGV2825 or GEVO3466 with the individual four-component DNA 40 OGV2831 fragment that contained the individual deletion allele of ald6A oGV284.0/oGV2680, oGV968/oGV2841, oGV2838/oGV2839 ALD2, ALD3, ALD4, ALD5 or HFD1 or with the four-com hf1A oGV2848/oGV2680, oGV968/oGV2849, oGV2846/oGV2847 ponent DNA fragment that contained the deletion allele of ALD2 and ALD3 together. Transformants were selected for TABLE 22 Genotype of Strains Disclosed in Example 4.

GEVO No. GEVO31.98 trp1

ter:Sc ILV3AN:Pris-Ec ilvC coSc' ilvD coSc:Priors:Ec ilvC coSc''':Peop:Ll adhA:Pe:Sc TRP1 {evolved for C2 supplement-independence, glucose tolerance and faster growth GEVO3466 MATaura3 leu2 his3 trp1

::LEU2:bla:P:Sc ILV3AN:P-Ec ilvC coSce'' ::Pter:Ll ilvD coSc:Priors:Ec ilvC coSc''':Pevo2:Ll adhA:Pe:Sc TRP1 {evolved for C2 supplement-independence, glucose tolerance and faster grow h: transformed with pCV2247 GEVO3567 MATaura leu2 his3 ::LEU2:bla:Per:Sc ILV3AN:Pros-Ec ilvC coSc' ::Pter:Ll ilvD coSc:Priors:Ec ilvC coSc''':Pevo2:Ll adhA:Pe:Sc TRP1 court:hph evolved for C2 supplement-independence, glucose tolerance and aster growth

US 9,012,189 B2 93 94 TABLE 23 TABLE 24

Plasmids Disclosed in Example 4. Shake Flask Fermentation Results Demonstrating Decreased Isobutyrate and Acetate Production by Deletion of ALD6 Plasmid Isobutanol Isobutanol Isobutyrate Acetate Name Genotype Titer Yield Produced Produced Strain g/L) % theoretical g/L g/L) pGV2247 P:Ll ilvD coSc Priors:Ec ilvC coSc'''. Peok:Ll kiv D2 coEc PENo.2: GEVO3466 2.60.1 44 + 2 O48 OO6 O.59 O.04 Ll adhA 21-ori, plJC-ori, bla, G418.R. (ALD6) 10 GEVO3714, 3.20.2 42 - 2 O.14 OO6 O.O8 OO1 GEVO3715 and Shake Flask Fermentations: Fermentations were per GEVO3716 formed to compare the performance of GEVO3466 to strains (ald6A) containing the ald2A, ald3A, ald2A ald3A, ald4A, ald5A, 15 hfd1A and ald6A deletion mutations. Yeast strains were The 72 h Shake flask fermentation results for GEVO3466 inoculated from cell patches or from purified single colonies and the ald2A, ald3A, ald2A, ald3A, ald4A, ald5A and hfd1A from YPD agar plates containing 0.2 g/L G418 into 3 mL of strains are summarized in Table 25 and Table 26. Strains with YPD containing 0.2 g/L G418 and 1% V/v ethanol medium in deletions in ALD3, ALD2 and ALD3 together or ALD4 had 14 mL round-bottom Snap-cap tubes. The cultures were incu no decrease in isobutyrate production compared with the bated overnight up to 24h shaking at an angle at 250 rpm at wild-type ALDH strain GEVO3466. Strains with deletions in 30° C. Separately for each strain, these overnight cultures ALD2, ALD5 or HFD1 had no appreciable decrease in isobu were used to inoculate 50 mL of YPD containing 0.2 g/L tyrate production compared with the wild-type ALDH strain 25 GEVO3466. Strains with deletions of both ALD2 and ALD3 G418 and 1% V/v ethanol medium in a 250 mL baffled flask together produced 19% less acetate than the wild-type ALDH with a sleeve closure to an ODoo of 0.1. These flask cultures strain GEVO3466 but strains with individual deletions of were incubated overnight up to 24h shaking at 250 rpm at 30° ALD2, ALD3, ALD4, ALD5 or HFD1 had no appreciable C. The cells from these flask cultures were harvested sepa decrease in acetate production compared with the wild-type rately for each strain by centrifugation at 3000xg for 5 min 30 ALD strain GEVO3466. utes and each cell pellet resuspended separately in 5 mL of YPD containing 80 g/L glucose, 1% V/v stock solution of 3 TABLE 25 g/L ergosterol and 132 g/L Tween 80 dissolved in ethanol, Shake Flask Fermentation Results Demonstrating No Decrease in 200 mM MES buffer, pH 6.5, and 0.2 g/L G418 medium. 35 Isobutyrate and Acetate production by Deletion of ALD2, ALD3, ALD4 Each cell suspension was used to inoculate 50 mL of YPD or ALD2 and ALD3 Together. containing 80 g/L glucose, 1% V/v Stock solution of 3 g/L Isobutanol Isobutanol Isobutyrate Acetate ergosterol and 132 g/L Tween 80 dissolved in ethanol, 200 Titer Yield Produced Produced mMMES buffer, pH 6.5, and 0.2 g/L G418 medium in a 250 Strain g/L) % theoretical g/L g/L) 40 GEVO3466 5.1 - 0.1 42 - 2 124 O15 O.95 O.O7 mL non-baffled flask with a vented Screw-cap to an ODoo of (wild-type) approximately 5. These fermentations were incubated shak GEVO3590, 5.2 0.2 45 - 2 1.21 OO6 O.85 O.O7 ing at 250 rpm at 30° C. Periodically, samples from each GEVO3591 and shake flask fermentation were removed to measure ODoo GEVO3592 and to prepare for gas chromatography (GC1) analysis, for 45 (ald2A) GEVO3593, 5.5 - 0.6 45 - 6 1.34 - 0.16 O.91 O.O7 isobutanol and other metabolites, and for high performance GEVO3594 liquid chromatography (LC1) analysis for organic acids and and GEVO3595 glucose. Samples of 2 mL were removed into a microcentri (ald3A) fuge tube and centrifuged in a microcentrifuge for 10 min at 50 GEVO3596, 6.8 O.1 51 - 1 141 O.O9 O.77 O.O8 maximum rpm. One mL of the Supernatant was analysis of GEVO3597 and extracellular metabolites by GC1 and LC1 as described. GEVO3598 (ald2A ald3A) Deletion of ALD6 decreased isobutyrate and acetate pro GEVO3579 5.6 0.7 46 6 1.34 - 0.13 O.89. O.15 duction in shake flask fermentations: The 52 h shake flask (aldAA) fermentation results for GEVO3466 and the ald6A strains 55 GEVO3714, GEVO3715 and GEVO3716 are summarized in Table 24. The ald6A strains GEVO3714, GEVO3715 and GEVO3716 produced 71% less isobutyrate than the ALD6 TABLE 26 strain GEVO3466. The ald6A strains GEVO3714, 60 Shake Flask Fermentation Results Demonstrating No Decrease in GEVO3715 and GEVO3716 also produced 86% less acetate Isobutyrate and Acetate Production by Deletion of ALDS or HFDl. than the ALD6 strain GEVO3466. Isobutanol yield in the Isobutanol Isobutanol Isobutyrate Acetate ald6A strains GEVO3714, GEVO3715 and GEVO3716 was Titer Yield 9% Produced Produced not appreciably different than the ALD6 strain GEVO3466. Strain g/L) theoretical g/L g/L) Isobutanol titer in the ald6A strains GEVO3714, GEVO3715 65 GEVO3466 4.0 + 0.4 447 O.47 O.O4 O.75 O.OS and GEVO3716 was 23% higher than the ALD6 strain (wild-type) GEVO3466. US 9,012,189 B2 95 96 TABLE 26-continued TABLE 27

Shake Flask Fermentation Results Demonstrating No Decrease in Benchtop Fermenter Fermentation Results Demonstrating Decreased Isobutyrate and Acetate Production and Increased Isobutanol Isobutyrate and Acetate Production by Deletion of ALD5 or HFD1. Yield by Deletion of ALD6. 5 Isobutanol Isobutanol Isobutyrate Acetate Isobutanol Isobutanol Isobutyrate Acetate Titer Yield 9% Produced Produced Titer Yield 9% Produced Produced Strain g/L) theoretical g/L) g/L) Strain g/L) theoretical g/L) g/L) GEVO3466 8.20.1 321 2.10.1 2.3 - 0.3 GEVO3708, 3.80.8 46 - 15 O41 - 0.04 O.64 O.08 10 (ALD6) GEVO3709 and GEVO3714 and 11.1 + 0.1 40 O 1.3 - 0.1 O.90.1 GEVO3710 GEVO3715 (ald6A) (ald5A) GEVO3720, 4.4 + 1.0 54 - 14 O4O O.O7 O.S60.18 GEVO3721 and 15 Example 5 GEVO3722 (hfd1A) Determination of ALD6 Activity in S. cerevisiae The following example illustrates that the isobutyralde Fermentations in benchtop fermenters: Fermentations in hyde oxidation activity is significantly decreased in an ald6A benchtop fermenters were performed to compare the perfor strain. mance of GEVO3466 (ALD6) to GEVO3714 and GEVO3715 (ald6A). Glucose consumption, isobutanol pro TABLE 28 duction, isobutyrate production, and ODoo were measured during the fermentation. For these fermentations, purified 25 Genotype of Strains Disclosed in Example 5. strains from streak plates were transferred to 500 mL baffled GEVO fi Genotype Source flasks containing 80 mL of YPD medium containing 1% V/v GEVO3527 MATC his3A-1 leu2A ATCC# 201389 (BY4742) ethanol, 100 uM CuSO4.5H20 and 0.2 g/L G418 and incu lys2A ura3A GEVO3940 MATC. his3A-1 OpenBiosystems cath YSC1054 bated for 32 hat 30° C. in an orbital shaker at 250 rpm. The 30 leu2Alys2A ura3A (Yeast MATalpha collection) flask cultures were transferred to individual 2 L top drive ald6A:kan motor fermenter vessels with a working volume of 0.9 L of YPD medium containing 80 g/L glucose, 1% v/v ethanol, 100 Yeast Strains GEVO3940 from which the ALD6 uM CuSO4.5H0 and 0.2 g/L G418 per vessel for a starting 35 (YPL061W) gene was deleted and its parent GEVO3527 ODoo of 0.5. Fermenters were operated at 30° C. and pH 6.0 were each cultured in triplicate by inoculating 3 mL of YPD controlled with 6N KOH and 2N HSO in a 2-phase aerobic medium in a 14 mL culture tube in triplicate for each strain. condition based on oxygen transfer rate (OTR). Initially, fer Cultures were started from patches on YPD agar plate for menters were operated at a growth phase OTR of 10 mM/h by GEVO3527 and onYPD agar plates containing 0.2 g/L G418 fixed agitation of 700 rpm and an air overlay of 5 sL/h. 40 plates for GEVO3940. The cultures were incubated overnight Cultures were grown for 24 h to approximately 9-10 ODoo at 30° C. and 250 rpm. The next day, the ODoo of the over night cultures were measured and the Volume of each culture then immediately switched to a production aeration OTR=2.0 to inoculate a 50 mL culture to an ODoo of 0.1 was calcu mM/h by reducing agitation from 700 rpm to 450 rpm for the lated. The calculated volume of each culture was used to period of 24 h to 86.5 h. Periodically, samples from each 45 inoculate 50 mL of YPD in a 250 mL baffled flask and the fermenter were removed to measure ODoo and to prepare for cultures were incubated at 30°C. and 250 rpm. The cells were gas chromatography (GC1) analysis, for isobutanol and other harvested during mid-log phase at ODs of 1.6-2.1 after 7 h of metabolites, and for high performance liquid chromatogra growth. The cultures were transferred to pre-weighed 50 mL phy (LC1) analysis for organic acids and glucose. Samples of Falcon tubes and cells were collected by centrifugation for 5 50 minutes at 3000xg. After removal of the medium, cells were 2 mL were removed into a microcentrifuge tube and centri washed with 10 mL MilliOH-0. After removal of the water, fuged in a microcentrifuge for 10 min at maximum rpm. One the cells were centrifuged again at 3000xg for 5 minutes and mL of the supernatant was submitted for GC1 and LC1 analy the remaining water was carefully removed using a 1 mL sis as described. pipette tip. The cell pellets were weighed and then stored at 55 -80° C. until they were lysed and assayed for isobutyralde Deletion of ALD6 decreased isobutyrate and acetate pro hyde oxidation activity as described. duction and increased isobutanol yield in benchtop fermenter As shown in Table 29, the specific activity of S. cerevisiae fermentations: The 86.5 h benchtop fermenter fermentation ALD6 in GEVO3527 lysates for the oxidation of 10 mM results are summarized in Table 27. The ald6A strains isobutyraldehyde was 13.9 mu/mg. The same strain with an GEVO3714 and GEVO3715 produced 38% less isobutyrate 60 ALD6 deletion had a specific activity of 0.6 mU/mg which is than the ALD6 Strain GEVO3466. The ald6A Strains 22-fold less. The specific activity of S. cerevisiae ALD6 in GEVO3714 and GEVO3715 also produced 61% less acetate GEVO3527 lysates for the oxidation of 1.0 mM isobutyral than the ALD6 strain GEVO3466. Isobutanol yield in the dehyde was 17.6 mU/mg. The same strain with an ALD6 ald6A strains GEVO3714 and GEVO3715 was 25% higher deletion had a specific activity of 2.1 mU/mg which is 8-fold than the ALD6 Strain GEVO3466. Isobutanol titer in the 65 less. The specific activity of S. cerevisiae ALD6 in ald6A Strains GEVO3714 and GEVO3715 was also 35% GEVO3527 lysates for the oxidation of 0.1 mM isobutyral higher than the ALD6 strain GEVO3466. dehyde was 6.7 mU/mg. The same strain with an ALD6 US 9,012,189 B2 97 98 deletion had a specific activity of 1.3 mU/mg which is 5-fold improved kinetic properties leads to a further decrease in less. These data demonstrate that the endogenous ALD6 isobutyrate production and to a further increase in isobutanol enzyme is responsible for the isobutyrate byproduct of the production. isobutanol pathway in S. cerevisiae Isobutyrate is a byproduct of isobutyraldehyde metabolism in yeast and can comprise a significant fraction of the carbon TABLE 29 yield. The following yeast strains were constructed: Specific Isobutyraldehyde Oxidation Activities of Strains GEVO3527 GEVO3466 was constructed by transforming strain and GEVO3940 Using Various Isobutyraldehyde Concentrations. GEVO3 198 with a 2L plasmid, pGV2247, carrying genes Specific Activities were Measured in Lysates From 3 Parallel Cultures encoding the following enzymes: KARI, DHAD, KIVD and of GEVO3527 and GEVO3940. Shown are the Averages and Standard 10 wild-type ADH (Ec ilvC coSc''P'', Ll ilvD coSc, Deviations of the Activities Measured in the Biological Replicate Cultures. Ll kiv D2 coEc, and L1 adhA, respectively). GEVO3 198 Activity ml J/mg total protein expresses a single copy of alcohol dehydrogenase (L. lactis measured with isobutyraldehyde ADH, L1 adhA) from its chromosomal DNA. The second O.1 mM 1.0 mM 10 mM 15 strain, of which biological replicates are termed GEVO3714 Strain Isobutyraldehyde Isobutyraldehyde Isobutyraldehyde and GEVO3715, was constructed by transforming two inde GEVO3527 6.7 0.4 17.6 1.2 13.9 O.4 pendent strains, GEVO3711 and GEVO3712, with a 2 plas GEVO3940 1.3 O2 2.1 - 0.2 O.6 0.1 mid pGV2247 carrying genes encoding the following enzymes: KARI, DHAD, KIVD and wild-type ADH (Ec ilvC coSc'''', Ll ilvD coSc, Ll kivD2 coEc, and L1 adhA, respectively). GEVO3711 and 3712 express a Example 6 single alcohol dehydrogenase (L. lactis ADH, Ll adhA) and have the ALD6 gene deleted from the chromosomal DNA. A Further Decreased Isobutyrate Production with third strain, of which biological replicates are termed Deletion of ALD6 Gene and Overexpression of an 25 GEVO3855 and GEVO3856, was constructed by transform Improved Alcohol Dehydrogenase in S. cerevisiae ing a strain, GEVO3711, with 2L plasmid pGV2602 carrying genes encoding the following enzymes: KARI, DHAD, The following example illustrates that the combination of KIVD and a mutant ADH (Ec ilvC coSc''''', Ll il an ALD6 deletion and overexpression of an ADH with vD coSc, Ll kivD2 coEc, and L1 adhA''', respectively). TABLE 30 Genotype of Strains Disclosed in Example 6. GEVO No. Genotype GEVO31.98 MATaura eu2 his3 trp1

LEU2:bla:Pref:ILV3AN:Priors:Ec ilvC coSc' PTEF:Ll {evolved for C2 Supplement-independence, glucose tolerance and faster growth

pdc5A::LEU2:bla:Perl:ILV3AN:Priors:Ec ilvC coSc' pdc6A::PTE:Ll ilvD:Priors:Ec ilvC coSc''':Pevo2:Ll adhA:Pe:Sc TRP1 Transformed wi hpGV2247 evolved for C2 supplement-independence, glucose olerance and faster growth GEVO3711, MATaura eu2 GEVO3712 bd2A::

dc6A::PTE:Ll d6A::PCC12: hph evolved for C2 supplement-independence, glucose tolerance and faster growth GEVO3714, MATaura eu2 GEVO3715 bd2A::

dc6A::PTEF:Ll ilvD:Priors:Ec ilvC coSc''':PNo.2:Ll adhA:Pe:Sc TRP1 d6A::P:hph Transformed with pCV2247 evolved for C2 supplement independence, glucose tolerance an faster growth GEVO3855, MATaura eu2 his3 trp1 gpd1A:Tk UR-43 GEVO3856 bd2A::

dc6A::Pter:Ll ilvD:Priors:Ec ilvC coSc''':Pevo2:Ll adhA:Pe:Sc TRP1 d6A::P:hph Transformed with pCV2602 evolved for C2 supplement independence, glucose tolerance an faster growth US 9,012, 189 B2 99 100 TABLE 31 matography (GC1) and liquid chromatography (LC1) analy sis. For GC1 and LC1, 2 mL sample was removed into an Plasmids Disclosed in Example 6. Eppendorf tube and centrifuged in a microcentrifuge for 10 Plasmid Name Genotype 5 minat maximum. One mL of the Supernatant was analyzed by pGV2247 Per:Ll ilvD coSc, Priors:Ec ilvC coSc''', GC1 (isobutanol, other metabolites) and one mL analyzed by PPok1:Ll kiv D2 coEc, PENo.2: Ll adh A. high performance liquid chromatography (LC1) for organic 2-ori, plJC-ori, bla, G418.R. pGV2602 P:Ll ilvD coSc, P:Ec ilvC coSc'''. acids and glucose. Peck:Ll kiv D2 coEc, Pevo2: Ll adh Af'. The 72.5 h data from two separate fermentation sets A and 2-ori, plJC-ori, bla, G418.R. " B are summarized in Tables 32 and 33. Fermentation set A compared GEVO3466 (WT ADH) to GEVO3714 and 3715 Two different sets of fermentations were performed. Fer- (WT ADH, ald6A) while the fermentation set B compared mentation set A was performed to compare the performance GEVO3714 (WT ADH, ald6A) to GEVO3855 and 3856 of GEVO3466 (L1 adhA) to GEVO3714-GEVO3715 (L1- 15 (L1 adhA', ald6A) adhA, ald6A). Fermentation set B was performed to compare The da the performances of GEVO3714 (L1 adhA, ald6A) to that i e b tat S.ref r1ng toGOs fi tatt t A (Table. e 32)) shS oW GEVO3855-GEVO3856 (L1 adhA', ald6A) respectively. RS A. s ere G. yield intnes N it.s Glucose consumption, isobutanol production, isobutyrate L a W1t t e ALD6 gene deletion was 1.4- an 1.3-fo production, and ODoo were measured during the fermenta- higher, respectively, compared to the stran carrying Ll adhA tion. For these fermentations, single isolate cell colonies without the ALD6 gene deletion. The strain carrying L1 adhA grown on YPD agar plates were transferred to 500 mL baffled without ALD6 gene deletion (GEVO3466) had an isobutyrate flasks containing 80 mL of YPD medium containing 1% V/v yield (gram isobutyrate produced/gram glucose consumed) Ethanol, 100 uM CuSO,5H.0, and 0.2 g/L G418 and incu- of 0.040 g/g while the strains carrying L1 adhA with the bated for 32hat 30°C. in an orbital shaker at 250 rpm. The 25 ALD6 gene deletion (GEVO3714, GEVO3715) had a lower flask cultures were transferred to individual 2 L top drive isobutVrate vield of 0.017 g/g. The strain carrving the L. lacti motor fermenter vessels with a working volume of 0.9 L of y y 99. rry1ng CCS YPD medium containing 80 g/L glucose, 1% v/v Ethanol, adhA without the ALD6 gene deletion produced 2.3 g/L 100 uM CuSO4.5H20, and 0.2 g/L G418 per vessel for a acetate while the strain carrying the L. lactis adhA with the starting ODoo of 0.5. Fermenters were operated at 30°C. and ALD6 gene deletion produced 0.6 g/L acetate. TABLE 32

