US 20150337275A1 (19) United States (12) Patent Application Publication (10) Pub. No.: US 2015/0337275 A1 Pearlman et al. (43) Pub. Date: Nov. 26, 2015

(54) BOCONVERSION PROCESS FOR Publication Classification PRODUCING NYLON-7, NYLON-7.7 AND POLYESTERS (51) Int. C. CI2N 9/10 (2006.01) (71) Applicant: INVISTATECHNOLOGIES S.a.r.l., CI2P 7/40 (2006.01) St. Gallen (CH) CI2PI3/00 (2006.01) CI2PI3/04 (2006.01) (72) Inventors: Paul S. Pearlman, Thornton, PA (US); CI2P 13/02 (2006.01) Changlin Chen, Cleveland (GB); CI2N 9/16 (2006.01) Adriana L. Botes, Cleveland (GB); Alex CI2N 9/02 (2006.01) Van Eck Conradie, Cleveland (GB); CI2N 9/00 (2006.01) Benjamin D. Herzog, Wichita, KS (US) CI2P 7/44 (2006.01) CI2P I 7/10 (2006.01) (73) Assignee: INVISTATECHNOLOGIES S.a.r.l., (52) U.S. C. St. Gallen (CH) CPC. CI2N 9/13 (2013.01); C12P 7/44 (2013.01); CI2P 7/40 (2013.01); CI2P 13/005 (2013.01); (21) Appl. No.: 14/367,484 CI2P 17/10 (2013.01); CI2P 13/02 (2013.01); CI2N 9/16 (2013.01); CI2N 9/0008 (2013.01); (22) PCT Fled: Dec. 21, 2012 CI2N 9/93 (2013.01); CI2P I3/04 (2013.01); PCT NO.: PCT/US2012/071.472 CI2P 13/001 (2013.01); C12Y 102/0105 (86) (2013.01) S371 (c)(1), (2) Date: Jun. 20, 2014 (57) ABSTRACT Embodiments of the present invention relate to methods for Related U.S. Application Data the biosynthesis of di- or trifunctional C7 alkanes in the (60) Provisional application No. 61/578.265, filed on Dec. presence of isolated or in the presence of a recom 21, 2011, provisional application No. 61/578,272, binant host cell expressing those enzymes. The di- or trifunc filed on Dec. 21, 2011, provisional application No. tional C7 alkanes are useful as intermediates in the produc 61/578,289, filed on Dec. 21, 2011. tion of nylon-7, nylon-7.X. nylon-X.7, and polyesters. Patent Application Publication Nov. 26, 2015 Sheet 1 of 42 US 2015/0337275 A1

F.G.

Ho-oh HON-N-N-N-OH Pinelic acid G 1,7-heptanedio carboxylic acid reductase Sphingomonas 4. aidehyde DH falcohol DH (CAR) EC 1.2.99.- macrogofitabida EC ...- alcoho H EC 1.1.1 - --- Ho-ohO

Pimelic acid semialdehyde 7-hydroxyheptanoate (7-oxoheptanoate) co-transaminase EC 2.6.1- Amidohydrolase EC 3.52 Hos--~~ -e O C. 7-aminoheptanoic acid eantholactam Aldehyde DH EC 111 acchio H ~-NH. O 7-aminoheptanal 7-aminoheptanol ()-transaminase EC 261

heptamethylenediamine (1,7-diaminoheptane) Patent Application Publication Nov. 26, 2015 Sheet 2 of 42 US 2015/0337275 A1

FG, 2.

Biotin biosynthesis pathway Benzoate degradation pathway carboxylate pathway biosynthesis pathway Condensation of 3 X malonyl-CoA - (reaction not assigned to pathway) - Hos---sco Hor-lap O O O O fatA fat ThioesterPimeloy-CoA hydrolase tigase (EC(EC 3.1.2.-)6.2.1.-) %t. CoA transferase (ECW2.83.-) H or-no O O Acyl-(acyl carrier ) Acyl-CoA reductase Pima lic acid reduct eductase EC 2.50 afooxylic acid reductase VE: 1299.- EC 2.18O G

Horn-inO O Pimelic acid semiaidehyde Tetratin degradation pathway CO2 Ec: 4.1.1 - o-keto acid decarboxylase O H* O OH HO (.-Ketosuberate4. O Coenzyme B pathway (Methanogenic archae) Patent Application Publication Nov. 26, 2015 Sheet 3 of 42 US 2015/0337275 A1

FG, 3 Ho-roO O O-ketoguterateO Acetyl-CoA, H2O EC 2.3.3.14 (akSA) EC 4.2.1.114 (aksDiaksE) or EC 4.2.1.36 CoA EC 4.2.1.114 (aksDiaksE) CO2 EC 1.1.1.87 (aksF) O O to-oh -X // > O-ketoadipate O Acetyl-CoA, H2O EC 2.3.3.14 (akSA) EC 4.2.1.114 (aksDfaksE) or EC 4.2.1.36 CoA EC 4.2.1.114 (aks faksE) CO2 EC 1.1.1.87 (aksF) Ho-n-rotO O O O-ketopimelate Acetyl-CoA, H2O EC 2.3.3.14 (akSA) Coenzyme B EC 4.2.1.114 (aksDfaksE) or EC 4.2.1.36 COA EC 4.2.1.114 (aksOfaksE) CO2 EC 1.1.1.87 (aksF)

O O OH O

-- CO HO O O-ketoSuberate O H 2 pimetic acid semialdehyde (7-oxoheptanoate)

Change: 4.1.3- to 2.3.3.14 (homocitrate synthase) AksD/AksE can be 4.2.1.114 (R)-homocitrate yase only or 4.2.1.114 + 4.2.1.36 (homoisocitrate hydro-iyase AksF hono-isocitrate 1.1.1.87 Patent Application Publication Nov. 26, 2015 Sheet 4 of 42 US 2015/0337275 A1

FG. 4

O O condensation O O O o'-'s-coa os---is-ca Malonyl-CoA O O 3-oxoglutaryl-CoA 1sul CO2 oxo group elimination O S-CoA CoA Malonyl-CoA

O O Condensation & ----- OXO group elimination O O HO S-CoA o'-'s-coa Pimeloyl-CoA Glutaryl-CoA CO2 CoA Malonyl-CoA Patent Application Publication Nov. 26, 2015 Sheet 5 of 42 US 2015/0337275 A1

F.G. S

PA5OBO HO J--- S-acp (CYP107H1)s R S-acp Pimeloy-acp (CH2)CH3 long-chain faty acid-acp O y O Ho-oh Pineic acid ATP, CoA BioW EC 62.4 AMP, PP, L-Aia CO2 O O H CoA

HO --~ul S-COA “HN O O Pimeoyl-CoA BioF OH EC 2.34f BioC 7-keto-8-aminopelagonate BioH BioGiEOKFBOY

2 Biotin Patent Application Publication Nov. 26, 2015 Sheet 6 of 42 US 2015/0337275 A1

F.G. 6

EC 2.1.19 EC 2.3.18O O O BioC O O FabH O. O. O. -e -> HO S-COA so-sco O acp Madon yW-COA Maory-CoAy newestery 3-OXO-Gutaw-acol9 y- plgety methwester is, EC 1.3.1.10 EC 4.2.1.59 Fag Fabi FaZ Yo is acp Yo usicS. acp - Yo if acp glutary-acp methyl ester enoylglutary-acp methyl ester 3-hydroxy-glutaryl-acp methyl ester EC 2.3141 FabB EC 1.1.1.100 EC 4.289 O O O FacG O CH C. Faby O O YO acp) YO ---sus aCp Yo ---s-s acp 3-oxopimelyl-acp methyester 3-hydroxypimely-acp) methyl ester enoylpinely-acp methyl ester EC 3.110 EC 2.3.47 EC 3.185 Fabi O O BigF O O BioH O O a "HN OH XHO 1s-s-s-s acp Sas-s-s-sO acp 7-keto-8-aminopelargonate pinely-acp pimelyl-acp methyl ester EC 3.12.4 aaaS Fata, FatB Lipase or Biotin O O esterase O O HO ----- OH N O --~us OH pimelic acid pimelate monomethyl ester EC 2.180 O Acyl-acp reductase -----> HO SO Pimelic acid semialdehyde Patent Application Publication Nov. 26, 2015 Sheet 7 of 42 US 2015/0337275 A1

F.G. 7 coO Benzoate EC 6.2.1 2. C’soO O OH Benzoyl-CoA

COA-S Cyclohexane carboxylate O

O

2-ketocyclohexane-1-carboxyl-CoA

H D. 2-ketocyclohexanecarboxyCoA hydrolase (Rp-bad) O O g O Ho-r-r-sca -X-> Ho-sca Pimeloyl-CoA B-Oxidation Glutary-CoA Patent Application Publication Nov. 26, 2015 Sheet 8 of 42 US 2015/0337275 A1

FG, 8 -/- CO -t, - O OH OH OH Naphtalene Tetrain O (benzocyclohexane) 2,4-dihydrodec-2-ene-1,10-dioic acid O OH h X-K O HC3 O

HO --~sSO Puruvate Pimelic acid semialdehyde ThG | Ho-n-n-oh Pinetic acid Patent Application Publication Nov. 26, 2015 Sheet 9 of 42 US 2015/0337275 A1

FEG, 9

O O O O O sy- os---sca O SCoA O OH O O O n’ 2H) O ---sul SCOA Crotorate -s-so O ls-s-sS-1 scoA N-H2O APCOA r -CO2 O O O O crotonyl-CoA -> "sca -s-s SCCA 2H ho Enoyl-CoA O r O 3-hydroxy- OH O hydrataseEC 4.2.1.17 to- O ----- SCOA butyryl-CoA -lul OH. O. Pimeoyl-CoA SCOA --scoA. /y 2H 3-hydroxybutyryl- 2H

EC 1157 t aceto- O O O O acetyl-CoA Y CoA--sca O --sca O OO O - sco O-. acetyl-CoAy 1 sco 1sca EC 239COA 'year w- carboxylate N- COASH ATP O Patent Application Publication Nov. 26, 2015 Sheet 10 of 42 US 2015/0337275 A1

F.G. 10A Diaminopimelate NH2 NH2 decarboxylase Ho-oh 6 Lysine O O meso-2,6-diaminopimelate Ammonia iyase NH Ho-s-s-to O O ------6-amino-2-heptenedioic acid Enoate Reductase y ------NH ,

Annonia Ho-oh ------A ------lyase O O O K s t O3 di p 3.t e 2-amino-heptanedioic acid ------Amino goes O as O HO OH ~~~ HO OH AkSA HO -oh O O -o- HO s 2-heptenedioic acid Oo-ketopimel ate? O CoA transferase - - - Hydro-lyase Carbonyl reductase HO YO acid-thiol ligase OH trihomocitrate

Hos---is-CoAs, Ho-oh O t AKSD/AKSE O O O d O 2-heptenedioic acid-CoA c-hydroxypimelate HO 2 Enoate Enoate , OH Reductase Reductase, HO. O. Enoyl-CoA , Cis-(homo)3-aconitate dehydrogenas w t Aks), AkSE HO S-CoA HO OH -e- O O O O O Pimeloyl-CoA hioesterase Pineic acid HO OH HO YO Fatty-acyl-CoA Carboxylic acid (-)-threo-iso(homo)scitrate reductase reductase AkSF Decarboxylase O HO 2O -- HO O O O Pimelic acid semialdehyde O-ketosuberate Amino transferase Anino tances O Decarboxylase HO Hos----NH. se- r-r-no O O NH2 7-aminoheptanoic acid C-aminOSuberate Patent Application Publication Nov. 26, 2015 Sheet 11 of 42 US 2015/0337275 A1

FG, 10B

CoA transferase OH acid-thiol gase CH Ho-oh - - Ho----> S-CoA O O O O CoA-SH O-hydroxypimelate C O-hydroxypimeoyl-CoA Hydro-lyase Hydro-lyase CoA transferase Ho-oh acid-thioi Eigase Hos---sca O O 7 O O 2-heptenedioic acid COA-SH 2-heptenedioic acid-CoA Enoate Enoate Reductase Reductase COA transferase Enoyl-CoA dehydrogenase acid-thiol ligase or Ho-oh hioesterase Hos---is-CoA O O O O Pinneic acid CoA-SH Pimeoyl-CoA

Fig. 10B is a schematic diagram showing the conversion of either O-hydroxypimetate of its CoA ester to 2-heptenedioic acid or its corresponding CoA ester, O-hydroxypine:oyl-CoA, respectively, by a hydro-lyase. Similarly, 2-heptenedioic acid or its corresponding CoA ester can be reduced to Pimelic acid or pimeoyl-CoA, respectively, by an enoate reductases that acts on the free acid or the CoA activated thioester. In a similar fashion, the activation of O-hydroxypimetate or 2-heptenedioic acid may not be to the CoA thioester, but instead to the acpthioester by a acyl-acp-synthetase, The hydro-lyase may catalyse the elimination of water from O-hydroxypimetoyl-acp, and the enoate reductases may reduce the double bond of 2-heptenedioic acid-acp). Pinetoyl-CoA or pimeloyl acp may subsequently be hydrolysed by a transferase or acid-thiot ligase, for recycling of CoA or acp, or by A to produce pimetic acid, Patent Application Publication Nov. 26, 2015 Sheet 12 of 42 US 2015/0337275 A1

F.G. 1A

Fig. 11 A SDS-PAGE analysis on soluble and insoluble fractions from expression of the Citrobacter amafonaticus ammonia lyase protein (14) (expected protein size: 45 kDa) Lane 1, Pre-induction soluble sample, Lane 2, 4 h post-induction soluble sample, Lane 3, 24 h post-induction soluble sample: Lane 4, 24h negative control soluble sample (17), Lane 5. Pre-induction insoluble sample, Lane 64 h post-induction insoluble sample, ilane 7. 24 h post-induction insoluble Sanpie, Lane 8. 24 h regative control insoluble sample (117. Lanes A. Aarker Patent Application Publication Nov. 26, 2015 Sheet 13 of 42 US 2015/0337275 A1

F.G. B.

S Fig. 11B: SDS-PAGE analysis on soluble and insoluble fractions from expression of the Clostridium tetanomorphum (5) and Aspergilius oryzae (6) ammonia lyase (expected protein size: 45 kDa each) lane Assembly 5: Pre-induction soluble sample, Lane 2. Assembly 5: 4 h post-induction soluble sample, Lane 3. Assembly 5: 24 h post-induction soluble sample: Lane 4, Assembly 6: Pre-induction soluble sample; Lane 5. Assembly 6: 4 h post induction soluble sample; lane 6. Assembly 6: 24 h post-induction soluble sample: Lane 7. Assembly 5: Pre-induction insoluble sample, Lane 8. Assembly 15.4 h post-induction insoluble sample, lane 9. Assembly 5: 24 h post-induction insoluble sample: Lane 10. Assembly 6: Pre-induction insoluble sample; Lane 11. Assembly 6: 4 h post-induction insolubie sample, Lane 2. Assembly 6: 24 h post-induction insoluble sample; anes M. Marker Patent Application Publication Nov. 26, 2015 Sheet 14 of 42 US 2015/0337275 A1

FG, 12

Fig. 12: SDS-PAGE analysis on soluble and insoluble fractions from expression of the Nocardia sp. carboxylic acid reductase . (Expected protein sizes. 61 kDa for each protein) Lane M. Marker, Lane 1. Pre-induction solubie sample, lane 2. Pre-induction insoluble sample, Lane 3.4 h post-induction soluble sample, Lane 4, 4 h post-induction insoluble sample. Patent Application Publication Nov. 26, 2015 Sheet 15 of 42 US 2015/0337275 A1

F.G. 13

Carboxylic acid reductase assay consumption of NADPH, over time, at 340mm)

sax-y\YYYYYYYYYYYR *Sassocii Bezosts CAR ('Saprelic Acid CAR *::::: CAR ^SSodiu Benzoate 8.21 (YSSYPiec Acid 2. Adipic Acid B21.

Fig. 13. Change in absorbance over time due to reduction of pimelic acid. adipic acid and benzoic acid to pimelic acid semialdehyde, adipic acid semialdehyde and benzaldehyde by Nocardict CAR reductase activity. Benzoic acid was used as positive control and reaction mixtures with biocatalyst harbouring the empty vector as negative control. Patent Application Publication Nov. 26, 2015 Sheet 16 of 42 US 2015/0337275 A1

FG, 14

Fig. 14: SDS-PAGE analysis on soluble fraction and insoluble fraction from expression of the Haemophilus influenzae YoiA thioesterase protein (E8) (expected protein size: 17 kDa) Lane 1. Pre-induction soluble sample, lane 2. 4 h post-induction soluble sample, Lane 3. 24 h post induction soluble sample, Lane 4, 24 h negative control soluble sample (17), Lane 5. Pre-induction insoluble sample, Lane 6. 4 h post-induction insoluble sample, lane 7. 24 h post-induction insoluble sample, Lane 8, 24 h negative control insoluble sample (17); Lanes M. Marker. Patent Application Publication Nov. 26, 2015 Sheet 17 of 42 US 2015/0337275 A1

F.G. 5

8s

Fig. 15: SDS-PAGE analysis of the aminotransferase v-Omega Vffusion (piNG2022) and livi-Ad optimized omega fusion (pinG2030) proteins Expected protein sizes. 50 kDa and 47 kDa respectively) Lane 1. Pre-induction negative control, Lane 2. 4 h post-induction negative control, Lane 3. 24 h post-induction negative control, Lane 4. plNG 2022 Pre-induction, Lane 5. piNG2022 4 h post induction Lane 6. piNG2022 24 h post-induction; Lane 7. piMG 2030 Pre-induction, Lane 8 plNG 20304 h post-induction, Lane 9 piMG 2030 24 h post-induction, Lanes M. Marker Patent Application Publication Nov. 26, 2015 Sheet 18 of 42 US 2015/0337275 A1

F.G. 16

WXSaat

wSwka

...a...... M.&S. samysacscs. Nëss- W &^ my s N sasssssss-SS: & ex::::::::::::::::y: : SNN

Fig. 16: Results of the biotransformation of sodium pyruvate and 7-aminoheptanoic acid with the measured concentrations of 7-aminoheptanoic acid substrate and -alanine product. Patent Application Publication Nov. 26, 2015 Sheet 19 of 42 US 2015/0337275 A1

F.G. 17

Fig. 17: Summary of the HPLC results from the L- and D-2-aminosuberic acid enantiomer Concentrations in the six biotransformations after 48 and 96 hours. Patent Application Publication Nov. 26, 2015 Sheet 20 of 42 US 2015/0337275 A1

F.G. 18

Fig. 18: SES-PAGE analysis on soluble and insoluble fractions from the bead lysis of expression of the E.coli GadA (29), E.coli LySA (E30), E.coli GadA ilvE (31) and E.coli LySA iv (32) proteins. (Expected protein sizes: 52 kDa, 46 kDa, 52 kDa and 46 kDa respectively) Lanes M. Marker, lane 1. Negative contro 24h post-induction soluble sample, lane 2. Assembly 29 24 h soluble sample, Lane 3. Assembly 30 4 h soluble sample, Lane 4. Assembly 31, 24h soluble sample, lane 5. Assembly 31 24 h post-induction soluble sample, lane 6. Assembly 29 24 h post induction insoluble sample, lane 7. Assembly 30 24 h post-induction insoluble sample, Lane 8. Assembly 31 24 h post-induction insoluble sample, Lane 9. Assembly 32 24 h post-induction insoluble sample, Lane 10. 24 hour negative control insoluble sample (17). Patent Application Publication Nov. 26, 2015 Sheet 21 of 42 US 2015/0337275 A1

rgi2S-- ric-f- ris 2-2-3. i-f-3 rc2-i-S, lf Shai I tail & trait itai tail Bistransfortation after 34 hour incusation

Fig. 19. Formation of 7-aminoheptanoic acid from the decarboxylation of 2-aminosubcratc. Il 7 (negative control); I29 (pT2/GadA); 30 (pET2} f ySA): 3 (pET2/IlvE-GadA), 32 (pET2: five-ySA), GadA. E. coli glutamate decarboxylase, LySA. E. coli diaminopinelate decarboxylase, ilv E-GadA: Fusion between E.coli lv. branched chain aminotransferase (N-terminus peptide: 5 residues) and E. coli glutamate decarboxylase, five-ySA: Fusion between E.coli Iive branched chain amino acid anninotransferase (N- terminus peptide. 5 residues) and E. coli diaminopimelate decarboxylase

Patent Application Publication Nov. 26, 2015 Sheet 24 of 42 US 2015/0337275 A1

FG, 22 LACLA (Lactococcus lactis) :

s: s 388 is . . . . s 5St suer seis s & i ŠN is sic s ------P-bitra iki ... if 8'E is Specific hits

Super-failies w Multi-donains Š:ks:

Hypothetical acetolactate synthase MJO277 (Methan Ocal docOCCUS iannaschii) s: 388 3. % s 39. ------E-bitik: ;

Specificits SuperFariies Hulti-domains

Com/conE (Methano Celiapaldicola)

s { 3. 88 3. 30 8s. 3.

ser: seas R-8s it is sits : Sixx: 's Specificits Superfatiies: tirdains

Fig. 22. Analysis of the functional domains in LACLA acetolactate synthese, com/comE from Methanocella paludicola and the hypothetica acetoiactate Synthase MJO277 showed that aii of them have a similar structure containing a thiamine phosphate binding site Patent Application Publication Nov. 26, 2015 Sheet 25 of 42 US 2015/0337275 A1

FG, 2.3

lac pMB4 ori N -

laco 5 promoter \s. piNG2022 N acO Eve fragment 7606 bp

Omega transaminase Wf codon optimized gene)

Fig. 23: Map of pi NG2022 vector, containing the vE-Omega Vffusion gene

FG, 24

Each pMB ori

\ 1

lacot T5 promoter Nilacoi N pi NG2030 live fragment 7567 by

Ad optimized omega transaminase

Ampr Fig. 24: Map of plMG2030 vector containing the liv-Adoptimized Omega fusion gene

Patent Application Publication Nov. 26, 2015 Sheet 30 of 42 US 2015/0337275 A1

FIG. 25 (continued)

at galaggttt tagt Caagg CC ggctgatt gg CCC agt aag catcaat Cg atttggaa 61 gatC goggtt at Cagttgg tgaCgg Cat C tatgaagttga tCagg gtgta Caaaggagta 21 ttgttctggct tacgtgag Ca tgcagagic git tttitt Cagaa gtgct gotga aat C ggaatt i81 to a Citg CC at t Cagtataga. agat Ct. Cag tggga CCt. GC aaaag Ct9t a Cagga aaat 24l gCgg to agitg agg gag Ciggit ataCattcag aca a Caagag gtgttgg CCCC gCQaaaa CaC 3C Cag tatgaag CCggCCt Cqa. QCCg Caga Cit act g c Citata Cgtttacggit gaaaaaa Cog 361 gag caa gag C agg catacgg agtggC ggC C atta Cagatg aggat citt C g Ctggitta aga 421 totgatat ca aaagttctgaa tt tact.gitat aat Ct catga Cg aag Caaag gg C C tatgaa. 48 g c cq gag Cat ttgaa.gc.cat it tact tagg CaCCCCCttg tta CCCaCCC la Cat CCiC. 54 a a C gtt tatg C C gtt at Caa Cgg Cacagtg cga a CaCatc Cggctaatcg gct Cattotc 6 C aatgga atta Cacggatgaa tattittagga. Ctgatt gaga. agaatgg gat Caaa Citggat 661 gag actCctg. tcagtgaaga agagttgaaa. Cagg CG galag agatctititat titcqtcaacg. 721 acgg Cagaaa ttatt Caggit cgtgacgctic gatgga caat Cg at C ggaag Coggaaac Co 78 g g accggtga C Caaa Cagct to aggctgct tittcaagaaa gCatt Caa Ca gg Ctgctago 84 attt Cataa SEQ D 9: : Bacillus subtilis (g 16078032) Gene sequence

l KWWNGRE GRSEASE DRGY QEGG YEVRWYKGV FGFREHAER EERSAAEG 6 SLPFSEDLE WICKLVEN AVSEGAVYIQ TRGVARKH QYEAGLEPQ TAYTFTWKKP :21 ECECAYGWAA TDECLRLR CKSERY JVMTKORAYE A3AFEAR GWVTEGTSS RHANR NGREENG LEKNGKD EWSEEEK AEEIFISSI 24 EAEPVV DGSEGSGKP GPWTKOQAA FCESIQQAAS S SEQ D 10: : Bacilius subtilis (gi16078032) protein sequence (Expected size 31182 Da)

Patent Application Publication Nov. 26, 2015 Sheet 42 of 42 US 2015/0337275 A1

FG, 27

ont VECR US 2015/0337275 A1 Nov. 26, 2015

BOCONVERSION PROCESS FOR alkanes ultimately proceeds through a common intermediate, PRODUCING NYLON-7, NYLON-7.7 AND pimelic acid semialdehyde (also known as 7-oxoheptanoate), POLYESTERS even though there are a number of different feedstocks that can be used and a number of ways to produce pimelic acid CROSS-REFERENCE TO RELATED semialdehyde. FIG. 1 is a schematic depicting the biotrans APPLICATIONS formation of pimelic acid semilaldehyde to various difunc 0001. This application claims the benefit of U.S. Provi tional C7 alkanes. sional Appl. Ser. No. 61/503,043, filed Jun. 30, 2011; U.S. 0006 For example, pimelic acid semialdehyde can be Provisional Appl. Ser. Nos. 61/578.265, 61/578,272, and derived from pimeloyl- (CoA), pimeloyl-acyl 61/578,289, all of which were filed Dec. 21, 2011; and Inter carrier protein (acp), pimelic acid (also known as heptane national Application PCT/US2012/44984, filed Jun. 29, 1,7-dioate), or O-ketosuberate. As described herein, 2012, the entireties of all of which are incorporated by refer pimeloyl-CoA, pimeloyl-acp and pimelic acid semialde ence as if fully set forth herein. hyde can be derived from a number of sources, including from a diverse number of naturally occurring metabolic pathways FIELD OF THE INVENTION that form pimeloyl-CoA, pimeloyl-acp or pimelic acid semialdehyde as an intermediate in the naturally occurring 0002 Embodiments of the present invention relate to metabolic pathway. methods for the biosynthetic production of di- or trifunctional 0007. In addition, pimeloyl-CoA and pimelic acid also can C7 alkanes in the presence of one or more isolated enzymes or be derived from the corresponding enoate 2-heptene-dioic in the presence of a recombinant host cell expressing one or acid or its corresponding CoA ester, 2-heptenedioyl-CoA. In more enzymes. The di- or trifunctional C7 alkanes are useful some embodiments, these C7 2-enoates are derived from as intermediates in the production, for example, of nylon-7. D.L-diaminopimelate or 2-oxopimelate. In some embodi nylon-7.X and/or nylon-X.7. ments, the production of difunctional C7 alkanes (e.g., 7 BACKGROUND OF THE INVENTION amino heptanoic acid) proceeds through O-ketosuberate or C.-aminosuberate, which can be derived from D.L-diami 0003. The promotion of sustainable practices by the nopimelate. In other embodiments, the C.-ketosuberate or chemical industry is focused on lowering energy usage, recy C.-aminoSuberate is derived from C.-ketopimelate. cling raw materials, and reducing by-products produced dur 0008 More specifically, the present document provides a ing the production of goods. However, the chemical industry variety of methods of converting compounds in order to pro Vitally depends upon petroleum and natural gas as basic feed duce intermediates useful for the manufacture of nylon-7. stocks. Given the state of optimization associated with major nylon-7.7, and polyester. It is not required that all possible petrochemical processes, a significant shift in feedstock use steps of any one method be performed and any of the methods and manufacturing designs will be required to lower energy, can be terminated when a desired compound is obtained. emissions, and costs to below their current level. Biotechnol Thus, any method can be terminated after the production of ogy offers a new approach for the production of industrial for example, pimelic acid, 7-amino-heptanoic acid, enan chemical intermediates and end-products from renewable tholactam, 1,7-diaminoheptane, or 7-hydroxy-heptanoic carbohydrates, such as plant derived Sugars, or other renew acid. able feedstocks such as fatty acids. In addition, biotechnology 0009. A first method of converting a compound can also offers away to utilize waste or low value streams such as include contacting a compound selected from 2.6 diami glycerol from biodiesel production, or to transform petro nopimelate (2.6 DAP) and C.-amino-pimelate (AAP) with an chemical-derived feedstocks such as benzene, toluene, and that catalyzes the reductive deamination of 2,6 DAP polycyclic aromatic hydrocarbons (PAHs) to useful products to 6-amino-2-heptenedioic acid (6A2HA) or the reductive in bioprocesses that are more benign to the environment than deamination of AAP to 2-heptenedioic acid (2 HDA), such existing chemical processes. that 6A2HA or 2 HDA is produced. 0004 US2010/0203600 and WO2011/031147 are incor porated by reference in their entirety. There is a need in the art 0010. The compound can be 2.6 DAP and the first method for cost-effective biosynthetic routes to chemical intermedi can further involve contacting the 6A2HA with an enzyme ates to nylon-7, nylon-7.X. nylon-X.7, and polyesters and the that catalyzes the enoate reduction of the 6A2HA to AAP. end-products nylon-7, nylon-7.X nylon-X.7, and polyesters such that AAP is produced. especially from a variety renewable and non-renewable feed 0011. The AAP can be contacted with an enzyme that stocks. catalyzes the reductive deamination of the AAP to 2 HDA, such that 2 HDA is produced. SUMMARY OF THE INVENTION 0012. The 2 HDA can be contacted with an enzyme that 0005. This disclosure is based at least in part of the devel catalyzes the enoate reduction of the 2 HDA to pimelic acid opment of enzymatic systems and recombinant hosts for bio (PA), such that PA is produced. synthesizing di- or trifunctional C7 alkanes that are useful 0013 The PA can be contacted with an enzyme that cata intermediates (monomers) for producing nylon-7, nylon-X.7. lyzes the carboxylic acid reduction of the PA to pimelic acid nylon-7.X., and polyesters. In particular, as described herein, semialdehyde (PAS), such that PAS is produced. intermediates useful in the production of nylon-7, and nylon 0014. The PAS can be contacted with an enzyme that 7.X. nylon-X.7, and polyesters can be biosynthetically pro catalyzes the semialdehyde amination of the PAS to 7-amino duced from renewable feedstocks, such as plant-derived Sug heptanoic acid (7 AHA), such that 7 AHA is produced. ars and glycerol, or from petrochemical-derived feedstocks 0015 The 7 AHA can be contacted with an enzyme that Such as benzene or cyclohexane carboxylate. In some catalyzes the amide hydrolysis of the 7 AHA to enantholac embodiments, production of the di- or trifunctional C7 tam (ENTL), such that ENTL is produced. US 2015/0337275 A1 Nov. 26, 2015