Data from Fermentation Set A. Isobutyrate Isobutanol Isobutyrate Acetate Isobutanol produced yield yield produced Strain ODoo produced g/L) g/L) % theoretical gig g/L GEVO3466 9.7 O.1 7.406 17 O.O 48.1 - 26 O.O4(OOOO4 2.3 0.1 (WTADH) GEVO3714, 10.0 + 0.7 10.4 + 0.1 0.8 + 0.1 55.3 + 0.6 0.017 + 0.003 0.6+ 0.0.1 GEVO3715 (WTADH, ALD6A) pH 6.0 controlled with 6N KOH in a 2-phase aerobic condi- 45 The data referring to fermentation set B (Table 33) show tion based on oxygen transfer rate (OTR). Initially, ferment that isobutanol titer and theoretical yield in the strain carrying ers were operated at a growth phase OTR of 10 mM/h by fixed L. lactis adhA''' with the ALD6 gene deletion was 1.2 and agitation of 700 rpm and an air overlay of 5 sL/h in both 1.1-fold higher, respectively, compared to the strain carrying experiments. Cultures were grown for 24 h to approximately L. lactis adhA with the ALD6 gene deletion. The strains 9-10 ODoo then immediately switched to production aera- 50 carrying L. lactis adhA''' with the ALD6 gene deletion tion conditions for 48.5 h. In the first experiment, an OTR of (GEVO3855, GEVO3856) had the lowest isobutyrate yield 2.5-3.0 mM/h was sustained by reducing agitation from 700 (gram isobutyrate produced/gram glucose consumed), 0.005 rpm to 425 rpm while in the second experiment, an OTR of g/g, and produced 0.0 g/L acetate compared to the strain 2.0-2.5 mM/h was sustained by reducing agitation from 700 carrying L. lactis adhA with ALD6 gene deletion rpm to 400 rpm. Periodically, samples from each fermenter 55 (GEVO3714) which had a higher isobutyrate yield of 0.014 were removed to measure ODoo and to prepare for gas chro g/g and a similar acetate titer of 0.0 g/L (Table 33). TABLE 33

Data from Fermentation Set B. Isobutanol Isobutyrate Isobutanol Isobutyrate produced produced yield yield Acetate Strain ODsoo g/L) g/L % theoretical gig produced gL GEVO3714, 9.7 + 0.2 10.3 + 0.1 O.8 O.O 46.5 - 16 OO14 O.OOO O.O.O.O (WTADH, ALD6A) US 9,012,189 B2 101 102 TABLE 33-continued

Data from Fermentation Set B. Isobutanol Isobutyrate Isobutanol Isobutyrate produced produced yield yield Acetate Strain OD6oo g/L) g/L) % theoretical gig produced gL GEVO3855, 9.9 + 0.3 12.0 + 0.0 0.3 + 0.0 51.5 + 0.8 OOOSOOOO GEVO3856 (LI adhARE, ALD6A)

Example 7 pressed conductivity, Suppressor type: ASRS 4 mm in Auto Suppression recycle mode, Suppressor current:300 mA). The Identification of DH2MB as a By-Product of 15 column temperature was 35°C. This method used the follow Isobutanol Fermentation ingelution profile: Hold at 0.25 mM for 3 min: linear gradient to 5 mM at 25 min: linear gradient to 38.5 mM at 25.1 min, Duringfermentation of isobutanol-producing yeast strains, hold at 38.5 mM for 4.9 min; linear gradient to 0.25 mM at it was found that an unknown peak, co-eluting with 2,3- 30.1 min; hold for 7 min to equilibrate. Flow was set at 2 dihydroxy isovalerate (DHIV) on method LC1, and quanti mL/min. Injection size is 5 uL and run time is 37.1 min. tated on this basis, was acting as a sink for a substantial GC-MS. Varian 3800CPGC system equipped with a single portion of the carbon being utilized. quad 320MS; DB-5 ms column: 1079 injection port at 250° Initially, it was believed this peak was solely 2,3-dihy C.; constant flow 1.0 mL/min at 100 split ration; oven profile: droxyisovalerate (DHIV), but subsequent studies indicated 25 initial temperature, 40°C., hold for 5 min, ramp of 20°C/min that KARI product inhibition would have occurred at these up to 235°C. and hold for 2 minutes; combiPAL autosampler levels of DHIV, making such concentrations impossible. delivering 0.5 LL of sample; collected masses of 35 to 100. Additional experiments showed that this recovered peak was BSTFA Derivation: (1) Evaporate sample to dryness under not reactive with DHAD in enzyme assays, thus eliminating nitrogen in a GC vial: (2) add 0.5 mL of Acetonitrile and 0.5 the possibility that significant amounts of DHIV were 30 mL of BSTFA reagent; (3) Incubate at 50° C. for 30 minutes: present. (4) Inject onto GC-MS. High Performance Liquid Chromatography LC1: Analysis LC-MS: For the LC-MS analysis of the LC1 peak fraction of organic acid metabolites was performed on an Agilent the sample was injected into an Agilent 1100 Series high 1200 High Performance Liquid Chromatography system performance liquid chromatographic (HPLC) system that equipped with two Rezex RFQ-Fast Fruit H+ (8%)150x4.6 35 was equipped with a multiple wavelength detector and an mm columns (Phenomenex) in series. Organic acid metabo LC/MSD Trap mass spectrometer (ion trap). The separations lites were detected using an Agilent-1 100 UV detector (210 were monitored by mass spectrometry to provide identifica nm) and refractive index (RI) detector. The column tempera tion for the component in the sample. The mass spectrometer ture was 60° C. This method was isocratic with 0.0128 N was operated in the atmospheric pressure chemical ionization HSO (25% 0.0512 NHSO in Milli-Q water) as mobile 40 (APCI) mode for sample injection. The analyses were con phase. Flow was set to 1.1 mL/min. Injection volume was 20 ducted using the positive and negative APCI modes. Detec uL and run time was 16 min. tion of the “unknown was only observed in the negative High Performance Liquid Chromatography LC3: For ionization mode. The analysis was conducted using MSn to samples containing a maximum of 10 mMaldehydes, ketones obtain fragmentation data on the sample analyte. Separations and ketoacid intermediates (combined), DNPH reagent was 45 were achieved using a 4.6x150 mm Agilent Zorbax SB C-18 added to each sample in a 1:1 ratio. 100 uL DNPH reagent (12 column with 5 um particles. The sample was run using an mM 2,4-Dinitrophenyl Hydrazine 20 mM Citric Acid pH 3.0 isocratic method which used an eluent of 90% HPLC water 80% Acetonitrile 20% MilliOHO) was added to 100 uL of and 10% methanol. A 10 uL injection was used for the analy each sample. Samples were incubated for 30 min at 70°C. in sis of the sample solution. The sample was also analyzed a thermo-cycler (Eppendorf, Mastercycler). Analysis of 50 bypassing the chromatographic column. acetoin, diacetyl, ketoisovalerate and isobutyraldehyde was DHIV and its isomer, DH2MB, elute at the same retention performed on an Agilent-1200 High Performance Liquid time on LC1. The peak related to these compounds is sepa Chromatography system equipped with an Eclipse XDB rated from other compounds in the fermentation samples. The C-18 150x4 mm; 5 um particle size reverse phase column peak was collected from the HPLC and used for further analy (Agilent) and a C-18 reverse phase guard column (Phenom 55 S1S. enex). All analytes were detected using an Agilent-1 100 UV The signal ratio of the RI detector signal to UV detector detector (360 nm). The column temperature was 50° C. This signal seen in LC1 for DHIV (and DH2MB) is characteristic method was isocratic with 60% acetonitrile 2.5% phosphoric of common organic acids (e.g. lactate, acetate, etc.); conju acid (0.4%), 37.5% water as mobile phase. Flow was set to 2 gated acids (e.g., pyruvate) have very different RI/UV signal mL/min. Injection size was 10 uL and run time is 10 min. 60 ratios. The recovered “peak DHIV had the characteristics of High Performance Liquid Chromatography LC4: Analysis a non-conjugated acid: of oxo acids was performed on a Agilent-1 100 High Perfor Ratio (RI/UV): Recovered DHIV/DH2MB peak (130): mance Liquid Chromatography system equipped with an Ion DHIV Std (150); Pyruvate (14). Pac AS11-HC Analytical, IonPac AG11-HC guard column The lack of a carbonyl moiety in the “mystery peak” was (3-4 mm for IonPac ATC column, Dionex) or equivalent and 65 confirmed by the complete lack of reaction between the an IonPac ATC-1 Anion Trap column or equivalent. Oxo acids recovered peak fraction from LC1 and DNPH: no adduct were detected using a conductivity detector (ED50-sup peaks were evident in the LC3 chromatographic system. US 9,012,189 B2 103 104 The recovered peak fraction from LC1 was then analyzed Ll ilvD, Ll kiv)2, and L1 adhA) was used to produce by method LC4, which runs under alkaline conditions, and is approximately 10 g/L DH2MB in a batch fermentation using capable of separating DHIV and acetolactate. That result is a 2 L top drive motor DasGip vessels filled with 1 L culture shown in FIG. 9, together with an overlay of standard mix medium (10 g/L yeast extract, 20 g/L peptone, 80 g/L glu tures. This clearly shows the separation between DH2MB (as 5 cose, 1% v/v Ethanol, 100 uM CuSO4.5H.0, 0.2 g/L G418) at it was subsequently identified), and DHIV. Some pyruvate 30°C., pH6.0, and an OTR of approximately 10 mmol/h. was also brought along in the collection of the DH2MB peak. NMR Analysis: The sample peak recovered from method The cell-free fermentation broth was acidified to pH 2 LC1 was neutralized and lyophilized and sent for NMR using concentrated HSO4. Acidified broth was concentrated analysis. The 2-D connectivity analysis by 1H-COSY NMR 10 to 350 mL under reduced pressure (0-100 mbar) using Bichi (FIG. 10) and the proton NMR spectrum (FIG. 11) yielded Rotovapor R-215. The flask containing broth was heated in good results. the water bath to 20-30° C. during evaporation. A 70 mL 2-D analysis of “mystery peak' eluting with DHIV (FIG. Volume of MeOH was added to concentrated broth and mix 10): One methyl group, shifted downfield, is not split by any ture was transferred to a 500 mL liquid-liquid extractor adjacent protons, where the methyl group at 0.95 ppm is split 15 (Sigma-Aldrich cat. # Z562432), which was set up according into a doublet by one proton adjacent to a hydroxyl. That to manufacturer's specifications for continuous extraction proton, in turn, is split into a quartet by the adjacent methyl with ethyl acetate (EtOAc). Continuous extraction was car group. Complex patterns between 3.1 and 3.7 ppm indicate ried out for 3 days replacing the EtOAc extract daily with the different anomers of glucose carried along during the peak fresh EtOAc. collection of “DHIV. Following extraction, the first two batches of DH2MB The assignments of the NMR peaks are shown in the spec extract in EtOAc were combined and dried with anhydrous trum below (FIG. 11), clearly indicating that the identity of MgSO, followed by filtration. Dry extract was concentrated the “mystery peak' is 2,3-dihydroxy-2-butyrate (DH2MB). under vacuum to 500 mL and was treated with 3 g of activated The "H NMR and COSY spectra support the presence of charcoal (Fluka cati 05105) for 30 min by stirring at room 2,3-dihydroxy-2-methylbutanoic acid, a structural isomer of 25 temperature. The decolorized solution was filtered and con dihydroxyisovaleric acid. Other signals in these spectra Sup centrated to approximately 50 mL under vacuum (0-100 mbar port the presence of anomeric proteins and, therefore, a Sugar using Büchi Rotovapor R-215). The Solution was incubated component. Furthermore, complex grouping of signals at 4° C. for two days. Obtained crystals were filtered and between 3.1-3.8 ppm are often observed with oligosaccha washed with ice-cold diethylether and acetone. Crystals were rides. The 13C NMR spectrum is very weak and appears to be 30 dried using lyophilizer under reduced pressure (0.05 mbar) an attached proton test (APT) experiment based on the signal for one day. at 45 ppm that falls below the base line. Isolated DH2MB was analyzed by 1H (FIG. 14) and 13C LC-MS was also carried out on the LC1 peak fraction. The (FIG. 15) NMR. 1H NMR (TSP) 1.1 (d. 6.5 Hz, 3H), 1.3 (s, LC-MS was sufficient to demonstrate that the compound had 3H), 3.9 (q, 6.5 Hz, 3H). A 13C spectrum indicated five a mass of 134 (both DHIV and DH2MB) (FIG. 12). 35 different carbon atoms present in the sample. Resonance at This analysis conclusively identified the unknown 181 ppm indicated carboxylic acid carbon present in the by-product as 2,3-dihydroxy-2-methylbutanoic acid (CAS if sample. In conclusion, based on NMR spectra one could 14868-24-7). This compound exists in 4 different stereoiso meric forms. 2,3-dihydroxy-2-methylbutanoic acid exists as estimate a 99% purity of isolated DH2MB. a set of cis and trans diastereomers, each of which exists as a 40 set of enantiomers. The four compounds are shown in FIG. Example 9 13. As described herein, DH2MB is derived from (2S)-2-hy Impact of DH2MB Production on Isobutanol Yield in droxy-2-methyl-3-oxobutyrate (acetolactate). The product of Fermentation this reaction would be either (2S,3R)-2,3-Dihydroxy-2-me 45 thylbutanoic acid, (2S,3S)-2,3-Dihydroxy-2-methylbutanoic acid or a mixture of the two diastereomers depending on the The purpose of this example is to demonstrate that Stereoselectivity of the endogenous enzyme(s) catalyzing this DH2MB accumulates to substantial levels in yeast strains conversion. comprising ALS and TMA29 activity. 50 Strains and plasmids disclosed in this example are shown Example 8 in Tables 34 and 35, respectively.