0016 A second method of converting compound can 0033. The PAS with can be contacted with an enzyme that include, after performing all the steps of the first method up to catalyzes the semialdehyde amination of the PAS to 7 AHA, the production of 7 AHA, contacting the 7 AHA with an such that 7 AHA is produced. enzyme that catalyzes the aldehyde dehydrogenation of the 7 0034. The 7 AHA can be contacted with an enzyme that AHA to 7-amino-heptanal (7 AHT), such that AHT is pro catalyzes the amide hydrolysis of the 7 AHA to ENTL, duced. wherein ENTL is produced. 0017. The 7 AHT can be contacted with an enzyme that 0035. A sixth method of converting a compound can catalyzes the transfer of an amino group to the 7 AHT to include, after performing all the steps of the fifth method up to produce 1,7-diaminoheptane (1.7 DAH), such that 1.7 DAH the production of 7 AHA, contacting the 7 AHA with an is produced. enzyme that catalyzes the aldehyde dehydrogenation of the 7 0018. A third method of converting compound can AHA to 7 AHT, such that 7AHT is produced. include, after performing all the steps of the first method up to 0036. The 7 AHT can be contacted with an enzyme that the production of 2 HDA, contacting the 2 HDA with an catalyzes the transfer of an amino group to the 7 AHT to enzyme that catalyzes the transfer of Coenzyme A (CoA) to produce 1.7 DAH, such that 1.7 DAH is produced. the 2 HDA to produce 2-heptene diacid-CoA (2 HDA-CoA), 0037. A seventh method of converting a compound can such that 2 HDA-CoA is produced. include, after performing all the steps of the fifth method up to 0019. The 2 HDA-CoA can be contacted with an enzyme the production of AHP, contacting the AHP with an enzyme that catalyzes the enoate reduction of the 2 HDA-CoA to that catalyzes the dehydration of the AHP to 2 HDA, such that pimeloyl-CoA (PCoA), such that PCoA is produced. 2 HDA is produced. 0020. The PCoA can be contacted with an enzyme that 0038. The 2 HDA can be contacted with an enzyme that catalyzes the thioester hydrolysis of the PCoA to PA, such catalyzes the enoate reduction of the 2 HDA to pimelic acid that PA is produced. (PA), such that PA is produced. 0021. The PA can be contacted with an enzyme that cata 0039. The PA can be contacted with an enzyme that cata lyzes the carboxylic acid reduction of the PA to PAS, such that lyzes the carboxylic acid reduction of the PA to PAS, such that PAS is produced. PAS is produced. 0040. The PAS can be contacted with an enzyme that 0022. The PAS can be contacted with an enzyme that catalyzes the semialdehyde amination of the PAS to 7 AHA, catalyzes the semialdehyde amination of the PAS to 7 AHA, such that 7 AHA is produced. such that 7AHA is produced. 0041. The 7 AHA can be contacted with an enzyme that 0023 The 7 AHA can be contacted with an enzyme that catalyzes the amide hydrolysis of the 7 AHA to ENTL, such catalyzes the amide hydrolysis of the 7 AHA to ENTL, such that ENTL is produced. that ENTL is produced. 0042 An eighth method of converting a compound can 0024. A fourth method of converting a compound can include, after performing all the steps of the fifth method up to include, after performing all the steps of the third method up the production of 7 AHA, contacting the 7 AHA with an to production of 7 AHA, contacting the 7 AHA with an enzyme that catalyzes the aldehyde dehydrogenation of the 7 enzyme that catalyzes the aldehyde dehydrogenation of the 7 AHA to 7 AHT, such that 7AHT is produced. AHA to 7 AHT, such that 7AHT is produced. 0043. The 7 AHT can be contacted with an enzyme that 0025. The 7 AHT can be contacted with an enzyme that catalyzes the transfer of an amino group to the 7 AHT to catalyzes the transfer of an amino group to the 7 AHT to produce 1.7 DAH, such that 1.7 DAH is produced. produce 1.7 DAH, such that 1.7 DAH is produced. 0044) A ninth method of converting a compound can 0026. A fifth method of converting a compound can include, after performing all the steps of the seventh method include, after performing all the steps of the first method up to up to the production of 2 HDA, contacting the 2 HDA with an the production of AAP contacting the AAP with an enzyme enzyme that catalyzes the transfer of CoA to 2 HDA to pro that catalyzes the transfer of an amino group to the AAP to duce 2 HDA-CoA, such that 2 HDA-CoA is produced. produce O-keto-pimelate (AKP), such that AKP is produced. 0045. The 2 HDA-CoA can be contacted with an enzyme 0027. The AKP can be contacted with one or more that catalyzes the enoate reduction of the 2 HDA-CoA to enzymes that catalyze the ketone reduction of the AKP to PCoA, such that PCoA is produced. O-hydroxy-pimelate (AHP), such that AHP is produced. 0046. The PCoA can be contacted with an enzyme that 0028. The AHP can be contacted with an enzyme that catalyzes the thioester hydrolysis of the PCoA to PA, such catalyzes the transfer of CoA to the AHP to produce C-hy that PA is produced. droxy-pimelate-CoA (AHP-CoA), such that AHP-CoA is 0047. The PA can be contacted with an enzyme that cata produced. lyzes the carboxylic acid reduction of the PA to PAS, such that 0029. The AHP-CoA can be contacted with an enzyme PAS is produced. that catalyzes the dehydration of the AHP-CoA to 2 HDA 0048. The PAS can be contacted with an enzyme that CoA, such that 2 HDA-CoA is produced. catalyzes the semialdehyde amination of the PAS to 7 AHA, 0030 The 2 HDA-CoA can be contacted with an enzyme such that 7 AHA is produced. that catalyzes the enoate reduction of the 2 HDA-CoA to 0049. The 7 AHA can be contacted with an enzyme that PCoA, such that PCoA is produced. catalyzes the amide hydrolysis of the 7 AHA to ENTL, such 0031. The PCoA can be contacted with an enzyme that that ENTL is produced. catalyzes the thioester hydrolysis of the PCoA to PA, such 0050. A tenth method of converting a compound can that PA is produced. include, after performing all the steps of the ninth method up 0032. The PA can be contacted with an enzyme that cata to the production of 7 AHA, contacting the 7 AHA with an lyzes the carboxylic acid reduction of the PA to PAS, such that enzyme that catalyzes the aldehyde dehydrogenation of the 7 PAS is produced. AHA to 7 AHT, such that 7AHT is produced. US 2015/0337275 A1 Nov. 26, 2015

0051. The 7 AHT can be contacted with an enzyme that 0068 A seventeenth method of converting a compound catalyzes the transfer of an amino group to the 7 AHT to can include, after performing all the steps of the ninth method produce 1.7 DAH, such that 1.7 DAH is produced. up to the production of PCoA, contacting the PCoA with an 0052 An eleventh method of converting a compound can enzyme that catalyzes the reduction of the PCoA to PAS, such include, after performing all the steps of the fifth method up to that PAS is produced. the production of AKP contacting the AKP with an enzyme 0069. The PAS can be contacted with an enzyme that that catalyzes the O-keto acid chain elongation of the AKP to catalyzes the semialdehyde amination of the PAS to 7 AHA, C.-keto-Suberate (AKS), such that AKS is produced. such that 7 AHA is produced. 0053. The AKS can be contacted with an enzyme that 0070 The 7 AHA can be contacted with an enzyme that catalyzes the transfer of an amino group to the AKS to pro catalyzes the amide hydrolysis of the 7 AHA to ENTL, such duce C.-amino suberate (AAS), such that AAS is produced. that ENTL is produced. 0054 The AAS can be contacted with an enzyme that 0071. An eighteenth method of converting a compound catalyzes the C-amino acid decarboxylation of the AAS to 7 can include, after performing all the steps of the seventeenth AHA, such that 7 AHA is produced. method up to the production of 7 AHA, contacting the 7AHA 0055. The 7 AHA can be contacted with an enzyme that with an enzyme that catalyzes the aldehyde dehydrogenation catalyzes the amide hydrolysis of the 7 AHA to ENTL, such of the 7 AHA to 7 AHT, such that 7 AHT is produced. that ENTL is produced. 0072 The 7 AHT can be contacted with an enzyme that 0056. A twelfth method of converting a compound can catalyzes the transfer of an amino group to the 7 AHT to include, after performing all the steps of the eleventh method produce 1.7 DAH, such that 1.7 DAH is produced. up to the production of 7 AHA, contacting the 7 AHA with an 0073. A nineteenth method of converting a compound can enzyme that catalyzes the aldehyde dehydrogenation of the 7 include contacting a compound selected from 6A2HA and 2 AHA to 7 AHT, such that 7AHT is produced. HDA with an enzyme that catalyzes the enoate reduction of 0057 The 7 AHT can be contacted with an enzyme that 6A2HA to AAP or for the enoate reduction of 2 HDA to PA, catalyzes the transfer of an amino group to the 7 AHT to such that AAP or PA is produced. It is understood that, fol produce 1.7 DAH, such that 1.7 DAH is produced. lowing the production of AAP or PA, any of the steps after the 0058. A thirteenth method of converting a compound can production of AAP or PA described above in the first to include, after performing all the steps of the eleventh method eighteenth methods can be performed. up to the production of AKS, contacting the AKS with an 0074. A twentieth method of converting a compound can enzyme that catalyzes the C-keto acid decarboxylation of the include contacting a compound selected from PCoA and AKS to PAS, such that PAS is produced. pimeloyl acp (PACP) with an enzyme that catalyzes the 0059. The PAS can be contacted with an enzyme that thioesterase hydrolysis PCoA or PACP to PA, wherein PA is catalyzes the semialdehyde amination of the PAS to 7 AHA, produced. such that 7AHA is produced. 0075. The PA can be contacted with an enzyme that cata 0060. The 7 AHA can be contacted with an enzyme that lyzes the carboxylic acid reduction of the PA to PAS, such that catalyzes the amide hydrolysis of the 7 AHA to ENTL, such PAS is produced. that ENTL is produced. 0076. The PAS can be contacted with an enzyme that 0061. A fourteenth method of converting a compound can catalyzes the semialdehyde amination of the PAS to 7 AHA, include, after performing all the steps of the thirteenth method such that 7 AHA is produced. up to the production of 7 AHA, contacting the 7 AHA with an 0077. The 7 AHA with an enzyme that catalyzes the amide enzyme that catalyzes the aldehyde dehydrogenation of the 7 hydrolysis of the 7 AHA to ENTL, such that ENTL is pro AHA to 7 AHT, such that 7AHT is produced. duced. 0062. The 7 AHT can be contacted with an enzyme that 0078. A twenty first method of converting a compound can catalyzes the transfer of an amino group to the 7 AHT to include, after performing all the steps of the twentieth method produce 1.7 DAH, such that 1.7 DAH is produced. up to the production of 7 AHA, contacting the 7 AHA with an 0063 A fifteenth method of converting a compound can enzyme that catalyzes the aldehyde dehydrogenation of the 7 include, after performing all the steps of the fifth method up to AHA to 7 AHT, such that 7AHT is produced. the production of PCoA, contacting the PCoA with an (0079. The 7 AHT can be contacted with an enzyme that enzyme that catalyzes the reduction of the PCoA to PAS, such catalyzes the transfer of an amino group to the 7 AHT to that PAS is produced. produce 1.7 DAH, such that 1.7 DAH is produced. 0064. The PAS can be contacted with an enzyme that 0080 A twenty second method of converting a compound catalyzes the semialdehyde amination of the PAS to 7 AHA, can include contacting AKP with one or more enzymes that such that 7AHA is produced. catalyze the ketone reduction of AKP to AHP, such that AHP 0065. The 7 AHA can be contacted with an enzyme that is produced. The AKP can have been produced by chain catalyzes the amide hydrolysis of the 7 AHA to ENTL, such elongation of O-keto adipate or O-ketoglutarate. The AHP that ENTL is produced. can then be converted to PA or to PCoA. It is understood that, 0066. A sixteenth method of converting a compound can following the production of PA or PCoA, any of the steps after include, after performing all the steps of the fifteenth method the production of PA or PCoA described above in the first to up to the production of 7 AHA, contacting the 7 AHA with an eighteenth methods can be performed. enzyme that catalyzes the aldehyde dehydrogenation of the 7 I0081. A twenty third method of converting a compound AHA to 7 AHT, such that 7AHT is produced. can include converting AKP to AKS by C.-keto acid chain 0067. The 7 AHT can be contacted with an enzyme that elongation. The AKP can have been produced by chain elon catalyzes the transfer of an amino group to the 7 AHT to gation of C.-keto adipate or C-ketoglutarate. It is understood produce 1.7 DAH, such that 1.7 DAH is produced. that, following the production of AKS, any of the steps after US 2015/0337275 A1 Nov. 26, 2015

the production of AKS described above in the first to eigh co-transaminase can be in EC 2.6.1, e.g., EC 2.6.1.18; EC teenth methods can be performed. 2.6.1.19; EC 2.6.1.11: EC 2.6.1.13: EC 2.6.1.39: EC 2.6.1.48: 0082. A twenty fourth method of converting a compound or EC 2.6.1.62. can include contacting PAS with an enzyme that catalyzes the 0090. In the described methods, an enzyme that catalyzes semialdehyde amination of PAS to 7 AHA, such that 7AHA thioester hydrolysis of CoA thioesters can include a is produced. The PAS can have been obtained by conversion thioesterase, an acid-thiol ligase, or a CoA transferase. The from 2,6 DAP, AKG, PCoA, or PACP. It is understood that, thioesterase can be in EC 3.1.2, e.g., Such as the gene product following the production of AHA, any of the steps after the of YciA, tesB or Acot13; EC 3.1.2.2: EC 3.1.2.18; EC 3.1.2. production of AHA described above in the first to eighteenth 19; or EC 3.1.2.20. The acid-thiol ligase/CoA synthetase can methods can be performed. be in EC 6.2.1, e.g., EC 6.2.1.3; EC 6.2.1.5, EC 6.2.1.14; or 0083. A twenty fifth method of converting a compound EC 6.2.1.23. The CoA transferase can be in EC 2.8.3, e.g., EC can include contacting a compound selected from PCoA and 2.8.3.12: EC 2.8.3.13 or EC 2.8.3.14, or or the geneproduct of PACP with an enzyme that catalyzes the reduction of the ThnH that catalyses the reversible transfer of CoA to pimelic compound to PAS, such that PAS is produced. It is understood acid. that, following the production of PAS, any of the steps after 0091. In the described methods, the enzyme that catalyses the production of PAS described above in the first to eigh thioester hydrolysis of acyl-carrier-protein (ACP) teenth methods can be performed. Moreover, any of the steps thioesters, can include a acyl-ACPI thioesterase from EC described in the first to eighteenth and the twenty second 3.1.2, e.g. EC 3.1.2.14 or EC 3.1.2.21, an acyl-ACPI-synth method leading to the production of PCoA can be performed tase from EC 6.2.1, e.g. EC 6.2.1.14, such as BioW; and EC prior to the twenty fifth method. 6.2.1.20. 0084. A twenty sixth method of converting a compound 0092. In the described methods, an enzyme that catalyzes can include contacting PAS with an enzyme that catalyzes the the transfer of an O-amino group can include an amino acid alcoholdehydrogenation of PAS to 7-hydroxy-heptanoic acid aminotransferase. The amino acid aminotransferase can be a (7HHA), such that 7 HHA is produced. It is understood that L- or D-selective aminotransferase in EC 2.6.1, e.g., EC 2.6. any of the steps described in the first to eighteenth and the 1.39; EC 2.6.1.42 or EC 2.6.1.21, or a 2-aminohexanoate twentieth, twenty second, twenty third, and twenty fifth meth aminotransferase from EC 2.6.1.67. The enzyme that cataly ods leading to the production of PAS can be performed prior ses the transfer of an alpha amino group can also be a dehy to the twenty sixth method. drogenase from the class EC 1.4.1.-. Such as glutamate dehy 0085. In any of the described methods the enzyme(s) can drogenase, which depending on the co-factor used belongs to be used in purified form. Alternatively, the enzyme(s) can be EC 1.4.1.2: EC 1.4.1.3; or EC 1.4.1.4, or a diaminopimelate used in the form of a cell lysate or a partially purified cell dehydrogenase from EC 1.4.1.16. lysate. In addition, the enzyme(s) can be in a cell recombi 0093. In the described methods, an enzyme that catalyzes nantly expressing it/them. ketone reduction can be a carbonyl reductase in EC 1.1.1.- or I0086. In the described methods, an enzyme that catalyzes 1.1.99. The carbonyl reductase can include EC 1.1.1.184: EC reductive deamination can include an ammonia lyase. The 1.1.1.79; EC 1.1.1.B3: EC 1.1.1.B4 or 2-hydroxyglutaryl ammonia lyase can be in EC 4.3.1, e.g., EC 4.3.1.1; EC dehydogenases/alpha ketogluterate reductases from EC 1.1. 4.3.1.2: EC 4.3.1.3; EC 4.3.1.9; EC 4.3.1.12: EC 4.3.1.13: EC 99.2. or EC 1199.6 4.3.1.14, EC 4.3.1.23 or EC 4.3.1.24. 0094. In the described methods, an enzyme that catalyzes 0087. In the described methods, an enzyme that catalyzes C.-keto acid decarboxylation can be an O-keto acid decar 2-enoate reduction can include an enoate reductase. The boxylase in EC 4.1.1, e.g. 4.1.1.1; 4.1.1.7; 4.1.1.72; or an enoate reductase can be in EC 1.3.1, e.g., EC 1.3.1.8; EC acetolactate synthase in the class EC 2.2.1.6 or EC 2.2.2.6. 1.3.1.9; EC 1.3.1.10, such as the gene product of Fabl; EC 0095. In the described methods, the enzyme that catalyses 1.3.1.31; EC 1.3.1.38; EC 1.3.1.39: EC 1.3.1.44 such as the C.-amino acid decarboxylation can bean C.-amino acid decar gene product ofter or tdter (Nishimaki et al., J. Biochem. boxylase from the class EC 4.1.1.-. The O.-amino acid decar 1984, 95, 1315-1321; Shen et al., Appl. Environ. Microbiol., boxylase can include EC 4.1.1.11: EC 4.1.1.15; EC 4.1.1.16; 2011, 77(9), 2905-2915; Bond-Watts et al., Biochemistry, EC 4.1.1.18; EC 4.1.1.20, EC 4.1.1.45; and EC 4.1.1.86. 2012, 51, 6827-6837); or pimeloyl-CoA dehydrogenase in 0096. In the described methods, an enzyme that catalyzes EC 1.3.1.62 such as PimC/PimD or ThnJ/Think. The enoate reduction of the CoA ester of a dicarbylic acid to the corre reductase may also be in the EC 1.3, such as EC 1.3.8.1; EC sponding semialdehyde (of, for example, PCoA to PAS) can 1.3.99.3; EC 1.3.99.B10 or Syntrophus aciditrophicus 2,3- be a fatty-acyl-CoA reductase. The fatty-acyl-CoA reductase didehydropimelyl-CoA reductase and homologs thereof. The can be in EC 1.2.1.-, e.g., EC 1.2.1.3; EC 1.2.1.10: EC 1.2.1. enoate reductases may act on trans-2-enoyl-dicarboxylic 22: EC 1.2.1.50; EC 1.2.1.57 and EC 1.2.1.76. acids, or thioesters of dicarboxylic acids such as 2-heptene 0097. In the described methods, an enzyme that catalyses dioic acid-CoA (trans-2,3-didehydropimlyl-CoA) or 2-hep the reduction of the ACPI thioester to the corresponding tene dioic acid-acp (trans-2,3-didehydropimlyl-acp). semialdehyde (of, for example PACP to PAS) can be a fatty 0088. In the described methods, an enzyme that catalyzes acyl-CoA reductase acting on CoA esters and ACPI esters carboxylic acid reduction can include a carboxylic acid described above, or an acyl-acyl-carrier-protein reductase reductase. The carboxylic acid reductase can be in EC 1.2.99, from e.g. EC 1.2.1.80. e.g., EC 1.2.99.6. The carboxylic acid reductase may also 0098. In the described methods, an enzyme that catalyzes belong to the NAD-dependent non-acylating aldehyde dehy aldehyde dehydrogenation to the carboxylic acid can be an drogenases, such as ThnG in Sphingomonas and Cupriavidus aldehyde dehydrogenase or an aldehyde oxidase. The alde spp. hyde dehydrogenase can be in EC 1.2.1, e.g., EC 1.2.1.3; EC 0089. In the described methods, an enzyme that catalyzes 1.2.1.4 or EC 1.2.1.63. The aldehyde dehydrogenase/car semialdehyde amination can include a co-transaminase. The boxylic acid reductase may also be in EC 1.2.99, e.g., EC US 2015/0337275 A1 Nov. 26, 2015

1.2.99.6. The carboxylic acid reductase may also belong to production of 7 AHA. Another aspect of the invention is a the NAD-dependent non-acylating aldehyde dehydrogena nylon 7.7 produced by a process comprising polymerizing 1.7 ses, such as ThnG in Sphingomonas and Cupriavidus spp. DHA and PA, where the 1.7 DHA is made by any of the The aldehyde oxidase can be in EC 1.2.3, e.g. 1.2.3.1. methods described above that lead to the production of 1.7 0099. In the described methods, an enzyme that catalyzes DHA and the PA is made by any of the methods described the ring closure of 7-aminoheptanoic acid to enantholactam above that lead to the production of PA. can be an amidohydrolase of EC 3.5.2, e.g. 3.5.2.11. 0106 The present document also provides a substantially 0100. In the described methods, an enzyme that catalyzes pure culture of host cells, a substantial number of which dehydration (of, for example, AHP or AHP-CoA) can be a comprise and express one or more exogenous nucleic acids hydro-lyase. The hydro-lyase can be in EC 4.2.1, e.g., EC encoding one or more enzymes involved in the biosynthesis 4.2.1.2: EC 4.2.1.59; 4.2.1.61; 4.2.1.17 or 4.2.1.18. The dehy of one or more polymer monomers selected from 7 AHA: PA; dratase may also be 2-hydroxyglutaryl-CoA dehydratase 1.7 DAH; ENTL: and 7 HHA. These enzymes include fatty described in clostrida and fusobacteria to which an EC num acyl-CoA reductases, acyl-acp reductases, thioester hydro ber had not yet been assigned, that is expressed with its lases, ammonia lyases, enoate reductases, amino acid ami activator (HgdCAB) or a 2-hydroxyacyl CoA dehydratase of notransferases, CoA transferases, acid-thiol ligases, hydro anaerobic bacteria. lyases, carbonyl reductases, carboxylic acid reductases, 0101. In the described methods, enzymes that catalyze ()-transaminases, C.-keto acid decarboxylases, aldehyde C.-keto acid chain elongation can include one of more of the , deaminating dehydrogenases, alcohol set of enzymes including AkSA, AksD, Aksh, and Aksh. The dehydrogenases, aldehyde oxidases, alcohol oxidases, ami AkSA can be in EC 2.3.3, e.g., EC 2.3.3.13 or 2.3.3.14. The dohydrolases, decarboxylating acetolactate synthases and AksD can be in EC 4.2.1, e.g., EC 4.2.1.3.6. The Aksh can be enzymes catalyzing O-keto acid chain elongation Such as in EC 1.1.1, e.g., EC 1.1.1.87. One or more (e.g., two or more, AkSA, AksD, Aksh, and Aks. As used herein, a “substantial three or more, four or more, five or more, six or more, seven number is at least 10% (e.g., at least:20%; 30%; 40%; 50%: or more, eight or more, nine or more, ten or more, 12 or more, 60%; 70%; 75%:80%; 85%:90%; 95%: 97%; 98%: 99%; or 14 or more, 18, or more, or 20 or more) of the C.-keto acid even 100%) of the cells in the culture. chain elongation enzymes can be from methanogenic bacte 0107 The cells of the culture can be prokaryotic cells, 18 Such as, without limitation, the genus Escherichia Such as the 0102. In the described methods, an enzyme that catalyzes species Escherichia coli: the genus Clostridia Such as the alcohol dehydrogenation can include an alcohol dehydroge species Clostridium ljungdahli, Clostridium autoethanoge nase. The alcohol dehydrogenase can include EC. 1.1.1.1 or num or Clostridium kluyveri; the genus Corynebacteria Such EC 1.1.1.2 or 1.1.1.21. It can also be, for example, adhA from as the species Corynebacterium glutamicum; the genus Zymomonas mobilis, adhE from Zymomonas mobilis, butanol Cupriavidus Such as the species Cupriavidus necator or dehydrogenase from Clostridium acetobutylicum, Saccharo Cupriavidus metallidurans; the genus Pseudomonas such as myces ADHIV, and ADH6 from S. cerevisiae. In the the species Pseudomonas fluorescens, Pseudomonas putida described methods, the enzyme that catalyzes alcohol dehy or Pseudomonas Oleavorans; the genus Delfia Such as the drogenation can be an alcohol oxidase. The alcohol oxidase species Delftia acidovorans; the genus Bacillus such as the can be in class EC 1.1.3.13. species Bacillus subtilis; the genus Lactobacillus Such as the (0103 PCoA or PACP used in the methods of the document species Lactobacillus delbrueckii; or the genus Lactococcus can be derived from acetyl-CoA or benzoyl-CoA or from any Such as the species Lactococcus lactis. metabolic pathway Such as biotin biosynthesis, benzoate deg 0108. Alternatively, the cells of the culture can be eukary radation, cyclohexane carboxylate pathway, or condensation otic cells, e.g., fungal cells such as yeast cells. The yeast cells of malonyl-CoA olrfatty acid co-oxidation. The acetyl-CoA can be of a yeast or fungus such as, without limitation, the can be derived from renewable feedstocks comprising cellu genus Aspergillus such as the species Aspergillus niger, the losic feedstocks, Sugars, glycerol, or fatty acids. Moreover, genus Saccharomyces such as the species Saccharomyces the acetyl-CoA can be derived from SynGas, methane or cerevisiae; the genus Candida Such as the species C. tropica methanol. The benzoyl-CoA can be derived from polycyclic lis, C. albicans, C. cloacae, C. guillermondii, C. intermedia, aromatic hydrocarbons. In addition to being derived from C. maltosa, C. parapsilosis, and C. zeylenoides, the genus PCoA or PACP or PA, PAS can be derived from the tetralin Pichia Such as the species Pichia pastoris; the genus Yar degradation pathway. rowia Such as the species Yarrowia lipolytica; the genus 0104. The 2.6 DAP used in the methods of the document Issatchenkia Such as the species Issathenkia Orientalis; the can be produced by a lysine-producing organism lacking genus Debaryomyces such as the species Debaryomyces diaminopimelate decarboxylase. The 2,6 DAP is derived hansenii; the genus Arxula Such as the species Arxula from any fermentable carbon Source, including Sugars, glyc adenoinivorans; the genus Kluyveromyces such as the species erol, fatty acids, Syngas, methane or methanol. Kluyveromyces lactis; the genus Exophiala, the genus Mucor, 0105. The present document features a bioderived nylon the genus Trichoderma, the genus Cladosporium, the genus 7, nylon-7. X nylon-X.7, and polyester. It also includes a Phanerochaete, the genus Cladophialophora, the genus Nylon-7 produced by a process that includes polymerizing 7 Paecilomyces, the genus Scedosporium, or the genus Ophios AHA, where the 7 AHA is derived from PAS or AAS. Also to ina. provided by the document is a nylon-7.7 produced by a pro 0109 Exogenous enzymes that can be expressed by the cess that includes polymerizing PA and 1.7 DHA, where the cells are as follows. The ammonia lyases can include an PA is derived from PCoA: PACP; or 2 HDA and the 17 DHA ammonia lyase in EC 4.3.1, e.g., EC 4.3.1.1; EC 4.3.1.2: EC is derived from 7 AHA. The document also embodies nylon-7 4.3.1.3; EC 4.3.1.9; EC 4.3.1.12: EC 4.3.1.13: EC 4.3.1.14, produced by a process that includes polymerizing 7 AHA EC 4.3.1.23 or EC 4.3.1.24. The enoate reductases can made by any of the methods described above that lead to the include an enoate reductase in EC 1.3.1, e.g. EC 1.3.1.8; EC US 2015/0337275 A1 Nov. 26, 2015

1.3.1.9; EC 1.3.1.10, EC 1.3.1.31; EC 1.3.1.38; EC 1.3.1.39; EC number had not yet been assigned, that is expressed with EC 1.3.1.44 or pimeloyl-CoA dehydrogenase in EC 1.3.1.62 its activator (HgdCAB) or a 2-hydroxyacyl CoA dehydratase such as PimC/Piml) or ThnJ/Think. The enoate reductase of anaerobic bacteria. may also be in the EC 1.3, such as EC 1.3.8.1 or EC 1.3.99.3: 0115 The enzymes catalyzing O-keto acid chain elonga EC 1.3.99.B10 or Syntrophus aciditrophicus 2,3-didehydro tion can include one or more enzymes of the set including pimelyl-CoA reductase and homologs thereof. The carboxy AkSA, AksD, Aksh, and Aks enzymes. The AkSA can lic acid reductases include a carboxylic acid reductase in EC include an AkSA enzyme in EC 2.3.3, e.g., EC 2.3.3.13 or 1.2.99, e.g. EC 1.2.99.6. The carboxylic acid reductase may 2.3.3.14. The AksD can include an AksDenzyme in EC 4.2.1. also belong to the NAD-dependent non-acylating aldehyde e.g., EC 4.2.1.3.6. The Aksh can include an Aksh enzyme in dehydrogenases, such as ThnG in Sphingomonas and Cupria EC 1.1.1, e.g., EC 1.1.1.87. The alcohol dehydrogenases can vidus spp. The co-transaminases can include a transaminase in include, e.g., E.C. 1.1.1.1 or EC 1.1.1.2 or EC 1.1.1.21. In EC 2.6.1, e.g., EC 2.6.1.18; EC 2.6.1.19; EC 2.6.1.11: EC addition the alcohol dehydrogenases can include, for 2.6.1.13: EC 2.6.1.39 EC 2.6.1.48 or EC 2.6.1.62. The example, adhA from Zymomonas mobilis, adhB from can include a thioesterase in EC 3.1.2, e.g., Such Zymomonas mobilis, butanol dehydrogenase from as the gene product of YciA, tesB or Acot13; EC 3.1.2.2: EC Clostridium acetobutyllicum, Saccharomyces ADHIV, and 3.1.2.18; EC 3.1.2.19; or EC 3.1.2.20. Thioesterases acting ADH6 from S. cerevisiae. The alcohol oxidase can be in class on acyl-acp thioesters include EC 3.1.2.14, such as FatA, EC 1.1.3.13. FatB; and EC 3.1.2.21. The acid-thiol ligases or CoA synth 0116. Another aspect of the present document is an iso tases can include an acid-thiol ligase in EC 6.2.1, e.g., EC lated host cell comprising and expressing one or more exog 6.2.1.3; EC 6.2.1.5; EC 6.2.1.14; or EC 6.2.1.23. The CoA enous nucleic acids encoding one or exogenous nucleic acids transferases can include a CoA transferase in EC 2.8.3, e.g., involved in the biosynthesis of one or more polymer mono EC 2.8.3.12: EC 2.8.3.13 or EC 2.8.3.14, or or the gene mers selected from 7 AHA: PA; 17 DAH; ENTL, and 7HHA. product of ThnH that catalyses the reversible transfer of CoA The enzymes include ammonia lyases, enoate reductases, to pimelic acid. carboxylic acid reductases, ()-transaminases, thioesterases, 0110. The acyl-acp-thioesterase include a thioesterase in acid-thiol ligases or CoA synthtases, CoA transferases, acyl EC 3.1.2, e.g. 3.1.2.14 and EC 3.1.2.21. The acyl-acp syn acpthioesterases, acyl-acp synthetases, amino acid ami thetase can include an acid-thiol ligase in EC 6.2.1, e.g. EC notransferases, deaminating dehydrogenases, carbonyl 6.2.1.14, such as BioW; and EC 6.2.1.20. The aminoacid reductases, C.-keto acid decarboxylases, acetolactate Syn aminotransferases can include an aminotransferase in EC thases, C.-amino acid decarboxylases, fatty-acyl-CoA reduc 2.6.1, e.g., EC 2.6.1.39; EC 2.6.1.21; EC 2.6.1.42; and EC tases, acyl-acp-reductases, aldehyde dehydrogenases, alde 2.6.1.67. The deaminating dehydrogenase can include dehy hyde oxidases, amidohydrolases, hydro-lyases, and enzymes drogenases from EC 1.4.1, e.g. 1.4.1.2: EC 1.4.1.3; EC 1.4. catalyzing O-keto acid chain elongation. The cell and 1.4; and EC 1.4.1.16. The carbonyl reductases can include enzymes can be any of those described above for a substan reductases in EC.1.1.1.184 EC 1.1.1.79, EC 1.1.1.B3: EC tially pure culture of host cells. 1.1.1.B4 or 2-hydroxyglutaryl dehydogenases/alpha ketoglu 0117. Other objects and advantages will become apparent terate reductases from EC 1.1.99.2. or EC 1.1.99.6 to those skilled in the art from a consideration of the ensuing 0111. The C-keto acid decarboxylase can include C.-keto detailed descriptions set forth herein. acid decarboxylases in EC 4.1.1, e.g. 4.1.1.1; 4.1.1.7; 4.1.1. 72; or an acetolactate synthase in the class EC 2.2.1.6. The BRIEF DESCRIPTION OF THE DRAWINGS C.-amino acid decarboxylase can include EC 4.1.1.11: EC 0118 FIG. 1 is a schematic showing the biotransformation 4.1.1.15; EC 4.1.1.16; EC 4.1.1.18; EC 4.1.1.20, EC 4.1.1.45: of pimelic acid semialdehyde, to various difunctional C7 and EC 4.1.1.86. alkanes useful for nylon 7, nylon 7.7 and nylon 7.x produc 0112 The fatty-acyl-CoA reductases can include a fatty tion, as well as polyesters. acyl-CoA reductase in EC 1.2.1, e.g. EC 1.2.1.3; EC 1.2.1.10: 0119 FIG. 2 is a schematic of the formation of pimelic EC 1.2.1.22; EC 1.2.1.50; and EC 1.2.1.76. The acyl-acp acid semialdehyde from pimeloyl-CoA, pimeloyl-acp, or reductase can include acyl-acyl-carrier-protein reductase C.-ketosuberate, all of which are intermediates in naturally from e.g. EC 1.2.1.80. occurring pathways as indicated, and from tetralin degrada 0113. The aldehyde dehydrogenases can include an alde tion. hyde dehydrogenase in EC 1.2.1, e.g., EC 1.2.1.3; EC 1.2.1.4 0120 FIG. 3 is a schematic of the series of reactions in or EC 1.2.1.63. The aldehyde dehydrogenase/carboxylic acid C.-ketoacid chain elongation. Three basic reactions are reductase may also be in EC 1.2.99, e.g., EC 1.2.99.6. The involved in chain elongation: a condensation between an carboxylic acid reductase may also belong to the NAD-de C.-ketodiacid and acetyl-CoA to form a citrate homolog, pendent non-acylating aldehyde dehydrogenases, such as isomerization of the citrate homolog to an isocitrate homolog, ThnG in Sphingomonas and Cupriavidus spp. The aldehyde and oxidative decarboxylation of the isocitrate homolog to an dehydrogenase/carboxylic acid reductase may also belong to C.-ketodiacid containing an additional methylene group the NAD-dependent non-acylating aldehyde dehydrogena Howel198: Howell DM, Harich K, Xu H, White RH (1998). ses, such as ThnG in Sphingomonas and Cupriavidus spp. Alpha-keto acid chain elongation reactions involved in the The aldehyde oxidase can be in EC 1.2.3, e.g. 1.2.3.1. The biosynthesis of coenzyme B (7-mercaptoheptanoyl threonine amidohydrolase can be an amidohydrolase from EC 3.5.2. phosphate) in methanogenic Archaea.” Biochemistry, 37(28); e.g. 3.5.2.11. 10108-17. Successive rounds of chain elongation (3 times) 0114. The hydro-lyases include a hydro-lyase in EC 4.2.1. from C.-ketogluterate to O-ketosuberate is followed by C.-keto e.g., EC 4.2.1.2: EC 4.2.1.59; 4.2.1.61; 4.2.1.17 or 4.2.1.18. acid decarboxylation to pimelic acid semialdehyde. These The dehydratase may also be 2-hydroxyglutaryl-CoA dehy reactions are analogous to the reactions catalyzed by citrate dratase described in clostrida and fusobacteria to which an synthase, aconitase, and isocitrate dehydrogenase in the TCA US 2015/0337275 A1 Nov. 26, 2015 cycle I (prokaryotic) and the reactions catalyzed by (R)- quently behydrolysed by a transferase or acid-thiolligase, for homocitrate synthase, homoaconitase and (-)-threo-iso recycling of CoA or acp, or by a thioesterase to produce homocitrate dehydrogenase in lysine biosynthesis VHowell, pimelic acid. 1998, supra. I0129 FIG. 11A shows an SDS-PAGE analysis on soluble 0121 FIG. 4 is a schematic of the formation of pimeloyl and insoluble fractions from expression of the Citrobacter CoA through condensation of three malonyl-CoAS. amalomaticus ammonia lyase protein. 0122 FIG. 5 is a schematic of the formation of pimelic I0130 FIG. 11B shows an SDS-PAGE analysis on soluble acid by oxidative cleavage of the acp thioesters of long and insoluble fractions from expression of the Clostridium chain fatty acids by Biol from Bacillus subtilis. Pimelic acid. tetanomorphum (I5) and Aspergillus Oryzae (16) ammonia or its CoA or acp thioesters are precursors in 7-keto-8- lyase proteins. aminopelargonate biosynthesis in biotin biosynthesis path I0131 FIG. 12 shows an SDS-PAGE analysis on soluble way II.7-keto-8-aminopelargonate is the key intermediate for and insoluble fractions from expression of the Nocardia sp. biotinbiosynthesis. Degradation of pimeloyl-CoA in the host carboxylic acid reductase gene. organism can be prevented by deletion or inactivation of I0132 FIG. 13 shows the change in absorbance over time BioF, the gene encoding 7-keto-8-aminopelargonic acid syn due to reduction of pimelic acid, adipic acid and benzoic acid thetase (E.C. 2.3.1.47). to pimelic acid semialdehyde, adipic acid semialdehyde and 0123 FIG. 6 is a schematic of the formation of pimeloyl benzaldehyde by Nocardia CAR reductase activity. Benzoic acp or its methyl ester in 7-keto-8-aminopelargonate bio acid was used as positive control and reaction mixtures with synthesis in biotin biosynthesis pathway I. The key interme biocatalyst harbouring the empty vector as negative control. diate in biotin biosynthesis. Pimelic acid or pimelic acid (0.133 FIG. 14 shows an SDS-PAGE analysis on soluble semialdehyde can be prepared from these precursors by the fraction and insoluble fraction from expression of the Hae action of aaaS (acyl-ACP synthetase), FatA (fatty acyl-ACP mophilus influenzae YciA thioesterase protein (I8) thioesteraseA). FatB (fatty acyl-ACP thioesterase B), or acyl I0134 FIG. 15 shows an SDS-PAGE analysis of the ami acp reductase as shown in bold text. notransferase IlvE-Omega Vffusion (plNG2022) and IlvE 0.124 FIG. 7 is a scheme showing formation of pimeloyl Adoptimized omega fusion (pNG2030) proteins. CoA from benzoate or cyclohexane carboxylate degradation. 0.135 FIG. 16 shows Results of the biotransformation of 0.125 FIG. 8 is a scheme showing the formation of pimelic Sodium pyruvate and 7-aminoheptanoic acid with the mea acid semialdehyde in the tetralin degradation pathway in Sph Sured concentrations of 7-aminoheptanoic acid substrate and ingomonas and Corynebacterium spp. L-alanine product. 0126 FIG. 9 is a scheme showing formation of pimeloyl (0.136 FIG. 17 shows a Summary of the HPLC results from CoA in the cyclohexane carboxylate pathway from crotonate the L- and D-2-aminoSuberic acid enantiomer concentrations by methanogens. See Mouttaki et al., Applied and Environ in the six biotransformations after 48 and 96 hours. mental Microbiology, pages 930-938 (2001). I0137 FIG. 18 shows an SDS-PAGE analysis on soluble 0127 FIG. 10A is a schematic showing examples of engi and insoluble fractions from the bead lysis of expression of neered pathways to pimelic acid semialdehyde and 7-amino the E. coli GadA (I29), E. coli LySA (I30), E. coli GadA ilvE heptanoic acid from D.L-diaminopimelate as a starting point (I31) and E. coli LySA ilvE (I32) proteins. via 2-aminoheptanedioic acid. There are 2 routes from 2-ami 0.138 FIG. 19 shows the formation of 7-aminoheptanoic noheptanedioic acid to pimelic acid semialdehyde, either via acid from the decarboxylation of 2-aminosuberate. reductive deamination by an ammonia lyase to 2-heptene 0.139 FIG. 20 shows protein alignment using the Clust dioic acid or via formation of the O.-keto-acid that can Subse alW algorithm showing significant homology between quently be converted to pimelic acid or its CoA ester. Alter natively, C.-keto-pimelate derived from 2-aminoheptanedioic MJ0277 and comD (SEQID 41)/comE (SEQID 42) proteins acid can undergo one round of chain elongation to O-keto (O140 FIG. 21 shows homology of MJ0277 to LACLA suberate, which can either be decarboxylated to form pimelic acetolactate synthase from Lactococcus lactis (SEQID 43). acid semialdehyde, or converted to C.-aminoSuberate, which 0141 FIG.22 shows an analysis of the functional domains will lead directly to 7-aminoheptanoic acid upon decarboxy in LACLA acetolactate synthase, comD/comE from Metha lation. C.-keto-pimelate also can be derived from C-keto-adi nocella paludicola and the hypothetical acetolactate synthase pate, which in turn can be derived from C-keto-glutarate, MJ0277 showed that all of them have a similar structure through two Successive rounds of C-keto-acid chain elonga containing a thiamine phosphate binding site. tion. 0.142 FIG. 23 depicts a map of plNG2022 vector, contain 0128 FIG. 10B is a schematic diagram showing the con ing the IlvE-Omega Vffusion gene. version of either C-hydroxypimelate or its CoA ester to 0.143 FIG. 24 depicts a map of plNG2030 vector contain 2-heptenedioic acid or its corresponding CoA ester, C-hy ing the IlvE-Adoptimized omega fusion gene. droxypimeloyl-CoA, respectively, by a hydro-lyase. Simi 014.4 FIG. 25 sets forth the SEQ. ID 1-40 larly, 2-heptenedioic acid or its corresponding CoA ester can 0145 FIG. 26 depicts a bar graph of the formation of be reduced to Pimelic acid or pimeoyl-CoA, respectively, by pimelic acid from pimeloyl-CoA by the acid thiol ligase (EC an enoate reductases that acts on the free acid or the CoA 6.2.1.5) of Thermococcus kodakaraensis (41/42), the CoA activated thioester. In a similar fashion, the activation of transferase (EC 2.8.3.12) of Acidominococcus fermentans C-hydroxypimelate or 2-heptenedioic acid may not be to the (43/44) and the acid-thiol ligase (EC 6.2.1.14) from CoA thioester, but instead to the acp thioester by a acyl Pseudomonas mendocina (45). acp-synthetase. The hydro-lyase may catalyse the elimina 0146 FIG. 27 depicts the formation of adipic acid from tion of water from C-hydroxypimeloy-acp, and the enoate trans-2-hexenedioic (A) and the formation of adipoyl-CoA reductases may reduce the double bond of 2-heptenedioic from trans-2,3-didehydroadipoyl-CoA by selected enoate acid-acp. Pimeloyl-CoA or pimeloyl-acp may Subse reductases. US 2015/0337275 A1 Nov. 26, 2015

DETAILED DESCRIPTION OF THE INVENTION 7-dioic acid (also known as pimelic acid or pimelate) with a diamine. The integer X indicates the number of carbon atoms 0147 In general, this disclosure relates to methods and in the diamine with which the heptane-1,7-dioic acid is materials for biosynthetically producing di- or trifunctional reacted. In some embodiments, the integer X is an integer C7 alkanes that are useful as intermediates in the production greater than 3, e.g., greater than 5, greater than 7, or greater of nylon-7, nylon-7.X and nylon-X.7, as well as the production than 9. For example, nylon-5.7 is produced by reacting hep of polyesters. tane-1,7-dioic acid and 1,5-pentanediamine Nylon 7.7 is pro 0148. In some embodiments, one or more (e.g., two or duced by reacting heptane-1,7-dioic acid and 1.7 diamino more; three or more; four or more; five or more; six or more; heptane. seven or more; eight or more; nine or more; ten or more; 0152 The term “nylon-7.x” refers to the family of polya eleven or more; twelve or more; thirteen or more; fourteen or mides that are produced by the polymerization reaction of more; fifteen or more; sixteen or more; seventeen or more; heptamethylenediamine with a . The integer eighteen or more; nineteen or more; twenty or more; or even x indicates the number of carbon atoms in the dicarboxylic more) isolated enzymes (e.g., an ammonia lyase, enoate acid with which the heptamethylenediamine is reacted. For reductase, aminotransferase, CoA transferase, thioestest example, nylon-7.5 is produced by the reaction of heptam erase, acid-thiol ligase, hydro-lyase, carbonyl reductase, ethylenediamine (also known as 1,7-diaminoheptane or 1.7- thioesterase, carboxylic acid reductase, fatty-acyl CoA heptanediamine) and pentane-1,5-dioic acid. In some reductase, ()-transaminase, C.-keto acid decarboxylase, embodiments, the integer X is an integer greater than 3, e.g., enzymes catalyzing O-keto acid chain elongation, acyl-ACP greater than 5, greater than 7 or greater than 9. synthetase, Fat A (fatty acyl-ACP thioesterase A), FatB (fatty 0153. The term “polyester” refers to a category of poly acyl-ACP thioesterase B), or acyl-acp reductase) can be mers which contains an ester functional group in their main used to biosynthetically produce the di- or trifunctional C7 chain. Polyesters can be homopolymers, for example polyca alkanes. Such enzymes can be isolated from recombinant prolactone, that is produced by ring-opening condensation of cells expressing exogenous nucleic acids encoding the . Copolymer polyesters are formed by the esteri enzyme(s) or non-recombinant cells naturally expressing the fication condensation of polyfunctional alcohols and acids. enzymes. For example, adipic acid and ethylene glycol are used to 0149. In some embodiments, a recombinant host that produce polyethylene adipate. Pimelic acid can be used in the includes one or more (e.g., two or more; three or more; four or place of adipic acid in Such copolymers. Alternatively, 7-hy more; five or more; six or more; seven or more; eight or more; nine or more; ten or more; eleven or more; twelve or more; droxyheptanoic acid can be used to produce polyenantholac thirteen or more; fourteen or more; fifteen or more; sixteen or tOne. more; seventeen or more; eighteen or more; nineteen or more; 0154 The foregoing and other aspects of the present twenty or more; or even more) exogenous nucleic acids invention will now be described in more detail with respect to encoding one or more (e.g., two or more; three or more; four embodiments described herein. It should be appreciated that or more; five or more; six or more; seven or more; eight or the invention can be embodied in different forms and should more; nine or more; ten or more; eleven or more; twelve or not be construed as limited to the embodiments set forth more; thirteen or more; fourteen or more; fifteen or more; herein. Rather, these embodiments are provided so that this sixteen or more; seventeen or more; eighteen or more; nine disclosure will be thorough and complete, and will convey the teen or more; twenty or more; or even more) enzymes (e.g., scope of the invention to those skilled in the art. one or more (as above) of an ammonia lyase, enoate reduc 11 DEFINITIONS tase, aminotransferase, CoA transferase, thioestesterase, acid-thiol ligase, hydro-lyase, carbonyl reductase, 0155 The terminology used in the description of the thioesterase, carboxylic acid reductase. ThnG, fatty-acyl invention herein is for the purpose of describing particular CoA reductase, co-transaminase, C.-keto acid decarboxylase, embodiments only and is not intended to be limiting of the C.-aminoacid decarboxylase, acetolactate synthase, amidohy invention. As used in the description of the embodiments of drolase, enzymes catalyzing O-keto acid chain elongation, the invention and the appended claims, the singular forms 'a. acyl-acp-thioesterase, acyl-acp synthetase, BioW, alcohol “an and “the are intended to include the plural forms as dehydrogenase, alcohol oxidase, aldehyde dehydrogenase, well, unless the context clearly indicates otherwise. Also, as aldehyde oxidase, FatA, FatB, or acyl-acp reductase), or a used herein, “and/or refers to, and encompasses, any and all lysate prepared from Such a recombinant host or a non-re possible combinations of one or more of the associated listed combinant cell naturally expressing the enzyme(s) can be items. Unless otherwise defined, all terms, including techni used to biosynthetically produce the di- or trifunctional C7 cal and Scientific terms used in the description, have the same alkanes. A recombinant host also may have a deficiency in meaning as commonly understood by one of ordinary skill in one or more endogenous enzymes in order to divert metabolic the art to which this invention belongs. intermediates towards the production of the di- or trifunc 0156. As used herein, the term “enzyme” or “enzymes s tional C7 alkanes. refers to a protein catalyst capable of catalyzing a reaction. 0150. As described herein, the biosynthetically produced Herein, the term does not mean only an isolated enzyme, but di- or trifunctional C7 alkanes can be used to produce nylon also includes a host cell expressing that enzyme. Accordingly, 7, nylon-7.X, or nylon-X.7, as well as polyesters. The term the conversion of A to B by enzyme C should also be con “nylon-7 is a polyamide produced by polymerization of the strued to encompass the conversion of A to B by a host cell monomer 7-aminoheptanoic acid or by ring-opening conden expressing enzyme C. sation enantholactam. Nylon-7 also is known as "polyenan 0157. As used herein, the term “difunctional C7 alkane s thamide' or “polyheptanamide.” refers to straight or branched, saturated or unsaturated, hydro 0151. The term “nylon-X.7” refers to the family of polya carbon having seven carbonatoms and two functional groups. mides that are produced by the polymerization of heptane-1, The functional groups may be positioned at any point along US 2015/0337275 A1 Nov. 26, 2015

the hydrocarbon chain. In some embodiments, the functional 0.165. In some embodiments with a trifunctional C7 groups are located at the terminal ends of the hydrocarbon alkane, two of the functional groups are carboxylic acid chain (e.g., one functional group at each terminus). The term groups and one functional group is a keto group or a hydroxyl is also intended to encompass cyclic compounds such as group or an amino group. enantholactam. Exemplary difunctional C7 alkanes include, 0166 The term "heterologous is used herein to refer to but are not limited to, pimeloyl-CoA, pimeloyl-acp, hep any nucleic acid or polypeptide that is not derived from a cell tane-1,7-dioic acid (pimelic acid or pimelate), 7-aminohep of the same species as the host cell. Accordingly, as used tanoic acid, 7-hydroxyheptanoic acid, heptamethylenedi herein, "homologous nucleic acids or proteins are those that amine, 1.7-heptanediol, 7-aminoheptanol, 7-aminoheptanal, occur in, or are produced by, a cell of the same species as the enantholactam, pimelic acid semialdehyde (also known as host cell. 7-oxoheptanoate), 2-heptenediacid, and 2-heptene diacid 0167. The term “exogenous” as used herein with reference CoA. to nucleic acid and a particular host cell refers to any nucleic 0158. As used herein, the terms “-oate' and “-oic acid”, acid that does not occur in (and cannot be obtained from) that for example heptanoate and heptanoic acid, are used inter particular cell as found in nature. Thus, a non-naturally-oc changeably. Moreover, it should be understood that "-ate'. curring nucleic acid is considered to be exogenous to a host “-oate' and “-oic acid can be used interchangeably through cell once introduced into the host cell. It is important to note out to refer to the compound in any of its neutral or ionized that non-naturally-occurring nucleic acids can contain forms, and so-called “Zwitterionic' forms. nucleic acid Subsequences or fragments of nucleic acid 0159. As used herein, the term “trifunctional C7 alkane' sequences that are found in nature provided that the nucleic refers to straight or branched, saturated or unsaturated, hydro acid as a whole does not exist in nature. For example, a carbon having seven carbon atoms and three functional nucleic acid molecule containing a genomic DNA sequence groups. The functional groups may be positioned at any point within an expression vector is non-naturally-occurring along the hydrocarbon chain. In some embodiments, two of nucleic acid, and thus is exogenous to a host cell once intro the three functional groups are located at the terminal ends of duced into the host cell, since that nucleic acid molecule as a the hydrocarbon (e.g., one functional group at each terminus). whole (genomic DNA plus vector DNA) does not exist in Exemplary trifunctional C7 alkanes include, but are not lim nature. Thus, any vector, autonomously replicating plasmid, ited to, 2-oxopimelate (C.-ketopimelate), 2-aminopimelate or virus (e.g., retrovirus, adenovirus, or herpesvirus) that as a (C-aminopimelate), 2-hydroxypimelate (C-hydroxypime whole does not exist in nature is considered to be non-natu late), and 2-hydroxypimelate-CoA (C-hydroxypimelate rally-occurring nucleic acid. It follows that genomic DNA CoA). fragments produced by PCR or restriction endonuclease 0160. As used herein, the term “functional group' refers to treatment as well as cDNAs are considered to be non-natu an amine group, a carboxylic acid group, an aldehyde group, rally-occurring nucleic acid since they exist as separate mol an alcohol group, a co-enzyme A group or a keto group. The ecules not found in nature. It also follows that any nucleic acid term “amine group' refers to the radical —NH2. The term containing a promoter sequence and polypeptide-encoding “carboxylic acid group' refers to the radical —COOH. The sequence (e.g., cDNA or genomic DNA) in an arrangement term "aldehyde group’ refers to the radical —C(O)H. The not found in nature is non-naturally-occurring nucleic acid. A term “alcohol group' refers to the radical —OH. The term nucleic acid that is naturally-occurring can be exogenous to a “keto group' refers to the radical C(O). particular cell. For example, an entire isolated 0161 The term “co-enzyme A group' refers to the organic from a cell of yeast X is an exogenous nucleic acid with cofactor, or prosthetic group (nonprotein portion of an respect to a cell of yeast y once that chromosome is intro enzyme), whose presence is required for the activity of many duced into a cell of yeasty. enzymes (the apoenzyme) to form an active enzyme system. 0.168. It will be clear from the above that “exogenous’ Coenzyme A functions in certain condensing enzymes, acts in nucleic acids can be “homologous' or "heterologous nucleic acetyl or other acyl group transfer and in fatty acid synthesis acids. In contrast, the term "endogenous” as used herein with and oxidation, pyruvate oxidation and in other acetylation. reference to nucleic acids or (or proteins encoded by the 0162 The term “acp' or “acyl carrier protein’ refers to a nucleic acids or genes) and a particular cell refers to any protein (which typically is active in fatty acid synthesis) that nucleic acid or gene that does occur in (and can be obtained picks up acetyl and malonyl groups from acetyl coenzyme A from) that particular cell as found in nature. and malonyl coenzyme A and links them by condensation to (0169. As used herein, the term “recombinant host” refer to form B-keto acid acyl carrier protein, releasing carbon diox a host, the genome of which has been augmented by at least ide and the sulfhydryl form of the acyl carrier protein. exogenous nucleic acid sequence. Such nucleic acid 0163 The functional groups on a di- or trifunctional C7 sequences include, but are not limited to, nucleic acids that alkane can be the same or different. For example, in some are not naturally present in the host (i.e., a heterologous embodiments with a difunctional C7 alkane, both functional nucleic acid), DNA sequences that are not normally tran groups can be amine groups, both functional groups can be scribed into RNA or translated into a protein (“expressed’), carboxylic acid groups, or both functional groups can be and other nucleic acid sequences which one desires to intro alcohol groups. duce into the non-recombinant host (e.g., a regulatory region 0164. In some embodiments, one of the functional groups to alter expression of a particular nucleic acid sequence). It on a difunctional C7 alkane is an amine group and the other is will be appreciated that typically the genome of a recombi a carboxylic acid group. In some embodiments, one of the nant host described herein is augmented through the stable functional groups on a difunctional C7 alkane is an alcohol introduction of one or more exogenous nucleic acids. Gener group and the other is a carboxylic acid group. In some ally, the exogenous nucleic acid is not originally resident in embodiments, one of the functional groups on a difunctional the host that is the recipient of the DNA (i.e., it is a heterolo C7 alkane is an alcohol group and the other is an amine group. gous nucleic acid), but it is within the scope of the invention US 2015/0337275 A1 Nov. 26, 2015 to isolate a nucleic acid from a given host, and to Subsequently alkanes in the presence of one or more isolated enzymes in the introduce one or more additional copies of that nucleic acid presence of a recombinant host cell expressing that/those into the same host, e.g., to enhance production of an encoded enzyme(s), or in the presence of a cell lysate (or a partially product or alter the expression pattern of a nucleic acid. In purified lysate) of cell (e.g., recombinant cell) that expresses Some instances, the exogenous nucleic acid will modify or the enzyme(s). In some embodiments, the production of the even replace an endogenous gene or DNA sequence by, e.g., di- or trifunctional C7 alkanes starts with pimelate or homologous recombination or site-directed mutagenesis. pimeloyl-CoA or pimeloyl-acp and proceeds through a Modification or replacement of an endogenous gene can common intermediate, namely, pimelic acid semialdehyde. result in a deficiency in a particular encoded product, e.g., an Specifically, in some embodiments, the invention relates to a enzyme. Suitable recombinant hosts are described below. method of producing the difunctional C7 alkane 7-aminohep Recombinant hosts can also contain nucleic acids that are not tanoic acid by using an enzyme that catalyzes the semialde incorporated into the genome of the host cell but exist (and hyde amination of pimelic acid semialdehyde to 7-aminohep preferably replicate) episomally in the cell. tanoic acid. See FIG. 1. A semialdehyde aminotransferase (0170. As used herein, a “promoter” refers to a DNA Such as a w-transanimase can be used. Exemplary ()-tran sequence that enables a gene to be transcribed. The promoter saminases include, but are not limited to transaminases clas is recognized by RNA polymerase, which then initiates tran sified under EC 2.6.1 such as EC 2.6.1.18; EC 2.6.1.19; EC Scription. Thus, a promoter contains a DNA sequence that is 2.6.1.1; EC 2.6.1.13; EC 2.6.1.48; and EC 2.6.1.62. In some either bound directly by, or is involved in the recruitment, of embodiments, the aminotransferase is B-alanine aminotrans RNA polymerase. A promoter sequence can also include ferase classified in EC 2.6.1.18 from V fluvialis, B. weihen "enhancer regions.” which are one or more regions of DNA Stephanensis, P. aureginosa, B. subtilis, or P. syringae. See that can be bound with proteins (namely, the trans-acting WO2011/031147, which is incorporated herein in its entirety. factors, much like a set of transcription factors) to enhance 0176). In some embodiments, 7-aminoheptanal can be pro transcription levels of genes (hence the name) in a gene duced using an enzyme that catalyzes the aldehyde dehydro cluster. An enhancer, while typically at the 5' end of a coding genation of 7-aminoheptanoic acid to 7-aminoheptanal. See, region, can also be separate from a promoter sequence and FIG. 1. An aldehyde dehydrogenases classified under EC can be, e.g., within an intronic region of a gene or 3' to the 1.2.1, such as EC 1.2.1.4 or EC 1.2.1.63, can be used to coding region of the gene. produce 7-aminoheptanal from 7-aminoheptanoic acid. The 0171 As used herein, “operably linked' means incorpo aldehyde dehydrogenase/carboxylic acid reductase may also rated into a genetic construct so that expression control be in EC 1.2.99, e.g., EC 1.2.99.6. The carboxylic acid reduc sequences effectively control expression of a coding tase may also belong to the NAD-dependent non-acylating sequence of interest. aldehyde dehydrogenases, such as ThnG in Sphingomonas 12 FEEDSTOCKS and Cupriavidus spp. 0177. In some embodiments, heptamethylenediamine 0172 A variety of different feedstocks can be used to (also known as 1,7-diaminoheptane) can be produced from produce di- or trifunctional alkanes. In some embodiments, a 7-aminoheptanal using an enzyme that catalyzes the transfer recombinant host cell, or a lysate prepared from a cell (e.g., a of an amino group. See FIG. 1. Suitable aminotransferases recombinant host cell), is contacted with a renewable feed include, but are not limited to, aminotransferases in EC 2.6.1 stock to produce a di- or trifunctional C7 alkane, such as such as EC 2.6.1.18; EC 2.6.1.19; EC 2.6.1.11: EC 2.6.1.13; 7-aminoheptanoic acid. Examples of renewable feedstocks EC 2.6.1.39 EC 2.6.1.48; and EC 2.6.1.62. In some embodi which may be used as a carbon Source include, but are not ments, the aminotransferase is B-alanine aminotransferase limited to, cellulosic feedstocks, plant-derived carbohydrates classified in EC 2.6.1.18 from V fluvialis, B. weihenstephan and fatty acids. In some embodiments, the feedstock with ensis, P. aureginosa, B. subtilis, or Psyringae. See WO2011/ which the recombinant host cell is contacted is glucose, 031147, which is incorporated herein in its entirety. Sucrose, Xylose, fatty acids or glycerol. 0178. In some embodiments, enantholactam can be pro 0173. In addition to renewable feedstocks, such as those duced from 7-aminoheptanoic acid using an enzyme that listed above, the recombinant host cell can be modified for growth on synthesis gas, also known as SynGas, as a source of catalyzes the amide hydrolysis of 7-aminoheptanoic acid. carbon. In this specific embodiment, the recombinant host See, FIG. 1. Suitable amidohydrolases include amidotrans cells of the invention are engineered to provide an efficient ferases classified under EC 3.5.2 such as EC 3.5.2.12 or EC metabolic pathway for utilization of SynGas, or other gaseous 3.5.2.11 can be used. carbon Sources, to produce di- or trifunctional C7 alkanes. 0179. In some embodiments, 7-hydroxyheptanoate is pro 0.174. Additionally, recombinant hosts described herein, duced from pimelic acid semialdehyde or 7-aminoheptanol is or a lysate prepared from a recombinant host cell, can be produced from 7-aminoheptanal using an enzyme that cata contacted with an aromatic hydrocarbon feedstock. The use lyzes the reduction of an aldehyde. See FIG. 1. For example, of aromatic hydrocarbon feedstocks, rather than renewable an alcohol dehydrogenase classified under EC 1.1.1 Such as feedstocks, provides away to convert Such petroleum derived EC 1.1.1.2 or EC 1.1.1.21 can be used. It will be understood materials to compounds that are more benign to the environ by those skilled in the art that a great many alcohol dehydro ment, reducing environmental impact. Examples of Suitable genases with broad Substrate ranges exist, and that other aromatic hydrocarbon feedstocks include, but are not limited dehydrogenases from EC 1.1.1.-, such as 1.1.1.B3 or EC 1.1.1.B4, or 1.1.1.79, or in EC 1.1.99, such as 1.1.99.2 or EC to, toluene, benzene, benzoic acid and shikimate. 1.1.99.6, may also be used to reduce an aldehyde to the 13. ROUTES TO DIFUNCTIONAL C7 ALKANES corresponding alcohol. Alternatively, an alcohol oxidase can COMPRISING ENZYME-CATALYZED STEPS be used. The alcohol oxidase can be in class EC 1.1.3.13. 0.175. As stated above, embodiments of the present inven 0180. In some embodiments, 1.7 heptanediol can be pro tion relate to methods for producing di- or trifunctional C7 duced from 7-hydroxyheptanoate using an aldehyde dehy US 2015/0337275 A1 Nov. 26, 2015