Production and Purification of DH2MB TABLE 34

The purpose of this example is to illustrate how DH2MB 55 Genotype of S. Cerevisiae Strain GEVO3160. was produced and purified. An engineered S. cerevisiae CEN.PK2 strain com Strain Genotype prising ALS activity (GEVO3160, S. cerevisiae CEN.PK2: GEVO3160 MATaura3 leu2 his3 trp1 gpd1A::Pt 12: MATa ura3 leu2 his3 trp1 gpd1A::P: Hph gpd2A::TK R4s of Przai : KI URA3: TK R 43 pdc1A: 60 Bs alsS coSc: Tcycl: Peck: Ll kiv D: PENo2: Pote: BS als.S. coSc: Toyo: Peck: Ll kiv D: Pevo2 s, Sp HIS5 pdc5A::LEU2: bla: P-ILV3AN: HIS5 pdc5A::LEU2: bla: P: ILV3 AN: P: ilvC coSc Priors: Ec ilvC coSc' pdc6A::Pter: Q11OV pacóA::P: Ll ilvD P: Ec ilvC coSc Ll ilvD Prs: Ec ilvC coSc'P'-': Pro2: Ll adhA: Pre: Sc TRP1){evolved for C2 P2D1-A1: P: Ll adhA: P: Sc TRP1 evolved for Supplement-independence, glucose tolerance and faster C2 Supplement-independence, glucose tolerance and faster 65 growth pCV2247 growth expressing plasmid pGV2247 (2-micron, G418 resistant plasmid for the expression of Ec ilvC P2D1-A1, US 9,012,189 B2 105 106 TABLE 35 This experiment was performed to determine whether ALS is required for the production of DH2MB. The strains used in Genotype of Plasmid pCV2247. this experiment were GEVO1187 (S. cerevisiae CEN.PK2: Plasmid Genotype MATa ura3-52 leu2-3 112 his3A1 trp1-289 ADE2) and GEVO2280 (S. cerevisiae CEN.PK2: MATaura3 leu2 his3 Ps (Eri: Ll ilvD coSc, Pse tota: Ec ilvC coS trp 1 ADE2 pdc1A::P:Bs alsS2:TRP1). Prior to fer Ps (P11: G418R, PS Pok1: Ll kiv D coEc, mentations, both strains were transformed with the 2 micron Ps No.2: Ll adhA, 21, AP, PMB1 plasmid pGV2082 (Prs: Ec ilvC coSc'''. Pre:Ll il VD coSc, P:Ll kiv D coEc, and P:Dm ADH, 2LL S. cerevisiae strain GEVO3160 was transformed with 10 ori, bla, G418R) as described. pGV2247 as described. A fermentation was performed to To measure ALS activity, yeast cell extracts from characterize the transformed Strain. A single isolate cell GEVO1187 and GEVO2280 were prepared. Cells were colony grown on a YPD agar plate containing 0.2 g/L G418 grown to an ODoo of about 1, induced with 1 mM CuSO for were transferred 5 mL of YPD medium containing 80 g/L 2 hours and then harvested. To prepare cells for assays, 50 ml glucose, 1% V/v ethanol, 100 uM CuSO4.5H20, and 0.2 g/L 15 of cells was collected by centrifugation at 2700xg. After G418 and incubated for 24 h at 30° C., 250 rpm. Next, this removal of the media, cells were resuspended insteriledHO, culture was transferred to 500 mL baffled flasks containing 80 centrifuged at 2700xg and the remaining media was carefully mL of the same medium and incubated for 24hat 30°C. in an removed with a 1 ml pipette tip. The cell pellets were weighed orbital shaker at 250 rpm. The flask culture was transferred to (empty tubes were preweighed) and then frozen at -80° C. a 2 L top drive motor fermenter vessel with a working volume until use. Cell lysates were made using the following SOP as of 0.9L of the same medium for a starting ODoo of 0.5. The described below. Cells were thawed on ice and resuspended in fermenter was operated at 30° C. and pH 6.0 controlled with lysis buffer (250 mMKPO pH 7.5, 10 mMMgCl, and 1 mM 6N KOH in a 2-phase aerobic condition based on oxygen DTT) such that the result was a 20% cell suspension by mass. transfer rate (OTR). Initially, the fermenter was operated at a A volume of 1000 ul of glass beads (0.5 mm diameter) were growth phase OTR of 10 mM/h by fixed agitation of 700 rpm 25 added to a 1.5 ml Eppendorf tube and 875 ul of cell suspen and an air overlay of 5 sI/h in both experiments. The cultures sion was added. Yeast cells were lysed using a Retsch MM301 was grown for about 20h to an ODoo of approximately 8, and mixer mill (Retsch Inc. Newtown, Pa.) by mixing 6x1 min then immediately switched to production aeration. An OTR each at full speed with 1 min icing steps between. The tubes of 1 mM/h was sustained by reducing agitation from 700 rpm were centrifuged for 10 min at 23,500xg at 4° C. and the to 350 rpm. After 93 h post inoculation, one replicate vessel 30 Supernatant was removed. Extracts were held on ice until from each strain was further reduced to an OTR=0.3 mM/h by assayed. The lysate protein concentration was determined decreasing the agitation from 350 rpm to 180 rpm. Periodi using the BioRad Bradford Protein Assay Reagent Kit (Cath cally, samples from each fermenter were removed to measure 500-0006, BioRad Laboratories, Hercules, Calif.) and using ODoo and to prepare for gas chromatography (GC1) and BSA for the standard curve as described. Briefly, all ALS liquid chromatography (LC1) analysis. For GC1 and LC1, 2 35 assays were performed in triplicate for each lysate, both with mL sample was removed into an Eppendorf tube and centri and without substrate. To assay each lysate, 100 uL of lysate fuged in a microcentrifuge for 10 min at maximum. One mL diluted 1:2 with lysis buffer was mixed with 900 uL of buffer of the supernatant was analyzed by GC1 (isobutanol, other (50 mM potassium phosphate buffer pH 6.0, 1 mM MgSO 1 metabolites) and one mL analyzed by high performance liq mM thiamin-pyrophosphate, 110 mM pyruvate), and incu uid chromatography (LC1) for organic acids and glucose. 40 bated for 15 minutes at 30°C. Buffers were prepared at room FIG. 16 depicts the product and by-product profiles of S. temperature. A no substrate control (buffer without pyruvate) cerevisiae GEVO3160 transformed with pGV2247. These and a no lysate control (lysis buffer instead of lysate) were profiles are representative for isobutanol producing Pdc-mi also included. After incubation 175 uL from each reaction nus, Gpd-minus yeast strains. Pdc-minus/Gpd-minus yeast was mixed with 25uL35% HSO and incubated at 37° C. for production strains are described in commonly owned and 45 30 min. Samples were submitted to analytics for analysis by co-pending publications, US 2009/0226991 and US 2011/ LC1. Using this method, it was determined that the wild-type 0020889, both of which are herein incorporated by reference strain GEVO1187 had no detectable ALS activity while the in their entireties for all purposes. FIG. 16 shows that isobu ALS-expressing strain GEVO2280 had 0.65 units/mg lysate tanol (13.9 g/L) and the unknown compound quantified as ALS activity. “DHIV” and now identified as DH2MB (8.4 g/L) are the 50 The performance of the two strains (with or without the primary products produced during microaerobic production heterologous ALS integrated expression construct) was com OTR. Assuming that the quantitation using the response fac pared using the following shake flask fermentation condi tor of DHIV leads to an accurate quantitation of DH2MB, tions. Strains were patched onto YPD plates containing 0.2 approximately 12-13% of the carbon consumed is diverted mg/mL G418. After overnight growth, cells were removed into production of DH2MB. If the acetolactate that is con 55 from the plate with a sterile toothpick and resuspended in 4 verted into DH2MB would instead be converted into isobu mL ofYPD with 0.2 g/L G418. The ODoo was determined for tanol then the isobutanol yield over the entire time of the each culture. Cells were added to 50 mL YP with 50 g/L fermentation shown in FIG.16 would be significantly higher. dextrose and 0.2 mg/mL G418 such that a final ODoo of 0.1 was obtained. To induce the CUP1 promoter driving ALS Example 10 60 expression, 1 mM copper sulfate was added at the 24 hour time point. Unused media was stored at 4°C. to act as medium ALS Expression is Necessary for DH2MB blank for GC and LC, and to act as the t-0 sample for the Production fermentation. At t=24, 48 and 72 hours samples were pre pared for analysis by GC1 and at 72 hours samples were The purpose of this example is to demonstrate that exog 65 additionally analyzed by LC1. At 24 and 48 hours a 1:10 enously expressed ALS activity is required for DH2MB accu dilution of the Supernatant of each culture was analyzed by mulation in S. cerevisiae. YSI. If needed 50% glucose containing 0.2 g/L G418 was US 9,012,189 B2 107 108 added to a final concentration of 100 g/L glucose. Fermenta TABLE 36-continued tions were performed at 30° C. shaking at 250 RPM. The DH2MB titer reached at 72 hours of a shake flask fermentation was determined using LC1 method for both the Genotype of Strains Disclosed in Example 12. WT strain (BUD1187) without ALS and the strain expressing the Pe:Bs alsS2 at PDC1 (BUD2280). Each strain was Strain Genotype transformed with the 4-component plasmid pGV2082. The fermentation was performed as described. Without exog pdc6A::URA3:bla: Pr: Ll kiv D2: Pr: Dm ADH enous ALS expression, the strain produced no DH2MB, {evolved for C2 Supplement-independence, glucose whereas the strain with ALS expression produced up to 1.4 10 tolerance and faster growth g/L DH2MB plus DHIV. Example 11 TABLE 37 Only ALS Expression is Necessary for DH2MB 15 Production Plasmids Disclosed in Example 12. Plasmid Genotype The purpose of this example is to demonstrate that ALS pGV2196 CEN, ARS, hph, bla, plJC-ori. activity alone is responsible for DH2MB accumulation in S. pGV2377 PTEF1: Ll ilvD coSc, PsPok1: Ll kiv) coEc, cerevisiae. Psy: Ll adhA, 21 ori, plJC ori, bla, G418R This experiment was performed to determine whether ALS pGV2466 Pter: Ll ilvD coSc, Pscripts: Ec ilvC coSc". alone or in combination with a KARI, DHAD, KIVD, ADH PsPok1: Ll kiv) coEc, PSENo2: Ll adhA, 21. ori, plJC ori, bla, G418R expressing plasmid is responsible for the production of pGV2398 Pter: Ll ilvD coSc, Pscripts: Ec ilvC coSc'''. DH2MB. The strain used in this experiment was GEVO2618 Pspok: Ll kiv) coEc, Psevo2:Ll adhA, (MATaura3 leu2 his3 trp 1 pdc1A::Pe: Bs alsS1 coSc: 25 2 ori, plJC ori, bla, G418R TRP1). The plasmids tested in this experiment were pGV2400 Pter: Ll ilvD coSc, Pscripts: Ec ilvC coSc''''', PsPok1: Ll kiv) coEc, PSENo2: Ll adhA, 21 ori, pGV2227 which contains the remaining four path pUC ori, bla, G418R way genes (-P:Ll ilvD coSc: Prs: EC ilvC pGV2406 Ps EC ilvC coSc', CEN, ARS, hph, bla, coSc':Ps te: G418: Pe: Ll kivD2 coEc:PDC1-3' pUC ori. region: P: Ll adh A2u bla, puC-ori), and pGV2020, the 30 empty vector control (Ps. 1, Ps 71, G418R, APr; 2L). Shake flask cultures of GEVO2618 transformed with S. cerevisiae strain GEVO2843 was transformed with 2. pGV2020 and GEVO2618 transformed with pGV2227 were plasmids pGV2377, pGV2466, pGV2398, and pGV2400 as started in YPD (15% glucose) containing 200 mM MES described to determine if expression of wild-type or engi pH6.5, and 0.4 g/L G418 at an OD600-0.1, and were run at 35 neered KARIs led to a greater accumulation of DH2MB. 30° C. and 75 rpm in a shaking incubator. Samples were taken Precultures of GEVO2843 transformed with the 2L plas at 24 hand 48 hand the samples were analyzed for metabolite mids (pGV2377, 2466, 2398, 2400) were started in YPD levels by HPLC (LC1) and GC (GC1). After 48 hours, all containing 1% ethanol and 0.2 g/L G418 and incubated over glucose was consumed from the media by both Strains. The night at 30° C. and 250 rpm. These precultures were used to strain containing the empty vector (GEVO2618+pGV2020) 40 produced 4.6 g/L of DHIV+DH2MB representing 3.8% inoculate 50 mL of the same medium in a baffled flask and yield. The strain containing the vector expressing additional incubated at 30° C. and 250 rpm until reaching an ODoo of four pathway genes (GEVO2618+pGV2227), produced a -5. They were pelleted in 50 mL Falcon tubes at 2700 rcffor similar titer of 5.6 g/L DHIV+DH2MB representing 3.1% 5 minutes at 25°C. Next, the cells from each 50 mL culture yield. 45 were resuspended in 50 mLYPD containing 8% glucose, 1% (v/v) ethanol, ergosterol, Tween-80, 0.2 g/L G418, and 200 Example 12 mM MES, pH6.5. The cultures were added to 250 mL unbaffled flasks and placed in an incubator at 30° C. and 75 Effect of Increased KARI Activity on DH2MB rpm. Samples were taken after 72 h to determine ODoo and to production 50 analyze the fermentation broth for extracellular metabolites via GC1 and LC1 analysis. The purpose of this example is to demonstrate that increased KARI activity results in decreased in DH2MB pro Table 38 shows that the strain transformed with pGV2377 duction in yeast comprising ALS activity. (Not overexpressing any KARI gene from plasmid) produced the highest carbon yield of 15% for combined DH2MB+ Strains and plasmids disclosed in this example are shown 55 DHIV, while the strains with pGV 2466 (containing in Tables 36 and 37, respectively. Ec ilvC coSc'), pGV2398 (containing Ec ilvC TABLE 36 coSc'''), and pCV2400 (containing Ec ilvC coSc'''''') had similar combined DH2MB+DHIV car Genotype of Strains Disclosed in Example 12. 60 bon yields of 8-10%. Likewise, the strain transformed with pGV2377 produced isobutanol at the lowest carbon yield of Strain Genotype 6%. The remaining strains comprising KARI genes on a GEVO2843 S. cerevisiae, MATaura3 leu2 his3 trp1 pdc1A::PCP1:BS alss1 coSc:Tcyc1: plasmid produced isobutanol at higher carbon yields. The Peak: Ll kiv D2: PNo: Sp HIS5 observation that decreased DH2MB production correlates pdc5A::LEU2: bla: P: ILV3AN: Ps: 65 with increased isobutanol production is consistent with the Ec ilvC coSc99. finding that DH2MB is produced from acetolactate via a reaction that does not involve KARI. US 9,012,189 B2 109 110 TABLE 38

Isobutanol and Combined DH2MB + DHIV Carbon Yields

Isobutanol carbon DH2MB + DHIV Strain Plasmid KARI yield (9%) carbon yield 9% GEVO2843 p6V2377 n/a 6 15 GEVO2843 p6V2466 Ec ilvC coScis 18 8 GEVO2843 p6V2398 Ec ilvC coSc910 his 15 8 GEVO2843 p6V2400 Ec ilvC coSc?--is 18 10

A second experiment was performed in which Strains TABLE 39-continued expressed either no KARI from a plasmid, a low level of KARI, or a high level of KARI. In this experiment the KARI KARI Activity, Isobutanol and Combined activity of cell lysates was measured. 15 DH2MB + DHIV Carbon Yields. S. cerevisiae strain GEVO2843 was transformed as Isobutanol DH2MB - described with combinations of plasmids as described in KARI activity carbon DHIV carbon Table 37: the no KARI strain contained pGV2377+pGV2196 Strain Plasmid Imol/min.img yield 96 yield .9% and had no plasmid-borne KARI, the low KARI strain con tained pGV2377+pGV2406 and expressed KARI from a low GEVO2843 p6V2377+ 0.0303 11: 16* copy plasmid, and the high KARI Strain contained pGV2406 pGV2398+pGV2 196 and expressed KARI from a high copy GEVO2843 p6V2398 + O.151.005 19 11 plasmid. Fermentations and sampling were performed as pGV2196 described. GC1 and LC1 methods were performed as 25 This data comprises only one sample described. Cells for KARI assays were lysed as described except that lysis buffer was 250 mM. KPO, pH 7.5, 10 mM MgCl, and 1 mM DTT. The protein concentration of lysates was determined as described. To measure in vitro KARI activity, acetolactate substrate 30 was made by mixing 50 ul of ethyl-2 acetoxy-2-methyl-ac Example 13 etoacetate with990ul of water. Next 10ul of 2N NaOH was sequentially added, with Vortex mixing between additions for 15 sec, until 260 ul of NaOH was added. The acetolactate was Effect of Increased DHAD Activity agitated at room temperature for 20 min and held on ice. 35 NADPH was prepared in 0.01N NaOH to a concentration of The purpose of this example is to demonstrate that 50 mM. The concentration was determined by reading the OD increased DHAD activity results in decreased in DH2MB of a diluted sample at 340 nm in a spectrophotometer and production in yeast comprising ALS activity. using the molar extinction coefficient of 6.22 M'cm to calculate the precise concentration. Three buffers were pre Strains and plasmids disclosed in this example are shown pared and held on ice. Reaction buffer contained 250 mM 40 in Tables 40 and 41, respectively. KPO, pH 7.5, 10 mM MgCl, 1 mM DTT, 10 mMacetolac GEVO2843 was transformed with different pairs of plas tate, and 0.2 mM NADPH. No substrate buffer was missing mids. Strain A contains pGV2227 plus pCV2196. Strain B the acetolactate. No NADPH buffer was missing the NADPH. Reactions were performed in triplicate using 10 ul of cell contains pGV2284 plus pCV2196. Strain C contains extract with 90 ul of reaction buffer in a 96-well plate in a 45 pGV2284 plus pGV2336. Single transformants of BUD2843 SpectraMax 340PC multi-plate reader (Molecular Devices, with one of the three 2-plasmid combinations were single Sunnyvale, Calif.). The reaction was followed at 340 nm by colony purified on YPD plates containing hygromycin, and measuring a kinetic curve for 5 minutes, with OD readings the patched cells were used to inoculate 3 mLYPD containing every 10 seconds at 30° C. The Vmax for each extract was 1% ethanol (v/v), 0.2 g/L G418, and 0.1 g/L hygromycin. The determined after Subtracting the background reading of the no 50 cultures were incubated at 30°C., 250 rpm overnight prior to substrate control from the reading in complete buffer. their use to inoculate 3 mLYPD containing 1% ethanol (v/v), Table 39 shows data for KARI activity, as well as carbon 0.2 g/L G418, and 0.1 g/L hygromycin. These cultures were yield in % for isobutanol and combined DH2MB+DHIV. As incubated at 30° C., 250 rpm overnight. The following day, KARI activity increased the isobutanol carbon yield the cultures were used to inoculate 50 ml YPD containing 8% increased and the combined DH2MB+DHIV carbon yield 55 glucose, 200 mMMES pH6.5, Ergosterol, and TWEENTM 80 decreased. to an ODoo of approximately 0.1. These cultures were incu TABLE 39 bated at 30° C., 250 rpm overnight. The following day the cultures were diluted in 50 mL of the same medium to an KARI Activity, Isobutanol and Combined 60 ODoo of -0.1. The cultures were incubated at 30° C. DH2MB + DHIV Carbon Yields. 250rpm, and 1.5 mL samples were removed after 0, 24, 47. Isobutanol DH2MB - 70, and 92 hours of incubation. The samples were prepared KARI activity carbon DHIV carbon for GC and LC analysis as described. After 92 hours, the Strain Plasmid Imol/min.img yield 96 yield .9% remainder of all samples was centrifuged and the pellets were GEVO2843 p6V2377+ O.O11 OO2 5 19 65 weighed and stored at -80°C. DHAD assays were performed pGV2196 with lysates prepared. from the frozen pellets as described. LC1 and GC1 analysis was performed as described. US 9,012,189 B2 111 112 TABLE 40 Genotype of Strains Disclosed in Example 13. Strain Genotype GEVO2843 MATaura3 leu2 his3 trp1

{evolved for C2 Supplement-independence, glucose tolerance and faster growth

TABLE 41 F contains pGV2196 plus pCV2485. The strain transformed with pGV2196+pGV2589 (no plasmid-borne DHAD) pro Plasmids Disclosed in Example 13. 15 duced 1.25 g/L isobutanol and 5.67 g/L DH2MB+DHIV. The Plasmids Genotype strain with DHAD expressed from a high-copy plasmid (p.GV2 196+pGV2485) produced 2.74 g/L isobutanol and pGV2227 PS (Er1: Ll ilvD coSc, Ps. Trots: Ec ilvC coSc''. Ps. Pri: G418, Ps. Pok: 3.71 g/L DH2MB+DHIV, indicating that an increase in Ll kivD coEc, Ps, ENO2: Ll adhA, 21, AP, PMB1 DHAD expression led to a decrease in DH2MB+DHIV accu pGV2284 Ps ter. Ps. Idita: Ec ilvC coSC''', mulation. The strain with DHAD expressed from a low-copy Ps (P11: G418, Ps. Pok1: Ll kiv D coEc, Pse ENo2: plasmid (pGV2529+pGV2485) produced an intermediate Ll adhA, 21, AP, PMB1 pGV2196 Ps PokiPs (EF1, PS (Pri: hph, CEN, AP, plJCORI level of both metabolites, consistent with an intermediate pGV2336 Pse ENO, TsopDC6 Pse PGK, Pse TEF1: level of DHAD activity. Ll ilvD coSc Ps. TDH3, Pse (P11: hph, CEN, AP, puC ORI 25 TABLE 43 Table 42 shows the DHAD activity, isobutanol yield and Additional Plasmids Disclosed in Example 13. the combined DHIV+DH2MB yield. The strain transformed with pGV2284+pGV2 196 (no DHAD expressed from a plas Plasmid Genotype mid) produced the highest carbon yield of 19% for combined 21.96 Pse PGK1: Pse TEF1: Ps (PrihPh, CEN, AP, 30 pUC ORI DH2MB+DHIV and the lowest carbon yield of isobutanol at 2529 PS Pok, Pse terLl ilvD coSc4, Ps. (Prihph, 9%. The strain transformed with pGV2227+pGV2.196 (high CEN, AP, puC ORI est DHAD expression from a plasmid) had the lowest carbon 2589 Ps, Ec ilvC coSc Q110V. P. P1G418R, yield of 9% for combined DH2MB+DHIV and the highest Ps vo2Ll adhA, 21, AP, PMB1 carbon yield for isobutanol at 18%. The strain transformed with pGV2284+pGV2336 (low copy DHAD expression from TABLE 44 DHAD activities, Isobutanol Titer and Yield, and Combined DH2MB + DHIV Titers at 72 hrs Fermentation.

Plasmid-borne Isobutanol Isobutanol DH2MB + DHIV Strain Plasmid(s) DHAD Titer (g/L). Yield (%) (g/L) D pGV2196+ pGV2589 None 1.25 + 0.27 16.1 S-67 0.29 E pGV2529 + pCV2589 Low-copy 2.15 0.05 24.8 S.OO 0.2O F pGV2196+ pGV2485 High-copy 2.74 - 0.22 31.0 3.71 - 0.11 a plasmid) had an intermediate carbon yield of 16% for com- Example 14 bined DH2MB+DHIV and of 12% for isobutanol. 50 Deletion of TMA29 in S. cerevisiae by Targeted TABLE 42 Deletion DHAD Activities, Isobutanol and Combined DH2MB + DHIV The following example illustrates that deletion of the Carbon Yields at 92 hrs Fermentation. TMA29 gene from the S. cerevisiae genome eliminates the DHAD Isobutanol carbon DH2MB + DHIV 55 production of DH2MB when acetolactate synthase is overex Strain Plasmids activity yield .9% carbon yield 96 pressed. Several reductase enzyme candidates that may catalyze the A poV2227+ O.29 O.OS 18 9 pGV2196 production of DH2MB were identified in the S. cerevisiae B pCV2284 + O.05 - O.OO 9 19 genome, including the TMA29 gene product. The genes pGV2196 60 encoding these reductases were deleted in the S. cerevisiae C pGV2284 + O.O8 O.O1 12 16 strain GEVO2618, a strain known to produce g/L quantities pGV2336 of DH2MB, using integration of a URA3 marker. Fermenta tions were performed with these strains to determine if delet In a second experiment, GEVO2843 was transformed with ing any of the candidate genes, including TMA29, reduced or different pairs of plasmid (Table 43) and assessed in a shake 65 eliminated the production of DH2MB. flask fermentation as above. Strain D contains pGV2 196 plus Strains, plasmids, and primer sequences are listed in Tables pGV2589. Strain E contains pGV2529 plus pGV2589. Strain 45, 46, and 47, respectively. US 9,012,189 B2 113 114 TABLE 45 Genotype of Strains Disclosed in Example 14. GEVO No. Genotype GEVO1187 S. cerevisiae CEN.PK2 MATaura3-52 leu2-3 112 his3A1 trp1-289 ADE2 GEVO2618 S. cerevisiae, MATaura3 leu2 his3 trp 1 podc1A::Pe: Bs alss 1 coSc: TRP1. GEVO3638 S. cerevisiae, MATaura3 leu2 his3 trp 1 podc1A::Pe: Bs alss 1 coSc: TRP1 tna29A:T K UR43 sher, PFB41: Kl URA3: TK1 UR43. GEVO3639 S. cerevisiae, MATaura3 leu2 his3 trp 1 podc1A::Pe: Bs alss 1 coSc: TRP1 tma29A:T K UR-43 shor:PFBA 1: Kl URA3: TK1 UR-43) GEVO3640 S. cerevisiae, MATaura3 leu2 his3 trp 1 podc1A::Pe: Bs alss 1 coSc: TRP1 tma29A:T K UR-43 shor: PFBA 1: Kl URA3:Tkl UR-43)