drogenase and an alcohol dehydrogenase. A Suitable alde ferases include CoA transferases classified under EC 2.8.3 hyde dehydrogenase can be classified under EC 1.2.1 (e.g., such as EC 2.8.3.6, EC 2.8.3.8, EC 2.8.3.12: EC 2.8.3.13 or EC 1.2.1.4 or EC 1.2.1.63 or EC 1.1.1.21). See, FIG. 1. A EC 2.8.3.14, or the gene product of Thnh that catalyses the suitable alcohol dehydrogenase can be classified under E.C. reversible transfer of CoA to pimelic acid. 1.1.1. Such as EC 1.1.1.1 or EC 1.1.1.2. It will be understood 0.184 Heptane-1,7-dioic acid can be obtained from by those skilled in the art that a great many alcohol dehydro pimeloyl-acp by the use of enzymes acting on acp genases with broad Substrate ranges exist, and that other thioesters, such as acyl-acpthioesterases in EC 3.1.2.14. dehydrogenases from EC 1.1.1.-, such as 1.1.1.B3 or EC such as fatA and fat3 or EC 3.1.2.21. Alternatively, acid-thiol 1.1.1.B4, or 1.1.1.79, or in EC 1.1.99, such as 1.1.99.2 or EC ligases acting on acp thioesters can be used. Such as EC 1.1.99.6, may also be used to reduce an aldehyde to the 6.2.1.14 and EC 6.2.1.20. corresponding alcohol. Alternatively, an alcohol oxidase can 0185. Alternatively, pimelic acid can be obtained form be used. The alcohol oxidase can be in class EC 1.1.3.13. pimeloyl-acpmethyl ester by hydrolysis of the ester bond by Similarly, a vast number of aldehyde dehydrogenases or car AaasaS (acyl-ACP synthetase), followed by removal of the boxylic acid reductases with broad substrate specificities methyl group by a lipase/esterase. exist. Suitable aldehyde dehydrogenases include EC 1.2.1.3: 0186 Pimelate can be transformed to pimelic acid semi EC 1.2.1.4; and EC 1.2.1.63. The aldehyde dehydrogenase/ aldehyde using an enzyme that catalyzes the reduction of a carboxylic acid reductase may also be in EC 1.2.99, e.g., EC carboxylic acid. Suitable reductases include, but are not lim 1.2.99.6. The aldehyde dehydrogenase/carboxylic acid ited to, carboxylic acid reductases classified in EC 1.2.99; reductase may also belong to the NAD-dependent non-acy The carboxylic acid reductase may also belong to the NAD lating aldehyde dehydrogenases, such as ThnG in Sphin dependent non-acylating aldehyde dehydrogenases, such as gomonas and Cupriavidus spp. Alternatively, and aldehyde ThnG in Sphingomonas and Cupriavidus spp. or fatty acyl oxidase may be used. Such as an aldehyde oxidase from EC CoA reductases classified in EC 1.2.1 such as EC 1.2.1.3; EC 1.2.3.1. 1.2.1.10: EC 1.2.1.22; EC 1.2.1.50; EC1.2.1.57; and EC 1.2. 0181. In embodiments in which the C7 di- or trifunctional 1.76. In some embodiments, a fatty-acyl-CoA reductase can alkanes may be toxic to the host (e.g., semialdehydes such as directly convert pimeloyl-CoA or pimeloyl-acp to pimelic 7-oxoheptanoate or 7-aminoheptanal), the conversion to the acid semialdehyde. Exemplary carbonyl reductases in EC toxic compound can be performed in vitro using isolated 1.2.99 include, but are not limited to EC 1.2.99.6 from Nor enzymes or a lysate from the recombinant host. In some cardia sp. (Aimin et al., Appl. Environ. Microbiol. 70: 1874 embodiments, the toxic compound is produced by the host 1881 (2004), incorporated by reference) or S. griseus (Suzuki and then Subsequently converted (using isolated enzymes, a et al., J. Antibiot. (Tokyo) 60:380-387 (2007) incorporated by recombinant host expressing an enzyme, or chemically con reference). Exemplary fatty-acyl-CoA reductases in EC 1.2.1 Verted) to enantholactam. include, but are not limited to EC 1.2.1.75 from, e.g., C. aurantiacus (Hugler, M. et al., J. Bacteriology 184: 2404 1.5 BIOSYNTHETIC PATHWAYS TO PIMELATE 2410 (2002), incorporated by reference), M. sedula (Kock AND/OR PIMELOYL-COA elkorn, D. and Fuchs, G., J. Bacteriology 191: 6352-6362 0182 Pimelate and/orpimeloyl-CoA can be obtained via a (2009), incorporated by reference), or S. tokodai (Alber, B. et number of different ways including: (i) O-ketosuberate pro al., J. Bacteriology 188: 8551-8559 (2006), incorporated by duced from the C.-keto acid chain elongation pathway, or (ii) reference); and 1.2.1.50 from, e.g., Acinetobacter sp., A. the biotin biosynthesis pathway I; (iii) biotin biosynthesis platyrhynchos (Ishige, T., et al., Appl. Envitl. Microbiology 68: pathway II; (iv) the condensation of three malonyl-CoA mol 1192-1195 (2002), incorporated by reference), A. thaliana ecules into a single pimeloyl-CoA molecule; (v) the benzoate (Doan, T. T., et al., Journal Plant Physiology 166: 787-796 degradation pathway; (vi) cyclohexane carboxylate pathway, (2009) and Hooks, M.A., et al., Plant J. 20: 1-13 (1999), both (vi) D.L-diaminopimelate pathway; and (vii) a biosynthetic of which are incorporated by reference); Homo sapiens pathway where the starting carbon Source is crotonate. See (McAndrew, R. P. et al., J. Biol. Chem. 283: 9435-9443 FIG. 2. These pathways to pimelate and/or pimeloyl-CoA or (2008), incorporated by reference), M. Musculis, P. leiog pimeloyl-acp are described in greater detail below. nathi, P phosphoreum, P. sativum, or S. chinesis. Exemplary 0183 In any of these pathways, pimelate can be produced acyl-acpreductases in EC 1.2.1 include EC 1.2.1.80, from from pimeloyl-CoA using an enzyme that catalyzes the e.g. Synecoccus elongates (Shirmer, A. et al., (2010). Micro hydrolysis of a thioester. For example, enzymes that can bila biosynthesis of alkanes. Science, 329 (5991: 559-562) hydrolyze a thioester include, but are not limited to, 1.5.1 C.-Keto Acid Chain Elongation thioesterases, acid-thiol ligases, and CoA transferases. Suit able thioesterases include thioesterases classified in EC 3.1.2 0187. Difunctional alkanes can be produced from O-keto such as such as the gene product of YciA, tesB or Acot13; EC glutarate via Successive chain elongation reactions to C-ke 3.1.2.2: EC 3.1.2.18; EC 3.1.2.19; or EC 3.1.2.20. toadipate, C.-ketopimelate or C-ketosuberate known in the art. Thioesterases acting on acyl-acp thioesters include EC 3.1. See, e.g., FIG. 3 and WO2010/068944. Decarboxylation of 2.14, such as FatA, FatB; and EC 3.1.2.21. For example, one the O.-keto dioic acid (Cn, n=5-8) or the corresponding of the following enzymes can be used: a thioester hydrolases/ 2-amino-dioic acid then provides precursors to C.()-difunc acyl-CoA thioesterase classified in EC 3.1.2 such as an ADP tional alkanes of Cn-1, e.g., C4-7. These Successive alpha dependent medium-chain-acyl-CoA hydrolase in EC 3.1.2. keto acid chain elongation reactions occur in the Coenzyme B 19 and 3.1.2.18, an acyl-CoA hydrolase in EC 3.1.2.20; or the biosynthesis pathway as elucidated in methanogenic bacteria gene product of YciA, tesB or Acot13: oleoyl-acyl-carrier Such as Methanocaldococcus jannaschii, Methanococcus protein hydrolases in EC 3.1.2.14. Suitable acid-thiol ligases voltae, and Methanosarcina thermophila. include acid-thiol ligases classified in EC 6.2.1 such as EC 0188 The three successive rounds of chain elongation 6.2.1.3; EC 6.2.1.14; and EC 6.2.1.23. Suitable CoA trans from C.-ketoglutarate (C5) to O-ketosuberate are catalyzed by US 2015/0337275 A1 Nov. 26, 2015

the same three enzymes acting three times (and in the case of Cronan, J. E. and Lin, S., Current Opinion in Chem. Biology EC 4.2.1.114 (AksD (3-isopropylmalate dehydratase large 15: 1-7 (2011), both of which are incorporated by reference subunit)/Aksh (3-isopropylmalate dehydratase small sub herein. FIG. 5 shows formation of pimeloyl-acp through unit)) six times) on Substrates, each time increasing chain oxidative cleavage of long chain fatty acid-acp thioesters by length. See FIG.3. These enzymes include, but are not limited Biol in Bacillus subtilis that is converted to pimelic acid or to: EC 2.3.3.14:homocitrate synthase/AkSA (alpha-isopropy pimeloyl-CoA which are precursors in biotin biosynthesis II lmalate synthase), which acts on C-ketoglutarate, C.-ketoadi (gram positive bacteria). FIG. 6 shows formation of pimelyl pate, and C.-ketopimelate; EC 4.2.1.114: homoaconitase or acp or its methyl esters in biotin biosynthesis I (gram nega AksD/Aksh, which catalyzes the dehydration reactions of tive bacteria). FIG. 6 also shows pathways for the conversion (R)-homocitrate, dihomocitrate and trihomocitrate, as well as of pimeloyl-acp methyl ester into pimelate monomethyl the hydration reactions of cis-homoaconitate, cis-(homo) ac ester. The pimelate monomethyl ester is Subsequently con onitate and cis-(homo)-aconitate; and EC 1.1.1.87: Verted to pimelic acid (heptane-1,7-dioic acid) in the presence homoisocitrate dehydrogenase or AksF (multifunctional of, e.g., a lipase in EC 3.1.1. FIG. 6 also shows the conversion 3-isopropylmalate dehydrogenase/D-malate dehydroge of pimeloyl-acp methyl ester into pimeloyl-acp in the nase , which catalyzes the NAD(P)+ dependent oxidative presence of an esterase (e.g., an enzyme in EC 3.1.1, Such as decarboxylation of homoisocitrate, threo-iso(homo) citrate EC 3.1.1.85, BioH). The pimeloyl-acp is subsequently con and threo-iso(homo) citrate. Verted into pimelic acid (heptane-1,7-dioic acid) in the pres 0189 Exemplary AkSA enzymes include, but are not lim ence of a acyl-acp-thioesterase, e.g., an enzyme in EC 3.1. ited to, those in EC 2.3.3 such as EC 2.3.13 or 2.3.3.14. 2.14, such as Fat A or FatB. In some embodiments, the Exemplary AksD enzymes include those in EC 4.2.1 such as invention contemplates a recombinant host cell capable of EC 4.2.1.36. Exemplary Akse enzymes include, but are not producing pimelic acid (heptane-1,7-dioic acid) that com limited to enzymes in EC 1.1.1 such as EC 1.1.1.87. prises one or all of the enzymes shown in FIG. 6. In some 0190. In some embodiments, the chain elongation enzyme embodiments, the recombinant host cell does has a deficiency is AksA MTH1630, AksD MTH 1631, Aksh MTHO829 or in 7-keto-8-aminopelargonic acid synthetase, an enzyme Aks MTH1388. See, e.g., WO2010/104391, which is incor capable of converting 6-carboxyhexanoyl-CoA (pimeloyl porated by reference herein. CoA) into 7-keto-8-aminopelargonic acid (KAPA), and 0191 The advantage of this pathway is that it only requires encoded by the BioF gene in host organisms with a native the recombinant expression of three heterologous proteins in biotinbiosynthesis pathway so that the pimeloyl-CoA cannot a host. Chain elongation from O-ketoglutarate to O-ketoadi be shuttled into biotin biosynthesis. pate also occur in the lysine biosynthesis V pathway of 0.195 Pimeloyl-acp methyl ester can also be produced as archae, including Aeropyrum permix, Deinococcus radiodu a result of the metabolism of long-chain acyl-acp and/or free rans, Pyrococcus abyssi, Pyrococcus horikoshii, Sulfolobus fatty acids by Biol (for example Biol from Bacillus subtilis). solfataricus, Sulfolobus tokodai, and Thermus thermophilus; and in the lysine biosynthesis IV pathway of yeast and fungi 1.5.5 Eukaryotic Biotin Biosynthesis Pathway including Euglena gracilis, Aspergillus, Penicillium 0.196 Pimelic acid and/or pimeloyl-CoA can also be chrysogenium and Saccharomyces cerevisiae. obtained from eukaryotic biotin biosynthesis pathways. See, 0.192 In the lysine biosynthesis IV and V pathways, the e.g., Roje, S., Phytochemistry 68: 1904-1921 (2007); and reactions from O-ketoglutarate to O-ketoadipate are catalyzed Charles R. Hall, The Contribution of Horizontal Gene Trans by the consecutive action of four enzymes, namely: EC 2.3. fer to the Evolution of Fungi (May 10, 2007) (unpublished 3.14: homocitrate synthase (hcs/LYS20/LYS21/nifV); EC Ph.D. dissertation, Duke University) (on file with Duke Uni 4.2.1.114: methanogen HACN; EC 4.2.1.36: homoaconitase versity Libraries), both of which are incorporated by refer hydratase; and EC 1.1.1.87-homoisocitrate dehydrogenase ence herein. In the eukaryotic pathway, acetyl-CoA is (hiclDH or LYS12). biotransformed to acetoacetyl CoA using an enzyme classi fied under EC 2.3.1.9; acetoacetyl-CoA is biotransformed 1.5.3 Malonyl-CoA Condensation Pathway into (S)-3-hydroxybutanoyl-CoA using an enzyme classified under E.C. 1.1.1.157: (S)-3-hydroxybutanoyl-CoA is 0193 In some embodiments, malonyl-CoA can be used as biotransformed to crotonyl-CoA using an enzyme classified the source of pimeloyl-CoA as certain organisms are able to under EC 4.2.1.17; crotonyl-CoA is biotransformed into condense three malonyl-CoA molecules into a single glutaconyl-1-CoA using an enzyme classified under EC 4.1. pimeloyl-CoA molecule. See FIG. 4. See also Lin, S. and 1.70; glutaconyl-1-CoA is biotransformed to glutaryl-CoA Cronan, J. E., Molecular Biosystems 7: 1811-21 (2011), using an enzyme classified under EC 1.3.99.7; glutaryl-CoA which is incorporated by reference herein. Accordingly, a is biotransformed into 3-ketopimelyl-CoA using an enzyme recombinant host can be produced in which pimeloyl-CoA is classified under EC 2.3.1.43: 3-ketopimelyl-CoA is biotrans produced from malonyl-CoA. In some embodiments, the host formed into 3-hydroxypimelyl-CoA using an enzyme classi cell is an organism other than Achromobacter. In some fied under EC 1.1.1.4259: 3-hydroxypimelyl-CoA is embodiments, the pimeloyl-CoA may be Subsequently con biotransformed into 2,3-didehydro-pimeloyl-CoA using an verted to pimelic acid or other difunctional C7 alkanes. enzyme classified under EC 4.2.1.-, and 2,3-didehydro pimeloyl-CoA is biotransformed into pimeloyl-CoA using an 1.5.4 Biotin Biosynthesis Pathways I (Gram enzyme classified under EC 1.3.1.62. The pimeloyl-CoA is Negative Bacteria) and II (Gram Positive Bacteria) Subsequently transformed into pimelic acid. 0194 In one embodiment, a recombinant host cell pro 0.197 Thus, in some embodiments, the invention provides duces pimelic acid and/or pimeloyl-acp or pimeloyl-CoA a recombinant host cell comprising the enzymes for convert from acetyl-CoA derived from glycerol and/or fatty acids. ing acetyl CoA into heptane-1,7-dioic acid. In some embodi See, e.g., Lin, S., Nature Chem. Biol. 6: 682-688 (2010); and ments, the recombinant host cell has a deficiency in 7-keto US 2015/0337275 A1 Nov. 26, 2015

8-aminopelargonic acid synthetase, an enzyme capable of decarboxylase. See, e.g., FIG. 10. The D.L diaminopimelate converting 6-carboxyhexanoyl-CoA (pimeloyl-CoA) into can be biotransformed into an unsaturated monoami 7-keto-8-aminopelargonic acid (KAPA), and encoded by the nopimelic acid by an enzyme that catalyzes the reductive BioF gene in the host, if the host organisms has a native biotin deamination of D.L diaminopimelate to 6-amino-2-heptene biosynthesis pathway, so that the pimeloyl-CoA or pimeloyl dioic acid. Examples of enzymes that catalyze reductive acp cannot be shuttled into biotin biosynthesis. In some deamination of D.L diaminopimilate include ammonia lyases embodiments, Biol is overexpressed to increase the pimeloyl classified in EC 4.3.1 such as EC 4.3.1.1; EC 4.3.1.2: EC acp formed via oxidative cleavage offatty acid-acp esters. 4.3.1.3; EC 4.3.1.9; EC 4.3.1.12: EC 4.3.1.13: EC 4.3.1.14: In some embodiments, acp-transferases/synthetases acti EC 4.3.1.23; and EC 4.3.1.24. In some embodiments, the Vating long chain fatty acids to their corresponding acp ammonia lyase is a methylaspartate ammonia lyase classified thioesters for cleavage by Biol is overexpressed in the host under EC 4.3.1.2 from C. amalomaticus, C. tetanomorphum cell. The embodiments of the invention also provide a method or A. oryzae. See, e.g., Botting, Biochemistry 27: 2953-2955 of producing heptane-1,7-dioic acid comprising the use of (1988); and Kato, Appl. Microbiol. Biotechnol. 50: 468-474 Such recombinant host cells. (1998), the entireties of which are incorporated by reference herein. 1.5.6 Benzoyl-CoA Degradation Pathway 0201 6-amino-2-heptenedioic acid can be biotransformed into 2-amino-2-heptenedioic acid using an enzyme that cata 0198 One additional source of pimeloyl-CoA and/or lyzes the enoate reduction of 6-amino-2-heptenedioic acid to pimelic acid contemplated herein is the benzoate degradation 2-amino-2-heptanedioic acid (also known as 2-aminopimelic pathway shown in FIG. 7. See, e.g., Bernsteinb, J. R., et al., acid or C.-aminopimelate). Examples of enzymes that cata Metabolic Engg 10: 131-140 (2008); Harwood, C. S., et al., lyze the enoate reduction of 6-amino-2-heptenedioic acid FEMS Microbiology Reviews 22:439-458 (1999); and Har include enoate reductases classified in EC 1.3.1 such as EC rison, F. H. and Harwood, C. S., Microbiology 151: 727-736 1.3.1.8; EC 1.3.1.9; EC 1.3.1.10, EC 1.3.1.31; EC 1.3.1.38; (2005), all of which are incorporated by reference herein. A EC 1.3.1.39; EC 1.3.1.44 or pimeloyl-CoA dehydrogenase in number of transformations are illustrated in FIG. 7, where EC 1.3.1.62 such as PimC/PimD or ThnJ/Think. The enoate benzoate is converted into benzoyl-CoA using an enzyme reductase may also be in the EC 1.3, such as EC 1.3.8.1 or EC classified under EC 6.2.1.25, and ultimately into pimeloyl 1.3.99.3; EC 1.3.99.B10 or Syntrophus aciditrophicus 2,3- CoA. Benzoyl-CoA is converted to cyclohexa-1,5-dienecar didehydropimelyl-CoA reductase and homologs thereof. In bonyl CoA using an enzyme classified under EC 1.3.99.15. In some embodiments, cyclohexa-1,5-dienecarbonyl CoA is some embodiments, the enoate reductase is Ya Mor OPR1/3 biotransformed to 6-hydroxycyclohex-1-ene-1-carboxyl in EC 1.3.1.31 from B. subtilis or L. esculentum, respectively. CoA using an enzyme classified under EC 4.2.1.100; 6-hy See, e.g., Stueckler, Org. Lett. 9: 5409-5411 (2007); Kitzing, droxycyclohex-1-ene-1-carboxyl-CoA is biotransformed to J. Biol. Chem. 280: 27904-27913 (2005); Hall, Angew. Chem. 6-ketoxycyclohex-1-ene-1-carboxyl-CoA using an enzyme Int. Ed. 46: 3934-3937 (2007); and Breithaupt, Proc. Natl. classified under EC 1.1.1-, 6-ketoxycyclohex-1-ene-1-car Acad. Sci USA 103: 14337-14342 (2006), the entireties of boxyl-CoA is biotransformed to 3-hydroxypimelyl-CoA which are incorporated by reference herein. using an enzyme classified under EC 3.7.1.-, 3-hydroxypime 0202 The 2-aminopimelic acid can be converted to the lyl-CoA is biotransformed to 6-carboxylhex-2-enoyl-CoA corresponding 2-heptenedioic acid (also known as 2.3-dide using an enzyme classified under EC 4.2.1.-, 3-hydroxypime hydropimelic acid) by an enzyme that catalyzes the reductive lyl-CoA is biotransformed to pomelyol-CoA using an deamination of 2-aminopimelic acid. Examples of enzymes enzyme classified under EC 1.3.1.62. The pimeloyl-CoA, in that catalyze reductive deamination of 2-aminopimelic acid turn, is converted into pimelic acid (heptane-1,7-dioic acid). include ammonia lyases classified in EC 4.3.1 such as EC In some embodiments, cyclohexa-1,5-dienecarbonyl CoA is 4.3.1.1; EC 4.3.1.2: EC 4.3.1.3; EC 4.3.1.9; EC 4.3.1.12: EC biotransformed to cyclohex-1-ene-1-carboxyl-CoA, cyclo 4.3.1.13; EC 4.3.1.14: EC 4.3.1.23; or EC 4.3.1.24 as hex-1-ene-1-carboxyl-CoA is biotransformed to 2-hydroxy described above. cyclohexane-1-carboxyl CoA using an enzyme classified 0203 The 2-heptene dioic acid can be transformed to under EC 4.2.1.-: 2-hydroxycyclohexane-1-carboxyl CoA is 2-heptene dioic acid-CoA using a CoA transferase (e.g., a biotransformed to 2-ketocyclohexane-1-carboxyl-CoA using CoA transferase classified in EC2.8.3 such EC 2.8.3.12: EC an enzyme classified under E.C. 1.1.1.-, and 2-ketocyclohex 2.8.3.13 or EC 2.8.3.14 or or the gene product of ThnH that anecarboxyl-CoA is biotransformed to pimeloyl-CoA using catalyses the reversible transfer of CoA to pimelic acid) and an enzyme classified under EC 3.1.2.- (e.g., a-ketocyclohex an acid thiol ligase (e.g., an acid-thiol ligases classified in EC anecarboxyl-CoA hydrolase Rip-bad I, an enzyme classified 6.2.1 such as EC 6.2.1.3; EC 6.2.1.5; EC 6.2.1.14; or EC under EC 3.1.2.-). 6.2.1.23). 0199. In some embodiments, the recombinant host cell 0204 2-heptene dioic acid-CoA is then converted to capable of producing heptane-1,7-dioic acid includes one or pimeloyl-CoA by an enzyme that catalyzes the enoate reduc tion of 2-heptene dioic acid such as an enoate reductase more of the enzymes listed in FIG. 7 or in this section. enzyme classified in in EC 1.3.1. Pimeloyl-CoA can be con 1.5.7 2,6-Diaminopimelate Pathway verted to pimelate or pimelic acid semialdehyde as described above. 0200 Yet another source of pimelate is D.L diaminopime (0205 As shown in FIG. 10 and in FIG. 3, a keto pimelate late, also known as 2,6-diaminopimelate. D.L-diaminopime also can undergo one round of chain elongation to O-keto late is an intermediate in the production of lysine. In some Suberate using chain elongation enzymes. Such as those embodiments, an endogenous lysine pathway can be modi present in methanogenic bacteria (e.g., Methanobacterium fied to preferentially produce pimelate by creating recombi autotrophicum). Such chain elongation enzymes include, but nant host cells that have a deficiency in diaminopimilate are not limited to AkSA, AksD, and Aksh or F as discussed US 2015/0337275 A1 Nov. 26, 2015

above. Exemplary AkSA enzymes include, but are not limited transfer of CoA (e.g., a CoA transferase described above). to, those in EC 2.3.3. Exemplary enzymes in EC 2.3.3 The O.-hydroxypimelate CoA can be converted to 2-heptene include, but are not limited to EC 2.3.13 or 2.3.3.14. Exem dioic acid CoA with an enzyme that catalyzes the dehydration plary AksD enzymes include those in EC 4.2.1. An exemplary of the C-hydroxypimelate-CoA (e.g., a hydro-lyase classified enzyme in EC 4.2.1 includes, but is not limited to EC 4.2.1.36. under EC 4.2.1. Such as EC 4.2.1.2, EC 4.2.1.17; EC 4.2.1.18: Exemplary Aks enzymes include, but are not limited to EC 4.2.1.59, and EC 4.2.1.61). The 2-heptenedioic acid CoA enzymes in EC 1.1.1. An exemplary enzyme in EC 1.1.1 can be converted to pimelic CoA using an enzyme that cata includes, but is not limited to EC 1.1.1.87. And, in this lyzes the enoate reduction of 2-heptenedioic acid CoA. Suit instance, the fourth set of enzymes comprises a decarboxy able enoate reductases are described above. lase. Exemplary decarboxylases include those in EC 4.1.1. 0210. In some embodiments, the invention provides a Exemplary decarboxylases in EC 4.1.1 include, but are not recombinant host cell comprising one or more enzymes nec limited to EC 4.1.1.11: EC 4.1.1.15; EC 4.1.1.17; EC 4.1.1. essary to convert D.L-diaminopimelate preferentially to 18; EC 4.1.1.19; EC 4.1.1.20 and EC 4.1.1.86. pimelate. 0206. In some embodiments, acetyl-CoA can be fed into the tricarboxylic acid (TCA) cycle, in which acetyl-CoA is 1.5.8 Crotonate Pathway transformed into (butan-1,4-dioic acid). The succinic acid is then diverted from the TCA cycle into a 0211. A source of pimeloyl-CoA, and ultimately a source pathway for synthesizing pimelic acid starting with the of pimelate, is a biosynthetic pathway where the starting biotransformation of 2-oxoglutarate into 2-hydroxy-1,2,4- carbon Source is crotonate. That biosynthetic pathway is illus butanetricarboxylic acid (homocitrate) (e.g., by an enzyme in trated in FIG. 9. See, e.g., Mouttaki, H., et al., Applied Envitl. EC 2.3.3.14); biotransformation of homocitrate to homo cis Microbiology 73: 930-938 (2001). The crotonate is broken aconitase and homo cis aconitase to homoisocitrate using an downto acetyl-CoA. It has been reported that two thirds of the enzyme classified under EC 4.2.1.36; biotransformation of acetyl-CoA produced in the biosynthetic pathway shown in homoisocitrate to oxaloglutarate using an enzyme classified FIG. 9 is transformed into acetate. In some embodiments, the under EC 1.1.1.87; biotransformation of oxaloglutarate to invention provides a recombinant host cell that is deficient in 2-oxoadipate using an enzyme classified under EC 1.1.1.87. the enzymes that would convert acetyl-CoA into acetate (e.g., and biotransformation of 2-oxoadipate to glutaryl-CoA using a phosphate acetyltransferase and acetate kinase such that the an enzyme classified under EC 1.2.4.2. Glutaryl-CoA can be acetyl-CoA is instead converted to acetoacetyl-CoA and Sub converted to pomeloyl-CoA as discussed above. The homoci sequently to glutaconyl-CoA in a number of metabolic steps trate is ultimately converted to pimeloyl-CoA and, subse as shown in FIG. 9. The glutaconyl-CoA is subsequently quently, into pimelic acid as described herein. converted to glutaryl-CoA. The glutaryl-CoA is chain-ex 0207. The O.-keto-suberate can either be decarboxylated tended to form 3-oxopimelyl-CoA. Following a reduction, by a C-keto acid decarboxylase to form pimelic acid semial dehydration, and a second reduction, pimelyl-CoA is dehyde, or converted to C.-aminoSuberate, using an amino obtained. acid aminotransferase which will lead directly to 7-amino 0212. In the biosynthetic pathway shown in FIG. 9, the heptanoic acid upon decarboxylation by an O-amino acid pimelyl-CoA is ultimately transformed into cyclohexane car decarboxylase. Alternatively, the pimelic acid semialdehyde boxylate. Such a biosynthetic pathway has been observed in can be converted to 7-aminoheptanoic acid using a (D-ami S. aciditrophicus. Thus, in Some embodiments, the invention notransferase. Exemplary C.-keto acid decarboxylases provides a recombinant host cell in which the genes encoding include those in EC 4.1.1 such as EC 4.1.1.1; EC 4.1.1.7; and the enzymes that would convert pimeloyl-CoA into cyclohex EC 4.1.1.72 or an acetolactate synthase in the class EC 2.2. 1-ene-1-carboxyl-CoA and ultimately into cyclohexane car 1.6. boxylate, are are “knocked out, such that the metabolic 0208 Exemplary C.-amino acid decarboxylase include pathway ends at pimeloyl-CoA. those in EC 4.1.1.11: EC 4.1.1.15; EC 4.1.1.17; EC 4.1.1.18: EC 4.1.1.19; EC 4.1.1.20; EC 4.1.1.45 and EC 4.1.1.86. In 1.5.9 Tetralin Degradation Pathway Some embodiments, the decarboxylase comprises ketoisov alerate decarboxylasekiv) in EC 4.1.1.1 from L. lactis; ben 0213. In some embodiments, pimelic acid semialdehyde is Zoylformate decarboxylasemdICA460I or BFD in EC4.1.1.7 produced using the tetralin (benzocyclohexane) degradation from Pputida; pyruvate decaroxylase isozymes 1 and 2Pdc1 pathway from Sphingomonas and Corynebacterium spp. in EC 4.1.1.1 from S. cerei visae; pyruvate decarboxylase Pdc 7-oxoheptanoic acid is produced as an intermediate is the (I472A) from Z. mobilis; or branched chain alpha-keto acid tetralin degradation pathway in Sphingomonas macrogolita decarboxylase kdcA in EC 4.1.1.72 from L. lactis, as bid. Lopez-Sanchez, A., et al., Appl. Environ. Microbiol. described in WO2011/031147, the entirety of which is incor 76:110-118 (2010). Tetralin is metabolized by cleavage of porated by reference herein. Exemplary amino acid ami both aromatic rings yielding pyruvate and pimelic acid semi notransferases include those in EC 2.6.1, such as EC 2.6.1.21; aldehyde. Tetralin, an organic solvent, is a complex starting EC 2.6.1.39; EC 2.6.1.42; and EC 2.6.1.67. Exemplary material that is derived from naphthalene. Naphthalene is ()-aminotransferases include those in EC 2.6.1, such as EC currently produced from coal tar, but biosynthetic pathways 2.6.1.11: EC 2.6.1.13; EC 2.6.1.18; EC 2.6.1.19; EC 2.6.1.48: for naphthalene synthesis are known to occur in termites, and EC 2.6.1.62. fungi and some types of plants. Chen, J. et al., Nature 392: 0209. In one embodiment, a keto pimelate can be con 558-559 (1998); Daisy, B. H., et al., Microbiology 148:3737 tacted with one or more enzymes that catalyze the ketone 3741 (2002); and Azuma, H., et al., Phytochemistry 42:999 reduction to a hydroxypimelate (e.g., a carbonyl reductase 1004 (1996). Pimelic acid semialdehyde can be converted to such as EC 1.1.1.184), which in turn can be converted to 7-amino heptanoic acid or 7-hydroxyheptanoic acid as dis C-hydroxypimelate CoA using an enzyme that catalyzes the cussed above. US 2015/0337275 A1 Nov. 26, 2015