TABLE 46 15 and 3' targeting sequences for reductase genes were designed with a 20 bp sequence homologous to a URA3 fragment. This Plasmids Disclosed in Example 14. was done so that SOE PCR could be used to create fragments Plasmid Name Genotype containing the URA3 marker and homologous regions flank ing the reductase gene of interest. PCR was performed on an pGV1299 Kl URA3, bla, plJC-ori. Eppendorf MastercyclerR (Cati 71086, Novagen, Madison pGV2129 KI URA3-5", bla. Wis.). The following PCR program was followed for primer sets used to generate SOE PCR fragments: 94° C. for 2 min then 30 cycles of (94° C.30 sec, 53° C. 30 sec, 72°C. 1.5 min) TABL E 47 then 72°C. for 10 min. The following primer pairs and tem 25 Oligonucleotide Sequences Disclosed plate were used for the first step of the SOE reactions. in Example la To generate the 5' URA3 fragment, oGV2232 and oGV2862 were used to amplify the 5' URA3 fragment using oGV # Sequence pGV2129 as template. The 1364 bp fragment was purified by 893 GGATGTGAAGTCGTTGACACAG gel electrophoresis. To generate the 3' URA3 fragment, (SEQ ID NO: 118) 30 oGV2231 and oGV893 were used to amplify the 3' URA3 22.31 TTGAAACGTTGGGTCCATAC fragment using pGV1299 as template. The 1115bp fragment (SEQ ID NO: 119) was purified by gel electrophoresis. To generate the 5' TMA29 fragment, oGV2867 and 2232 TTCACCGTGTGCTAGAGAAC (SEQ ID NO: 12O) 35 oGV2891 were used to amplify the 5'TMA29 fragment using S. cerevisiae S288c genomic DNA as template. The S. cer 2862 TTATACAGGAAACTTAATAGAACAAATC evisiae S288c strain was purchased from ATCC (SEQ ID NO: 121) (ATCC#204508). The 412 bp fragment was purified by gel 2867 TGAAACAGCATGGCGCATAG electrophoresis. To generate the 3' TMA29 fragment, (SEQ ID NO: 122) 40 oGV2869 and oGV2870 were used to amplify the 3'TMA29 2869 CTGTGTCAACGACTTCACATCCGAGGTAACGAGGAACAAGCC fragment using S. cerevisiae S288c genomic DNA as tem (SEQ ID NO: 123) plate. The 305bp fragment was purified by gel electrophore S1S. 287O TTTCGCCGGTATATTCCGTAG The following primer pairs and templates were used to (SEQ ID NO: 124) 45 generate the SOE PCR products. To generate the 5' TMA29 2891 GTTCTATTAAGTTTCCTGTATAACGGCATTGTTCACCAGAATGTC SOE PCR product, oGV2232 and oGV2867 were used. The (SEQ ID NO: 125) 5' URA3 fragment and the 5'TMA29 fragment were used as template. To generate the 3' TMA29 SOE PCR product, 29 O2 TCCCGACGGCTGCTAGAATG (SEQ ID NO: 126) oGV2231 and oGV2870 were used. The 3' URA3 fragment 50 and the 3' TMA29 fragment were used as template. 29 O4 CGCTCCCCATTAATTATACA Transformation of S. cerevisiae strain GEVO2618 with the (SEO ID NO: 127) bipartite integration SOE PCR products was performed as

291.3 GAAAGGCTCTTGGCAGTGAC described. Following transformation, the cells were collected (SEQ ID NO: 128) by centrifugation (18,000xg, 10 seconds, 25°C.) and resus 55 pended in 400 uL SCD-HLWU media. Integrative transfor 2914 GCCCTGGTGCAATTAGAATG mants were selected by plating the transformed cells on SCD (SEQ ID NO: 129) Ura agar medium. Once the transformants were single colony 291.5 TGCAGAGGGTGATGAGTAAG purified they were maintained on SCD-Ura plates. (SEQ ID NO: 13 O) Colony PCR was used to verify correct integration. To 60 screen for the correct 5'-end, the URA3: TMA295' junction 2916 GGCCAAAGGTAAGGAGAACG primers oGV2915 and oGV2902 were used to give an (SEQ ID NO: 131) expected band at 991 bp. To screen for the correct 3'-end, the URA3: TMA293'junction primers oGV2904 and oGV2916 Strain Construction: S. cerevisiae strains GEVO3638, were used to give an expected band at 933 bp. To screen GEVO3639, and GEVO3640 were constructed by transform 65 deletion of the TMA29 gene primers oGV2913 and oGV2914 ing GEVO2618 with bipartite integration SOE PCR products were used, expecting a lack of a 288 bp if the CDS was to replace TMA29 with a URA3 marker. Primers to amplify 5' deleted. US 9,012,189 B2 115 116 Fermentations: Fermentations were conducted with TABLE 50 tma29A strains GEVO3638, GEVO3639, and GEVO3640 and the parent TMA29 strain GEVO2618. Cultures were ORF Deletion Disclosed in Example 15. started in YPD shaking at 30° C. and 250 rpm. After four doublings, the ODoo was determined for each culture. Cells 5 ORF deletion Gene name Source were added to 50 mLYPD with 15% glucose such that a final YMR226C TMA29 Deletion library was obtained from Open ODoo of 0.05 was obtained. At t=24h, 2 mL of media was Biosystems, cat #YSC 1054 removed and 25 uL used at a 1:40 dilution to determine ODoo. The remaining culture was centrifuged in a microcen 10 trifuge at maximum speed for 10 min and 1 mL of Supernatant TABLE 51 was removed and submitted for LC1 and LC4 analysis. At t=48 h, 2 mL of media was removed and 25uL used at a 1:40 Plasmid Disclosed in Example 15. dilution to determine ODoo. 1 mL of Supernatant was Sub Plasmid Relevant Genes mitted for LC1 analysis. In addition, 14 mL was collected by 15 centrifugation at 2700xg. After removal of the media, cells pGV2435 Psect P1: BS alss1 coSc:Pset Pri:hph:Tscyc1. were resuspended in sterile dH20, centrifuged at 2700xg and CEN/ARS, bla, plJC-ori the remaining medium was carefully removed with a 1 mL pipette tip. The cell pellets were weighed (empty tubes were A commercial library of S. cerevisiae strains which has one preweighed) and then frozen at -80°C. until thawed for ALS gene/ORF deleted per strain was used to screen for a deletion assays as described. that might catalyze the production of DH2MB. The candidate The production of DH2MB is dependent on heterologous strain containing the deletion of the TMA29 (i.e., YMR226C) ALS expression, for instance the Bs alss1 coSc gene. The ORF was selected. Since exogenous ALS expression is ALS activity of cell lysates was measured as described to required for production of DH2MB, a CEN plasmid demonstrate that the TMA29 deletion had no impact on ALS 25 expression and/or activity. The ALS activity of extracts from (p.GV2435) containing the Bs alsS1 coSc gene driven by the the strains carrying the TMA29 deletion is not less than, and CUP1 promoter was transformed into the strains as described. is slightly more than the activity of extracts from the parent Transformations were recovered overnight at 30°C.,250 rpm strain. The results at 24h (48 h for ALS activity) are summa before plating onto YPD plates containing 0.2 g/L hygromy rized in Table 48 and clearly demonstrate the lack of DH2MB 30 cin. Transformants were then patched onto YPD plates con production in the strain with the TMA29 deletion. LC4 analy taining 0.2 g/L hygromycin and incubated at 30° C. sis confirmed that GEVO3527 did not produce DHIV. Fermentations were performed with these strains to deter mine if deleting TMA29 (YMR226C) reduced or eliminated TABLE 48 the production of DH2MB. Three independent transformants 35 of each strain were used to inoculate fermentation precultures Production of DH2MB in Strain with TMA29 Deletion. which were grown overnight to saturation in YPD containing Glucose DH2MB by ALS 0.2 g/L hygromycin at 30° C. and 250 rpm. The next day, the consumed by LC1 activity ODoo of the precultures was measured and the Volume of Strain ODoo LC1 g/L g/L Umg overnight culture needed to inoculate a 50 mL culture to an GEVO2618 9.2 - 0.9 6156 - 12.O 1510.1 O44- 0.06 40 ODoo of 0.1 was calculated for each culture. 50 mL of YPD GEVO3638, 12.5 SO 68.44 - 12.5 OOOO.O O57 0.04 GEVO3639, containing 150 g/L glucose, 200 mM MES, pH 6.5, and 0.2 GEVO3640 g/L hygromycin in a 250 mL non-baffled flask were inocu (tma29A) lated with the calculated amount of overnight culture. Cells 45 were incubated at 30° C. and 75 rpm in an orbital shaker. At 24 h, all cultures were fed an additional 75 g/L of glucose by Example 15 addition of 8.8 mL of a 50% glucose solution to each flaskand then returned to incubation at 30°C. and 75 rpm. At 72 h, 1.5 Deletion of TMA29 in S. cerevisiae by Deletion mL was sampled from each flask (750 uL divided between Library 50 two Eppendorf tubes). The ODoo was measured for each culture (1:40 dilution in HO). The cells were removed from The following example illustrates that deletion of the samples by centrifugation at >14000xg for 10 minutes in a TMA29 gene from the S. cerevisiae genome eliminates the microcentrifuge. The Supernatants from the samples were production of DH2MB when acetolactate synthase is overex collected and stored at 4°C. until analysis by LC1, and the pressed. 55 cell pellets were stored at -80°C. until thawed for ALSassays Strains, ORF deletions, and plasmids are listed in Tables as described. 49, 50, and 51. There was some variation in the growth between the two TABLE 49 strains, with ODoo values of 13.7 for GEVO3527 and 15.7 60 for the TMA29 deletion strain at 72 h (Table 52). The strains Genotype of Strains Disclosed in Example 15. consumed the same amount of glucose of around 223 g/L by GEVO fi Genotype/Source 72 h (Table 52). GEVO3527 produced 2.8 g/L of DH2MB by 72h. The YMR226C deletion strain (tma29A) did not produce GEVO3527 S. cerevisiae BY4742: MATa his3A1 leu2AO lys2AO ura3AO/ATCC #201389, purchased from ATCC 10801 detectable levels of DH2MB. The specific DH2MB titer for University Boulevard Manassas, VA 20110-2209 65 GEVO3527 was 0.2 g/L/OD; the YMR226C deletion strain (tma29A) did not produce detectable levels of DH2MB. LC4 analysis confirmed that GEVO3527 did not produce DHIV. US 9,012,189 B2 117 118 TABLE 52 TABLE 54 Cell Growth, Glucose Consumed, and DH2MB Production at 72 h. Plasmids Disclosed in Example 16. Glucose DH2MB consumed titer by Specific DH2MB 5 Plasmid Strain ODoo by LC1 g/L LC1 g/L titer gLOD Name Genotype GEVO3527 13.7 O.3 223.30.6 2.8. O.1 O2 O.O1 pGV1299 Kl URA3, bla, plJC-ori. TMA29A 15.7 SS 223.90.2 O.O.O.O O.O.O.O pGV2129 Kl URA3-5", bla, puC ori pGV2550 Pster:Ll ilvD coS, Pstors:Ec ilvC coSc''''':- 10 Pseok:Ll kiv D2 coEc:Psevo2:Ll adhA coSc'''. Example 16 2-ori, plJC-ori, bla, G418.R. Improved Isobutanol Rate, Yield, and Titer with Deletion of TMA29 Gene in S. cerevisiae Yeast strain construction: GEVO3663 was constructed by 15 transforming GEVO3351 with the bipartite integration SOE The following example illustrates that deletion of the PCR products described in Example 14 to replace TMA29 TMA29 gene from the S. cerevisiae genome leads to an with a URA3 marker as described, except after transforma increase in productivity, yield, and titer of the desired product, tion the cells were resuspended in 350 u, SCD-Ura media isobutanol. In addition, it leads to a decrease in DH2MB before being spread to SCD-Ura plates. productivity, yield and titer. DH2MB is a byproduct of acetolactate metabolism in S. cerevisiae strains GEVO3690, GEVO3691, and yeast. In isobutanol fermentations, DH2MB can comprise GEVO3692 were constructed by transforming GEVO3351 10% or greater of the carbon yield. Strains with wild-type with plasmid pGV2550. S. cerevisiae strains GEVO3694, TMA29 produce DH2MB in the presence of expressed aceto 25 GEVO3695, and GEVO3697 were constructed by transform lactate synthase (ALS), encoded by Bs alsS1 coSc (SEQID ing GEVO3663 with plasmid pGV2250 Briefly, competent NO:23). Strains deleted for TMA29 do not produce DH2MB in the presence of expressed BS alsS1 coSc. A yeast strain cells were prepared by removing cells from a fresh plate into deleted for all PDC and GPD genes that expresses ALS 100 uL 100 mM lithium acetate. The cell suspension was (Bs alsS1 coSc) from the chromosome was deleted for incubated at room temperature for 30 min. Plasmid DNA was TMA29 and transformed with a high copy four-component 30 transformed as described. After transformation, the cells were isobutanol pathway plasmid, pGV2550 with genes for resuspended in 400 uLYPD containing 1% ethanol and incu DHAD (L1 ilvD coSc), KARI (Ec ilvC coSc'P'-''') bated at 30° C. for 6hshaking at 250 rpm. The cells were then KIVD (Ll kivD2 coEc) and ADH (L1 adhA coSc''') spread onto YPD plates containing 0.2 g/L G418. Transfor Isobutanol titer, yield and productivity of this strain were compared to that of the parent strain that was not deleted for 35 mants were single colony purified onto YPD plates containing the TMA29 gene, in both a shake flask fermentation and in 0.2 g/L G418 plates. Once the transformants were single fermenters. Strains and plasmids are listed in Tables 53 and colony purified they were maintained on YPD plates contain 54, respectively. ing 0.2 g/L G418. TABLE 53 Genotype of Strains Disclosed in Example 16. GEVO No. Genotype GEVO1187 S. cerevisiae CEN.PK2 MATaura3 leu2 his3 trp 1 ADE2 GEVO3351 MATaura leu.2 bdc1 A::P alss1 pdc5 A:: :ILV3AN: Ps: ilvC coSce'' bdc6 A::P ilvD:PTDH3:Ec ilvC coSc''':Peo2:Ll adh A:Pre:Sc TRP1 {evolved for C2 supplement-indepen ence, glucose tolerance and faster growth GEVO3663 MATaura3 leu2 his3 trp1 gpd1::T& R 43 gpd1A::Te R43 bdc1 A::P alss1 pdc5 A:: bdc6 A::P :PENo2:Ll adh A:PBA1:Sc TRP1 Ema29A:: TK UR-43 sher, PFBA 1:KI URA3: Tkl UR-43) evo ved for C2 supplement indepen aster growth GEVO3690, MATaura GEVO3691, bdc1 A::P GEVO 3692 pdc5 A:: o110V bdc6 A::P PENo2:Ll adh A:PB41:Sc TRP1 Transform ith pCV2550 evolve for C2 supplement-independence, glucose olerance GEVO3694, MATaura GEVO3695, bdc1 A::P GEVO3696 pdc5 A:: ; bla: Pter: GEVO3697 bdc6 A::P TEA-Ll ilvD:PTIts:Ec ilvC coSc PENo2:Ll adh A:PB41:Sc TRP1 Ema29A:: TK1 UR43 sher, PFB.41:KI URA3: TK UR43 Trans ormed with pCV2550 {evolve for C2 supplement-indepen ence, glucose tolerance and faster growth US 9,012,189 B2 119 120 Fermentations: A shake flask fermentation was performed flask cultures were transferred to individual 2 L top drive comparing performance of GEVO3690-GEVO3692 motor fermenter vessels with a working volume of 1.2 L of 80 (TMA29) to GEVO3694-GEVO3695 and GEVO3697 mL of YP medium with 20 g/L glucose, 1% V/v Ethanol, 100 (tma29A). Cultures (3 mL) were started in YPD containing uM CuSO4.5H20, and 0.2 g/L G418 for a starting ODoo of 1% ethanol and 0.2 g/L G418 and incubated overnight at 30 5 0.2. Fermenters were operated at 30° C. and pH 6, controlled C. and 250 rpm. The ODoo of these cultures was measured with 6N KOH in a two-phase aerobic fermentation. Initially, after about 20 h. An appropriate amount of each culture was fermenters were operated at a growth phase oxygen transfer used to inoculate 50 mL of YPD containing 1% ethanol and 0.2 g/L G418 in a 250 mL baffled flask to an ODoo of rate (OTR) of 10 mM/h by fixed agitation of 850 rpm and an approximately 0.1. These precultures were incubated at 30° 10 air overlay of 5 sL/h. Cultures were grown for 31 hto approxi C. and 250 rpm overnight. When the cultures had reached an mately 6-7ODoo then immediately switched to a production OD of approximately 5 they were centrifuged at 2700 rcf aeration OTR of 0.5 mM/h by reducing agitation from 850 for 5 minat 25°C. in 50 mL Falcon tubes. The cells from each rpm to 300 rpm for the remainder of the fermentation of 111 50 mL culture were resuspended in 50 mL of fermentation h. Periodically, samples from each fermenter were removed media as described. The cultures were then transferred to 250 15 to measure ODoo and to prepare for gas chromatography mL unbaffled screw-cap flasks with small vents and incu- (GC1) analysis. For GC, 2 mL sample was removed into an bated at 30° C. and 75 rpm. At 24 and 48 h, samples from each Eppendorf tube and centrifuged in a microcentrifuge for 10 flask were removed to measure ODoo and to prepare for GC1 analysis. For GC1, 2 mL sample was removed into an Eppen minat maximum. One mL of the Supernatant was analyzed by dorf tube and centrifuged in a microcentrifuge for 10 min at 20 GC1 (isobutanol, other metabolites). At 72 h the same proce maximum. One mL of the Supernatant was analyzed by GC1. dures were used to collect cells for ODoo and GC analysis At 72 h the same procedures were used to collect cells for and in addition the samples were analyzed by high perfor ODoo and GC analysis and in addition the samples were mance liquid chromatography (LC1) for organic acids and analyzed by high performance liquid chromatography (LC1) glucose. for organic acids, including DH2MB and DHIV, and glucose. 25 The results at 111 h are summarized in Table 56. Isobutanol The results at 72 hare summarized in Table 55. Isobutanol titer, yield, and rate increased with deletion of the TMA29 titer, yield and rate increase with deletion of the TMA29 gene, gene. DH2MB production decreased to undetectable while DH2MB production decreases. levels. TABLE 55

Isobutanol Titer. Yield, and Rate Increase at 72 h.

Glucose Isobutanol consumed Isobutanol yield Isobutanol rate DH2MB Strain ODsoo g/L) produced g/L 96 theoretical g/L/h produced g/L GEVO3690, 8.3 + 29.8 + 1.3 5.5 + 0.4 45.1 4 O.08 3.1 GEVO3691, O.3 GEVO3692 GEVO3694, 8.3 + 33.4 + 1.0 7.6 0.2 55:12 O.11 O.O3 GEVO3695, 0.7 GEVO3697 (TMA29A)

In addition, the performance of GEVO3690-GEVO3691 (TMA29) to GEVO3694-GEVO3696 (tma29A) was also 45 compared in fermentations performed in fermenter vessels. Plated cultures were transferred to 500 mL baffled flasks containing 80 mL of YP medium with 20 g/L glucose, 1% V/v Ethanol, 100 uM CuSO4.5H20, and 0.2 g/L G418 and incu bated for 34.5 hat 30° C. in an orbital shaker at 250 rpm. The TABLE 56

Isobutanol Titer. Yield, and Rate Increase at 111 h.