111 HEXAMETHYLENEDIAMINE FROM pimelate to produce alpha keto pimelate; an enzyme that 2-OXOPIMELATE catalyzes the carbonyl reduction of alpha keto pimelate to alpha hydroxy pimelate; an enzyme that catalyzes the transfer 0214. Some embodiments of the invention also provides a of CoA to alpha hydroxy pimelate to produce alpha hydrox method of producing hexamethylenediamine from 2-ox ypimelate CoA, an enzyme that catalyzes the reduction of opimelate comprising using recombinant host cells. 2-ox alpha hydroxypimelate to 2-heptenedioic acid; an enzyme opimelate is a known intermediate in the Coenzyme B bio that catalyzes the alpha keto acid chain elongation of alpha synthesis pathway in methanogenic Archae. ketopimelate to alpha ketosuberate, an enzyme that catalyzes 0215 Thus the invention provides a recombinant host cell the transfer of an amino group to alpha ketoSuberate to pro comprising: an enzyme capable of converting 2-oxopimelate duce alpha-aminosuberate, an enzyme that catalyzes the into 2-aminopimelate (e.g., an aminotransferase enzyme in alpha keto acid decarboxylation of alpha aminoSuberate to EC 2.6.1.67), an enzyme capable of converting 2-aminopime 7-aminoheptanoic acid; an enzyme that catalyzes the alpha late to 2-amino-7-oxoheptanoate (e.g., a reductase enzyme in keto acid decarboylation of alpha ketosuberate to produce EC 1.4.1), an enzyme capable of converting 2-amino-7-OXo pimelic acid semialdehyde; an enzyme that catalyzes the heptanoate to 2,7-diaminoheptanoate (e.g., a 1-aminotrans enoate reduction of 6-amino-2-heptendioic acid to produce ferase enzyme in EC 2.6.1), and an enzyme capable of con pimelic acid; or an enzyme that catalyzes the thioester Verting 2,7-diaminoheptanoate to hexamethylenediamine hydrolysis of pimeloyl acp to pimelic acid. Recombinant (e.g., a decarboxylase enzyme in EC 4.1.1). In some embodi host cells can contain any subgroup consisting of one or more ments, the recombinant host cell capable of producing hex (as above) nucleic acids encoding one or more of the above amethylenediamine comprises any or all of these enzymes. listed enzymes. Embodiments of the invention also provide a method of pro 0218. The host cells that are used typically possess a num ducing hexamethylenediamine comprising the use of these ber of properties: they may be easily genetically modified, are recombinant host cells. tolerant of the conditions used in the method of the invention, 0216. In some embodiments, the invention provides a and grow to cells densities which are industrially useful. method for producing hexamethylenediamine from renew 0219. Optionally the host cell may be a single celled able feedstocks such as Sugars, fatty acids, glycerol, and microorganism, or may be a cell of a cell line. The host cell SynGas. In some embodiments, the non-naturally occurring may be of a wild type genotype. In this instance, the enzyme host cell capable of producing hexamethylenediamine com that is used to catalyze one or more steps in the method of the prises any or all of the enzymes required to convert renewable invention is naturally present in the host cell and is expressed feedstocks to hexamethylenediamine. at a level that has industrial use in the methods of the inven tion. In an alternative, the host cell has been genetically modi 22 RECOMBINANT HOST CELLS fied to express the enzyme at a level that has industrial use in 0217. This disclosure features recombinant host cells that the methods of the embodiments of invention. The enzyme recombinantly express one or more (e.g., two or more; three may be sourced from the cell in which it is expressed. In an or more; four or more; five or more; six or more; seven or alternative, the enzyme is sourced from a different strain or more; eight or more; nine or more; ten or more; eleven or species of cell. more; twelve or more; thirteen or more; fourteen or more; 0220. In one alternative, the host cell is a prokaryote. In fifteen or more; sixteen or more; seventeen or more; eighteen another alternative it is a eukaryote. Typically single celled or more; nineteen or more; twenty or more; or even more) of microorganisms are used. the enzymes used for producing a compound in one of the 0221) The term prokaryotic cell includes gram positive pathways described herein and methods of using such host and gram negative bacteria. Examples of suitable gram nega cells to produce di- and trifunctional C7 alkanes. For tive bacteria include the genus Escherichia such as Escheri example, a host cell can include one or more (as above) chia coli, from the genus Clostridia Such as Clostridium exogenous nucleic acids encoding one or more (as above) of liungdahli, Clostridium autoethanogenium or Clostridium the following: an enzyme that catalyzes the reductive deami kluyveri; from the genus Corynebacteria Such as Corynebac nation of D.L diaminopimelate or alpha-aminopimielate; an terium glutamicum; from the genus Cupriavidus Such as enzyme that catalyzes the enoate reduction of 2-heptenedioic Cupriavidus necator or Cupriavidus metallidurans; from the acid to pimelic acid; an enzyme that catalyzes the carboxylic genus Pseudomonas Such as Pseudomonas fluorescens, acid reduction of pimelic acid to pimelic acid semialdehyde, Pseudomonas putida or Pseudomonas Oleavorans; from the an enzyme that catalyzes the semialdehyde amination of genus Delftia Such as Delftia acidovorans; from the genus pimelic acid semialdehyde to 7-aminoheptanoic acid, an Bacillus such as Bacillus subtillis; from the genus Lactoba enzyme that catalyzes the amidohydrolysis of 7-aminohep cillus such as Lactobacillus delbrueckii; or from the genus tanoic acid to enantholactam, an enzyme that catalyzes the Lactococcus such as Lactococcus lactis. aldehyde dehydrogenation of 7-aminoheptanoic acid to 0222 Eukaryotic host cells include those from yeast and 7-aminoheptanal; an enzyme that catalyzes the transfer of an other fungi, as well as, for example, insect, mouse, rat, pri amino group to 7-aminoheptanal to produce 1,7-diaminohep mate, or cells. Examples of Suitable eukaryotic host tane; an enzyme that catalyzes the transfer of CoA to 2-hep cells include yeasts and fungi from the genus Aspergillus Such tenedioic acid to produce 2-heptene diacid-CoA, an enzyme as Aspergillus niger; from the genus Saccharomyces such as that catalyzes the enoate reduction of 2-heptene dioic acid Saccharomyces cerevisiae; from the genus Candida Such as CoA to pimeloyl-CoA; an enzyme that catalyzes the thioester C. tropicalis, C. albicans, C. cloacae, C. guillermondii, C. hydrolysis of pimeloyl-CoA to produce pimelic acid; an intermedia, C. maltosa, C. parapsilosis, and C. zeylenoides enzyme that catalyzes the carboxylic acid reduction of Pichia Such as Pichia pastoris; from the genus Yarrowia Such pimelic acid to pimelic acid semialdehyde; an enzyme that as Yarrowia lipolytica; from the genus Issatchenkia Such as catalyzes the transfer of an amino group from alpha amino Issathenkia Orientalis; from the genus Debaryomyces such as US 2015/0337275 A1 Nov. 26, 2015

Debaryomyces hansenii; from the genus Arxula Such as cally 2-5% w/w. Stabilising additives such as salts (e.g. KCl). Arxula adenoinivorans; or from the genus Kluyveromyces Sugars, proteins and the like may be included to improve Such as Kluyveromyces lactis or from the genera Exophiala, thermal tolerance or improve the drying characteristics of the Mucor, Trichoderma, Cladosporium, Phanerochaete, Cla recombinant cells during the drying process. dophialophora, Paecilomyces, Scedosporium, and Ophios 0226. In some embodiments, partially or wholly purified to ina. enzymes may be used instead of or in combination with 0223) The host cells may be provided in a variety of dif recombinant hosts or lysates of recombinant or non-recom ferent forms. The cells may be resting cells. That is to say that binant cells. Wholly or partially purified enzymes can be the cells are grown up in cultures and removed from the wholly or partially purified lysates of any of the recombinant culture medium before being employed as the biocatalyst. cells described here or non-recombinant cells. Methods of The cells may be used directly following growth, or they may purification are well known in the art. Partial or complete be stored before use. Typical methods of storage include purification has the advantage that the enzyme of interest is freezing. In an alternative, the cells are lyophilized prior to separated from other cellular components which may inter use. In a further alternative, the cells are growing cells. That is fere with the reaction catalyzed by that enzyme (either by also to say that the cells perform their biocatalytic action while reacting with the Substrate or by converting intermediates or being cultured. If the substrate for a particular biocatalytic the desired product to an unwanted compound). Purification reaction cannot cross the cell membrane to be transformed by does add further steps to any biocatalyst preparation, how host cells, then in one alternative, a crude lysate may be used. ever, and therefore may not be suitable in all instances. The A crude lysate is the initial Suspension of cellular components determination of the suitability of the use of partially or produced following lysis of the cells. Lysis of the cells may be wholly purified enzymes will be well within the grasp of the performed by any means, including chemical or mechanical. skilled person following the teaching herein. Further methods of lysis will be evident to the skilled person in light of their general knowledge and the teachings herein. 0227. In some embodiments one or more of the conver In another alternative, a clarified lysate may be used. A clari sions in the method of the invention will be performed by the fied lysate may be prepared by centrifuging the crude lysate to biocatalyst under aerobic conditions. In an alternative one or pellet unlysed cells and other cellular debris. more of the conversions in the method of the invention will be 0224. In some embodiments, substantially pure cultures of performed under anaerobic conditions. In a further alterna recombinant host cells are provided. As used herein, a “sub tive, some steps of the method of the invention will be per stantially pure culture of a recombinant cell is a culture of formed under aerobic conditions and some steps will be per that cell in which less than about 40% (i.e., less than about: formed under anaerobic conditions. 35%; 30%; 25%; 20%; 15%: 10%; 5%; 2%; 1%: 0.5%; 0.25%: 0.1%; 0.01%; 0.001%; 0.0001%; or even less) of the 2.3 MODIFICATION OF WHOLE CELL total number of viable cells in the culture are viable cells other BIOCATALYSTS than the recombinant cells, e.g., bacterial, fungal (including yeast), mycoplasmal, or protozoan cells. The term “about in 0228. The biocatalysts used in the methods of the inven this context means that the relevant percentage can be 15% tion may be unmodified host cells of the species in which the percent of the specified percentage above or below the speci enzyme naturally occurs. Typically, however, it is necessary fied percentage. Thus, for example, about 20% can be 17% to to modify genetically the host cell to produce a non-naturally 23%. Such a culture of recombinant cells includes the cells occurring (i.e., recombinant or engineered) host cell. and a growth, storage, or transport medium. Media can be liquid, semi-solid (e.g., gelatinous media), or frozen. The 24 CHROMOSOMAL MODIFICATION OF THE culture includes the cells growing in the liquid or in?on the HOST CELL GENOME semi-solid medium or being stored or transported in a storage or transport medium, including a frozen storage or transport 0229. In some embodiments, the chromosome of the medium. The cultures are in a culture vessel or storage vessel whole cell biocatalyst (i.e., the host cell) has been modified. or Substrate (e.g., a culture dish, flask, or tube or a storage vial This modification may be an insertion, a deletion or a Substi or tube). tution of a nucleic acid in the chromosome. Methods of effect 0225. In one embodiment, the recombinant cells of this ing modifications to the chromosome of a cell are well known, disclosure can be harvested from a fermentation process by e.g., transposon mutagenesis, Cre-LOX mediated recombina conventional methods such as filtration or centrifugation and tion, lambda Red and RecET mediated recombination. formulated into a dry pellet or dry powder formulation while maintaining high activity. Processes for production of a dry 0230 Stable integration into the chromosome of a nucleic powder whole recombinant cell composition exhibiting one acid is advantageous because it allows the maintenance of the or more of the activities disclosed herein can include spray nucleic acid without the need for a selectable marker. drying, freeze-drying, fluidised bed drying, vacuum drum 0231. By performing repeated insertions, it is possible to drying, or agglomeration and the like. The compositions can integrate stably a number of different nucleic acids into a host be a shelf-stable, dry biocatalyst composition suitable for the cell biocatalyst’s chromosome. Stable integration means that biosynthetic methods described herein. Drying methods such the host cell can be cultured for greater than five generations as freeze-drying, fluidised bed drying or a method employing without loss of the alteration. Generally, stable genetic alter extrusion/spheronisation pelleting followed by fluidised bed ations include modifications that persist greater than 10 gen drying can be particularly useful. Temperatures for these erations, particularly stable modifications will persist more processes may be <100° C. but typically <70° C. to maintain than about 25 generations, and more particularly, stable high residual activity and stereoselectivity. The dry powder genetic modifications will be greater than 50 generations, formulation should have a water content of 0-10% w/w, typi including indefinitely. US 2015/0337275 A1 Nov. 26, 2015

25 EPISOMAL MODIFICATION OF THE HOST enzymes used in the methods of the invention. In this CELL BIOCATALYSTS GENOME instance, the multiple enzymes may be encoded by the same introduced nucleic acid. In an alternative, the enzymes may 0232. In some embodiments, the episomal composition of be encoded on separate introduced nucleic acid fragments. the host cell genome is modified. By episome is meant any The enzymes may all be expressed from a single promoter autonomously replicating element, for example a plasmid (for example by arranging the enzymes in the form of an (which may be linear or circular), a cosmid, a yeast artificial operon). In an alternative, the enzymes may be expressed chromosome (YAC), etc. Episomal elements frequently place from multiple separate promoters. In some instances, the a metabolic load on their host cell, and accordingly, in the multiple separate promoters may be induced by the same absence of any active partitioning mechanism to ensure that at chemical (for example, each of the multiple enzymes may be least one copy of the episome is present in each daughter cell expressed from the yeast GAL promoter, thus meaning that following cell division it is necessary to include a selectable each gene is inducible with galactose). Other Suitable pro marker. moters are known in the art. In an alternative, each of the genes encoding an enzyme used in the method of the inven 2.6 INTRODUCED NUCLEICACID SEQUENCES tion is under the control of a different promoter. Thus, differ 0233. The nucleic acids introduced into the cell may com ent enzymes can be introduced individually through the use of prise one or more (e.g., two or more; three or more; four or different inducing compounds. In another alternative, an more; five or more; six or more; seven or more; eight or more; intermediate approach is used, wherein a number of enzymes nine or more; ten or more; eleven or more; twelve or more; are under the control of the same promoter, and a number of thirteen or more; fourteen or more; fifteen or more; sixteen or enzymes are under the control of different promoters. This more; seventeen or more; eighteen or more; nineteen or more; alternative may be particularly advantageous when numerous twenty or more; or even more) of a number of elements. enzyme pathways have been generated in a host cell and it is Typically one of the elements encodes an enzyme that is used desirous to control each member of a pathway in concert with to produce di- or trifunctional C7 alkanes or precursors to the other pathway members, but to control each pathway Such molecules. In some alternatives, the nucleic acid separately. encodes a protein which is not an enzyme that functions in the 0237 Further considerations can be made with regard to method of the invention. whether the multiple enzymes are expressed from the chro 0234 Typically, a promoter is operably linked to a nucleic mosome or from the plasmid. In instances where the enzyme acid. The choice of promoter will depend on the intended is expressed from the plasmid, advantageously each plasmid application, and can readily be determined by the skilled comprises a different origin and/or a different selectable person. For example, if an enzyme catalyzed reaction may be marker. detrimental to a particular host cell, it may be desirable to use 0238. The expression of multiple enzymes in a single cell a regulated or inducible promoter, Such that gene expression is advantageous because, if the co-expressed enzymes act can be turned on or off as necessary. Alternatively, it may be directly after one anotherina reaction pathway, it ensures that preferred to have expression driven by either a weak or strong the product of the upstream enzyme reaction is immediately constitutive promoter to ensure that the enzyme is expressed available to be acted upon by the downstream enzyme. This at all stages of growth. Exemplary promoters suitable for avoids the need for either (i) purification of the intermediate eukaryotic cell systems include the SV40 early promoter, the and the setting up of second reaction vessel to perform the cytomegalovirus (CMV) promoter, the mouse mammary second reaction; or (ii) the need for the product to be exported tumor virus (MMTV) steroid-inducible promoter, and the from the cell comprising the upstream enzyme into the cul Moloney murineleukemia virus (MMLV) promoter. Exem ture medium where it may be acted upon by a cell comprising plary yeast promoters include 3-phosphoglycerate kinase the downstream enzyme. promoter, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) promoter, galactokinase (GAL1) promoter, galac 2.8 CHAPERONE SYSTEMS toepimerase promoter, and alcohol dehydrogenase (ADH) 0239 When a cell has been engineered to express a protein promoter. Other Suitable promoters in yeast are known to under non-natural conditions (for example, when a protein those skilled in the art. Exemplary promoters suitable for that is native to a species is expressed at levels above the bacterial cell systems include but are not limited to: T7, T3, natural level, or in an alternative, when a protein from a SP6, lac and trp promoters. different species is expressed in a host cell) in some instances 0235 If necessary to ensure expression, the introduced that protein will not be expressed in an active form. Instead it nucleic acid also can be operably linked to comprise 5’ may fold incorrectly, and accumulate as a non-functional untranslated region (UTR), 3' UTR, enhancer and/or termi “inclusion body aggregates. In this instance, the cells used to nator regions, as required. Such elements will be knownto the express the protein may be subjected to genetic modification skilled person. to further express chaperone proteins which are able either to prevent the mis-folding of the protein, or are able to refold it 2.7 EXPRESSION OF MULTIPLE ENZYMES IN from the aggregated State. The inclusion of Such chaperone A HOST CELL proteins is advantageous because it increases the quantity of 0236. In some embodiments, a single strain of host cell is active protein per cell, and therefore increases the overall used to express more than one (e.g., two or more; three or efficiency of the method of the invention. The expressed more; four or more; five or more; six or more; seven or more; chaperone may be a chaperone protein of the host cell. In an eight or more; nine or more; ten or more; eleven or more; alternative, the chaperone may be from the same species/ twelve or more; thirteen or more; fourteen or more; fifteen or strain as the protein. Typical chaperone proteins for expres more; sixteen or more; seventeen or more; eighteen or more; sion include members of the GroEL/GroES family, and mem nineteen or more; twenty or more; or even more) of the bers of the DnaJ/DnaK/GrpE family. Homologs of the US 2015/0337275 A1 Nov. 26, 2015 archetypal E. coli GroEL/GroES and DnaJ/DnaK/GrpE pro or any intermediates produced in the reaction pathway pro teins have been identified in other prokaryotic species (see, ducing difunctional C7 alkanes, to a different, unwanted, end for example.), and eukaryotic homologs are also known product. In an alternative, the enzyme is not deleted, but is (GroEL and GroES correspond to the eukaryotic proteins instead altered so that it can acts against the Substrate, inter Hsp60 and Hsp10, and DnaJ. DnaK and GrpE correspond to mediates, or product, at a rate that is less than the rate of the the eukaryotic proteins Hsp70, Hsp40 and Hsp24, respec wildtype enzyme. In the instance where the enzyme catalyses tively). These proteins have been identified in a number of a reversible reaction, then the enzyme should be altered so species of yeast (for example, Saccharomyces cerevisiae). that it acts only in the desired direction of reaction. The choice of appropriate chaperone proteins for coexpres 0243 For example, in some embodiments, a recombinant sion with an enzyme used in the method of the invention will host cell can have a deficiency in an enzyme capable of be evident to the skilled person following the teachings converting pimeloyl-acp into 7-keto-8-aminopelargonic herein. acid (KAPA) or converting 6-carboxyhexanoyl-CoA (pimeloyl-CoA) into KAPA (e.g., a deletion or inactivation of 29 METABOLICENGINEERING OF HOST BioF, the gene encoding 7-keto-8-aminopelargonic acid syn CELLS thetase (E.C. 2.3.1.47). 0240 Metabolic engineering is the process of optimizing 0244. In some embodiments, a recombinant host cell can the parameters in a host cell in order to increase the ability of have a deficiency in diaminopimilate decarboxylase, which a cell to produce a compound. The host cells used in the creates alysine auxotroph. A lysine auxotroph can be particu method of the present invention optionally have been engi larly usefulforderegulating carbon flux to D.L diaminopime neered to optimize the output of the difunctional C7 alkanes late, the immediate precursor to lysine. A lysine auxotroph discussed above. would require fed-batch fermentation. 0241 Metabolic engineering to increase the ability of a cell to produce a compound is principally performed via two 2.10 GROWING WHOLE CELL BIOCATALYSTS avenues. The first is to optimize the enzymes in the pathway 0245. In some embodiments of the invention, host cells are producing the desired product from the starting material. In a used which are growing (i.e., dividing) at the time the cells multi-enzyme pathway resulting in the production of a perform the conversions in the method of the invention. In difunctional C7 alkanes (as shown in the figures and these embodiments the cells are cultured under conditions described in the preceding sections), it is possible to deter which optimize the production of desired difunctional C7 mine the concentration of each intermediate in the pathway alkane. Any of the recombinant host cells described herein using techniques known to the skilled person (for example, can be cultured to produce and/or secrete the biosynthetic two dimensional electrophoresis, the use of isotopically products of the invention. As used herein, the term culture is labeled precursors, and nuclear magnetic resonance (NMR) equivalent with fermentor and bioreactor. spectroscopy), and therefore determine which of the enzyme conversions is the rate limiting Step—that is to say which step 2.10.1 Media in the reaction scheme is the slowest. This can be determined by observing a build up of an intermediate, which indicates 0246. In some instances, the carbon source used for that the enzyme acting upon this intermediate is limiting the growth will be provided in the form of a nutrient broth, for overall rate of conversion. In this instance, the rate at which example Luria medium or yeast extract medium. In other this intermediate is reacted should therefore be increased. instances, a defined medium (that is to say, a medium in which This can be performed by a number of means. Firstly, the the concentration of each component is known) appropriate expression level of the limiting enzyme may be increased. for growth of the host cell may be used. In one alternative, the Optionally this may beachieved by placing the gene encoding growth medium comprises glucose as a carbon Source. In the enzyme under the control of a strong promoter, e.g., the T7 another alternative, the carbon source used by the host cells in promoter if the enzyme is being expressed in E. coli or the the medium is glucose, Sucrose, Xylose, fatty acids and glyc TEF promoter if the enzyme is being expressed in yeast. The erol. second option is to increase the number of copies of the gene encoding the enzyme that are present in cell, for instance by 2.10.2 Growth of Multiple Strains in the Same placing the gene on a multicopy plasmid, or by incorporating Culture multiple copies of the gene into the chromosome of the host 0247. In some embodiments, the enzymes which catalyze cell (these copies may be incorporated at the same location in the conversions in the method of the invention are present in the chromosome or in different locations in the chromosome). more than one strain/species of cell, wherein those more than 0242. The production of difunctional C7 alkanes (as one strain/species of cell are used simultaneously. In the shown in the figures and described in the preceding sections) instance where the multiple strains/species are growing when can also be increased by inactivating or reducing the activity the conversions in the method of the invention are performed of any enzymes which are capable of diverting the Substrate, (i.e., the cells are in co-culture), then the Strains/species that any of the intermediates, or the product into a metabolic are chosen must be selected Such that one strain does not pathway other than that which is the aim of the method. outcompete the other strain. Such a relationship can be Therefore, the activity of the enzyme may be lowered to obtained by introducing an artificial symbiosis into the co increase the yield of the difunctional C7 alkanes. Thus in culture. That is to say that the two stains/species that are used Some embodiments, the recombinant host cells disclosed are each auxotrophic for an essential nutrient, but for different herein can have a deficiency in one or more enzymes (e.g., by nutrients. Furthermore, the other strain should be engineered deleting the nucleic acid encoding an enzyme of interest or to produce an excess of that nutrient so that both strains can reducing expression of the nucleic acid) that are capable of survive together in a culture when the growth medium of the diverting the starting material of the method of the invention, culture does not comprise the two essential nutrients. Selec US 2015/0337275 A1 Nov. 26, 2015 tion of appropriateauxotrophies will be well within the grasp culture and sequentially subjected to chemical or biocatalytic of the skilled person following the teachings herein. conversions to convert the product to other compounds, if desired. 2.10.3 Fermentations 2.11 COMPOSITIONS OF THE INVENTION 0248. The culture conditions described herein can be scaled up and grown continuously for manufacturing of hep 0251. The invention also provides compositions compris tane-1,7-dioic acid, 7-oxoheptanoic acid, 7-hydroxyhep ing recombinant host cells according to the invention and tanoic acid, 7-hydroxyheptanal, 7-aminoheptanoic acid, hep heptane-1,7-dioic acid, 7-oxoheptanoic acid, 7-hydroxyhep tamethylenediamine, 1.7-heptandiol, 7-aminoheptanal, tanoic acid, 7-hydroxyheptanal, 7-aminoheptanoic acid, hep 7-aminoheptanol or enantholactam. Exemplary growth pro tamethylenediamine, 1,7-heptandiol, 7-aminoheptanal, cedures include, for example, fed-batch fermentation and 7-aminoheptanol and enantholactam. The invention also pro batch separation; fed-batch fermentation and continuous vides a composition comprising a recombinant host cell separation, or continuous fermentation and continuous sepa according to the invention and a feedstock. In some embodi ration. All of these processes are well known in the art. Fer ments the feedstock is a renewable feedstock, for example mentation procedures are particularly useful for the biosyn wherein the feedstock is selected from the group consisting of thetic production of commercial quantities of di- and glucose, Sucrose, Xylose, fatty acids and glycerol. In some trifunctional C7 alkanes. Generally, and as with non continu embodiments the feedstock is a polyaromatic hydrocarbon, ous culture procedures, the continuous and/or near-continu for example benzene, toluene or shikimate. ous production of heptane-1,7-dioic acid, 7-oxoheptanoic acid, 7-hydroxyheptanoic acid, 7-hydroxyheptanal, 7-amino 3.1 EXAMPLES heptanoic acid, heptamethylenediamine, 1,7-heptandiol. 0252 Having now generally described the invention, the 7-aminoheptanal, 7-aminoheptanol or enantholactam same will be more readily understood by reference to the described previously include culturing a recombinant host following examples, which are provided by way of illustra cell producing heptane-1,7-dioic acid, 7-oxoheptanoic acid, tion and are not intended as limiting. It is understood that 7-hydroxyheptanoic acid, 7-hydroxyheptanal, 7-aminohep various modifications and changes can be made to the herein tanoic acid, heptamethylenediamine, 1,7-heptandiol, 7-ami disclosed exemplary embodiments without departing from noheptanal, 7-aminoheptanol or enantholactam described the spirit and scope of the invention. herein in the presence of Sufficient nutrients and medium to Sustain and/or nearly Sustain growth in an exponential phase. 3.1.1 Conversion of 2,6-Diaminopimelate to Continuous culture under Such conditions can include, for 2-Amino-5,6-Dehydro Pimelic Acid (6-Amino-2- example, 1 day, 2, 3, 4, 5, 6 or 7 days or more. Additionally, Heptene Dioic Acid) and of 2-Aminoheptanedioic continuous culture can include 1 week, 2, 3, 4 or 5 or more Acid to 2-Heptenedioic Acid by Ammonia Lyases weeks and up to several months. Alternatively, recombinant (EC 4.3.1.-) (See, FIG. 10) cells can be cultured for hours, if suitable for a particular application. It is to be understood that the continuous and/or (0253 3.1.1.1 Selection of MAL Enzyme Gene Targets near-continuous culture conditions also can include all time 0254. Upon identification of the pathways involved in intervals in between these exemplary periods. It is further conversion of 2,6-diamino pimelic acid into pimelic acid, understood that the time of culturing the host cell of the which is used in the production of nylon 7.7, and conversion invention is for a sufficient period of time to produce a suffi of pimelic acid into 7-amino heptaonic acid, gene targets cient amount of product for a desired purpose. from these pathways were selected. In particular, enzyme 0249 Fermentation procedures are well known in the art. targets were selected from the ammonia lyase family (EC Briefly, fermentation for the biosynthetic production of hep 4.3.1), including methylaspartate ammonia lyase (MAL, EC tane-1,7-dioic acid, 7-oxoheptanoic acid, 7-hydroxyhep 4.3.1.2), based on its broad Substrate specificity, presence in a tanoic acid, 7-hydroxyheptanal, 7-aminoheptanoic acid, hep wide range of hosts, ability to be cloned and expressed exog tamethylenediamine, 1.7-heptandiol, 7-aminoheptanal, enously, and available crystal structure for rational design 7-aminoheptanol or enantholactam described previously can protein engineering. The MAL genes from Citrobacter ama be utilized in, for example, fed-batch fermentation and batch lomaticus, Clostridium tetanomorphum and Aspergillus separation; fed-batch fermentation and continuous separa oryzae were selected. The MAL enzymes encoded by C. tion, or continuous fermentation and continuous separation. amalonaticus (GeneBank AB005294; fragment 1641 Examples of batch and continuous fermentation procedures 2882878596 . . . 879060NM NM) and C. tetanomorphum are well known in the art. (GeneBank 548141; fragment 756-1997) have significant 0250 In addition to the above fermentation procedures dissimilarity (57% identity). The Aspergillus oryzae MAL using the heptane-1,7-dioic acid, 7-oxoheptanoic acid, 7-hy gene (GeneBank XM 001827609) is evolutionarily diver droxyheptanoic acid, 7-hydroxyheptanal, 7-aminoheptanoic gent (44% identity with Citrobacter amalonaticus MAL, acid, heptamethylenediamine, 1,7-heptandiol, 7-aminohep 39% identity with Clostridium tetanomorphum MAL), which tanal, 7-aminoheptanol or enantholactam producers may show different catalytic properties and/or selectivities. described herein for continuous production of substantial Another unusaual ammonia lyase that is a potential enzyme quantities of these, the heptane-1,7-dioic acid, 7-oxohep for the reductive deamination of 2,6-diaminopimelate and tanoic acid, 7-hydroxyheptanoic acid, 7-hydroxyheptanal, 2-aminoheptanedioic acid, is the D-glucosaminate ammonia 7-aminoheptanoic acid, heptamethylenediamine, 1.7-heptan lyase (EC 4.3.1.9) from Pseudomonas fluorescens (accession diol, 7-aminoheptanal, 7-aminoheptanol or enantholactam number BAD69624). This is an unusual ammonia lyase that producers also can be, for example, simultaneously subjected catalyzes an O.f3-elimination: to further procedures to convert the product to other com D-glucosaminate->2-dehydro-3-deoxy-D-gluconate+ pounds or the product can be separated from the fermentation NH US 2015/0337275 A1 Nov. 26, 2015 20