Isobutanol DH2MB Isobutanol Glucose produced produced yield Strain ODoo consumed gL g/L) g/L) % theor. Isobutanol rate g/L/h) GEVO3690 7.2 29.7 11 8.60.1 2.9 62.43 O.09 GEVO3691, 0.7 (TMA29+) GEVO3694, 7.4 + 35.73.9 12.3 1.2 O 7S.O. O.O1 O.14 GEVO3695, 1.3 GEVO3696 (TMA29A) Glucose, isobutanol, and DH2MB titers are the final titers, i.e. at 111 h offermentation. Isobutanol yield and rate are calculated based on the production phase only, i.e. from 31 to 111 h offermentation, US 9,012,189 B2 121 122 Example 17 Cell pellets were thawed on ice and resuspended in lysis buffer (10 mM sodium phosphate pH7.0, 1 mM dithiothreitol, Determination of TMA29 Activity in S. cerevisiae 5% w/v glycerol) such that the result was a 20% cell suspen sion by mass. One mL of glass beads (0.5 mm diameter) was The following example illustrates that the (S)-2-acetolac 5 added to a 1.5 mL Eppendorf tube for each sample and 850LL tate reduction activity is significantly decreased in atma29A of cell Suspension were added. Yeast cells were lysed using a strain. Retsch MM301 mixer mill (Retsch Inc. Newtown, Pa.), mix ing 6x1 min each at full speed with 1 min incubation on ice TABLE 57 between. The tubes were centrifuged for 10 min at 21,500xg 10 at 4°C. and the supernatant was transferred to a fresh tube. Genotype of Strains Disclosed in Example 17. Extracts were held on ice until they were assayed using the TMA29 assay as described. GEVO fi Genotype Source The specific activity of S. cerevisiae TMA29 in GEVO3527 MATC. his3A-1 leu2A ATCC# 201389 (BY4742) GEVO3527 lysates, a wild-type MATa S. cerevisiae strain, lys2A ura3A GEVO3939 MATC. his3A-1 leu2A OpenBiosystems cathi 15 for the reduction of (S)-2-acetolactate was 6.9+0.2 mU/mg. lys2A ura3A YSC1054 (Yeast MATalpha The tima29A strain GEVO3939 had a specific activity of tma29:kan collection) 0.7+0.3 mU/mg. The wild-type GEVO3527 strain had about a 10-fold higher specific TMA29 activity than the deletion strain. Yeast Strains GEVO3939 from which the TMA29 (YMR226C) gene was deleted and its parent GEVO3527 Example 18 were each cultured in triplicate by inoculating 3 mL of YPD in a 14 mL culture tube in triplicate for each strain. Cultures Determination of TMA29 Activity in Kluyveromyces were started from patches on YPD agar plate for GEVO3527 lactis and on YPD plates containing 0.2 g/L G418 for GEVO3939 25 and GEVO3940. The cultures were incubated overnight at The following example illustrates that the (S)-2-acetolac 30° C. and 250 rpm. The next day, the ODoo of the overnight tate reduction activity is significantly decreased in atma29A cultures were measured and the volume of each culture to strain. inoculate a 50 mL culture to an ODoo of 0.1 was calculated. The calculated volume of each culture was used to inoculate TABLE 58 50 mL of YPD in a 250 mL baffled flaskand the cultures were 30 incubated at 30° C. and 250 rpm. Genotype of Strains Disclosed in Example 18. The cells were harvested during mid-log phase at ODs of 1.6-2.1 after 7 h of growth. The cultures were transferred to GEVO fi Genotype pre-weighed 50 mL Falcon tubes and cells were collected by GEVO1287 Kluyveromyces lactis, MATC. uraA1 trp 1 leu2 lysA1 centrifugation for 5 minutes at 3000xg. After removal of the 35 ade1 lac4-8 pKD1 medium, cells were washed with 10 mL MilliO H0. After GEVO1742 Kluyveromyces lactis, MATalpha uraA1 trp 1 leu2 lysA1 removal of the water, the cells were centrifuged again at ade1 lac4-8 pKD1 pdc1A::kan 3000xg for 5 minutes and the remaining water was carefully GEVO4458 Kluyveromyces lactis, MATalpha uraA1 trp 1 leu2 lysA1 removed using a 1 mL pipette tip. The cell pellets were weighed and then stored at -80°C. until further use. TABL E 59 Oligonucleotide Sequences Disclosed in Example 18. oGV # Sequence

821 CGGGTAATTAACGACACCCTAGAGG (SEQ ID NO: 132)

232O GGCTGTGTAGAAGTACTCGCCGATAG (SEQ ID NO: 133)

3 O65 AAAAAGGAGTAGAAACATTTTGAAGCTATGCGTTGATAAGGGCAACAACGTTA. GTATC (SEQ ID NO: 134)

3 O 66. ATACTAACGTTGTTGCCCTTATCAACGCATAGCTTCAAAATGTTTCTACTCCTT TTTTAC (SEQ ID NO: 135)

306.7 TCAAATTTTTCTTTTTTTTCTGTACAGTTACCCAAGCTGTTTTGCCTATTTTCAA AGC (SEQ ID NO: 136)

3 O 68 GCTTTGAAAATAGGCAAAACAGCTTGGGTAACTGTACAGAAAAAAAAGAAAAA TTTG (SEO ID NO : 137) US 9,012,189 B2 123 124 TABLE 59 - continued Oligonucleotide Sequences Disclosed in Example 18. oGV # Sequence

3 O69 AGTTCAAATCAGTTCGAGGATAATTTAAG (SEQ ID NO: 138)

3. Of O TTAATAAATGCTCAAAAGAAAAAAGGCTGGCG (SEQ ID NO: 139)

31.03 ACCGGTGCTTCTGCAGGTATTG (SEQ ID NO: 14 O)

31O6 ATGCTTGGTTGGAAGCAAATAC (SEQ ID NO: 141)

The K. lactis strain GEVO4458 was constructed from pulses. The transformation was incubated for 30 min at 30° GEVO1742 as follows. DNA constructs were made to delete C., followed by incubation for 22 min at 42°C. The cells were the TMA29 locus of K. lactis using SOE PCR. The 5' target collected by centrifugation (18,000xg, 10 sec, 25°C.). The ing sequence was amplified by PCR using GEVO1287 cells were resuspended in 400 uLYPD medium and allowed genomic DNA as template with primers oGV3103 and to recover overnight at 30° C. and 250 rpm. The following oGV3065. The 376 bp fragment was purified by gel electro morning, the cells were spread onto YPE plates 1% (w/v) phoresis. The 3' targeting sequence was amplified by PCR 25 yeast extract, 2% (w/v) peptone, 25 mL/L ethanol) supple using GEVO 1287 genomic DNA as template with primers mented with 0.1 g/L Hygromycin. Transformants were single oGV3106 and oGV3067. The 405 bp fragment was gel puri colony purified onto YPE plates supplemented with 0.1 g/L fied. The Hph marker was amplified by PCR using pGV2701 Hygromycin. (P-Hph, CEN/ARS. pUC-ori, bla) as template with prim The single colony isolates were patched onto YPE supple ers oGV3066 and oGV3068. The 1,165 bp fragment was gel 30 mented with 0.1 g/L Hygromycin plates and the patches were purified. Next the 5' targeting sequence and the hph marker were joined together using PCR products described as tem screened for the correct integration by colony PCR. Presence plate. The reaction was amplified using primers oGV3068 of the correct PCR product was confirmed using agarose gel and oGV3103. The 1984 bp fragment was gel purified. Next electrophoresis. To screen for the internal TMA29 coding the 5' targeting sequence plus Hph marker PCR fragment was 35 region, primers oGV3103 and oGV3106 were used. To screen joined with the 3' targeting sequence using PCR with primers the 5' integration junction, primers oGV3069 and oGV821 oGV3103 and oGV3106. The 2,331 bp was gel purified and were used. To Screen the 3' integration junction, primers used for transformation. Yeast DNA was isolated using the oGV2320 and OGV3070 were used. Zymo Research ZR Fungal/Bacterial DNA Kit (Zymo Yeast cells were cultured by inoculating 3 mL of YPD Research Orange, Calif.; Catalog #D6005). GEVO1287 was 40 medium (1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v) grown to saturation in 12.5 mL of YPD in baffled 125 mL glucose) in a 14 mL culture tube in triplicate for each strain. flasks. The entire culture was collected in 15 mL Falcon tubes Cultures were started from patches on aYPD plate 1% (w/v) and cells collected at 2700 rcf for 5 min. Genomic DNA was yeast extract, 2% (w/v) peptone, 2% (w/v) glucose, 2% agar). isolated according to the manufacturers instructions. The The cultures were incubated overnight at 30° C. and 250 rpm. DNA concentration was measured and all genomic DNA 45 The next day, the ODoo of the overnight cultures were mea preps were diluted to a final concentration of 25 ng/uL. sured and the volume of each culture to inoculate a 50 mL GEVO1742 was transformed as follows. 50 mL YPD culture to an ODoo of 0.1 was calculated. The calculated medium in 250 mL baffled flasks were inoculated with Volume of each culture was used to inoculate 50 mL of YPD GEVO1742 cells from a fresh plate. The cultures were incu in a 250 mL baffled flask and the cultures were incubated at bated overnight at 30° C. and 250 rpm. The next morning the 50 30° C. and 250 rpm overnight. Cells were harvested during culture was diluted 1:50 inYPD medium and allowed to grow mid-log phase at ODs of 1.8-2.2. The cultures were trans for 6 h. Cells were collected by centrifugation at 2700 rcffor ferred to pre-weighed 50 mL Falcon tubes and cells were 2 minat30°C. Cells were washed by fully resuspending cells collected by centrifugation for 5 min at 3000xg. After with 50 mL sterile MilliO water. Cells were collected by removal of the medium, cells were washed with 10 mL MilliO centrifugation at 2700 rcf for 2 min at 30° C. Cells were 55 H0. After removal of the water, the cells were centrifuged washed by resuspending with 25 mL sterile MilliO water. again at 3000xg for 5 min and the remaining water was Cells were collected by centrifugation at 2700 rcf for 2 minat carefully removed with a 1 mL pipette tip. The cell pellets 30° C. Cells were resuspended in 1 mL 100 mM lithium were weighed and then stored at -80C. acetate, transferred to an Eppendorf tube and collected by Cell pellets were thawed on ice and resuspended in lysis centrifuging at 14,000 rcf for 10 seconds. The supernatant 60 buffer (10 mM sodium phosphate pH7.0, 1 mM dithiothreitol, was removed and the cells were resuspended with 4x the 5% w/v glycerol) such that the result was a 20% cell suspen pellet volume in 100 mM LiOAc. A mixture of DNA (15 L sion by mass. One mL of glass beads (0.5 mm diameter) was of PCR product), 72 uL 50% PEG, 10u L 1 M lithium acetate, added to a 1.5 mL Eppendorf tube for each sample and 850LL and 3 uI of denatured salmon sperm DNA (10 mg/mL) was of cell Suspension were added. Yeast cells were lysed using a prepared for each transformation. In a 1.5 mL tube, 15 uL of 65 Retsch MM301 mixer mill (Retsch Inc. Newtown, Pa.), mix the cell suspension was added to the DNA mixture (170 uL), ing 6x1 min each at full speed with 1 min incubation on ice and the transformation Suspension was Vortexed for 5 short between. The tubes were centrifuged for 10 min at 21,500xg US 9,012,189 B2 125 126 at 4°C. and the supernatant was transferred to a fresh tube. MgSO, and 0.2 g/L G418 and incubated for 30 hat 30°C. in Extracts were held on ice until they were assayed using the an orbital shaker at 250 rpm. The flask cultures were trans TMA29 assay as described. ferred to four individual 2 L top drive motor fermenter vessels The specific activity of Gevo 1742 with the TMA29 gene with a working volume of 0.9 L of YPD containing 80 g/L for the reduction of (S)-2-acetolactate was 0.0043+0.0005 glucose, 5 g/L ethanol, 0.5 g/L MgSO, and 0.2 g/L G418 per umol/min/mg lysate. The specific activity of Gevo4459 vessel for a starting ODoo of 0.3. Fermenters were operated deleted for the TMA29 gene was 0.0019+0.0003 umol/min/ at 30° C. and pH 6.0 controlled with 6N KOH in a 2-phase mg lysate. aerobic condition based on oxygen transfer rate (OTR). Ini tially, fermenters were operated at a growth phase OTR of 10 Example 19 10 mM/h by fixed agitation of 700 rpm and an air overlay of 5 Increased Isobutanol Yield in Strains Comprising an sL/h. Cultures were grown for 22.5 h to approximately 10-11 ALD6 Deletion, a TMA29 Deletion and an Alcohol ODoo then immediately switched to production aeration Dehydrogenase with Increased k, and Decreased conditions for 40.7 h. Cell density during production phase K, in S. cerevisiae 15 approached 13-14 ODoo. The production phase was operated at an OTR of 0.5 mM/h by fixed agitation of 300 rpm. Peri The following example illustrates that the combination of odically, samples from each fermenter were removed to mea an ALD6 deletion, TMA29 deletion and overexpression of a Sure ODoo and to prepare for gas chromatography (GC) and gene encoding an ADH with improved kinetic properties liquid chromatography (LC) analysis. For GC and LC, 2 mL leads to increased isobutanol production and theoretical sample was removed into an Eppendorf tube and centrifuged yield. in a microcentrifuge for 10 min at maximum. One mL of the A S. cerevisiae CEN.PK2 strain, GEVO3991, was con supernatant was analyzed by GC1 (isobutanol, other metabo structed by transforming a S. cerevisiae CEN.PK2 strain, lites) and one mL analyzed by high performance liquid chro GEVO3956, which expresses an improved alcohol dehydro matography (LC1) for organic acids and glucose as genase (L. lactis ADH, L1 ADH) and a decarboxylase (L. 25 described. lactis KIVD, Ll kiv)2) from its chromosomal DNA with a GEVO3991 achieved a cell density of 13.8 during the 22.5 2u plasmid, pCV2603 (P:Ec ilvC coSc''''', h growth phase. The isobutanol produced during the entire Per:Ll ilvD coSc, Pco2:L1 adhA'.2u-ori, puC-ori, duration of the experiment (63.2 h) was 18.6+0.9 g/L with bla, G418R), expressing genes encoding enzymes: KARI, 0.84+0.10 g/L isobutyrate and 0.15+0.02 g/L acetate pro DHAD, and the improved ADH (EC ilvC coSc'P''' duced. The theoretical isobutanol yield achieved during the Ll ilvD coSc, and L1 adhA', respectively). production phase of the experiment (22.5-63.5 h) was TABLE 60 Genotype of Strains Disclosed in Example 19. GEVO No. Genotype GEVO3991 MATaura3 leu2 his3 trp1 ald6A::Pevo2:Ll adh Af':Pe:Sc TRP1 gpd1A:T K UR43 gpd2A:T K UR43 ma29A:T K UR-43 pdc1A::Pepci:Ll kiv D2 coSciS:Pir41:LEU2:Ter:Part1:BS alss1 coSc:Tcycl:Peck:Ll kiv D2 coEc: Po2:Sp HIS5 pdcSA:T K UR43 pdc6A::Priders:Sc AFT1:PeNo2:Ll adhA'':T-ki or 43 a?: Pe:KI URA3:TK, as evolved for C2 Supplement-independence, glucose tolerance and faster growth, pGV2603) GEVO3956 MATaura3 leu2 his3 trp 1 ald6A::Peo:Ll adhA':Pe:Sc TRP1

Pe:KI URA3:TK, as evolved for C2 Supplement-independence, glucose tolerance and faster growth

A fermentation was performed to determine the perfor- to 80.3+1.1% while the isobutyrate yield was only 0.013+0.001 mance of GEVO3991 (L1 adhA''', ALD6A, TMA29A) in g/g glucose. The production of DH2MB was not detected. four replicate fermenters. Glucose consumption, isobutanol In addition, three independent transformants of production, isobutyrate production, acetate production and GEVO3991 were also characterized in shake flasks. The ODoo were measured during the fermentation. For these strain was grown overnight in 3 mL of YPD containing 1% fermentations, single isolate cell colonies grown onYPDagar 65 ethanol and 0.2 g/L G418 at 30°C. at 250 rpm. These cultures plates were transferred to 500 mL baffled flasks containing 80 were diluted to an ODoo of 0.1 in 50 mL of the same medium mL of YPD containing 80 g/L glucose, 5 g/L ethanol, 0.5 g/L in a baffled 250 mL flask and grown overnight. The ODoo US 9,012,189 B2 127 128 was measured and a Volume of cells approximately equal to Strains, plasmids, and oligonucleotide sequences dis 250 OD was collected for each culture b centrifugation at closed in this example are listed in Tables 61, 62, and 63, 2700 refor 2 minutes and the cells were resuspended in 50 respectively. mL of fermentation medium (YPD containing 80 g/L glu cose, 0.03 g/L ergosterol, 1.32 g/L TWEENTM 80, 1% v/v ethanol, 200 mM MES, pH6.5), and transferred to an TABLE 61 unbaffled vented screw cap 250 mL flask. The ODoo was checked and the cultures were placed at 30° C. at 75 rpm to Genotype of Strains Disclosed in Example 20. initiate the microaerobic fermentation. Samples for liquid chromatography (LC), gas chromatography (GC) analysis GEVO No. Genotype and ODoo were taken at roughly 24h intervals. The samples 10 1947 ura,3-delta2, derived from strain NRRL-Y-7571 (2 mL) were centrifuged at 18,000x g for 10 min and 1.5 mL, Kluyveromyces marxianus (E. C. Hansen) of the clarified supernatant was used for analysis by GC 1 and van der Walt (1971) LC1. 2348 ura3-delta2 pdc1A::G418R, Ps. 1:31COX4 Fermentations started at an ODoo of about 4. The cells MTS:Bs alsS:Ps 1:URA3 grew to an ODoo of about 8 by 72 h of microaerobic fermen ura-delta2 tation. After 72 h, the isobutanol titer was 12.3 g/L and the 15 6403, 64.04 ura3-delta2 pdc1A::G418R, Ps. 1:31COX4 isobutanol yield was 67.2% of theoretical. Isobutyrate titer MTS:alsS: Ps. 1:URA3 ura,3- and yield were low: 0.6 g/L isobutyrate was produced at a delta2 tma29A::Ps 11-hph yield of 0.013 g/g, glucose. The production of DH2MB was not detected. Example 20 TABLE 62 Plasmid Disclosed in Example 20. Effect of TMA29 Deletion in K. marxianus Plasmid Name Relevant Genes/Usage Genotype The purpose of this example is to demonstrate that the 25 pGV2701 For SOE PCR to give the P:hph, CEN, puC ori, deletion of TMA29 in a Kluyveromyces marxianus strain hph fragment bla comprising ALS activity results in reduced DH2MB produc tion.

TABLE 63

Oligonucleotide Sequences Disclosed in Example 20.

Primer Sequence

3498 ATGTCTCAAGGTAGAAGAGCTG (SEQ ID NO: 142)

3.137 GGAGTAGAAACATTTTGAAGCTATGTATATCTTCTGAATCAATTGCACCGAC (SEQ ID NO: 143)

314 O CAAATTTTTCTTTTTTTTCTGTACAGAGAGGTATGATTAATACCAATGTCTTGGG (SEQ ID NO: 144)

3499 TCATTCACCACGGTAAATGTGG (SEQ ID NO: 145)

3.138 GTCGGTGCAATTGATTCAGAAGATATACATAGCTTCAAAATGTTTCTACTCC (SEQ ID NO: 146)

3139 GTATTAATCATACCTCTCTGTACAGAAAAAAAAGAAAAATTTGAAATATAAATAACG (SEO ID NO : 147)

35 O1 GAAGGAAATTCCAGTCTCCTAGTTCCTTTGAACAC (SEQ ID NO: 148)

232O GGCTGTGTAGAAGTACTCGCCGATAG (SEQ ID NO: 149)

35 OO CAGAACAATCAATCAACGAACGAACGACCCACCC (SEQ ID NO: 15O)

821 CGGGTAATTAACGACACCCTAGAGG (SEQ ID NO: 151)

3.141 AAGGAGATGCTTGGTTTGTAGCAAACACC (SEQ ID NO: 152) US 9,012,189 B2 129 130 Strain Construction: The K. marxianus TMA29 gene tation cultures were incubated at 30° C. and 75 rpm in homolog encoding the K. marxianus TMA29 protein (SEQ unbaffled 250 mL flasks. One 15 mL aliquot of medium was ID NO. 23) was deleted from parent K. marxianus strain also collected to use as a blank for LC4 analysis and was kept GEVO2348 as follows, resulting in strains GEVO6403 and at 4°C. until sample submission. After 72 h, 1.5 mL of culture GEVO64O4. was removed and samples were prepared as above for ODoo Genomic DNA was isolated from GEVO1947 as and LC4 analysis. In addition, samples for enzyme assays described. Constructs were made to integrate the E. coli hph were harvested at 72 h by transferring 80 OD's of the appro (hygromycin resistance) cassette into the TMA29 locus of priate sample to two 15 mL Falcon tubes centrifuged at GEVO2348 by SOE PCR as described. PCR step #1 consisted 3000xg for 5 min at 4°C. Pellets were resuspended in 3 mL of three reactions resulting in the 5' TMA29 targeting 10 cold, sterile water and were centrifuged at 5000xg for 2 min sequence, the 3' TMA29 targeting sequence, and the hph at 4°C. in a Swinging bucket rotor in the tabletop centrifuge. marker. The 5' targeting sequence was amplified from pre The water was removed by vacuum aspirator. The conical pared GEVO1947 genomic DNA with primers oGV3498 and tubes were stored at -80° C. oGV3137. The 385bp fragment was purified by gel electro The in vitro ALS enzymatic activities of the lysates were phoresis. The 3' targeting sequence was amplified from pre 15 measured as described. Table 64 shows the average in vitro pared GEVO1947 genomic DNA with primers oGV3140 and ALS enzymatic activity of lysates from the strains after 72 h. oGV3499. The 473 bp fragment was gel purified. The Pe: ALS activity is measurable in GEVO2348 (average of 3.14 hph:Tcl cassette was amplified from pCV2701 with Units/mg lysate) as well as in both tima29A strains primers oGV3138 and oGV3139. The 1,651 bp fragment was GEVO6403 and GEVO6404 (averages of 1.63 and 1.58 gel purified. The final SOE PCR step joined the 3 products Units/mg lysate respectively). from step #1 (5' targeting sequence/hph marker/3' targeting Table 64 also shows the DH2MB and DHIV titers by LC4 sequence). The reaction was amplified using primers for these strains. GEVO2348 (TMA29) strains produced oGV3498 and oGV3499. The 2,414 bp fragment was gel average DH2MB titers of 0.89 g/L while DHIV was not purified as described and used for transformation of 25 detected. The DH2MB titers were significantly decreased in GEVO2348 as described. Medium used to grow the cells for the tima29A Strains GEVO6403 and GEVO6404 which mea the transformation was YPE. Following the transformation, sured at 0.16 and 0.15g/L respectively. While the ALS activ 150 uL of the transformation culture was spread onto YPE ity is decreased in the tima29A Strains, this does not account plates containing 0.1 g/L hygromycin. The plates were incu for the D-80% decrease in DH2MB titers in the deletion bated at 30° C. and transformed colonies were single colony strains. For example, one technical replicate of GEVO2348 isolated and then patched for colony PCR on YPE plates 30 exhibited an ALS activity of 2.5 Units/mg lysate and pro containing 0.1 g/L hygromycin. duced 0.83 g/L DH2MB while one of the technical replicates Yeast Colony PCR was used to screen for the appropriate 3' of the tima29A strain GEVO6404 has similar activity of 1.9 integrationjunction, 5' integration junction, as well as lack of Units/mg lysate and produced only 0.16 g/L DH2MB. the TMA29 coding region as described. The proper 3' inte gration junction was confirmed using primers oGV3501 and 35 2320. The proper 5' integrationjunction was confirmed using TABLE 64 primers oGV3500 and oGV0821 were used. Finally, to screen ALS Activity, DH2MB and DHIV titers, and Percent DH2MB for deletion of the TMA29 internal coding region, primers Decrease intma29A Strains After 72h Fermentation. oGV3500 and OGV3141 were used. 40 ALS Fermentation: Shake flask fermentations was performed in Activity DH2MB DHIV by DH2MB triplicate for each of the strains GEVO2348 (TMA29), (Umg by LC4 LC4 decrease GEVO6403 (tma29A), and GEVO6404 (tma29A) as Strain TMA29 lysate) (gL) (gL) (%) described to determine if deletion of TMA29 in strains GEVO2348 -- 3.10.5 0.89 0.07 n.d. expressing BS alsS would result in diminished production of GEVO6403 A 1.6 + 0.2 0.16 - 0.02 n.d. 82% DH2MB. Single colony isolated transformants of tima29A 45 GEVO6404 A 1.6-0.3 0.15 + 0.01 n.d. 83% strains were patched to YPE plates containing 0.1 g/L hygro mycin, while parent strains were patched to YPE plates. Cells n.d. = not detected from the patches were used to inoculate 3 mL cultures of YPE. Cultures were incubated overnight at 30° C. and 250 rpm. Example 21 After overnight incubation, the ODoo of these cultures was 50 determined by diluting 1:40 in water. The appropriate amount Effect of TMA29 Deletion in Kluyveromyces lactis of culture was added to 50 mL of YPE to obtain an ODoo of 0.1 in 250 mL baffled flasks and incubated at 30° C. and 250 The purpose of this example is to demonstrate that the rpm. After a 24h incubation, the ODoo of these cultures was deletion of TMA29 in a Kluyveromyces lactis strain compris determined by diluting 1:40 in water. The appropriate amount 55 ing ALS activity results in reduced DH2MB production. of culture was added to 50 mL of YPD containing 8% glucose Strains, plasmids, and oligonucleotide primers disclosed in and 200 mM MES, pH 6.5 to obtain an ODoo of 5. Fermen this example are listed in Tables 65, 66, and 67, respectively. TABLE 65 Genotype of Strains Disclosed in Example 21.

GEVO Number Genotype

1742 MATalpha uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 podc1:kan, derived from K. lactis strain ATCC 200826 (Kluyveromyces lactis (Dombrowski) van der Walt, teleomorph) US 9,012,189 B2 131 132 TABLE 65-continued Genotype of Strains Disclosed in Example 21. GEVO Number Genotype 4458 MATalpha uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 podc1:kantma29:hph 6310, 6311, 6312 MATalpha uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 pdc1::kan pCV1429 6313, 6314, 6315 MATalpha uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 pdc1::kan pCV1645 6316,6317 MATalpha uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 podc1:kan + random integration of Bs alss:TRP1 6318, 6319, 6320 MATalpha uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 pdc1::kan tima29::hph pGV1429) 6321, 6322, 6323 MATalpha uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 pdc1::kan tima29::hph pGV1645) 6324, 6325 MATalpha uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 podc1: :kan tima29::hph + random integration of Bs alss:TRP1

TABLE 66 pended in 400 uL of 1.25xSC-HWLU and spread over SCD-W plates to select for transformed cells. Random inte Plasmids Disclosed in Example 21. gration of Ahd I linearized pGV1726 in both GEVO1742 and Plasmid Name Relevant Genes/Usage Genotype tma29A strain GEVO4458 was confirmed by colony PCR with primers oGV 1321 and oGV1324 that are specific to the pGV1429 High copy 1.6 L empty 1.6-ori, PMB1 vector containing TRP1 ori, bla, TRP1 internal BS alsS coding region as described. Strains pGV1645 High copy 1.6 L vector 1.6-ori, PMB1 GEVO6316, GEVO6317, GEVO6324, and GEVO6325 were containing TRP1 and Bs alss ori, bla, TRP1, Bs alss positive for the gene integration. pGV1726 Vector containing TRP1 and PMB1 ori, bla, TRP1, 25 (linearized Bs alss Bs alss Fermentation: A shake flask fermentation was performed with Ahdl) on the various GEVO strains (Table 65) as described to deter mine if deletion of TMA29 in strains expressing Bs alsS would result in diminished production of DH2MB. Single TABLE 67 Oligonucleotide Secuences Disclosed in Example 21. Primer Sequence dGW3 O65 AAAAAGGAGTAGAAACATTTTGAAGCTATGCGTTGATAAGGGCAACAACGTTAG TATC (SEQ ID NO: 153) oGV3 O66 ATACTAACGTTGTTGCCCTTATCAACGCATAGCTTCAAAATGTTTCTACTCCTTTT TTAC (SEQ ID NO: 154) oGV3 O67 TCAAATTTTTCTTTTTTTTCTGTACAGTTACCCAAGCTGTTTTGCCTATTTTCAAA GC (SEO ID NO: 155) dGW3 O68 GCTTTGAAAATAGGCAAAACAGCTTGGGTAACTGTACAGAAAAAAAAGAAAAATT TG (SEQ ID NO: 156) oGW3 1.03 ACCGGTGCTTCTGCAGGTATTG (SEO ID NO: 157) oGW3 1.06 ATGCTTGGTTGGAAGCAAATAC (SEQ ID NO: 158) dGW1321 AATCATATCGAACACGATGC (SEO ID NO: 159) oGW1324 AGCTGGTCTGGTGATTCTAC (SEQ ID NO: 16O)

Strain Construction: The K. lactis TMA29 gene homolog colony isolated transformants were patched to SCD-W plates, encoding the K. lactis TMA29 protein (SEQID NO: 7) was 60 non transformed parents were patched onto YPD. Cells from deleted from parent K. lactis strain GEVO1742 as follows, the patches were used to inoculate 3 mL cultures in either resulting in strain GEVO4458 as described in Example 18. YPD (parent strains and integrated strains) or 3 mL SCD-W. K. lactis strains GEVO1742 (parent, TMA29) and Cultures were incubated overnight at 30° C. and 250 rpm. GEVO4458 (tma29A) were transformed with plasmid After overnight incubation, the ODoo of these cultures was pGV1429 (empty control vector), pGV1645 (expressing 65 determined by diluting 1:40 in water. The appropriate amount Bs alsS) or with Ahd I linearized plasmid pGV 1726 (result of culture was added to 50 mL of YPD containing 5% glucose ing in random integration of Bs alss) as described, resus or SCD-W containing 5% glucose to obtain an ODoo of 0.1 US 9,012,189 B2 133 134 in 250 mL baffled flasks and incubated at 30° C. and 250 rpm. Example 22 After 24h incubation, the ODoo of these cultures was deter mined by diluting 1:40 in water. The appropriate amount of Effect of TMA29 Deletion in I. Orientalis culture was added to 50 mL of YPD containing 8% glucose, 200 mM MES pH 6.5 or SCD-W containing 8% glucose to 5 The following example illustrates that deletion of the I. obtain an ODoo of 5. When 250 OD's were not available to orientalis TMA29 gene results in decreased TMA29 activity start the fermentation, the entire 50 mL culture was used. and also results in decrease in DH2MB production in strains Fermentation cultures were incubated at 30°C. and 75 rpm in comprising ALS activity. unbaffled 250 mL flasks. A 15 mL conical tube was also collected for media blanks for LC1 and LC4 analysis as 10 TABLE 69 described and kept at 4°C. until sample submission. At the 72 h timepoint, 1.5 mL of culture was collected. ODoo values Genotype of Strains Disclosed in Example 22. were determined and samples were prepared for LC1 and LC4 analysis by centrifuging for 10 min at 14,000 rpm and GEVO fi Relevant Genotype removing 1 mL of the Supernatant to be analyzed. In addition 15 GEVO44SO ura Aura A samples for enzyme assays were harvested at the 72 h time pdc1-1A::Ll kiv D: Ts: loXP: Po: BS alss. pdc1-2A::Ll kiv D: Tsco: loXP: Po: BS alss point. 60 OD's of the appropriate sample were transferred TMA29,TMA29 into a 15 mL Falcon tube and centrifuged at 3000xg for 5 min GEVO12425 ura Aura A at 4°C. Pellets were resuspended in 3 mL cold, sterile water pdc1-1A:: Ll kiv D: Tscycl: loXP: PENo.1: Bs alss pdc1-2A :: Ll kiv): loxP and transferred to 3, 1.5 mL Eppendorf tubes (1 mL each) to TMA29,TMA29 make 3x20 OD replicates. The tubes were centrifuged at GEVO6155 ura Aura A 5000xg for 2 min at 4°C. in a swinging bucket rotor in the pdc1-1A::Ll kiv D: Tscyc1: loXP: PENo.1: BS alss tabletop centrifuge. The water was removed by vacuum aspi pdc1-2A::Ll kiv D: loxP rator. The Eppendorf tubes were stored at -80° C. TMA29, ma29A::Pe:Ll adhA: Prs: Ec ilvC'P'-': loxP: The in vitro ALS enzymatic activities of the lysates were 25 URA3: loXP: Pay: Ll ilvD measured as described. Table 68 shows the average in vitro GEVO6158 ura Aura A ALS enzymatic activity of lysates from the strains after 72 h. pdc1-1A:: Ll kiv D: Tscycl: loXP: PENo.1: BS alss ALS activity was measurable only in strains with Bs alsS pdc1-2A:: Ll kiv D: OxP adh A: Priors: randomly integrated (GEVO6316, GEVO6317, GEVO6324, coCB: 6325) or expressed from plasmid (GEVO6313-6315, 30 oxP:URA3: loxP. PENo.1: Ll ilvD/ GEVO6321-6323). ALS activity in strains with Bs alsS inte Ema29A:: PPC: L adh AF: grated is lower than in strains expressing Bs alsS from plas P:Ec ilvC'P''': loxP: URA3: loxP: PENo.1: Ll ilvD mid. However, the activity of 0.25 Units/mg lysate in the GEVO12473 ura Aura A TMA29 strains with integrated Bs alsS (GEVO6316, pdc1-1A:: Ll kiv D: Tscycl: loXP: PENo.1: BS alss GEVO6317) was still enough to produce a titer 1.06 g/L of 35 pdc1-2A:: Ll kiv D: OxP combined DHIV+DH2MB. ma29A:: OxP:URA3: OxP. ma29A::loxP:MEL5: loxP Table 68 shows the combined DHIV+DH2MB titers for the GEVO12474 ura Aura A various strains after 72 h offermentation based on LC1 analy pdc1-1A:: Ll kiv D: Tscycl: loxP: PENo.1: BS alss sis. Strain GEVO1742 (parent, TMA29) strains produced pdc1-2A:: Ll kiv D: OxP measurable combined DHIV+DH2MB titers only when 40 ma29A:: OxP:URA3: OxP. Bs alsS was randomly integrated (1.06 g/L) or expressed ELS: OxP from plasmid pGV1645 (0.45 g/L). These DHIV+DH2MB titers were abolished in the tima29A strain GEVO4458 when Strain Construction: Issatchenkia orientalis strains derived expressing Bs alsS via random integration (GEVO6324, from PTA-6658 were constructed that were wild-type for the GEVO6325) or plasmid (GEVO6321-6323). LC4 analysis 45 TMA29 gene (GEVO4450, GEVO12425), heterozygous for indicated that the majority of the combined DHIV+DH2MB deletion of one copy of the TMA29 gene (GEVO6155), or titer was in fact DH2MB. completely deleted for the TMA29 gene (GEVO6158, TABLE 68 ALS Activity, Combined DHIV + DH2MB Titer, and Percentage of DH2MB of Combined DHIV + DH2MBTiter.

Plasmid Integrated (I), DHIV - DH2MB Parent plasmid (P), ALS Activity by LC1 DH2MB - DHIV Strain Strain or control (C) TMA29 ALS (U/mg lysate) (gL) by LC4

GEVO1742 Ole -- OOOOOO OOOOOO GEVO6316,6317 GEVO1742 pGV1726 (I) -- -- O.25 OO6 1.06 O.23 GEVO4458 GEVO1742 Ole A OOOOOO OOOOOO GEVO6324, 6325 GEVO4458 pGV1726 (I) A -- O.86 O.28 OOOOOO GEVO631O-6312 GEVO1742 pCV1429 (C) -- OOOOOO OOOOOO GEVO6313-6315 GEVO1742 pGV1645 (P) -- -- 6.12 1.09 O45 O.O2 GEVO6318-6320 GEVO4458 pCV1429 (C) A OOOOOO OOOOOO GEVO6321-6323 GEVO4458 pGV1645 (P) A -- 1.23 O.45 OOOOOO nia = not applicable, samples had no detectable peak by LC1 so were not analyzed by LC4 US 9,012,189 B2 135 136 GEVO 12473, GEVO 12474) using standard yeast genetics flask were removed, the ODoo was measured and samples and molecular biology methods. These strains also carry a prepared for LC4 analysis by transferring 1 mL sample to an copy of the Bacillus subtilis alsS gene. Eppendorf tube and centrifuging at 18,000xg, 10 seconds, TMA29 Enzyme Assay: For the TMA29 in vitro assay, I. 25°C. After centrifugation, 0.75 mL of supernatant was trans orientalis strains GEVO4450 (TMA29/TMA29), ferred to a microtiter plate and analyzed by LC4. Also at 72h GEVO6155 (tma29A/TMA29), and GEVO6158 (complete cells for enzyme assays were collected by transferring 80 tma29A/tma29A) were grown by inoculating 25 mLYPD in ODs to 15 mL Falcon tubes as described. Cells for ALSassays 125 mL baffled flasks with cells from a fresh YPD plate. were resuspended, lysed, and assayed as described. Cultures were grown overnight at 30° C. and 250 rpm. These Table 71 shows the DH2MB production and ALS activities cultures were used to inoculate 50 mL of YPD in 250 mL. 10 for GEVO12425, 12473, and 12474 at 72h. The DH2MB titer baffled flasks to an ODoo of 0.05. The cultures were grown at was determined by LC4. The ALS activity was similar in all 30° C. and 250 rpm until they had reached an ODoo of strains. approximately 5-8 (late log phase). Cells were harvested by collecting 80 ODs of cells in a 50 mL Falcon tube and cen TABLE 71 trifuging at 2,700xg for 3 min. After removal of supernatant, 15 cells were placed on ice and washed with 5 mL cold water. DH2MB Production and ALS Activity in I. orientalis Strains at 72h Cells were centrifuged at 2,700xg for 3 minand the water was Fermentation. removed. The cell pellets were stored at -80° C. until use. DH2MB by LC4 ALS activity Additionally, the same strains were grown by inoculating 3 STRAIN g/L) Umg mL of YPD from fresh plates and growing for 8h at 30°C. and GEVO1242S 1870.60 4.6+ 1.1 250rpm. These cultures were used to inoculate 50 mL ofYPD GEVO12473 O.O8 O.O1 4.O. O.1 in 250 mL baffled flasks to an ODoo of 0.01 and the cultures GEVO12474 O.O7 O.OO 3.11.1 were grown at 30° C. and 250 rpm until they reached an ODoo of approximately 4-8. This culture was used to inocu 25 late 50 mL ofYPD containing 8% glucose, 200 mMMES pH Example 23 6.5 to a final ODoo of 4-5 by centrifuging an appropriate amount of culture at 2,700xg for 3 minin a 50 mL Falcon tube and then resuspending the cell pellet in 50 mL of the stated Effect of TMA29 deletion in S. pombe medium. Cells were incubated in 250 mL non-baffled flasks at 30° C. and 75 rpm for 48 h (fermentation phase). Eighty OD 30 The following example illustrates that the (S)-2-acetolac cell pellets were harvested as described. Cells were resus tate reduction activity is significantly decreased in an S. pended, lysed and assayed for TMA29 activity as described. pombe tima29A strain compared to an S. pombe TMA29 Table 70 shows the specific TMA29 activity of lysates of I. strain. orientalis strains GEVO4450, 6155, and 6158 in U/mg of total protein. Specific TMA29 activity is reduced in 35 TABLE 72 GEVO6155 (tma29/TMA29) and GEVO6158 (complete Genotype of strains disclosed in Example 23. tma29 deletion) as compared to GEVO4450 (TMA29/ TMA29). GEVO fi Genotype Source 40 GEVO6444 hade6-M216, ura4-D18, Bioneer strain BG 0000H8 TABLE 70 leu1-32 GEVO6445 h+ SPAC521.03A::kanMX4, Bioneer strain BG 1772H TMA29 Activity in I. orientalis Strains. ade6-M216, ura4-D18, leu1-32 TMA29 activity TMA29 activity TMA29 homolog Late log phase 48 h fermentation phase 45 (SEQ ID NO: 22) STRAIN Umg total protein Umg total protein deleted GEVO44SO O.OO48 OO1O OO27 OOO3 GEVO6155 O.OO25 OOO8 OO1OOOO1 Yeast strains GEVO6444 which has an intact TMA29 gene GEVO6158 O.OO23 OOO3 OO1OOOO3 (SEQ ID NO: 161) and GEVO6445 which has the TMA29 50 gene deleted, were grown overnight in 12 mLYPD in 125 mL Fermentation: For the fermentation, I. Orientalis strains baffled flasks at 250 rpm and 30° C. The next day, ODoo GEVO 12425 (TMA29/TMA29), GEVO12473 (tma29/ values were determined and technical triplicate cultures were tma29), and GEVO 12474 (tma29/tma29) were grown by started in 50 mL YPD with 5% glucose at an ODoo of inoculating 12 mLYPD in 125 mL baffled flasks with cells approximately 0.3. Cultures were allowed to grow at 250 rpm from a freshYPD plate. Cultures were grown overnight at 30° 55 and 30° C. throughout the day. At the end of the day, the C. and 250 rpm. The ODoo of the 12 mL overnight cultures cultures were diluted inYPD with 5% glucose to an ODoo of were determined and the appropriate amount was used to approximately 0.15 and incubated overnight at 250 rpm and inoculate 50 mL YPD containing 5% glucose in 250 mL 30° C. The cells were harvested upon reaching an ODoo of baffled flasks to an ODoo of 0.1. The flasks were incubated at between 4 and 6. To harvest pellets for enzyme assays 80ODs 30° C. and 250 rpm overnight. The ODoo of the 50 mL 60 of the appropriate sample were transferred into two 15 mL cultures was determined. The appropriate amount of culture Falcon tube (for duplicate samples) and centrifuged at was centrifuged at 2700 rcf for 5 min at 25°C. in 50 mL 3000xg for 5 min at 4°C. Pellets were resuspended in 3 mL Falcon tubes and the supernatant removed. The cells from cold, sterile water and were centrifuged at 5000xg for 2 min each 50 mL culture were resuspended in 50 mL YPD con at 4°C. in a Swinging bucket rotor in the tabletop centrifuge. taining 8% glucose, 200 mMMES, pH 6.5. The cultures were 65 The water was removed by vacuum aspirator. The pellets then transferred to 250 mL unbaffled screw-cap flasks and were stored at -80° C. Lysates were prepared and TMA29 incubated at 30° C. and 75 rpm. At 72 h samples from each enzyme assays were performed as described. US 9,012,189 B2 137 138 The specific activity of S. pombe GEVO6444 lysates for TABLE 73-continued the reduction of (S)-2-acetolactate was 0.018+0.002 U/mg -- total protein. Lysates of the tima29A strain GEVO6445 had a - Genotype of K. marxianus Strains Disclosed in Example 24, specific activity of 0.001 +0.002 U/mg total protein. 5 GEVO Number Genotype GEVO6271 MTS:Ps apti:ADH7:Ps FB41:URA3 ald6A:: PTEF1-hph Example 24 10 Effect of ALD6 Deletion in K. marxianus TABLE 74 The purpose of this example is to demonstrate that the Plasmids Disclosed in Example 24. deletion of ALD6 in a Kluyveromyces marxianus strain Plasmid Name Relevant Genes/Usage Genotype results in reduced isobutyraldehyde oxidation activity and is pGV2701 For SOE PCR to give the P:hph, CEN, plJC ori, bla isobutyrate production. hph fragment Strains, plasmids, and oligonucleotide primers disclosed in this example are listed in Tables 73, 74, and 75, respectively TABLE 75 Oligonucleotide Sequences Disclosed in Example 24. Primer Sequence

oGW.349 O GTCAAGATTGTTGAACAAAAGCC (SEQ ID NO: 162)

oGW.3492 GAGTAAAAAAGGAGTAGAAACATTTTGAAGCTATGGTTTAGTGGGGTTGGGGAA GCTGGC (SEQ ID NO: 163)

oGW.3493 CAAATTTTTCTTTTTTTTCTGTACAGGCCAACATCAAGAAGACTATTCCAAACTTG GTC (SEQ ID NO: 164)

oGW.349 TGTATGATTCGAAAGCTTCTTCACC (SEQ ID NO: 165)

oGW.3491 GCCAGCTTCCCCAACCCCACTAAACCATAGCTTCAAAATGTTTCTACTCCTTTTT TACTC (SEQ ID NO: 166)

oGW.3494 GACCAAGTTTGGAATAGTCTTCTTGATGTTGGCCTGTACAGAAAAAAAAGAAAAA TTTG (SEO ID NO : 167)

oGW.3497 TTACTCGAGCTTGATTCTGAC (SEQ ID NO: 168)

oGW232O GGCTGTGTAGAAGTACTCGCCGATAG (SEQ ID NO: 169)

oGW.3496 ATGTCTTCATCACTAGCAGAG (SEO ID NO : 17 O)

oGWO821 CGGGTAATTAACGACACCCTAGAGG (SEO ID NO : 171)

oGWOf O6 GGTTGGTATTCCAGCTGGTGTCG (SEO ID NO : 172)