formed in a total volume of 50 uL and were incubated in a thermocycler (Eppendorf Mastercycler Gradient). Cycles of Earl digestion/ligation were achieved by alternating the tem HOil II. perature between 37°C. and 16°C., which correspond to the optimum temperatures for the Earl digestion and for the liga tion using the T4DNA ligase respectively. The samples were loaded on a 0.7% agarose gel, and the part linker was purified from the residual backbone through agarose gel separation HO and gel extraction (QIAGEN QIAquick Gel Extraction Kit). Expected size fragments were observed after agarose gel O electrophoresis for the preparation of part linker fragments. Ligation of the digested fragment with the annealed oligo 0255 3.1.1.2 Cloning Expression Vectors Containing nucleotides displaying compatible 3 bp overhangs resulted in Gene Targets the formation of a non-cleavable part/linker fragment. 0256 The selected MAL gene targets from Citrobacter amalonaticus, Clostridium tetanomorphum, and Aspergillus 0259. The gene part/linker fusions were then combined oryzae were then cloned into the IPTG inducible plT21-a with their respective vector part/linker fusion using a 2-part backbone usinginABLE technology to generate 14 (pET21-a assembly to construct an E. coli expression vector using the with the Citrobacter amalonaticus MAL gene), I5 (pET21-a part linkers previously generated. The gene and vector part with the Clostridium tetanomorphum MAL gene), and 16 linker fusions were mixed together in an equimolar amount (pET21-a with the Aspergillus Oryzae MAL gene). First, the (0.1 pmol each), which resulted in the expected DNA frag gene and vector DNA are split into truncated parts using ment assembly due to the specific complementary overhangs Software analysis. Potential restriction enzyme, e.g., Earl. formed in the annealed part oligonucleotide and the annealed sites in the selected MAL genes and plT21 vector were linker oligonucleotide from each part. Reactions were incu identified and disrupted by introducing mutations. Earl bated at room temperature for 30 min prior to transformation restriction sites located within coding sequences were dis of high efficiency chemically competent NEB10B E. coli cells rupted, e.g., by incorporation of a silent mutation, as to avoid using 2 ul of the assembly reaction and 10ul of the competent altering the encoded protein sequence. The resulting DNA cells. Transformed cells were plated on LB-Amp-agar and was free of Earl sites, and was used the input sequence for incubated at 37° C. overnight. Five hundred clones were software analysis to design corresponding truncated parts obtained for each assembly. Two random clones were then flanked by Earl (ordered through DNA 2.0), long and short picked from each assembly and the corresponding vectors linker oligonucleotides (ordered through Sigma-Aldrich), were isolated using the QIAGEN QIAprep Miniprep Kit. and long and short part oligonucleotides (ordered through 0260 Restriction analysis was performed on the previ Sigma-Aldrich). The 5'-end of the part and linker oligonucle ously isolated vectors to confirm the construction of the cor otides were phosphorylated by incubating the primer in the rect assembly, e.g., assembly of the Citrobacter amalonaticus present of T4 kinase under suitable reaction conditions (6M gene into the plT21-a backbone. Clones obtained from the oligo, 1 PNK buffer, 1 mM ATP, 10 UT4 kinase, 5 mM DTT, assembly of the Citrobacter amalomaticus gene into the E. 5% (w/v) PEG8000) at 37° C. for 30 minutes, followed by coli expression vector were analyzed using Pvul and BmgBI inactivation of the enzyme at 65° C. for 20 minutes. to identify positive clones for the insertion. Clones obtained 0257 Next, the phosphorylated linker oligos and phos from the assembly of the Clostridium tetanomorphum gene phorylated part oligos were annealed by mixing equimolar into the E. coli expression vector were analyzed using PstI amounts of each oligo (50 uL at 6 LM) and then heating the and AlwNI (FIG. 30, Table 10) to identify positive clones for mixture to 65°C. in athermocycler before gradually lowering the insertion. Clones obtained from the assembly of the the temperature down to 20° C. to form partially double Aspergillus Oryzae gene into the E. coli expression vector Strand part oligos and partially double-strand linker oligos were analyzed using SphI and EcoRV to identify positive having specific 16 bp overhangs (melting temperature above clones for the insertion. The restriction products from each 75° C.). The resulting annealed part oligos (POA) and sample were analyzed usignagarose gel electrophoresis. The annealed linker oligos (LOA) were then diluted to a final expected band pattern was observed for both clones from concentration of 1 M with TE buffer (Tris-EDTA). each assembly confirming the construction of E. coli expres 0258. Then, the part/linker fragments corresponding to the sion vectors harbouring the Citrobacter amalonaticus, selected genes (Citrobacter amalonaticus, Clostridium teta Clostridium tetanomorphum and Aspergillus Oryzae MAL nomorphum and Aspergillus Oryzae MAL) were generated by genes. Assembly of the 17 negative control vector (pET21-a Successive cycles of Earl digestion of the cloned truncated control) was performed using the part/linker fusion pET21-a parts followed by ligation of their respective annealed part part/pET21-a linker was performed as a negative control, and oligos (POA28, POA29 or POA30), their truncated part clones from the generation of the negative control gene-free (TP28, TP29 or TP30), and the annealed linker oligos from construct were analyzed using XmnI and AlwNI. To further the vector backbone (LOA31). Part linker fusions corre confirm proper assembly of the expression vectors, the part sponding to the vector backbone were prepared by ligation of junctions between the 3'-end of the vector backbone and the its annealed part oligos (POA31), its truncated part (TP31) 5'-end of the genes as well as between the 3'-end of the gene and the annealed linker oligos from either genes (LOA28, and the 5'-end of the vector backbone were sequenced from LOA29 or LOA30). Ten and twenty-fold molar excesses of one of the two clones from each assembly. LOA and POA compared to the truncated part were used in 0261 3.1.1.3 Expression of Exogenous MAL Genes in E. the gene and vector reactions respectively to favor ligation coli between the truncated part and the oligonucleotides during 0262 The E. coli constructs previously obtained were each Earl digestion/ligation cycle. The reactions were per used to study the expression of the Citrobacter amalonaticus, US 2015/0337275 A1 Nov. 26, 2015

Clostridium tetanomorphum and Aspergillus Oryzae MAL scaled up to increase production of the soluble proteins for genes in E. coli. The vectors (10 ng DNA from each assem use in enzymatic assays. The cultures were grown, induced, bly) were used to transform electrocompetent BL21 (DE3) E. and harvested as previously described. Briefly, a single clone coli cells, and transformation samples were plated on LB was picked from each transformation plate previously pre Amp-Agar. Transformants were obtained after overnight pared, and 20 ml of LB-Amp medium were inoculated as incubation at 37° C. starting cultures. The ODoo was measured after overnight 0263. A single clone was picked from each assembly incubation at 37° C. and 250 rpm. 16 mL of each starting plate, and was used to inoculate 5 ml of LB-Amp medium as culture (approximately 4.4x10 cells) were used to inoculate a starting culture. The ODoo was measured after overnight 800 mL of LB-Amp, and the cultures were incubated in a 2 L incubation at 37° C. and 250 rpm. Two mL of each starting baffled shake flaskat 37° C. and 250 rpm until an ODoo of 0.6 culture (approximately 4.5x10-6.2x10 cells) were used to to 0.8 was reached. Protein expression was induced by addi inoculate 100 mL of LB-Amp, and the cultures were incu tion of 1 mM IPTG (final concentration), and cultures were bated in a 500 mL baffled shake flask at 37° C. and 250 rpm further incubated at 37° C. and 250 rpm. After 24 h post until an ODoo of 0.6 to 0.8 was reached. Protein expression, induction incubation, the remaining culture was transferred which is controlled by the T7 promoter in pRT21-a, was to four 50 mL falcon tubes, harvested by repeated centrifu induced by addition of 1 mM IPTG (final concentration) and gation at 5000 rpm for 10 minutes (4 rounds of centrifugation cultures were further incubated at 37° C. and 250 rpm. 1 mL per tube; 50 mL of medium per round) and the cell pellets of each sample was taken prior to induction, 4h post-induc stored at -20°C. tion, and 24 h post-induction. Cell growth was checked by 0267 Glass bead lysis was used to disrupt the cells for measuring ODoo. The remaining culture left after 24h post collecting the Citrobacter amalonaticus, Clostridium tetano induction incubation was transferred to a 50 mL falcon tube, morphum and Aspergillus Oryzae MAL protein. Cell pellets harvested by centrifugation at 5000 rpm for 10 minutes and from the 24 hour post-induction samples were re-suspended the cell pellets stored at -20°C. in bead lysis buffer consisting of 0.1 mM Tris, 1 mg/ml 0264. Next, the time-point samples from each assembly Pepstatin A and 200 mM PMSF protease inhibitors. Acid were processed for SDS-PAGE analysis. The samples were washed 212-300 um glass beads (0.4 g per 1 mL of lysis centrifuged at 13000 rpm for 2 min before the supernatant buffer) were added to each tube, and lysed in a Sonicating was removed and the cells were lysed using the Bugbuster water bath in parallel. 250 ul samples from each of the 4tubes protein extraction reagent Supplemented with lysozyme (15 were taken prior to sonication and after 5 minutes and 10 mg/mL) and benzonase (3.4U/LL). The lysis reactions were minutes of cell lysis. The 250 ul samples from each time point then centrifuged at 13000 rpm for 2 min, and the soluble were pooled to give a total sample of 1 mL. Cell free extracts fraction was transferred to a new tube and the insoluble frac were prepared by centrifugation for 10 minutes at 5000 rpm tion was re-suspended in water. 20 Jul of each fraction was and transfer of the supernatants to new tubes. Insoluble mixed with 80 ul SDS-Sample buffer (SDS-Loading Buffer, samples were prepared after re-suspension of the cell pellet 9% DTT and water) and the mixture was heated in a heatblock isolated after 10 min of sonication with an equivalent volume for 5 minutes at 95°C. 10 ul of each sample was then loaded of water. The remaining pellets and Supernatant were stored at on SDS-PAGE 4-20% Tricene gels to analyze the protein -20°C. 20 ul of each pooled fraction was mixed with 80 ul content of the soluble and insoluble fractions. A protein band SDS-Sample buffer (SDS-loading buffer, 9% DTTandwater) at the expected size (45 kDa) was clearly visible in the soluble and the mixture was heated in aheatblock for 5 minutes at 95° fraction after IPTG induction. No corresponding protein was C. 10 ul of each SDS-sample preparation was then loaded on observed in the negative control experiment suggesting that SDS-PAGE4-20% Tricene gels to analyze the protein content the expected 45kDa Citrobacter amalonaticus MAL protein of the soluble and insoluble fractions. Protein bands were and 45 kDa Clostridium tetanomorphum MAL protein were clearly detected in the samples from expression of the Citro successfully expressed as soluble proteins in E. coli from the bacter amalomaticus MAL and the Clostridium tetanomor assembled expression vector constructs No significant pro phum MAL. tein was observed in the insoluble fraction of any sample after 0268 3.1.1.4 Functionality of Exogenously Expressed induction. The 45 kDa Aspergillus Oryzae MAL protein was MAL also expressed in E. coli from the assembled expression vec 0269. The activity of the methyl aspartate ammonia lyase tor construct, but it was expressed as an insoluble protein (a derived from Citrobacter amalonaticus, Clostridium tetano 45 kDa band was visible in the insoluble fraction after induc morphum, and Aspergillus Oryzae was assayed using the I4. tion) (See, FIGS. 11A and 11 B). I5, and I6 strains towards beta-methyl aspartate, in the first 0265. In order to improve the solubility of the Aspergillus place. In Subsequent studies described herein, the activity of oryzae MAL protein, the temperature of induction and/or the these strains were assayed in the context of different sub concentration of inducer was reduced, which in some cases strates, namely, 2-aminopimelate, D.L-lysine, and, finally, has been shown to result in increased protein solubility due to 2,6-. A negative control biotransforma a slower rate in protein synthesis and therefore a higher effi tion of 2-aminopimelate with I17 was also conducted. ciency in protein folding. These parameters were reduced 0270 3.1.1.4.1 Biotransformation Screening of I4. I5, and using the same expression and induction procedure described I6 Strains Toward Beta-Methyl Aspartate above, but SDS-PAGE analysis shows that the Aspergillus oryzae MAL protein was still insoluble (a 45 kDa band was visible in the insoluble fraction after induction). A yeast NH2 expression vector containing the Aspergillus Oryzae MAL CO2H protein may be used to express the protein in soluble form. HOC 0266 Cultures of E. coli expressing the Citrobacter ama lomaticus or Clostridium tetanomorphum MAL genes were US 2015/0337275 A1 Nov. 26, 2015 22

-continued 0280 I5 strain 5 g/L: 2-aminopimelate concentration 5 CO2H NH mM 0281 I5 strain 5 g/L: 2-aminopimelate concentration 10 r 3 mM 0282. I5 strain 20 g/L: 2-aminopimelate concentration 5 0271 Six biotransformations were prepared at either 5 g/L mM or 20 g/L whole cell equivalent of the I4. I5 and I6 lyase 0283 I5 strain 20 g/L: 2-aminopimelate concentration 10 strains. The enzyme concentration was calculated based on mM whole cell equivalent which is the pelleted cell mass after 0284. The biotransformations were placed in a falcon tube centrifugation (wet cell weight) divided by the volume of at 30°C. in a shaker incubator. Aliquots (1 ml) were removed buffer used to resuspend the cells during lysis. The biotrans from the biotransformations after 7, 14 and 21 days. Pro formations were conducted in 3 ml volumes with 10 mM longed incubation was operated to provide best possible DL-threo-beta methyl aspartate. chance of observing lyase activity. The 1 ml aliquots were 0272 Reactions were incubated at 30° C. in an orbital prepared for analysis by heat treatment to remove heat labile incubator and samples removed for analysis by HPLC after 1, proteins. The samples were Subsequently centrifuged at 5 and 18 hours. The 1 ml aliquots were prepared for analysis 10,000 rpm for 2 minutes and filtered through a 0.2 micron by heat treatment to remove heat labile proteins. The samples filter. Analysis of the samples was carried out by HPLC using were subsequently centrifuged at 10,000 rpm for 2 minutes a precolumn derivatisation method. This method comprises and filtered through a 0.2 micron filter. Analysis of the forming a chiral adduct with a Boc-cysteine and ortho-ph samples was carried out by HPLC using a Phenomenex Rezex thaladehyde reagent to form a diastereomeric pair of L- and column with 5 mM sulfuric acid mobile phase using RID and D-2-aminopimelate derivatives. The diastereomers (repre UV detection. Confirmation of lyase activity was affirmed by senting the enantiomers of 2-aminopimelate) can then be detection of mesaconic acid in the samples. The concentra separated by HPLC using a C18 column and a 5 mM potas tion of mesaconic acid formed in each biotransformation was sium phosphate buffer pH 7.0 and acetonitrile gradient at 338 determined by measurement of an external standard mesa nm. The peak areas for L- and D-2-aminopimelate signals in conic acid from Aldrich. each sample were recorded and observed decrease in the L-2-aminopimelate standard is expressed as % relative to the 0273 Mesaconic acid was detected in all the I4 and I5 D-2-aminopimelate. strain biotransformations, confirming functional expression of the lyase enzyme was present in these cell extracts. No 0285. The data show L-2-aminopimelate being degraded activity was observed in the I6 strain which is consistent with in all the biotransformations. The increase in enantiomeric the results observed by SDS PAGE analysis of the expressed excess is consistent with the natural L-enantioselectivity protein which showed the protein was expressed in insoluble expected for these lyase enzymes which preferentially acts on form. The yield of mesaconic acid in all samples analyzed L-beta-methyl aspartate. The lower substrate loadings (5 from the I4 and I5 strains was ~5 mM, which is consistent mM) also showed a greater enantio-enrichment (ie higher with the anticipated activity on 10 mMDL-threo-beta methyl conversion of the available substrate) compared to the higher aspartate, since the wildtype enzyme is L-enantioselective. In Substrate loadings which is consistent with the anticipated the I4 and I5 strains complete conversion was observed after lyase activity. just 1 hour incubation. 0286 The presence of unreacted 2-aminopimelate was confirmed via mass spectrometry. The M+H equal to 176.09 0274 3.1.1.4.2 Biotransformation Screening of I4 and I5 was observed in the sample and matched the simulated mass Strains Toward 2-Aminopimelate spectrum for CHNO. On the full Sweep mass spectrum, the enoic acid product (M--H=159.06) was not observed. However, when the MS accumulated ions for 20 seconds over a 20 m/z, wide mass isolation the correct product signal was observed and matched with the simulated mass spectrum for HOC ~s CO2H C7H10O4. 0287. A sample of the rac-2-aminopimelic acid starting material was also analyzed. Using the same MS methods carried out on this standard clearly displayed the expected 0275 Eight biotransformations were prepared varying molecular ion for 2-aminopimelate. However, repeating the levels of enzyme and Substrate concentrations (2 variable, 2 MS ion accumulation for 20 seconds over a 20 m/z, wide mass level stat design) to examine I4 and I5 enzyme activity isolation (as had been carried out on the previous biotrans towards 2-aminopimelate. The biotransformations were con formation sample) revealed the presence of the enoic acid ducted in 3 ml Volumes with varying loadings as indicated product as a mass fragment. below: 0288 3.1.1.4.3 Negative Control Biotransformation of 0276 I4 strain 5 g/L: 2-aminopimelate concentration 5 2-Aminopimelate with I17 Strain mM 0289 Four biotransformations were prepared with vary 0277 I4 strain 5 g/L: 2-aminopimelate concentration 10 ing level of cell extract and substrate to match the levels used mM in the experiment described in 3.1.1.4.1. The biotransforma 0278 I4 strain 20 g/L: 2-aminopimelate concentration 5 tions were conducted in 3 ml reaction Volumes as indicated mM below: 0279 I4 strain 20 g/L: 2-aminopimelate concentration 10 0290 I7 strain 5 g/L, 2-aminopimelate concentration 5 mM mM US 2015/0337275 A1 Nov. 26, 2015

0291 I7 strain 5 g/L: 2-aminopimelate concentration 10 general degradation of L-lysine from DL-lysine was observed across the eight biotransformations which is consistent with riz I7 strain 20 g/L, 2-aminopimelate concentration 5 the expected stereoselectivity of potential lytic activity aris ing from lyase functioning in the I4 and I5 strains. Fa I7 strain 20 g/L; 2-aminopimelate concentration 10 0308. In order to gather further evidence to support the functioning of lyase activity in these biotransformations, a to:- The biotransformations were incubated at 30°C. in sample was analyzed by MS to attempt to detect the desired a shaker incubator and samples taken after 8, 13 and 21 days. enoic acid product in the sample. The MS signal correspond Analysis of the reaction was carried out by HPLC following ing to the unreacted lysine (M+H=176.09) was observed in the method described in the experiment described above in the sample and matched the simulated mass spectrum for 3.1.1.4.1. CHNO. 0295 The data show there is some loss of the L-enanti 0309. On the full sweep mass spectrum, the enoic acid omer in the I17 strain. However the relative decrease in the product (M+H=130.09) was not observed. However, as for L-enantiomer is less than the equivalent biotransformations the amino-pimelic acid reactions, when the MS accumulated with the I4 and I5 strains, particular when the second time ions for 20 seconds over a 20 m/z, wide mass isolation the points incubating for 2 weeks are compared. The trend correct product signal was observed and matched with the depicted below clearly shows the I4 and I5 strains in the simulated mass spectrum for CHNO. biotransformations have a consistently higher level of L-2- 0310. A sample of the DL-lysine starting material was also aminopimelate degradation compared to the I17 negative analyzed. Using the same MS methods carried out on this control strain across the 2x2 statistical design experiment standard clearly displayed the expected molecular ion for examining cell extract and Substrate loadings. DL-lysine. However, repeating the MS ion accumulation for 0296 3.1.1.4.4 Biotransformation Screening of I4 and I5 20 seconds over a 20 m/z, wide mass isolation (as had been CFE with DL-Lysine carried out on the previous biotransformation sample) revealed the presence of the enoic acid product as a mass fragment. NH2 0311 3.1.1.4.5 Biotransformation Screening of I4 and I5 CFE with Meso-2,6-Diaminopimelic Acid --~~ Hs oc-1S-1N1 SNL, + NH NH2 NH2

0297 Eight biotransformations were prepared in a similar HOC ---> CO2H fashion as used to test for lyase activity towards 2-ami NH2 NH2 nopimelate. The I4 and I5 lyase strains were tested at varying concentrations of cell extract and DL-lysine substrate as fol HOC ---> CO2H + NH lows: 0298 I4 strain 5 g/L, DL-lysine concentration 5 mM 0299 I4 strain 5 g/L, DL-lysine concentration 10 mM 0312 Six biotransformations were prepared to test for 0300 I4 strain 20 g/L, DL-lysine concentration 5 mM lyase activity towards meso-2,6-diaminopimelic acid. The I4. 0301 I4 strain 20 g/L, DL-lysine concentration 10 mM I5 and I7 strains were tested with 5 mM meso-2,6-diami 0302 I5 strain 5 g/L, DL-lysine concentration 5 mM nopimelic acid. Varying concentrations of cell extract were 0303 I5 strain 5 g/L, DL-lysine concentration 10 mM examined as follows: 0304 I5 strain 20 g/L, DL-lysine concentration 5 mM 0313 I4 strain 5 g/L; meso-2,6-diaminopimelic acid con 0305 I5 strain 20 g/L, DL-lysine concentration 10 mM centration 5 mM 0306 The biotransformations were placed in a falcon tube 0314. I4 strain 20 g/L: meso-2,6-diaminopimelic acid at 30°C. in a shaker incubator. Aliquots (1 ml) were removed concentration 5 mM from the biotransformations after 7, 14 and 21 days. Pro 0315 I5 strain 5 g/L; meso-2,6-diaminopimelic acid con longed incubation was operated to provide best possible centration 5 mM chance of observing lyase activity. The 1 ml aliquots were 0316. I5 strain 20 g/L: meso-2,6-diaminopimelic acid prepared for analysis by heat treatment to remove heat labile concentration 5 mM proteins. The samples were Subsequently centrifuged at 0317. I17 strain 5 g/L: meso-2,6-diaminopimelic acid 10,000 rpm for 2 minutes and filtered through a 0.2 micron concentration 5 mM filter. DL-lysine samples were analyzed using a chiral HPLC 0318. I17 strain 20 g/L: meso-2,6-diaminopimelic acid column (Chirobiotic T2 Astec) to monitor enantiomer ratios concentration 5 mM to determine lyase activity. Samples were removed from the 0319. The biotransformations were placed in a falcon tube biotransformations after 7, 14 and 21 days. at 30°C. in a shaker incubator. Aliquots (1 ml) aliquots were 0307 The results indicate loss of L-lysine in all the removed from the biotransformations after 4 days, further biotransformations with 5 mM lysine. Some variations in sample acquisitions are pending prior to HPLC analysis. enantioselective action on DL-lysine was observed, where 0320 A sample acquired after 4 days was analyzed by MS. the concentration of L-lysine exceeded the concentration of The analysis confirmed the presence of meso-2,6-diami D-lysine. Owing to the poor UV absorbance (detector set to nopimelic acid which matched the simulated mass signal for 210 nm) and low signal to noise ratio of lysine in the HPLC C7H1N2O4. chromatogram the possibility of a false positive L-lysine 0321. The desired enoic acid signal was not observed in detection cannot be excluded from these results. However, in this mass spectrum in the biotransformation sample. How US 2015/0337275 A1 Nov. 26, 2015 24 ever, when ion accumulation was introduced for 0.5 sec then 27904-27913; Asymmetric bioreduction of activated alkenes the corresponding enoic acid signal (M+H=174.08) was using cloned 12-oxophtodienoate reductase isoenzymes observed in the sample. OPR-1 and OPR-3 from Lycopersicon esculentum (Tomato): A striking switch of stereopreference: Hall et al. Angew. 3.1.2 Conversion of 2-Amino-5,6-Dehydro Pimelic Chem. Int. Ed. 2007, 46, 3934-3937; Crystal structure of Acid (6-Amino-2-Heptenedioic Acid) to 2-Amino 12-oxophytodienoate reductase 3 from tomato—Self inhibi Heptane Dioic Acid and 2-Heptenedioic Acid or its tion by dimerization: Breithaupt et al. PNAS 2006, 103, CoA Ester to Pimelic Acid or its CoA Ester by 14337-14342). In addition, XenA (EC 1.3.1.31) from Enoate Reductases (EC 1.3.1.31) or Enoyl-CoA Pseudomonas putida, TER (EC 1.3.1.44) from Euglena gra Dehydrogenases (EC 1.3.1.62) (See, FIGS. 10A and cilis, PECR and MECR from Homo sapiens, acdA (EC 1.3. 10B) 99.-) from Pseudomonas putida, ACADS (EC 1.3.99.-) from 0322 3.1.2.1 Selection of Enoate Reductase Targets Bos Taurus and PimC/pimD (EC 1.3.1.62) from 0323. Upon identification of the pathways involved in the Rhodopseudomonaspalustris were selected for expression in conversion of two unsaturated intermediates of pimelic acid, E. coli and tested for the reduction of 2-heptene-dioic acid i.e., 2-amino-5,6-dehydro pimelic acid and 2-heptenedioic and 2-heptenedioic acid-CoA.

Target gene GenBank Accession Bacilius subtilis (YoM) D84432 fragment 242791-243807 Solanum lycopersicum (OPR1) NM OO1247852, fragment 108-1238 Solanum lycopersicum (OPR3) NM OO1246944; fragment 94-1284 Pseudomonas putida Xenobiotic reductase A (XenA, 39.8 kDa, 363 AA) AAF02538.1 Eugiena gracilis mitochondrial Trans-2-enoyl-CoA reductase (TER, 43.8 AAW 66853.1 kDa, 405AA) Homo sapiens peroxisomal trans-2-enoyl-CoA reductase (PECR, 32.5 NP 060911.2 kDa, 303AA) Homo sapiens mitochondrial trans-2-enoyl-CoA reductase (MECR, 40.5 NP 057095.2 kDa, 373AA) Pseudomonas puttida Acyl-CoA dehydrogenases (acdA, 41.6 kDa, NP 745629.1 383AA) Bos taurus acyl-CoA dehydrogenase (ACADS, 44.6 kDa, 412AA) NP 001029573.1 Rhodopseudomonas palustris Pimeloyl-CoA dehydrogenase (PimC, NP 94.9051.1 large subunit, 44.5 kDa, 396AA) Rhodopseudomonas palustris Pimeloyl-CoA dehydrogenase (Pim D, NP 949050.1 Small subunit, 40.5 kDa,389AA) acid, into their corresponding Saturated compounds 2-amino 0325 Other enoate reductases that are potential candi heptanedioic acid and pimelic acid, enzyme targets were dates for the reduction of the 2,3-double bond are those from selected from the enoate reductase family (EC 1.3.1), such as Clostridium species due to their broad substrate specificity EC 1.3.1.31, EC 1.3.1.8; EC 1.3.1.9; EC 1.3.1.10, EC 1.3.1. (Chiral compounds synthesized by biocatalytic reductions: 31; EC 1.3.1.38; EC 1.3.1.39; EC 1.3.1.44 or pimeloyl-CoA Simon et al. Angew. Chem. Ed. Engl. 1985, 24, 539-553; dehydrogenase in EC 1.3.1.62 such as PimC/Piml) or ThnJ/ Properties and mechanistic aspects of newly found redox ThnK. The enoate reductase may also be in the EC 1.3, such enzymes from anaerobes Suitable for bioconversions on pre as EC 1.3.8.1 or EC 1.3.99.3; EC 1.3.99.B10 or Syntrophus paratory scales: Simon et al. Pure & Appl. Chem. 1992, 64. aciditrophicus 2,3-didehydropimelyl-CoA reductase and 1181-1186). However, those enzymes were not cloned and homologs thereof. based on to their different substrate speci expressed for initial proof of concept due to their high sensi ficity towards C.B-unsaturated carboxylic acids and their sta tivity towards oxygen (Enoate reductases of Clostridia— bility (Stereocomplementary bioreduction of C.B-unsatur Cloning, sequencing and expression: Rohdich et al. J. Biol. ated dicarboxylic acids and dimethyl esters using enoate Chem. 2001,276,5779-5787), but may be useful catalysts for the final pathway construction in an anaerobic host. In par reductases—Enzyme- and Substrate-based stereocontrol: ticular, the following enoate reductases should be useful for Stueckler et al. Org. Lett. 2007, 9,5409-5411). the reduction of the double bonds in the target 2-enoyl C7 0324. In particular, the enoate reductase genes from Bacil compounds: lus subtilis (YoM) (GeneBank D84432; fragment 242791 0326 enr2-Enoate reductase (Y16137) EC 1.3.1.31 from 243807) and Solanum lycopersicum (OPR1 and OPR3) Clostridium kluyveri: Whole cells of Clostridium kluyveri (GeneBank NM 001247852, fragment 108-1238 and were found to reduce a broad range of 2-enoates in the pres GeneBank NM 001246944; fragment 94-1284, respec ence of molecular hydrogen. The Substrates were: tiglic acid, tively) were selected because they had been previously cloned acrylic acid, methacrylic acid, dimethylacrylic acid, (E)-pen and Successfully expressed in E. coli, crystal structures of the tenoic acid, (E)-hexenoic acid, Sorbic acid, cinnamic acid YoM and OPR3 enzymes were reported, which enables pro 19, (E)-2-butenoic acid and (E)-2-methyl-2-butenoic acid, tein engineering through rational design (The 1.3 A crystal (E)- and (Z)-3-methyl-2-pentenoic acid, p-methoxy-cin structure of the flavoproteinYoM reveals a novel class of old namic acid and p-nitro-cinnamic acid Bihler, M., Giesel, H., yellow enzyme: Kitzing et al. J. Biol. Chem. 2005, 280, Tischer, W. and Simon, H. (1980) Occurrence and the pos US 2015/0337275 A1 Nov. 26, 2015

sible physiological role of 2-enoate reductases. FEBS Lett. carboxylate formation pathway indicate that enoyl-CoA 109, 244-246. A partial sequence of the gene encoding the reduction occurs in this organism. Other hypothetical enzyme was identified Rohdich, F. Wiese, A., Feicht, R., pimeloyl-CoA dehydrogenases found in databases include Simon, H. and Bacher, A. (2001) Enoate reductases of ThnJ (D9PTNO) and ThnK (D9PTM9) from Sphingomonas Clostridia. Cloning, sequencing, and expression. J. Biol. macrogolitabida—found in the gene cluster for tetralin deg Chem. 276,5779-5787). Since the genome of C. kluyveri has radation via the beta-oxidation pathway, PimC (Q6N3IO) and been sequenced, we found the full sequence of the gene Piml) (Q6N3I1) from Rhodopseudomonas palustris and (YP 001394144.1, GENEBANK). PimC (A5ES10) and Piml) (A5ES09) from Bradyrhizobium 0327 enr2-Enoate reductase (Y09960) EC 1.3.1.31 from sp. PimC/Piml) were included in the selection for expression C. tyrobutyricum (previously Clostridium sp. La 1): 2-Enoate and biotransformation to evaluate reduction of 2-heptene reductase was found in Clostridium sp. growing on (E)- dioic acid and its CoA ester. butenoate and purified to homogeneity. It accepted (E)-2- 0330. Other dehydrogenases useful to the invention methylbutenoate and cinnamate Elshahed M. S. Bhu include GDH Glutaryl-Coenzyme A Dehydrogenase from pathiraju V. K. Wofford N Q. Nanny M A. McInerney M.J. Desulfococcus multivorans, a non-decarboxylating glutaco (2001) Metabolism of benzoate, cyclohex-1-ene carboxylate, nyl-coenzyme A-forming glutaryl-coenzyme A dehydroge and cyclohexane carboxylate by “Syntrophus aciditrophicus' nase was characterized in the obligately anaerobic bacteria strain SB in syntrophic association with H(2)-using microor Desulfococcus multivorans Wischgoll S. Demmer U. ganisms. Appl Environ Microbiol. 67: 1728-1738), (E)-al Warkentin E, Günther R, Boll M, Ermler U. (2010) Structural pha-formylamino-cinnamate, and 4-methyl-2-pentenoate basis for promoting and preventing decarboxylation in glu Bühler, M., Giesel, H., Tischer, W. and Simon, H. (1980) taryl-coenzyme A dehydrogenases. Biochemistry. 49: 5350 Occurrence and the possible physiological role of 2-enoate 5357. The enzyme was overexpressed and its crystal struc reductases. FEBS Lett. 109,244-246. The gene encoding enr ture was determined which makes it a useful target for protein reductase was identified and the enzyme was overexpressed engineering to catalyse reversible oxidation and reduction of in E. coli under anaerobic conditions Rohdich, F. Wiese, A., the C7 equivalents (2,3-didehydropimeloyl-CoA and Feicht, R., Simon, H. and Bacher, A. (2001) Enoate reduc pimeloyl-CoA) of its native C5 substrates glutacolnyl-CoA tases of Clostridia. Cloning, sequencing, and expression. J. and glutaryl-CoA Biol. Chem. 276,5779-5787. 0331 3.1.2.2 Cloning Expression Vectors Containing 0328 enr2-Enoate reductase (Y16136) EC 1.3.1.31 from Gene Targets Moorella thermoacetica (previously Clostridium thermoace ticum): An antiserum against enoate reductase from C. 0332 The selected YajM, OPR1, and OPR3 gene targets tyrobutyricum cross-reacted with a protein of the thermo from Bacillus subtilis and Solanum lycopersicum were then philic Clostridium thermoaceticum. The gene encoding the cloned into the IPTG inducible plT21-a backbone using reductase was identified and the enzyme was successfully inABLE technology, as described in detail in Section 3.1.1.2) overexpressed in E. coli Rohdich, 2001. to generate I1 (Bacillus subtilis YoM gene in pRT21-a). I2 0329. In addition to enoate reductases, pimeloyl-CoA (Solanum lycopersicum OPR1 gene in pRT21-a), and I3 dehydrogenases (EC 1.3.1.62) should catalyse the reduction (Solanum lycopersicum OPR3 gene in pRT21-a). of the double bond in 2-heptenedioic acids and the corre 0333 Briefly, potential Earl sites in the selected genes and sponding CoA esters. Pimeloyl-CoA dehydrogenase activity the vector backbone were disrupted to prevent interference was found in cell extracts of Syntrophus aciditrophicus grown with the part/linker fusion preparation involving Earl diges in a co-culture with Desulfovibrio sp. strain G11 with ben tion/ligation cycles. No Earl sites were observed in the Bacil Zoate or in a pure culture with crotonate and is involved in the lus subtilis Yd Menoate reductase gene, but one and two Earl benzoate degradation pathway (Elshahed MS, Bhupathiraju sites were detected in the Solanum lycopersicum OPR1 and V. K., Wofford N Q. Nanny M A. McInerney M.J. (2001) OPR3 genes, respectively, which were disrupted by incorpo Metabolism of benzoate, cyclohex-1-ene carboxylate, and rating silent mutation(s) in the restriction site. Disruption of cyclohexane carboxylate by “Syntrophus aciditrophicus' the Earl sites 1 and 2 in the Solanum lycopersicum OPR3 gene strain SB in syntrophic association with H(2)-using microor sequence resulted in an inadvertent Earl site being created, ganisms. Appl Environ Microbiol. 67: 1728-1738). It is which was only discovered after the sequences were synthe believed that a pathway for cyclohexane carboxylate forma sized. The residual Earl site was removed using the tion from crotonate involves reduction of glutaconyl-CoA to Quikchange single-site directed mutagenesis kit from Strate glutaryl-CoA and 2,3-didehydropimelyl-CoA to pimeloyl gene. Forward and reverse primers were designed to disrupt CoA. The enzymes responsible for this activity have not been the residual Earl site, and the primers were phosphorylated as identified but the S. aciditrophicus but a BLAST search using described in the section 3.1.1.2 prior to the Quikchange reac the Desulfococcus multivorans nondecarboxylating glutaryl tion. The Quikchange and the negative control reactions were CoA dehydrogenases (EC 1.3.99.B10) as a query identified prepared as follows: six sequences annotated as acyl-CoA dehydrogenases in S. aciditrophicus:YP 462067.1 (73% identity), YP 460269.1 (39%), YP 461117.1 (36%), YP 460428.1 (34%), TP27 Negative YP 460268.1 (32%), and YP 463031.1 (28%). Wischgoll Reagent Quikchange control et al., (2009) compared sequences of a number of glutaryol 10x Pfuturbo Reaction 5 ul 5 ul Co-enzyme A dehydrogenases and identified primary struc Buffer TP27 1 Il 1 Jul ture similarities between YP 462067.1 and the Desulfococ Forward Primer 3 ul 3 ul cus multivorans gene that indicated that the former is also Reverse Primer 3 ul 3 ul nondecarboxylating. Although the dehydrogenase reaction is dNTP Solution 5 ul 5 ul the reverse of what is required, the proposed cyclohexane US 2015/0337275 A1 Nov. 26, 2015 26