TABLE 73 55 Strain Construction: The K. marxianus ALD6 gene homolog encoding the K. marxianus ALD6 protein (SEQID - Genotype of K. marxianus Strains Disclosed in Example 24. NO: 39) was deleted from parent K. marxianus strains GEVO Number Genotype GEVO1947 and GEVO2087 as follows, resulting in strains GEVO6264/GEVO6265, and GEVO6270/GEVO6271 GEVO6264, ura,3-delta2 ald6A::P-hph respectively. GEVO626S Genomic DNA was isolated from GEVO1947 as GEVO2O87 ura,3-delta2, PDC1, Psi Pic:31COX4 described. Constructs were made to integrate the E. coli hph MISS's RSE CO ERA, (hygromycin resistance) cassette into the ALD6 locus of GEVO6270 Sicia, SC, P.I.,TCOX 6s GEVO1947 and GEVO2087 by SOE PCR as described. PCR MTS:alsS:Ps. Its:kivD co HMI1 step #1 consisted of three reactions: the 5' ALD6 targeting sequence, the 3' ALD6 targeting sequence, and the hph US 9,012,189 B2 139 140 marker. The 5' targeting sequence was amplified from pre g/L/OD, respectively. These total and specific isobutyrate pared GEVO1947 genomic DNA with primers oGV3490 and titers were significantly decreased in the ald6A strain oGV3492. The 635 bp fragment was purified by gel electro GEVO6264 (0.06 g/L and 0.004 g/L/OD respectively), and phoresis. The 3' targeting sequence was amplified from pre also in the ald6A strain GEVO6265 (0.05 g/L and 0.003 pared GEVO1947 genomic DNA with primers oGV3493 and g/L/OD respectively). The ALD6 parent strain GEVO2087 oGV3495. The 645bp fragment was gel purified. The Pr: produced total and specific isobutyrate titers of 0.15 g/L and hph:To cassette was amplified from pCV2701 with 0.008 g/L/OD, respectively. The total and specific isobutyrate primers oGV3491 and oGV3494. The 1,665 bp fragment was titers were significantly decreased in the ald6A strain gel purified. The final SOE PCR step joined the 3 products GEVO6270 (0.05 g/L and 0.003 g/L/OD), and also in the from step #1 (5' ALD6 targeting sequence/hph/marker/3' 10 ald6A strain GEVO6271 (0.08 g/L and 0.005 g/L/OD, respec ALD6 targeting sequence). The reaction was amplified using tively). primers oGV3490 and oGV3495. The 2,826 bp fragment was gel purified and used for transformations of GEVO1947 and TABLE 76 GEVO2087 as described. Medium used to grow cells for the Isobutyrate Production of ALD6 Parent Strains and ald6A Strains transformation was YPD. Following the transformation, 150 15 Derived From Said ALD6 Parent Strains. uL of each transformation culture was spread onto YPD plates Isobutyraldehyde Feed Supplemented with 0.2 g/L hygromycin. The plates were Fermentation (48 hr incubated at 30°C. Transformed colonies were patched for initial colony PCR screening, then single colony isolated and Isobutyrate repatched on YPD plates supplemented with 0.2 g/L hygro Parent Titer Isobutyrate mycin. Strain Strain ALD6 (g/L) Decrease (%) Yeast Colony PCR was used to screen for the appropriate 3' GEVO1947 -- O.19 O.OS integrationjunction, 5' integration junction, as well as lack of GEVO6264 GEVO1947 O.O60.02 68% GEVO626S GEVO1947 O.05 - 0.02 74% the ALD6 coding region as described. The proper 3' integra GEVO2O87 -- O.15 0.03 tion junction was confirmed using primers oGV3497 and 25 GEVO627O GEVO2O87 O.05 - 0.03 67% oGV2320. The proper 5' integration junction was confirmed GEVO6271 GEVO2O87 O.O8 O.O2 47% using primers oGV3496 and OGVO821. Finally, deletion of the ALD6 internal coding region was confirmed using prim ers OGV3495 and OGVO706. Fermentation: A shake flask fermentation with 2 g/L isobu 30 Example 25 tyraldehyde was performed as described using technical trip licates of the ald6A strains GEVO6264/GEVO6265 and Effect of ALD6 Deletion in K. lactis GEVO6270/GEVO6271 and their corresponding ALD6 par ent Strains GEVO1947 and GEVO2087. The purpose of this example is to demonstrate that the Single colony isolated transformants of confirmed ald6A 35 deletion of ALD6 in a Kluyveromyces lactis strain results in strains were patched to YPD plates supplemented with 0.2 g/L reduced isobutyraldehyde oxidation activity and isobutyrate hygromycin plates and parents were patched to YPD plates. production. Cells from the patches were used to inoculate technical trip Strains, plasmids, and oligonucleotide primers disclosed in licate 3 mL cultures of YPD. Cultures were incubated over this example are listed in Tables 77, 78, and 79, respectively. night at 30° C. and 250 rpm. After overnight incubation, the 40 ODoo of these cultures was determined by diluting 1:40 in TABLE 77 water. The appropriate amount of culture was added to 50 mL of YPD with 5% glucose to obtain an ODoo of 0.1 in 250 mL Genotype of K. lactis Strains Disclosed in Example 25. baffled flasks and cultures were incubated at 30° C. and 250 GEVO Number Genotype 45 rpm. After 24h incubation, the ODoo of these cultures was GEVO1287 MATC. uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1, determined by diluting 1:40 in water. The appropriate amount Kluyveromyces lactis (Dombrowski) van der Walt, of culture was added to 50 mL of YPD containing 8% glucose, teleomorph, ATCC 200826 200 mMMES pH 6.5, and 2 g/L isobutyraldehyde to obtain GEVO6242 MATC. uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 ald6A::P1-hph an ODoo of 5. Fermentation cultures were incubated at 30°C. GEVO1830 MATC. uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 and 75 rpm in unbaffled 250 mL flasks. Unused media was 50 pdc1::kan:Ec ilvC AN:Ec ilvDAN coKl:Sc LEU2 collected as a media blank for LC analysis and kept at 4°C. integrated}{Ll kivD; Sc Adh7:Km URA3 until sample Submission. At 48 h, samples from each of the randomly integrated flasks were taken as follows. 1.5 mL of culture was removed {Ps cup-1:Bs alsS:TRP1 random integrated GEVO6244, MATC. uraA1 trp 1 leu2 lysA1 ade1 lac4-8 pKD1 into 1.5 mL Eppendorf tubes. ODoo values were determined GEVO6245 pdc1::kan:Ec ilvC AN:Ec ilvDAN coKl:Sc LEU2 and samples were prepared for LC1 analysis. Each tube was 55 integrated centrifuged for 10 min at 14,000 rpm and the supernatant was {Ll kiv.D, Sc Adh7:Km URA3 integrated analyzed by LC1. In addition samples for enzyme assays {Ps cup-1:Bs alsS:TRP1 were harvested after 48 h. 80 ODs of the appropriate sample random integrated ald6A::Pre-hph were transferred into two 15 mL Falcon tube (for duplicate samples) and centrifuged at 3000xg for 5 min at 4°C. Pellets 60 were resuspended in 3 mL cold, sterile water and were cen TABLE 78 trifuged at 5000xg for 2 min at 4°C. in a swinging bucket Plasmid Disclosed in Example 25. rotor. The water was removed by vacuum aspirator. The coni cal tubes were stored at -80° C. Plasmid Name Genotype Table 76 shows the isobutyrate titer after 48 h of fermen 65 pGV2701 P:hph, CEN, plJC ori, bla tation. The ALD6 parent strain GEVO1947 produced average total and specific isobutyrate titers of 0.19 g/L and 0.013 US 9,012,189 B2 141 142 TABL E 79 Oligonucleotide Sequences Disclosed in Example 25. Primer Sequence oGW35O2 GAAACACAGTGGATTAGTGCTGTC (SEO ID NO: 173) oGW3504 GAAGAGTAAAAAAGGAGTAGAAACATTTTGAAGCTATGCTCTTTGTAATTGTTGTT GGTG (SEO ID NO: 174) oGV3s Os CAAATTTTTCTTTTTTTTCTGTACAAACAGAGTCCATCCGTTTGAAACTGATTGCAT GTC (SEO ID NO: 175) oGV3s Of TCAAATTCTATTATCGCGCGGG (SEO ID NO: 176) oGV3 O3 CACCAACAACAATTACAAAGAGCATAGCTTCAAAATGTTTCTACTCCTTTTTTACT CTTC (SEO ID NO: 177) dGW3506 GACATGCAATCAGTTTCAAACGGATGGACTCTGTTTGTACAGAAAAAAAAGAAAA ATTTG (SEO ID NO: 178) oGV3 is O9 CTCCTCCGTTGCAGAACAAGGCTTTG (SEO ID NO: 179) oGW232O GGCTGTGTAGAAGTACTCGCCGATAG (SEQ ID NO: 180) oGW.3508 CGGTGTTAAGTGCCAGAAATTGGTTG (SEQ ID NO: 181) dGWO821 CGGGTAATTAACGACACCCTAGAGG (SEQ ID NO: 182) oGW351 O CGGCGTACTCGACGTCTTGAGAAGTAG (SEQ ID NO: 183)

Strain Construction: The K. lactis ALD6 gene homolog oGV3509 and oGV2320. The proper 5' integration junction encoding the K. lactis ALD6 protein (SEQ ID NO: 29) was was confirmed using primers oGV3508 and oGV 0821. deleted from parent K. lactis strains GEVO1287 and Finally, deletion of the ALD6 internal coding region was GEVO1830 as follows, resulting in strains GEVO6242 and 40 confirmed using primers oGV3508 and oGV3510. Fermentation: A first shake flask fermentation with 2 g/L GEVO6244/GEVO6245, respectively. isobutyraldehyde in the medium was performed using tech Genomic DNA was isolated from GEVO1287 as nical triplicates of the ald6A strain GEVO6242 and the ALD6 described. Constructs were made to integrate the E. coli hph wild-type parent strain GEVO1287. Single colony isolated (hygromycin resistance) cassette into the ALD6 locus of 45 transformants of confirmed ald6A deletion strains were GEVO1287 and GEVO1830 by SOE PCR as described. PCR patched to YPD plates supplemented with 0.1 g/L hygromy step #1 consisted of three reactions: the 5' ALD6 targeting cin plates, parent strains were patched onto YPD. Cells from sequence, the 3' ALD6 targeting sequence, and the hph the patches were used to inoculate technical triplicate 3 mL marker. The 5' targeting sequence was amplified from pre cultures ofYPD. Cultures were incubated overnight at 30° C. pared GEVO1287 genomic DNA with primers oGV3502 and 50 and 250 rpm. After overnight incubation, the ODoo of these oGV3504. The 639 bp fragment was purified by gel electro cultures was determined by diluting 1:40 in water. The appro phoresis. The 3' targeting sequence was amplified from pre priate amount of culture was added to 50 mL of YPD with 5% pared GEVO1287 genomic DNA with primers oGV3505 and glucose to obtain an ODoo of 0.1 in 250 mL baffled flasks and oGV3507. The 628 bp fragment was gel purified. The Pe: cultures were incubated at 30° C. and 250 rpm. After 24h hph:Tcl cassette was amplified from pCV2701 with 55 incubation, the ODoo of these cultures was determined by primers oGV3503 and oGV3506. The 1,663 bp fragment was diluting 1:40 in water. The appropriate amount of culture was gel purified. The final SOE PCR step joined the 3 products added to 50 mL of YPD containing 8% glucose, 200 mM from step #1 (5' targeting sequence/hph marker/3' targeting MES pH 6.5, and 2 g/L isobutyraldehyde to obtain an ODoo sequence). The reaction was amplified using primers of 5. Fermentation cultures were incubated at 30° C. and 75 oGV3502 and oGV3507. The 2,810 bp fragment was gel 60 rpm in unbaffled 250 mL flasks. Unused media was collected purified and used for transformations of GEVO1287 and as a media blank for LC1 analysis and kept at 4° C. until GEVO 1830 as described. Colonies were selected for hygro sample Submission. At 24 h, samples from each of the flasks mycin resistance on YPD plates supplemented with 0.1 g/L were taken as follows. 1.5 mL of culture was removed into 1.5 hygromycin. Yeast Colony PCR was used to screen for the mL Eppendorf tubes. ODoo values were determined and appropriate 3' integration junction, 5' integration junction, as 65 samples were prepared for LC1 analysis as described. Each well as lack of the ALD6 coding region as described. The tube was centrifuged for 10 min at 14,000 rpm and the super proper 3' integration junction was confirmed using primers natant was collected for analysis by LC1 as described. US 9,012,189 B2 143 144 A second shake flask fermentation with 2 g/L isobutyral culture to an ODoo of 0.1 was calculated. The calculated dehyde was performed as described using the ald6A deletion Volume of each culture was used to inoculate 50 mL of YPD strains GEVO6244/GEVO6245 and their corresponding in a 250 mL baffled flask and the cultures were incubated at ALD6 parent strain GEVO1830. This fermentation was 30° C. and 250 rpm. sampled at 24 and 48 h as described. Table 80 shows the 5 The cells were harvested during mid-log phase at ODs of isobutyrate titer for both of these fermentations. Isobutyrate 2.2-2.7 after 8 h of growth. The cultures were transferred to titers are significantly decreased in the ald6A strains com pre-weighed 50 mL Falcon tubes and cells were collected by pared to the ALD6 parent strains. centrifugation for 5 minutes at 3000xg. After removal of the medium, cells were washed with 10 mL MilliO H0. After TABLE 8O 10 removal of the water, the cells were centrifuged again at Isobutyrate Production of ALD6 Parent Strains and ald6A Strains 3000xg for 5 minutes and the remaining water was carefully Derived From Said ALD6 Parent Strains. removed using a 1 mL pipette tip. The cell pellets were Isobutyraldehyde Feed Isobutyraldehyde Feed weighed and then stored at -80° C. until further use. Fermentation (24 hr Fermentation (48 hr 15 Cell pellets were thawed on ice and resuspended in lysis buffer (10 mM sodium phosphate pH7.0, 1 mM dithiothreitol, Isobutyrate Isobutyrate Isobutyrate Titer Decrease Titer Isobutyrate 5% w/v glycerol) such that the result was a 20% cell suspen Strain (gL) (%) (g/L) Decrease (%) sion by mass. One mL of glass beads (0.5 mm diameter) was added to a 1.5 mL Eppendorf tube for each sample and 850LL GEVO1287 O.19 O.O3 n.d. n.d. of cell Suspension were added. Yeast cells were lysed using a GEVO6242 O.12 O.O2 36.8% n.d. n.d. GEVO1830 O.16 OOO O.12 O.O1 Retsch MM301 mixer mill (Retsch Inc. Newtown, Pa.), mix GEVO6244 O.O6 O.O2 62.5% O.04 O.O1 66.7 ing 6x1 min each at full speed with 1 min incubation on ice GEVO6245 O.O7 OOO 56.3% OOOOOO 79.2* between. The tubes were centrifuged for 10 min at 21,500xg n.d. = not determined in this experiment at 4°C. and the supernatant was transferred to a fresh tube. *based on LOO for isobutyrate of 0.025 g/L 25 Extracts were held on ice until they were assayed using the TMA29 assay as described to determine TMA29 activity towards (R/S)-AHB and (R/S)-AL. Example 26 The specific activity of S. cerevisiae TMA29 in GEVO3527 lysates, a wild-type MATa S. cerevisiae strain, TMA29 Activity Towards 2-aceto-2-hydroxybutyrate 30 for the reduction of (R/S)-AHB was 10.5+0.6 mU/mg. The tma29A strain GEVO3939 had a specific activity of 4.8+0.1 The following example illustrates that the S. cerevisiae mu/mg. The wild-type GEVO3527 strain had about a 2-fold TMA29 protein is active towards (S)-2-acetolactate ((S)-AL) higher specific TMA29 activity than the deletion strain. and 2-aceto-2-hydroxybutyrate (AHB). The specific activity of S. cerevisiae TMA29 in 35 GEVO3527 lysates, a wild-type MATa S. cerevisiae strain, TABLE 81 for the reduction of (R/S)-AL was 12.3+0.2 mU/mg. The tma29A strain GEVO3939 had a specific activity of 2.9-0.3 Genotype of Strains Disclosed in Example 26. mu/mg. The wild-type GEVO3527 strain had about a 4-fold GEVO fi Genotype Source higher specific TMA29 activity than the deletion strain. 40 GEVO3527 MATC his3A-1 leu2A ATCC# 201389 (BY4742) lys2A ura3A General Methods for Examples 27-30 GEVO3939 MATC. his3A-1 leu2A OpenBiosystems cathi YSC1054 lys2A ura3A (Yeast MATalpha collection) Strains, plasmids, gene/amino acid sequences, and primer tma29:kan sequences described in Examples 27-30 are listed in Tables 45 82.83, 84, and 85, respectively. Yeast Strains GEVO3939 from which the TMA29 (YMR226C) gene was deleted and its parent GEVO3527 TABLE 82 were each cultured in triplicate by inoculating 3 mL of YPD Genotype of Strains Disclosed in Examples 27-30. in a 14 mL culture tube in triplicate for each strain. Cultures Genotype or reference were started from patches on YPD agar plate for GEVO3527 50 E. coli BL21 (DE3) (Lucigen Corporation, Middleton, WI) and on YPD plates containing 0.2 g/L G418 for GEVO3939. E. coli DH5C. (Novagen, Gibbstown, NJ) The cultures were incubated overnight at 30° C. and 250 rpm. S. cerevisiae CEN.PK2 (Euroscarf, Frankfurt, Germany) The next day, the ODoo of the overnight cultures were mea sured and the volume of each culture to inoculate a 50 mL TABLE 83 Plasmids Disclosed in Examples 27-30.

Gevo No. Genotype or reference pET22(b)+ Novagen, Gibbstown, NJ pGV1102 Ps (E-HA-tag-MCS-Tcci, URA3, 2-micron, bla, puC-ori pGV1662 Ps (EF1-L. lactis kiv-Tse cyc1, bla, ColE1 ori, URA3, 21 ori. pGV1947 Ps (E-Ll adhA-Ts cyclbla URA3 pMB1 ori 21 ori pGV1947his Ps tri-L1 adhA"-Ts cyc bla URA3 pMB1 ori 21 ori pET1947 P:Ll adh A', bla, oripBR322, lacI US 9,012,189 B2 145 146 TABLE 83-continued Plasmids Disclosed in Examples 27-30. Gevo No. Genotype or reference pGV2274 Cloning vector containing Ll adhA coSc sequence (synthesized by DNA2.0, Menlo Park, CA) pGV2475 Ps tri-Ll adhA coSc'''-Ts cyc, bla, URA3, pMB1 ori, 2.1 ori pGV2476 Ps tri-Ll adhA coSci-Ts cyc1, bla, URA3, pMB1 ori, 2 ori pGV2477 Ps tri-L1 adhA coSci-Ts cyc, bla, URA3, pMB1 ori, 21 ori pGV30C11 Ps, refl-L1 adhA coSc''''-Ts cyc, bla, URA3, plMB1 ori, 21 ori

TABLE 84 Nucleic Acid and Protein Sequences Disclosed in Examples 27-30. Source Gene (SEQID NO) Protein (SEQ ID NO) iaciis L hA (SEQID NO: 184) hA (SEQ ID NO:185) iaciis L hA coSci (SEQID NO:186) hAisé (SEQID NO:187) iaciis L hA coSc? 7-i (SEQID NO:188) hA287-hise (SEQID NO:189) iaciis L hA coSci-fi (SEQID NO:190) hA30c -hise (SEQID NO: 191) iaciis L hA coSci-hise (SEQID NO:192) hAREl-his6 (SEQ ID NO: 193) iaciis L hA coSc744hise (SEQID NO: 194) hA744 his6 (SEQID NO: 195) iaciis L hA coSc'--lis (SEQID NO:196) hA43-hise (SEQID NO: 197) iaciis L hAF/his (SEQID NO: 198) hAF/his (SEQ ID NO: 199) iaciis L hAOFlo-hise (SEQID NO:200) hAOFlo-hise (SEQID NO: 201) iaciis L hASF11-his6 (SEQID NO: 202) hASF11-his6 (SEQID NO: 203) iaciis L hASPliohise (SEQID NO. 204) hASPliohise (SEQID NO: 205) iaciis L hA coSc (SEQID NO: 206) hA (SEQ ID NO:185) iaciis L hA coSc7 (SEQID NO: 207) hA287 (SEQID NO. 208) iaciis L hA coSc''' (SEQID NO: 209) hAC (SEQID NO:210) iaciis L hA coSc (SEQID NO: 211) hARE (SEQID NO: 212) lactis Ll hA coSc' (SEQID NO:213) hA' (SEQID NO:214) lactis Ll a hA coSc' (SEQID NO:215) hA' (SEQID NO: 216) lactis Ll a hAIF (SEQID NO:217) hAIF (SEQID NO: 218) lactis Ll a hAloo (SEQID NO: 219) hAloo (SEQID NO:220) iaciis L 8. hA.F (SEQID NO: 221) hA.F (SEQID NO: 222) iaciis L 8. hASPO (SEQID NO: 223) hASPO (SEQID NO:224)

TABL E 85

Primer Sequences (shown from 5' to 3') Disclosed in Examples 27-30. Primer Name Sequence:

GGAGAAAACCCATATGTCGTTTAC (SEQ ID NO: 225)

GCAGCCGAACGCTCGAGGGCGGCCG (SEQ ID NO: 226) His Not1 1947 rev CTCGAGCGGCCGCTTAGTGGTGGTGGTGGTGGTGTTTAGTAAA ATCAA (SEQ ID NO: 227) Sal1 for GAAAGCATAGCAATCTAATCTAAGTT (SEQ ID NO: 228) adhAcoSc Sallin for GTTTGTCGACATGAAGGCTGCAGTTGTCCGT (SEQ ID NO: 229) adhAcoSC Notlin his rev TCGAGCGGCCGCTTAGTGGTGGTGGTGGTGGTGCTTCGTGAAG TCTATAACCATTCTACC (SEQ ID NO: 230) pGV1994ep for CGGTCTTCAATTTCTCAAGTTTCAGTTTCATTTTTCTTGTTCTATT ACAAC (SEQ ID NO: 231) pGV1994ep rev CTAACTCCTTCCTTTTCGGTTAGAGCGGATGTGGG (SEQ ID NO: 232)

US 9,012,189 B2 149 150 TABLE 85- continued

Primer Sequences (shown from 5' to 3') Disclosed in Examples 27- 3 O. Primer Name Sequence: Recomb2S77Gen5 rev8 CGATCACCAACAGAAAGCGAGCTTACATCA (SEO ID NO: 258) Recomb2Y113 GenS for9 TTAAAAATGCAGGATATTCAGTTGATGGCG (SEO ID NO: 259) Recomb2Y113 Gen5 rev10 CGCCATCAACTGAATATCCTGCATTTTTAA (SEQ ID NO: 26O) Recomb2F113 Gen5 for11 TTAAAAATGCAGGATTTTCAGTTGATGGCG (SEQ ID NO: 261) Recomb2F113 Gen5 rev12 CGCCATCAACTGAAAATCCTGCATTTTTAA (SEQ ID NO: 262) Recomb2G113 Gen5 for13 TTAAAAATGCAGGAGGGTCAGTTGATGGCG (SEQ ID NO: 263) Recomb2G113 Gen5 rev14 CGCCATCAACTGACCCTCCTGCATTTTTAA (SEQ ID NO: 264) Recomb2T212 Mini for15 GAGCTGATGTGRYAATCAATTCTGGTGATG (SEQ ID NO: 265) Recomb2T212 Mini rev16 CATCACCAGAATTGATTRYCACATCAGCTC (SEQ ID NO: 266) Recomb2V264 Mini for17 TGGTTGCTGTGGCAKTACCCAATACTGAGA (SEO ID NO: 267) Recomb2V264 Mini rev18 TCTCAGTATTGGGTAMTGCCACAGCAACCA (SEQ ID NO: 268)

*A (Adenine), G (Guanline), C (Cytosine), T (Thymine), U (Uracil), R (Purine - A or G) Y (Pyrimidine - C or T), N (Any nucleotide), W (Weak - A or T), S (Strong - G or C), M (Amino - A or C) K (Ketc - G or T) B (Not A - G or C or T) H (Not G - A or C or T) D (Not C - A or G or T), and V (Not T - A or G or C)

Media and Buffers: Transformation of S. cerevisiae: In the evening before a SC-URA: 6.7 g/L DifcoTM Yeast Nitrogen Base, 14 g/L planned transformation, aYPD culture was inoculated with a SigmaTM Synthetic Dropout Media supplement (includes single S. cerevisiae CEN.PK2 colony and incubated at 30°C. amino acids and nutrients excluding histidine, tryptophan, 40 and 250 rpm over night. On the next morning, a 20 mLYPD and leucine), 10 g/L casamino acids, 20 g/L glucose, 0.018 culture was started in a 250 mL Erlenmeyer flask without g/L adenine hemisulfate, and 0.076 g/L tryptophan. baffles with the overnight culture at an ODoo of 0.1. This SD-URA: Commercially available at MP Biomedicals (Ir culture was incubated at 30° C. and 250 rpm until it reached vine, Calif.). Composition: 1.7 g/L yeast nitrogen base an ODoo of 1.3-1.5. When the culture had reached the desired (YNB), 5 g/L ammonium sulfate, 20 g/L glucose, with 45 ODoo, 200 uL of Tris-DTT were added, and the culture was casamino acids without uracil CSM-URA. allowed to incubate at 30° C. and 250 rpm for another 15 min. YPD (yeast peptone dextrose) media: 10 g/L yeast extract, The cells were then pelleted at 4°C. and 2,500xg for 3 min. 20 g/L peptone, 20 g/L glucose. After removing the Supernatant, the pellet was resuspended in Tris-DTT: 0.39 g 1,4-dithiothreitol per 1 mL of 1 M 50 10 mL of ice-cold buffer E and spun down again as described TrishC1, pH 8.0, filter sterilized. above. Then, the cell pellet was resuspended in 1 mL of Buffer A: 20 mM Tris, 20 mMimidazol, 100 mMNaCl, 10 ice-cold buffer E and spun down one more time as before. mM MgCl, adjusted to pH 7.4, filter sterilized. After removal of the supernatant with a pipette, 200 uL of Buffer B: 20 mM Tris, 300 mM imidazol, 100 mM NaCl, ice-cold buffer E were added, and the pellet was gently resus 10 mM MgCl, adjusted to pH 7.4, filter sterilized. 55 pended. The 6LL of insert/backbone mixture was split in half Buffer E: 1.2 g Tris base, 92.4 g glucose, and 0.2 g MgCl, and added to 50 uL of the cell suspension. The DNA/cell per 1 L of deionized water, adjusted to pH 7.5, filter sterilized. mixtures were transferred into 0.2 cm electroporation Construction of plT1947: The L. lactis adhA (L1 adhA) cuvettes (BIORAD) and electroporated without a pulse con gene was cloned out of pCV 1947 using primers His Not1 troller at 0.54 kV and 25 LF. Immediately, 1 mL of pre 1947 fivd and Sall rev and ligated into pET22b(+), yielding 60 warmed YPD was added, and the transformed cells were plasmid pET1947. allowed to regenerate at 30° C. and 250 rpm in 15 mL round Construction of pGV2476: Plasmid pGV2274 served as bottom culture tubes (Falcon). After 1 hour, the cells were template for PCR using forward primeradhacoSc Sallin for spun down at 4°C. and 2,500xg for 3 min, and the pellets and reverse primer adhAcoSC Notlin his rev. The PCR were resuspended in 1 mL pre-warmed SD-URA media. Dif product was purified, restriction digested with NotI and SalI. 65 ferent amounts of transformed cells were plated on SD-URA and ligated into pGV1662, which had been cut with Notland plates and incubated at 30°C. for 1.5 days or until the colonies SalI and purified. were large enough to be picked with sterile toothpicks. US 9,012,189 B2 151 152 Plasmid Mini-Preparation ofYeast Cells: The ZymoprepTM trifuged for 10 min at 23,500xg and 4°C., and the supernatant II Yeast Plasmid Miniprep kit (Zymo Research, Orange, was removed. Extracts were stored at 4°C. Calif.) was used to prepare plasmid DNA from S. cerevisiae Purification of ADH: The ADH was purified by IMAC cells according to the manufacturer's protocol for liquid cul (Immobilized metal affinity chromatography) over a 1 mL tures, which was slightly altered. An aliquot of 200 uL of Histrap High Performance (histrap HP) column pre-charged yeast cells was spun down at 600xg for 2 min. After decanting with Nickel (GE Healthcare) using an Akta purifier FPLC the supernatant, 200 uL of Solution 1 were added to resus system (GE Healthcare). The column was equilibrated with pend the pellets. To the samples, 3 uI of ZymolyaseTM were four column volumes (cv) of buffer A. After injecting the added and the cell/enzyme Suspensions were gently mixed by crude extracts onto the column, the column was washed with flicking with a finger. After incubating the samples for 1 hour 10 buffer A for 2 cvs, followed by a linear gradient to 100% elution buffer B for 15 cvs and collected in 96-well plates. The at 37°C., Solutions 2 and 3 were added and mixed well after fractions containing the protein were pooled and at Stored at each addition. The samples were then spun down at maximum 40 C. speed and 4° C. for 10 min. The following clean-up over ADHCuvette Assay: ADH activity was assayed kinetically Zymo columns was performed according to the manufactur 15 by monitoring the decrease in NADH concentration by mea er's instructions. The plasmid DNA was eluted with 10 uL of suring the absorbance at 340 nm. A reaction buffer was pre PCR grade water. Half of this volume was used to transform pared containing 100 mM Tris/HCl pH 7.0, 1 mM DTT, 11 E. coli DH5O. mM isobutyraldehyde, and 200 uMNADH. The reaction was Heterologous ADH expression in E. coli: Flasks (500 mL initiated by addition of 100 uL of crude extract or purified Erlenmeyer) containing 50 mL of Luria-Bertani (LB) protein in an appropriate dilution to 900 uL of the reaction medium (10g, tryptone, 10 g NaCl, 5g yeast extract per liter) buffer. with amplicillin (final concentration 0.1 mg/mL) were inocu ADH Microtiter Plate Activity Assay: The activity mea lated to an initial ODoo of 0.1 using 0.5 mL overnight LB, Surement in microtiter plates is a downscaled cuvette assay. culture of a single colony carrying plasmid plT1947. The 50 The total volume was 100 uL. Ten LL of crude lysates or mL LB expression culture was allowed to grow for 3-4 hat 25 purified enzyme, appropriately diluted, were placed in assay 250 rpm and 37°C. Protein expression was induced at ODoo plates. The reaction buffer was prepared as described above of about 1 with the addition of IPTG to a final concentration (isobutyraldehyde substrate only) and 90 uL thereof were of 0.5 mM. Protein expression was allowed to continue for 24 added to the enzyme solutions in the plates. The consumption hat 225 rpm and 25°C. Cells were harvested at 5300xg and of NADH was recorded at 340 nm in an infinite M200 plate 4°C. for 10 min, and then cell pellets were frozen at -20°C. 30 reader (TECAN Trading AG, Switzerland). until further use. ADH High-Throughput Activity Assay: Frozen yeast cell Heterologous Expression in S. cerevisiae CEN.PK2: pellets in 96-well plates were thawed at room temperature for Flasks (1000 mL Erlenmeyer) filled with 100 mL of SC-URA 20 min, and then 100 uL ofY-Per (Pierce, Catil 78990) were were inoculated with 1 mL overnight culture (5 mL SC-URA added. Plates were vortexed briefly to resuspend the cell inoculated with a single CEN.PK2 colony, grown at 30° C. 35 pellets. After a 60-min incubation period at room temperature and 250 rpm). The expression cultures were grown at 30°C. and 130 rpm, 300 uL of 100 mM Tris-HCl (pH 7.0) were and 250 rpm for 24 hours. The cells were pelleted at 5300xg added to the plates to dilute the crude extract. Following a for 5 min. The supernatant was discarded and the pellets were centrifugation step at 5,300xg and 4°C. for 10 min, 40 uL of spun again. The residual Supernatant was then taken off with the resulting crude extract were transferred into assay plates a pipette. The pellets were frozen at -20°C. until further use. 40 (flat bottom, Rainin) using a liquid handling robot. The assay Heterologous Expression in CEN.PK2 in 96-Well Plates plates were briefly spun down at 4,000 rpm and room tem for High Throughput Assays Shallow 96-well plates, 1 mL perature. Twelve mL assay buffer per plate were prepared capacity per well, filled with 300 uL of SC-URA were inocu (100mM Tris-HCl, pH 7.0, 1 mM, 0.5 mM, 0.25 mM or 0.125 lated with single CEN.PK2 colonies carrying plasmids cod mM isobutyraldehyde, 1 mM DTT, 200M NADH) and 100 ing for L1 adhA' or variants thereof. Deep 96-well plates, 45 uL thereof were added to each well to start the reaction. The 2 mL capacity per well, filled with 600 uL of SC-URA per depletion of NADH was monitored at 340 nm in an infinite well were inoculated with 50 uL of these overnight cultures. M200 plate reader (TECAN Trading AG, Switzerland) over 2 The plates were grown at 30° C. and 250 rpm for 24 h, and 1. were then harvested at 5300xg for 5 min and 4°C. and stored Determination of Specific Activity Based on Data at -20° C. 50 Obtained from the Activity Assays The protein concentra Preparation of ADH-Containing Extracts from E. coli: E. tions of samples containing heterologously expressed L. lac coli cell pellets containing expressed ADH were thawed and tis Adha. Such as crude extract and purified proteins, were resuspended (0.25 g wet weight/mL buffer) in buffer A. The measured using the Quick StartTM Bradford Kit (Bio-Rad, resuspended cells were lysed by sonication for 1 min with a Hercules, Calif.) following the manufacturers instructions. 50% duty cycle and pelleted at 11000xg and 4°C. for 10 min. 55 One unit of enzyme activity (1U) is defined as the amount of Extracts were stored at 4°C. enzyme that catalyzes the conversion of one micromole of Preparation of ADH-Containing Extracts from S. cerevi substrate per minute under the specified conditions of the siae CEN.PK2: S. cerevisiae CEN.PK2 cell pellets contain assay method. ing expressed ADH were thawed and weighed to obtain the Thermostability Measurements: Tso values (temperature, wet weight of the pellets. Cells were then resuspended in 60 at which 50% of the enzyme activity is retained after an buffer A such that the result was a 20% cell suspension by incubation time of 15 min) of the parent L1 adhA and variants mass. Glass beads of 0.5 mm diameter were added to the 1000 thereof were measured to obtain thermostability data. Thirty uL-mark of (0.5 mm diameter) of a 1.5 mL Eppendorf tube, uL aliquots of purified enzyme were transferred into PCR before 875ul of cell suspension were added. Yeast cells were tubes. Each tube was assigned to a specific incubation tem lysed by bead beating using a Retsch MM301 mixer mill 65 perature, which corresponded to a slot on the block of a (Retsch Inc. Newtown, Pa.), mixing 6x1 min each at full MastercyclerRep PCR machine (Eppendorf, Hamburg, Ger speed with 1-min icing steps between. The tubes were cen many) programmed with a gradient covering a 20° C.-tem US 9,012,189 B2 153 154 perature range. The tubes were incubated for 15 min in their PCR grade water, and used to transform electrocompetent S. slots. Then, the reaction was quenched on ice. The residual cerevisiae cells as described in General Methods. activity was determined with the ADH microtiter plate activ A total of 88 clones from each of the 100, 200, and 300 uM ity assay as described above. MnCl libraries were picked into 96-well plates along with Use of His-Tags for Purification: In each of the examples four clones containing parent plasmid pGV2476 and three described below, reference is made to an ADH enzyme com prising a his-tag. As is understood in the art, Such his-tags clones containing pCV1102 as no-ADH control. One well facilitate protein purification. As would be understood by one was left empty and served as a sterility control. After screen skilled in the art equipped with the present disclosure, ADH ing these libraries as described under General Methods (Het enzymes lacking said his-tags are equally or better Suited for 10 erologous expression in CEN.PK2 in 96-well plates for high the conversion of isobutyraldehyde to isobutanol. Examples throughput assays, ADH high-throughput activity assay), the of the modified ADH enzymes described herein which lack 300 uM library was chosen and an additional 4,000 clones the purification-enabling his-tags are found in SEQID NOs: were screened in the same way. A total of 24 variants had a 206-224. 15 more than 1.5-fold improvement compared to wild type and were chosen for a re-screen in triplicate. The top ten variants thereof were grown and expressed in 100 mL cultures as Example 27 described under General Methods (Heterologous expression in S. cerevisiae CEN.PK2), and their specific activities in Directed Evolution Via Random Mutagenesis crude yeast extracts were determined as described under Gen eral Methods (ADH microtiter plate assay). Two variants, The following example illustrates a method for improving Ll AdhA27' and L1 AdhA''' exhibited a more kinetic properties of an ADH and also describes the kinetic than 2-fold improvement in activity (0.3 and 0.25 U/mg total properties of such improved ADH enzymes. 25 lysate protein, respectively) compared to the wild-type Plasmid pGV2476, a derivative of plasmid pGV1662, car enzyme Ll Adh A" (0.1 U/mgtotallysate protein) and were rying the L1 adhA coSc' gene served as template for error characterized in greater detail. prone PCR using forward primer pGV 1994ep for and reverse primer pGV 1994 rev. These primers are specific to Plasmid DNA from these two variants was extracted as 30 described under General Methods (Plasmid mini-preparation the backbone pGV1662 and bind 50 bp upstream and down out of yeast cells) and Subjected to DNA sequencing stream of the ADH insert to create an overlap for homologous (Laragen, Los Angeles, Calif.), which revealed two mutations recombination in yeast. The compositions of the three error per variant as listed in Table 87. Two of these mutations prone PCR reactions are summarized in Table 86. The tem (Y50F and L264V) are localized in close proximity to the perature profile was the following: 95°C. 3 min initial dena 35 active site which is a gap between the Substrate binding turation, 95°C.30s denaturation, 55° C. 30s annealing, 72° domain (cyan) and the cofactor binding domain (green). C. 2 min elongation, 25 cycles, 5 min final elongation at 72° Mutations I212T and N219Y are located on the surface of the C. cofactor binding domain (as shown in FIG. 17). In order to highlight the location of the cofactor binding site mutations, TABLE 86 40 FIG. 17 entails two views on the structure alignment. PCR Conditions for the Error Prone Libraries. TABLE 87 final MinCl conc. LM 45 List of Mutations Found in Two Improved Variants of the 1OO 200 3OO First Error Prone Library (Generation 1). Templateng 2 2 2 primer forward M O.2 O.2 O.2 Variant Mutations primer reverse IM O.2 O.2 O.2 dNTP's IM) 400 400 400 Ll adhA28E7-hisé Taq buffer (10X stock) L 10 10 10 50 Ll adh AOC I-iisés MgCl2 LM 7 7 7 Taq polymerase U 8 8 8 MnCl2 (1 mM stock) M. 1OO 200 3OO PCR grade water L. 41.4 31.4 21.4 The two enzyme variants, Ll Adha?''' and Ll Adh A''''', were expressed from plasmids pGV2475 55 The PCR products were checked on a 1% analytical TAE and pGV30C11, respectively on larger scale (100 mL cultures agarose gel, DpnI digested for 1 h at 37°C. to remove traces each), purified, and characterized in greater detail as oftemplate DNA, and then cleaned up using a 1% preparative described under General Methods (Heterologous expression TAE agarose gel. The agarose pieces containing the PCR in S. cerevisiae CEN.PK2. Preparation of ADH-containing products were cleaned using Freeze n Squeeze tubes (BIO 60 extracts from S. cerevisiae CEN.PK2, Purification of ADH). RAD, Hercules, Calif.; catalog #732-61.66) followed by pel The wild-type Ll Adh A' enzyme was expressed from plas let paint procedure (Novagen, catalog #69049-3) according to mid pGV2476 and purified in the same way. The enzymes manufacturers’ protocols. In the meantime, plasmid were characterized for the kinetic properties as described pGV1662 was restriction digested with Not and SalI before under General Methods (ADH cuvette assay). Table 88 shows running out the digestion mixture on anagarose gel and pellet 65 the kinetic parameters measured with isobutyraldehyde and painting. Plasmid and insert, 500 ng each, were mixed NADH. A decreased K-value was observed for both vari together, precipitated with pellet paint, resuspended in 6 LL of ants, while the k,egg was only improved for Ll AdhA'.