-continued either genes. A negative control gene-free assembly was also prepared by ligation of the vector backbone annealed part TP27 Negative oligos, its truncated part, and its annealed linker oligos, Reagent Quikchange control resulting in self-assembly of the vector backbone. Molecular Biology Grade 32 ul 33 ul 0340. A ten and twenty-fold molar excesses of linker and Water part oligos to the truncated part was used in the gene and Pfuturbo enyme 1 Il vector reactions to favor ligation between the truncated part and the oligonucleotides during each Earl digestion/ligation Total Volume 50 ul 50 ul cycle. The 50LL Earl digestion/ligation reactions were incu bated in a thermocycler (Eppendorf Mastercycler Gradient), 0334) The reactions were run in athermocycler (95°C. 0.5 alternating the temperature between 37° C. and 16°C. Upon minutes; 35 cycles of 95°C. 0.5 minutes, 55° C. 1 minute, 68° completion of the cycles, samples were loaded on a 0.7% C. 10 minutes; 68°C. 1 minutes. 4°C. hold), and then treated agarose gels, and the expected size fragments were observed with DpnI (1 uL) to digest the methylated parental DNA at for preparation of the part/linker fusions, e.g., Bacillus sub 37°C. for 2 hours. The digestion reactions (3 uL) were used tilis YajM part/pBT21-a linker and Solanum lycopersicum to transform TOP10 electrocompetent E. coli cells, and the OPR1 part/pT21-a linker as well as the 3 kb vector back transformation samples (100 uL) were plated onto LB-Kan bone; or plT21-apart/Solanum lycopersicum OPR3 linker— Agar. About 100 clones were obtained on the Quikchange 5.3 kb and Solanum lycopersicum OPR3 part/pET21-a reaction plate compared to none on the negative control plate linker—1.2 kb as well as a 1.8 kb band corresponding to the after overnight incubation at 37°C. Ten clones were picked vector backbone. The correct size bands were then excised from the plate, the corresponding plasmids were purified, and from the gel, and the DNA was gel extracted using the analyzed by Earl digestion. The absence of a 300 bp band on QIAGEN QIAquick Gel Extraction Kit. The DNA concen the agarose gel indicated that the Earl site located in the OPR3 tration ranged from 14.4 ng/ul to 49.6 ug/ul in a total Volume gene was deleted. Earl digestion showed that several clones of 30 ul (DNA conc. range 5.4 nM to 35.6 nM). were positive for the removal of the residual Earl site. DNA 0341 Combination of the gene part/linker fusions with sequencing confirmed the clone's sequence. their respective vector part/linker fusion was then carried out 0335 Vector from clone 3 (a clone positive for removal of through a 2-part assembly as described in section 3.1.1.2. Earl) was used to transform electrocompetent TOP10 E. coli Briefly, the gene part linkers and the vector part linkers were cells. QIAGEN. Hi-Speed Plasmid Purification Midi Kit was incubated at room temperature for 30 min prior to transfor used to prepare a large DNA stock from the culture, and the mation of high efficiency OPR3 chemically competent DNA sample was Subsequently concentrated using the NEB10B E. coli cells using 2 ul of the assembly reaction and QIAGEN PCR purification kit. The final DNA concentration 10 ul of the competent cells. Transformed cells were plated on and the volume of the sample was 470.0 ng/ul in 120 ul (DNA LB-Amp-agar, and incubated at 37° C. overnight. Five hun conc. 199.8 nM). dred clones were obtained for each assembly. One or two 0336 Three Earl sites were detected in the vector back random clones were selected from each assembly, and the bone pET21-a, which were disrupted by mutating the last corresponding vectors were isolated using the QIAGEN guanidine base of the restriction site into a thymine (corre QIAprep Miniprep Kit. sponding to the first cytosine into an adenine on the reverse 0342. Restriction was performed to confirm vector con complement sequence). struction, as described in section 3.1.1.2. Briefly, clones 0337 The Earl free sequences were then used as input obtained from the assembly I1 were analyzed using PstI and sequences in the part designer Software to design the corre HincII to identify positive clones for the insertion of the sponding truncated parts flanked by Earl sites, as discussed in Bacillus subtilis YoM gene into the E. coli pET21-1 expres Section 3.1.1.2. The synthesized Bacillus subtilis YdM sion vector, clones obtained from the assembly 12 were ana (TP25), Solanum lycopersicum OPR1 (TP26), and Solanum lyzed using PVul and Psil, and clones obtained from the lycopersicum OPR3 (TP27) truncated parts were inserted into assembly 13 were analyzed using BglII and Mlul. The restric the DNA2.0 pJ201 vector. The pET21-a truncated part vector tion products were run on an agarose gel, and the expected was fully synthesized with introduction of a fragment includ band pattern was observed for the clones tested from each ing the chloramphenicol resistance geneto produce the vector assembly, confirming the construction of E. coli pET21-a TP31. expression vectors harbouring the Bacillus subtilis YajM, the 0338 Phosphorylation and subsequent annealing of the Solanum lycopersicum OPR1 and Solanum lycopersicum inABLE oligonucleotides were performed as described in OPR3 enoate reductase genes. In addition, the part junctions section 3.1.1.2. between the 3'-end of the vector backbone and the 5'-end of 0339 Next, part linker fusions were prepared by ligating the genes as well as between the 3'-end of the gene and the the 5'-end of the truncated part with its corresponding part 5'-end of the vector backbone were sequenced to confirm the oligo annealed fragment, and the 3'-end of the truncated part construction. with the linker oligo annealed fragment, as previously 0343 3.1.2.3 Expression of Exogenous Enoate Redu described in section 3.1.1.2. Part linker fusions correspond catase Genes in E. coli ing to the Bacillus subtilis YajM, Solanum lycopersicum 0344) Initially, the cloned enoate reductase genes were OPR1 and Solanum lycopersicum OPR3 genes were prepared expressed in E. coli using 1 mM IPTG and 37 induction by ligation of their respective annealed part oligos, their trun temperature, as described above in Section 3.1.1.3. Briefly, cated part, and the annealed linker oligos from the vector the OD600 was measured for each culture pre-induction and backbone. Part linker fusions corresponding to the vector at various time points after induction. The remaining culture backbone were prepared by ligation of its annealed part oli left after 24h post-induction incubation was transferred to a gos, its truncated part, and the annealed linker oligos from 50 mL falcon tube, harvested by centrifugation at 5000 rpm US 2015/0337275 A1 Nov. 26, 2015 27 for 10 minutes, and the protein content of the samples were 36 uM dichlorophenolindophenol, 0.3 mM N-ethylmaleim analyzed using SDS-PAGE. A large protein band at the ide, 1.5 mM phenazine methosulfate, 60 uMacyl-CoA and expected size (35 kDa) was clearly visible in the soluble enzyme (25-55ug/ml). Reactions were started by the addition fraction from the expression study of the Bacillus subtilis of phenzine methosulfate (Le et al., 2000). YcMenaoate reductase after IPTG induction. A large band 0353 Reactions were started by the addition of substrate was visible in the insoluble fraction from the expression of the (2,3-didehydropimely-CoA or 2-heptenedioic acid) (60 uM). Solanum lycopersicum OPR1 enoate reducatase after IPTG Reactions were carried out at 30° C. for 4 hours, and then induction, Suggesting that the expected 42 kDa OPR1 protein stopped by addition of 2 MHCl at 10% final volume. Products expressed from 12 was insoluble in E. coli. A protein band at were analyse by LC-MS. Controls used were reactions with the expected size (44 kDa) was clearly detected in the soluble no Substrate and reactions with boiled enzymes. portion after IPTG induction of Solanum lycopersicum OPR3 0354) Results: ACADS, PimlD, TER and XenA catalyzed enoate reductase 13, as well as in the insoluble fraction, the reduction of 2-hexenedioic acid to adipic acid (See, FIG. Suggesting that the expected 44 kDa Solanum lycopersicum 27A). MECR, PER, TER and XenA catalyzed the reduction OPR3 enoate reductase protein was expressed in E. coli as of trans-2,30didehydroadipoyl-CoA to adipoyl-CoA (See, partial soluble form. FIG. 27B). No formation of the corresponding C7 substrates 0345 Based on reports of expression functional expres by these enzymes were observed under the reaction condi sion of the Solanum lycopersicum OPR1 enoate reductase in tions tested, indicating that the enzymes have lower activity E. coli ptimized in the literature, the conditions for IPTG for the C7 2-enoyl Substrates and that enzyme engineering concentration (range of 0.1-1 mM IPTG) and induction tem will be required to improve the activity of the enzymes to be perature (range of 16-37°C.) were optimized in an attempt to useful. improve the solubility of Solanum lycopersicum OPR1 enoate reductase in E. coli. (Strassner et al. JBC 1999, 274, p. 3.1.3 Conversion of Pimeloyl-CoA and Pimeloyl 35067-35073; Hall et al. Angew. Chem. Int. Ed. 2007, 46, p. Acp to Pimelic Acid Semialdehyde by Fatty Acyl 3934-3937). The same expression protocol and the same CoA Reductases (Aldehyde Forming FAR) (EC 1.2. SDS-PAGE sample preparation as described in section 3.1.1. 1.50) or Acyl-Acp Reductases (EC 1.2.1.80) (See, 3. was used for the solubility optimization experiments. FIG. 2) 0346. The time-point samples from each assembly were 0355 3.1.3.1 Selection of Target Enzymes then processed for and analyzed using SDS-PAGE analysis, 0356. Fatty acyl CoA reductases are involved in wax ester as described in section 3.1.1.3, which showed that lowering synthesis in plants and other higher eukaryotes for cuticle the induction temperature to 16°C. and the inducer concen wax formation. In Archaea and bacteria, these enzymes are tration to 0.1 mM was effective to partially solubilize the involved in long-chain alkane biosynthesis, although their Solanum lycopersicum OPR1 enoate reductase expressed in physiological function in these organisms is still not always E. coli. clear. The formation of wax esters in plants begins with the 0347 E. coli strains expressing the 3 enoate reductases elongation of acyl CoAS to form long-chain fatty acids were scaled up to 800 ml cultures and assayed for activity. (LFCA) (or fatty acyl-CoAs), which are then reduced via the 0348. The other enoate reductases, XenA, TER, PECR, decarbonylation pathway or acyl CoA reduction pathways MECR, ACADS, PimC/Piml) were expressed using standard (Cheesbrough and Kolattukudy, 1984). techniques in E. coli, and purified on Histrap columns, 0357 Hydrocarbons are formed via the decarbonylation 0349 Biotransformation Reactions pathway. This pathway is initiated by a fatty acyl-CoA reduc 0350 Reduction of the double bond of both C6 (2-hexene tase-catalysed reaction to produce the aldehyde, followed by dioic acid and 2,3-didehydroadipolyl-CoA) and C7 sub decarbonylation catalysed by an aldehyde decarbonylase to strates (2-heptenedioic acid and 2,3-didehydropimeloyl yield the Cn-1 alkane (Cheesbrough and Kolattukudy, CoA) were evaluated. 1984). 0351. Enzyme assays for enoyl-CoA reductase activity 0358. The acyl CoA reduction pathway is also initiated by were based on published methods (Bergler H. et al., (1993). a reduction of LCFAs to aldehydes, followed by a further Protein EnvM is the NADH-dependent enoyl-ACP reductase reduction of aldehydes by an aldehyde reductase to produce (Fabi) of Escherichia coli. Reactions were performed in even-chained primary alcohols (Millar et al., 1999) (Kolat phosphate buffer (100 mM, pH7.5), containing 300 uM tukudy, 1971). Similar pathways are thought to exist in NADPH, NADH or FAD depending on the cofactor require prokaryotes (Schirmer et al., 2010). ment of the enzyme, and purified enzyme (25-55 ug/ml). 0359 Both pathways are initiated by reduction of long Reactions were started by addition of the substrates, 2.hep chain acyl-CoAs to aldehydes catalysed by fatty acyl-CoA tane dioic acid or 2,3-didehydropimloyl-CoA. Reactions reductase (FAR; EC 1.2.1.50) in a reversible reaction. In the were incubated at 30° C. for 4 hours, and stopped by the acyl CoA reduction pathway, the aldehydes are reduced fur addition of 2M HCL (10% v/v). Products were analysed by ther to the alcohols. This can beachieved in either one or two LC-MS. Controls were comprised of reaction mixtures with steps, depending on the organism. The two step reduction is no substrate and reaction mixtures with inactivated (boiled initiated using the acyl-CoA reductase-catalysed reaction, for 5 min) enzymes. which releases the fatty aldehyde for a second reduction 0352 Enzyme assays for acyl-CoA dehydrogenase activ catalysed by a fatty aldehyde reductase. The one step reduc ity were based on published methods (Le W. Abbas A S. tion is catalysed by a bi-functional acyl CoA reductase which Sprecher H. Vockley J and Schulz H (2000). Long-chain reduces the acyl CoA directly to the alcohol without releasing acyl-CoA dehydrogenase is a key enzyme in the mitochon the aldehyde intermediate. Thus, fatty acyl CoA reductases drial B-Oxidation of unsaturated fatty acids. Biochimica et can be broadly divided into two types: (1) alcohol-forming Biophysica Acta. 1485: 121-128). Assays were performed in acyl-CoA reductases and (2) aldehyde generating acyl-CoA 1 ml containing 50 mM potassium phosphate buffer (pH 7.6), reductases. There are examples of both types in prokaryotes US 2015/0337275 A1 Nov. 26, 2015 28 and eukaryotes. Since the target is pimelic semialdehyde, the TABLE 1-continued aldehyde-generating fatty acyl CoA reductases are the most Candidate aldehyde forming Acyl-CoA reductases promising candidates for reduction of pimeloyl CoA. from various Organisms. 0360 Preliminary literature and database searches Shortest revealed that the overwhelming majority of acyl-CoA reduc Protein substrate No. Level tases studied or identified by were alco name tested (no. of Reduction of hol-forming FARs in both eukaryotes and prokaryotes. How accession carbon amino to Evi ever, one of the early examples of an aldehyde-generating number Organism atoms) acids alcohol dence reductase came from Acinetobacter calcoaceticus—Acrl. AAB35106 Botryococcus 16 26 No Enzyme Therefore, the primary sequence of this enzyme was used in (partial) bratini activity database searches for other aldehyde-generating enzymes. *Level of evidence-Enzyme activity means enzyme has been purified either heterologously 0361. A general rule for acyl-CoA reductases seems to be or homologously, or activity has been measured in crude extracts or other cellular fractions that the shorter proteins (295-350 aa) tend to be aldehyde generating enzymes, whereas longer polypeptides (500+ aa) 0364 Acinetobacter calcoaceticus Acrl (P94129) (Re are alcohol producers. However, most of the shorter enzymes iser, S., and Somerville, C. (1997): Isolation of mutants of deposited in databases had putative or hypothetical annota Acinetobacter calcoaceticus deficient in wax ester synthesis tions with no further information beyond sequence data and, and complementation of one mutation with a gene encoding a for this reason, they were not considered. fatty acyl coenzyme A reductase. J Bacteriol 179, 2969 2975): An aldehyde-generating acyl-CoA reductase has been 0362. The candidates were chosen using the following purified and contains 295 amino acids, with a molecular criteria:—(1) The enzyme must generate aldehyde only or weight of 32.5 kDa. This enzyme has been expressed in E. there must be scope to eliminate alcohol formation by protein coli and cell-free extracts were shown to contain acyl-CoA engineering; (2) The must be evidence of existence of the reductase activity. Acyl-CoA substrates ranging from C12 to protein i.e. protein has been identified and purified or isolated C24 in length were assayed (but not the C-odd substrates). in cellular fractions; (3) The enzyme was tested using acyl The corresponding aldehyde was generated in all cases except CoA or acyl-ACP substrates; (4) The enzyme must not be part for the C12 substrate. It would be worth examining the sub of multi-enzyme complex unless activity has been demon strate range of this enzyme further and testing its activity on strated independently of the complex. Once candidates were shorter length acyl-CoAS. Even if pimeloyl CoA is not a identified, further bioinformatics and literature searches were substrate, there is scope for directed evolution to extend the performed to provide additional data to rank the selections. Substrate range. This enzyme has been named directly in a 0363 Table 1 shows the results of database searches car patent about enhancing production of fatty acid derivatives ried out against a known aldehyde-forming enzyme (Acrl). (US20110256599). Each of the candidates has been purified and assayed in vitro. 0365 AAR from Synechococcus elongatus PCC 7942 The candidates were ranked and grouped according to their (YP 400611) (Schirmer, A., Rude, M.A., Li, X., Popova, E., Suitability. The top ranking enzymes (1-3) are considered the and Del Cardayre, S. B. (2010). Microbial Biosynthesis of best candidates for further study, the intermediate ranked Alkanes. Science 329,559-562): This acyl-ACP reductase is enzymes (4 and 5) are considered to be good candidates but in the cyanobacterium S. elongatus and designated AAR. The may require further investigation and the lower ranked enzyme is thought to be involved in alkane biosynthesis. It enzyme (6) would require more effort to obtain enzyme activ has been expressed and purified in E. coli MG 1655. Both ity data, due to the absence of the protein sequence and the CoA and ACP derivatives are substrates. Thus, octadecanal is lack of a sequenced genome. formed from either oleoyl-CoA and oleoyl-ACP without fatty alcohol formation. However, oleoyl-ACP was the preferred TABLE 1. substrate, with a KM of 8 uM compared to 130 uM for oleoyl-CoA. The enzyme is NADPH-dependent and requires Candidate aldehyde forming Acyl-CoA reductases magnesium for activity. No other substrates have been tested. from various organisms. The selectivity for aldehyde formation and the activity with Shortest CoA derivatives makes this enzyme a good candidate for Protein substrate No. Level testing for pimeloyl-CoA reduction. Furthermore, there area name tested (no. of Reduction of accession carbon amino to Evi number of orthologues of this gene in closely related organ number Organism atoms) acids alcohol dence isms (Schirmer et al., 2010), although AAR is the best stud ied. The following patents are associated with the work from P94129 Acinetobacter 14 295 No Enzyme Caicoaceticus activity (Schirmer et al., 2010): WO 2009/140695 and WO 2009/ YP 400611 Synechococcus 18 341 No Enzyme 140696. elongatus PCC activity 0366 LuxC (BAF92773) from Photobacterium phospho 7942 BAF92773 Photobacterium 14 488 No Enzyme reum (Boylan, M., Miyamoto, C. Wall, L., Graham, A., and LuxC phosphoretin activity Meighen, E. (1989). Lux C, D and E genes of the Vibrio EC 1.2.1.50 fischeri luminescence operon code for the reductase, trans YP 95.9769 Marinobacter 8 661 Yes, but may Enzyme ferase, and synthetase enzymes involved in aldehyde biosyn aquaoeiei VT8 be possible activity thesis. Photochemistry and photobiology 49, 681-688; Lee, to truncate YP 959486 Marinobacter ND 513 Yes, but may Enzyme C.Y., and Meighen, E. A. (1997). Cysteine-286 as the site of aquaoeiei VT8 be possible activity acylation of the Lux-specific fatty acyl-CoA reductase. Bio to truncate chim Biophys Acta 1338, 215-222; Wang, X., and Kolat tukudy, P. E. (1995). Solubilization and purification of alde US 2015/0337275 A1 Nov. 26, 2015 29 hyde-generating fatty acyl-CoA reductase from green alga region as described previously. It is not clear whether or not Botryococcus braunii. FEBSLett370, 15-18): LuxC is part of FAR and FAcoR are entirely different enzymes, since two a multi-enzyme complex (LuxCDE) consisting of a fatty separate (and competing) research groups have studied them. acyl-CoA reductase, fatty acyl synthetase and fatty acyl 0369 Aldehyde-generating acyl-CoA reductase from thioesterase, and is involved in bioluminescence. LuxC is a Botryococcus braunii (AAB35106) (Wang, X., and Kolat well-studied enzyme and has been cloned, expressed and tukudy, P. E. (1995). Solubilization and purification of alde purified in E. coli. The enzyme exhibits acyl-CoA reductase hyde-generating fatty acyl-CoA reductase from green alga activity with the C14 substrate, tetradecanoyl-CoA (myris Botryococcus braunii. FEBS Lett 370, 15-18): This enzyme toyl-CoA), which is reduced to the corresponding aldehyde, is an aldehyde-generating acyl-CoA reductase from the tetradecanal. No other substrates have been tested. Although microalga, Botryococcus braunii. The enzyme has been puri the LuxC is part of a multisubunit complex in vivo, the fied from the wild-type organism and assayed using labeled enzyme has been successfully purified from the wild-type activated fatty acid, 1-14C-palmitoyl-CoA. The enzyme organism and has been expressed and partially purified from was shown to produce the corresponding aldehyde and was E. coli. In both cases, LuxC was active independently of the NADH-dependent. The substrate range has not been studied other subunits. This makes LuxC a good candidate to convert and this would need to be addressed. Furthermore, the protein pimeloyl-CoA to pimelic acid semialdehyde. needs to be cloned, expressed and purified in E. coli. Acces 0367 FA.coAR from Marinobacter aquaeolei VT8 (YP sion number AAB35106 refers to a partial 26 amino acid 95.976.9.1) (Willis, R. M., Wahlen, B. D., Seefeldt, L. C., and N-terminal sequence from the purified enzyme. The full pro Barney, B. M. (2011). Characterization of a fatty acyl-CoA tein sequence is not deposited on UNIPROT BRENDA or reductase from Marinobacter aquaeolei VT8: a bacterial NCBI databases. Sequencing of the genome is underway enzyme catalyzing the reduction of fatty acyl-CoA to fatty (http://www.jgi.doe.gov/). There might be scope to isolate the alcohol. Biochemistry 50, 10550-10558): This fatty acyl CoA gene using primers for the N terminal sequence, but the gen reductase (designated FAcoAR) was recently discovered in eral lack of information about the enzyme may complicate its the gamma proteobacterium Marinobacter aquaeolei VT8. It exploitation for metabolic engineering. is the first prokaryotic example of a bi-functional acyl-CoA 0370 Thus, a number of acyl-CoA reductases have been reductase, catalyzing reduction of acyl-CoA derivatives to identified that are suitable candidates to be tested for reduc alcohol. Crucially it is the only enzyme that has been tested tion of pimeloyl-CoA to pimelic semialdehyde. The enzyme with substrates below C12, and shows activity with octanoyl from Acinetobacter calcoaceticus, Acrl, appears to be the CoA in vitro (Willis et al., 2011). Although this enzyme best candidate, as it is the best characterized of the enzymes generates alcohol from acyl-CoA, the rationale for its inclu identified even though there is no information on activity with sion is that there is clear evidence that it has abroad substrate short chain, C-odd substrates. If initial tests turn out to be range from octanoyl-CoA (C8) to arachidonoyl-CoA (C20). negative, the enzyme would still be a good target for directed Furthermore, the C-terminal end shows high sequence evolution to obtain pimeloyl CoA reduction. homology to the Acrl enzyme, so the N-terminal region may 0371 AAR from S. elongatus and its orthologues as well be the alcohol-forming domain. Therefore, it would be worth as LuxC are also promising candidates. Although these while to test the effect of removing the portion of the gene enzymes operate on relatively long chain fatty acyl CoA coding for the N-terminal region of the polypeptide, with the substrates, they have not been tested with short chain sub aim of generating a functional, aldehyde-generating enzyme. strates. Again, directed evolution may provide a promising A potential problem with this approach is that the enzyme route to obtain the desired substrate selectivity if the initial might be inactive or the aldehyde may not be released from tests are negative. the truncated enzyme complex. This enzyme has not been 0372. The two enzymes from M. aquaeolei VT8 may also tested for activity against acyl ACP derivatives. be promising, although some protein engineering is likely to 0368 FAR from Marinobacter aquaeolei VT8 (YP be required to prevent oxidation of the aldehyde. These 959486) (Hofvander, P., Doan, T. T. P., and Hamberg, M. enzymes are Suitable candidates because they are known to (2011). A prokaryotic acyl-CoA reductase performing reduc act on the shortest chain fatty acyl CoAs out of all the fatty tion of fatty acyl-CoA to fatty alcohol. FEBS Letters 585, acyl CoA reductases identified. The gene encoding the 3538-3543): This fatty acyl CoA reductase (designated FAR) enzyme from B. braunii has not yet been identified. This was originally identified as a fatty aldehyde reductase makes it the most difficult of the enzymes to work with, but it (Wahlen, B.D., Oswald, W. S. Seefeldt, L. C., and Barney, B. could be investigated if other options are not successful. M. (2009). Purification, characterization, and potential bac Similar enzymes are also known in higher plants (e.g. Pisum terial wax production role of an NADPH-dependent fatty sativum) and provide further options for testing, but, again, aldehyde reductase from Marinobacter aquaeolei VT8. Appl are not well characterised. Environ Microbiol 75,2758-2764), but has since been shown 0373 Thus, the above six enzymes have been selected to have acyl-CoA reductase activity (Hofvander, P., Doan, T. because there are literature reports that they have been pre T. P., and Hamberg, M. (2011). A prokaryotic acyl-CoA pared, purified (partially of fully) either recombinantly or reductase performing reduction of fatty acyl-CoA to fatty natively in cellular fractions and assayed in vitro. Their nucle alcohol. FEBS Letters 585, 3538-3543). It is very similar to otide and primary amino acid sequences are deposited on the the fatty acyl-CoA reductase described above, but it is also main protein databases BRENDA and UNIPROT, except for known to act on acyl ACP derivatives Activity was observed the enzyme from Botyrococcus braunii. Results from the with C16-CoA up to C20-CoA, but has not been tested with studies have been published in peer-reviewed journals. The shorter Substrates. The enzyme is 513 amino acids long and Substrates tested range from octanoyl-CoA (C8) up to arachi has the same C-terminal domain associated with aldehyde nadoyl-CoA (C20) and also included acyl-ACP as substrate generation as Acrl. It is possible that this enzyme could also in the case of AAR from Synechococcus elongates. The cor be truncated, which would mean cleaving off the N-terminal responding aldehydes are formed as products. However, to US 2015/0337275 A1 Nov. 26, 2015 30 date, none of the enzymes have been tested with pimeloyl well as with hexanoyl-CoA. The SucD enzyme from CoA or any other C-odd number substrates, presumably Clostridium kluyveri was obtained in the insoluble fraction because the metabolism of these substrates is invariably and activity could not be detected in the soluble fraction, assumed to follow the same pattern as the C-even number hence a broken cell Suspension of Clostridium kluyveri was Substrates and, until now, there has been no reason to test them used in the Subsequent biotransformation assays. against these substrates. Therefore, it is highly likely that the 0377 Biotransformations to assay for activity of the enzymes will catalyse the reduction of pimeloyl-CoA. Even if enzymes with pimeloyl-CoA with purified (Histrap columns) pimeloyl CoA is not a substrate, the enzymes would provide enzymes produced in E. coli were performed under aerobic a good starting points for directed evolution or targeted pro conditions. A total reaction volume of 250 uL made of DTT (5 tein engineering. mM), MgSO4 (5 mM), NADH (1.2 mM), adipoyl-CoA or 0374. Other candidates amongst the long fatty acyl-CoA pimeloyl-coA (1 mM) in tris buffer (50 mM, pH7.5) was reductase in EC 1.2.1.50 for the conversion of pimeloyl-CoA prepared in a 96-well plate. The reaction was started by add to pimelic acid semialdehyde include, but is not limited to, ing 10 to 15ull of the soluble fraction of the proteins prepared reductases from the following organisms: Pisum sativum in E. coli. Each reaction was done in triplicate. A “control (Vioque et al., 1997. Resolution and purification of an alde buffer was prepared by replacing the enzymatic preparation hyde-generating and an alcohol-generating fatty acyl-CoA with Tris (50 mM, pH7.5). This control was prepared in reductase from Pea leaves (Pisum sativum L.), Archives of triplicate. A "control enzyme’ was prepared using the mix Biochemistry and Biophysics, 340 (1), 64-72); Vibrio fischeri ture described above but replacing the hexanoyl-CoA with (P12748) (Boylanet al., 1985. Functional identification of the water for each enzymatic preparation. This control was done fatty acid reductase components encoded in the luminescence in duplicate. operon of Vibrio fischeri, Journal of Bacteriology, 163(3), 0378 All the mixtures (“Assay”, “Control buffer”, “Con 1186-1190); Brassica oleracea and B. napus (Q39342) trol enzyme') were prepared in parallel and were incubated in (Kolattukudy, 1971. Enzymatic synthesis of fatty alcohols in a 96-well plate incubated overnight at 37 or 32°C. under 200 Brassica oleracea. Archives of Biochemistry and Biophysics, rpm shaking before being spinned down. The Supernatants 142(2), 701-709); Clostridium butyricum (Day et al., 1978. were added into an appropriate LC/MS 96-well plate for Partial purification and properties of acyl-CoA reductase analysis by LC/MS. from Clostridium butyricum, Archives of Biochemistry and 0379 Biotransformations to assay for the reduction of Biophysics, 190(1), 322-331); Anas platyrhynchos (Ishige, pimeloyl-CoA to pimelic acid semialdehdye using enzyme T.: Tani, A.; Takabe, K. Kawasaki, K.; Sakai, Y.: Kato, N. fractions of Clostridium kluyveri, were performed as follows: Wax Ester Production from n-Alkanes by Acinetobacter sp. Culture of Clostridium kluyveri and Enzymatic Fraction Strain M-1: Ultrastructure of Cellular Inclusions and Role of Preparation Acyl Coenzyme A Reductase. Appl Environ Microbiol 2002, 0380 2 strains of C. kluyveri (DSM555 and DSM563) 68, 1192-1195). Arabidopsis thaliana (Doan, T. T. P.; Carls were grown in flask on ethanol (20 mL/L culture) and Succi son, A. S.; Hamberg, M.; Bülow, L.; Stymine, S.; Olsson, P. nate (5 g/L) in 40 mL of the modified DSM-52 medium Functional expression of five Arabidopsis fatty acyl-CoA containing: K2HPO4 (0.31 g/L), KH2PO4 (0.23 g/L), NH4Cl reductase genes in Escherichia coli. Journal of plant physiol (0.25 g/L), MgSO4x7H2O (0.20 g/L), trace element solution ogy 2009, 166,787-796; Hooks, M.A.; Kellas, F.; Graham, I. SL-10 (1 mL) (10 mL HCl (25%, pH7.7M)/L solution; 1.5 A. Long-chain acyl-CoA oxidases of Arabidopsis. Plant J. g/L. FeC12x4H2O: 70 mg/L ZnCl2: 100 mg/L MnCl2x4H2O: 1999, 20, 1-13); Homo sapiens, R. P.; Wang, Y.; Mohsen, 6 mg/L H3BO3; 190 mg/L CoCl2x6H2O: 2 mg/L CuCl2x A.-W.; He, M.; Vockley, J.; Kim, J-J P Structural basis for 2H2O: 24 mg/L. NiCl2x6H2O: 36 mg/L Na2MoC)4x2H2O), Substrate fatty acyl chain specificity: crystal structure of Selenite-tungstate solution (1 mL) (0.5 g/L NaOH, 3 mg/L human very-long-chain acyl-CoA). Na2SeO3x5H2O: 4 mg/L Na2WO4x2H2O), yeast extract (1 0375 From these aldehyde-forming dehydrogenases in g/L), resaZurin 90.5 g/L), NaHCO3 (2.5 g/L), seven vitamins EC 1.2.1.50 described above, LuxC, the NADPH specific solution (1 mL) (100 mg/L vitamin B12: 80 mg/L p-aminoen acyl-CoA reductase from Photobaterium phosphoreum (Ac Zoic acid; 20 mg/L D(+)-biotin; 200 mg/L nicotinic acid; 100 cession number BAF92773.1) was chosen as an example to mg/L calcium pantothenate;300 mg/L pyridoxine hydrochlo demonstrate the formation of pimelic acid semialdehyde ride; 200 mg/L thiamine-HClx2H2O), cysteine-HClxH2O form pimeloyl-CoA. Other interesting dehydrogenases are (0.25 g/L), Na2Sx9H2O (0.25 g/L). The medium was pre found in EC 1.2.1.76 and EC 1.2.1.10. For example pared strictly anaerobically. Clostridium kluyveri DSM555 has a CoA dependent succi (0381. After 4-5 days of growth at 30° C., the culture was nate semialdehyde dehydrogenase (SucD, EC 1.2.1.76. harvested and the pellet weighed. The pellet was lysed using Accession number AAA92347.1), as well as 2 otheraldehyde the Y-perRyeast extraction reagent flushed during 1 h with N dehydrogenases (Accession number EDK34221.1) and an gas. All procedures were performed in an anaerobic cabinet. acetaldehyde dehydrogenase (EC 1.2.1.10, Accession num The soluble fraction obtained from the C. kluyveri cultures ber EDK33116.1) that may be used to convert pimloyl-CoA were used in the exemplification reactions. to pimelic acid semialdehyde. An additional enzyme in EC 0382 Biotransformation reactions with enzyme fractions 1.2.1.10, PduB, a CoA-dependent propionaldehyde dehydro from C. kluyveri Strains were performed in an anaerobic genase from Propionibacterium feudenreichi Subssp. Sher cabinet as follows: The reactions were carried out in Tris manii (Accession number D7GD28) was selected. buffer (50 mM, pH7.5), DTT (5 mM), MgSO4 (5 to 10 mM), 0376. These 5 selected enzymes were cloned in the NADH (1 mM), NADPH (1 mM), adipoyl coenzyme A or pET151/Topo vector and expressed in E. coli BL21 (DE3) or pimeloyl coenzyme A (0.5 to 1 mM) in a total volume of 200 Rosetta2 (DE3) using standard techniques. Activity was con uL. The reaction was started by adding 15 uL of enzymatic firmed by monitoring the reaction at 340 nm for disappear preparation (soluble or insoluble fraction). These reaction ance of NADH or NADPH with acetyl-CoA as substrate, as were done in triplicate. A “control buffer was prepared by US 2015/0337275 A1 Nov. 26, 2015 replacing the enzymatic preparation with Tris (50 mM, pH7. carboxylic acids by Nocardia aldehyde oxidoreductase 5). This control was prepared in triplicate. All the mixtures requires a phosphopantetheinylated enzyme. J Biol Chem (Assay”, “Control buffer') were prepared in parallel and 2007, 282, 478-485.) were incubated overnight at 30°C. (or 22°C. for P phospho 0388 Expression study was then initiated using standard reum) under 200 rpm shaking before being spinned down. conditions (1 mMIPTG, 37° C. pre- and post-induction tem The supernatants were added into an appropriate LC/MS peratures) and was expressed in E. coli under IPTG induction 96-well plate for analysis by LC/MS. according to the previously reported protocol, and Scaled up 0383. The detection of the adipic-(C6-) and pimelic-(C7-) to produce enough protein for the activity assay. semi-aldehydes was performed using the 6538 Accurate (0389) 3.14.3 Activity Assays mass Q-TOF coupled with the 1290 Infinity LC with HTC/ HTS autosampler (Agilent). A Synergi-2.5u FusionRP col 0390 The activity assay was based on spectrophotometric umn (Phenomenex, Ref: 00B-4423-B0) was used for the detection of CAR activity towards sodium benzoate as posi analysis. The authentic standards of the C6- and C7-semi tive control and Sodium pimelate. The conditions for the assay aldehydes were detected in negative mode using ultrapure were taken from Applied and Environmental Microbiology water supplemented with 0.1% (v/v) of formic acid and aceto 2004, 70, p.1874-1881. nitrile supplemented with 5% (v/v) of ultrapure water as 0391) Each assay comprised of final concentrations of 50 running buffers. The method used to detect the semi-alde mM substrate (100 ul), 100 mM magnesium chloride (100 ul) hydes authentic standards was used to analyse the reactions. 50 mM tris-hydrochloric acid, 1 mM ethylenediaminetet raacetic acid, 1 mM dithiothreitol added together from one 0384 Of the enzymes expressed in E. coli and obtained in stock solution (200 ul), 10 mMadenosine tri-phosphate (100 the soluble fraction that were assayed under aerobic condi ul), 1.5 mM nicotinamide-dinucleotide phosphate hydrate tions, all enzymes displayed detectable activity with acetyl (reduced) (100 ul) and the biocatalyst, which was charged as CoA and hexanoyl-CoA as determined spectrophotometri cell free extract to give a final concentration of 4.3 g/L whole cally. However, only the CoA-dependent propionaldehyde cell equivalents. The reactions were carried out at ambient dehydrogenase from Propionibacterium feudenreichii sub temperature (-20 degC.) in a 2 mL cuvette at pH 7.5. Each ssp. Shermanii (Accession number D7GD28) produced assay was charged first with magnesium chloride (100 ul), 50 pimelic acid semialdehyde from pimeloyl-CoA as deter mMtris-hydrochloric acid, 1 mMethylenediaminetetraacetic mined by LC-MS, despite the fact that only a small fraction of the enzyme was obtained in the soluble fraction. The enzymes acid, 1 mM dithiothreitol (200 ul), 10 mM adenosine tri from C. kluyveri are known to be oxygen sensitive, so the lack phosphate (100 ul) and 1.5 mM nicotinamide-dinucleotide of activity observed by the 3 enzymes from C. kluyveri could phosphate hydrate (reduced) (100 ul) and absorbance mea be due to the assay conditions. Thus, C. kluyveri cell frac surements recorded at thirty second (30s) intervals from time tions, produced under anaerobic conditions with Succinate as Zero (0 s) to two minutes (120 s). Biocatalyst was then inducer in the medium, and assayed under anaerobic condi charged and absorbance recorded at thirty second (30s) inter tions in the native strains, were used to determine activity of vals from time Zero (0s) to two minutes (120 s). Lastly the SucD (EC 1.2.1.76). Enzyme fractions of both strains of C. Substrate was charged and absorbance recorded at thirty sec kluyveri (DSM555 & DSM563) used, produced detectable ond (30s) intervals from time Zero (0s) to three minutes (180 amounts of pimelic acid semialdehyde from pimeloyl-CoA as S). analysed by LC-MS. The following assays were 3.1.4 Conversion of Pimelic Acid to Pimelic Acid performed: Experiment ID substrate biocatalyst comment Semialdehyde by Carboxylic Acid Reductases (EC KG1112-021-20-CAR-1SB Sodium rCAR Positive control 1.2.99.-) benzoate KG1112-021-20-CAR-2PA Pimelic acid rCAR 0385 3.1.4.1 Selection, Cloning, and Expression of Car KG 1112-021-20-CAR-3AA Adipic acid rCAR boxylic Acid Reducatse Targets KG 1112-021-2O-W-1SB Sodium Water Negative control benzoate using water 0386 Enzyme targets for the reduction of pimelic acid KG1112-021-37-BL21-1SB Sodium BL21 Negative control into pimelic acid semialdehyde were selected from the car benzoate using host cells with empty vector boxylic acid reductase family (EC 1.2.99.-, such as EC 1.2. KG1112-021-37-BL21-2PA Pimelic acid BL21 Negative control 99.6). In particular, the Nocardia carboxylic acid reducatse is using host cells a monomeric enzyme that displays a large Substrate specific with empty vector ity towards aromatic Substrates and was successfully cloned KG 1112-021-37-BL21- Adipic acid BL21 Negative control and expressed in E. coli (Nocardia sp. Carboxylic acid reduc 3AA using host cells tase: Cloning, expression, and characterization of a new alde with empty vector hyde oxidoreductase family: He et al. Appl. Environ. Micro biol. 2004, 70, 1874-1881). The enzyme was therefore a good 0392 Results from Carboxylic Acid Reductase Assays candidate for an initial Screening to monitor conversion of 0393. The results of the spectrophotometric assay measur pimelic acid to pimelic acid semialdehyde. ing NADPH consumption over time for each of the assays 0387. The Nocardia sp. carboxylic acid reductase (CAR) above is given in FIG. 13. When corrected for background gene (GeneBank AY495697) was expressed in E. coli BL 21 rate of NADPH consumption, (negative control experiments along with Nocardia phosphopantetheline transferase using the Substrate and biocatalyst that harboured the empty (PPTase) and used as a cell-free extract following cell lysis vector, Table 2), it is clear that both adipic acid and pimelic and centrifugation to remove insoluble fraction. (See Venki acid is reduced at a higher rate than the model Substrate, tasubramanian, P.; Daniels, L.; Rosazza, J. P. N. Reduction of benzoate. Thus, it was successfully demonstrated that car US 2015/0337275 A1 Nov. 26, 2015 32 boxylic acid reductase enzyme has consumed NADPH and náez, and Eduardo Santero, 2010. Tetralin-Induced and has activity towards sodium benzoate, adipic acid and pimelic ThnR-Regulated Aldehyde Dehydrogenase and B-Oxidation acid. Genes in Sphingomonas macrogolitabida Strain TFA. Appl. Environ Microbiol. 76: 110-118). Amongst those three fami TABLE 2 lies, thioesterases present the best candidates for the hydroly sis of pimeloyl-CoA due to the large number of characterized NADPH NADPH consumption enzymes and their broad substrate specificity in terms of consumption (corrected for chain length (C-3 to C-26) and in terms of functional groups (uncorrected) background) (Dicarboxylyl-CoA substrate). Sodium benzoate -OOO172 -OOOO91 0396 3.1.5.1 Acid-Thiol Ligases (EC 6.2.1.-) and CoA Adipic acid -OOO125 -OOO18O Transferases (EC 2.8.3.-) Pimelic acid -OOOO96 -OOO103 0397 Acid thiol ligases catalyse the formation of carbon sulfur bonds with concomitant hydrolysis of adenosine triph 0394 Thus, CAR from Nocardia catalysed the reduction osphate (ATP). The CoA ligases have not been tested rigor of pimelic acid to pimelic acid semialdehyde with >100% of ously or systematically for reversibility. No crystal structures the activity for benzoic acid. Other suitable enzymes for the are available, so the active site structure is not known. There reduction of pimelic acid from EC 1.2.99.6 include for fore, it is not clear whether AMP. pyrophosphate or magne example that of Streptomyces griseus (Suzuki et al., 2007. sium ions would be needed for the reverse reaction, and in GriC and Gril) constitute a carboxylic acid reductase vitro assays should be done in the presence and absence of involved in grixaZone biosynthesis in Streptomyces griseus. J combinations of these cofactors. Of course, the ultimate goal Antibiot (Tokyo). June; 60(6):380-7). In addition, ThnG, a is to obtain the reverse reaction in whole cells, and this might non-acylating aldehyde dehydrogenase in Sphingomonas and be more challenging. The later reactions in the pathway will, Cupriavidus species, converts pimelic acid semialdehyde to of course, pull the equilibrium in the thiolytic direction. Nev pimelic acid during tetralin degradation (See, FIG. 8) and ertheless, this might be a difficult reaction to reverse because may be reversible. the hydrolysis of ATP and subsequent hydrolysis of pyro phosphate make the forward reaction exergonic. Therefore, it 3.1.5 Conversion of Pimeloyl-CoA and/or Pimeloyl is important to measure the effect of different ATP:AMP Acp to Pimelic Acid by Thioester Hydrolases (EC ratios on reversibility of the reaction in vitro before attempt 3.1.2.-), Pimeloyl-CoA Ligases (EC 6.2.1.-), or CoA ing to engineer the cells. If the ATP:AMP ratio is critical, it Transferases, (EC 2.8.3.-) might be necessary to manipulate the cellular energy charge to make the intracellular ATP concentration low enough to 0395. Enzymes that are able to catalyse the hydrolysis or allow the reverse reaction to proceed. This, in turn, may cause transfer of CoASH or acyl-acyl-carrier protein thioesters further difficulties with obtaining good productivity, since it include thioester hydrolases from EC 3.2.1.-, such as EC might be necessary to use ATP synthase inhibitors or uncou 3.1.2.2, EC 3.1.2.14 EC 3.1.2.18; 3.1.2.19, 3.1.2.20 EC 3.1. plers to achieve low enough ATP concentrations. In principle, 2.21, acid-thiol ligases from EC 6.2.1.-, such as EC 6.2.1.3. none of these problems is insurmountable, but development EC 6.2.1.5; EC 6.2.1.14, EC 6.2.1.20; EC 6.2.1.23 and CoA of the engineered biocatalysts might become complicated. transferases from EC 2.8.3.-, such as EC 2.8.3.12 and EC Hydrolysis of Acyl-CoA would therefore require the genera 2.8.3.13 or or the gene product of ThnH that catalyses the tion of ATP during the acid thiol ligase-catalysed reaction reversible transfer of CoA to pimelic acid (Aroa López which is thermodynamically disfavoured compared to the Sánchez, Belén Floriano, Eloisa Andujar, Maria José Her reverse reaction:

O O O O

HO--~. S-CoA + AMP + PP, --> HO--~~ OH + COASH + ATP

0398. However, these enzymes are still useful if found to be reversible. Several CoA ligases are known to act on pimeloyl-CoA:

Over express- Purifi Enzyme EC Source Reaction Gene Size ion cation

bioW 6.2.1.14 Lysinibacdius pimelic acid -> Yes 28

-continued

Over express- Purifi Enzyme EC Source Reaction Gene Size ion cation pauA 6.2.1.14 Pseudomonas pimelic acid -> Yes 74 Yes Yes, (AJO12480) catabolic mendocina35 pimeloyl-CoA kDa E. coi homo pimeloyl-CoA adipic acid -> geneity synthase adipoyl-CoA -> aZelaoyl-CoA SCS 6.2.1.5 Thermococcus Succinyl-CoA Yes Succinyl-CoA kodakaraensis -> Succinate synthetase 9. 6.2.1.n Bacilius pimelic acid -> No No Yes, 34 pimeloyl-CoA negaterium pimeloyl-CoA fold synthetase 9. pimeloyl-CoA Lavandula vera pimelic acid -> No No No synthase L. pimeloyl-CoA 9. plant soluble Pisum sativum pimelic acid -> No 53 No No acyl-CoA L. pimeloyl-CoA kDa synthetase azelaic acid -> aZelaoyl-CoA 9. plan Pisum sativum pimelic acid -> No No No peroxisomal L. pimeloyl-CoA acyl-CoA synthetase 9. 6.2.1.23 Rattus CS-C16 No No No microsomal norvegicus dicarboxylates dicarboxylate- ->CoA esters CoA ligase 9. 6.2.1.23 LP-1 strain pimelic acid -> No No No pimeloyl-CoA pimeloyl-CoA synthetase

BioW (Uniprot:P22822)—from Lysinibacillus sphaericus Berthod C, Frutiger-Hughes S. Hughes G, Shaw NM. (1999) (Bacillus sphaericus) Purification, characterization, DNA sequence and cloning of 0399 Pimelic acid CoA synthase (6-carboxyhexanoate a pimeloyl-CoA synthetase from Pseudomonas mendocina CoA ligase) was discovered in Bacillus sphaericus using cell 35. Biochem J. 340:793-801. The enzyme activates dicar free extracts. It is involved in the biotinbiosynthesis pathway. boxylic acids which are subsequently metabolized via the The enzyme activity was determined indirectly by a reaction beta-Oxidation pathway. The enzyme was purified to homo coupled with 7-keto-8-aminopelargonic acid synthetase and geneity and the partial amino acid sequence was determined. microbiological assay using Saccharomyces cerevisiae/1. PauA showed a broader substrate range than bioW, and The gene encoding pimelic acid CoA synthase (bioW) was accepts pimelate (100% relative activity), adipate (72%) and identified together with 7-keto-8-aminopelargonic acid syn aZelate (18%). The enzyme was between one and two orders thetase (bioF) by complementation studies of E. coli mutants of magnitude more active than bioW involved in the biotin with the library carrying sequences from the B. sphaericus biosynthesis pathways. The gene pauA was identified and the genome Gloeckler R, Ohsawa I, Speck D. Ledoux C, Ber ligase was successfully overexpressed in E. coli. The ligase nard S. Zinsius M. Villeval D. Kisou T. Kamogawa K. Lem pauA from Pseudomonas mendocina was selected as an oine Y. (1990) Cloning and characterization of the Bacillus example from this class because of its Substrate range and sphaericus genes controlling the bioconversion of pimelate specific activity (preferred substrate is pimelic acid, Binieda into dethiobiotin. Gene. 87:63-70. BioW was overexpressed et al., 1999. Purification, characterization, DNA sequence in E. coli and purified to homogeneity Ploux O, Soularue P. and cloning of a pimeloyol-CoA synthetase from Pseudomo Marquet A, Gloeckler R. Lemoine Y. (1992) Investigation of nas mendocina 35. Biochem. J. 340: 793-801). the first step of biotin biosynthesis in Bacillus sphaericus. O HO Purification and characterization of the pimeloyl-CoA syn thase, and uptake of pimelate. Biochem J. 287: 685-690. The enzyme was active in the presence of disodium pimelate, ATP, --~'s + CoA + ATP -e- CoASH and MgCl2. The products of the reaction were CO2656 pimeloyl-CoA and AMP. no ADP was detected. The enzyme Pimelic acid was specific only for ATP and pimelate. Other nucleotides O O (such as GTP AMP) and alternative substrates (succinate, glutarate, adipate. Suberate) were not accepted. --~~. -- PauA (GenBank: AJO12480.1) from Pseudomonas men CO)1063 docina 35 (EC 6.2.1.14) Pimeloyl-CoA 04.00 Catabolic pimeloyl-CoA ligase, pauA, is involved AMP + diphosphate in degradation of dicarboxylic acids (C6-C10) in Pseudomo nas mendocina 35 Binieda A. Fuhrmann M. Lehner B. Rey US 2015/0337275 A1 Nov. 26, 2015 34

04.01. An alternative might be the Rhodopseudomonas found to be a precursor in biotin biosynthesis, there are no palustris pimA, which likewise has been expressed in E. coli data available about the enzyme or the gene. and shown to have ligase activity with a broad range of Plant Soluble Acyl-CoA Synthetase from Pisum sativum L. Substrates (C7-Cadicarboxylic acids), although pimelic acid 04.05 The activity was found in the soluble fraction of pea was not the preferred substrate (Harrison and Harwoo, 2005. protein extracts (7). The enzyme required ATP/Mg" and The pimEABCDE operon from Rhodopseudomonas palus CoASH. The molecular weight of the ligase was determined tris mediates dicarboxylic acid degradation and participates by western blottechnique using polyclonal antibodies against in anaerobic benzoate degradation. Microbiol. 151:727-736). the ligase from Lysinibacillus sphaericus. The enzyme For this reason the pauA gene from Pseudomonas mendocina accepted pimelic acid as well as azelaic acid, but not nonanoic seems to be a better choice and was chosen for exemplifica acid. GTP could be used instead of ATP, but no reaction was tion. observed for ADP. The gene encoding the soluble pimeloyl SCS (Uniprot Q5JEN7) from Thermococcus kodakaraensis CoA synthetase was not identified. (EC 6.2.1.5) Plant Peroxisomal Acyl-CoA Synthetase from Pisum sativum 0402. The succinyl-CoA synthetase from the hyperther L. mophilic Archaea Thermococcus kodakaraensis (SCS) is a 0406. The enzyme was located in the peroxisomal protein heteromeric enzyme (CB) encoded by TK1880 (O-subunit) fraction Gerbling H. Axiotis S. Douce R. (1994) A new and TKO943 (B-subunit). SCS catalyzes the reversible con acyl-CoA synthetase, located in higher plant cytosol.JPlant version of Succinyl-CoA to Succinate and CoA concomitant Physiol. 143: 561-564... Pimelic acid was activated to with substrate level phosphorylation of ADP/GDP. SCS is pimeloyl-CoA and Subsequently degraded to acetyl-CoA and relatively specific, displaying relevant levels of activity for malonyl-CoA by the beta-oxidation pathway. The substrate only a few acids. The activity levels and kinetic parameters of range of this enzyme was not tested. It showed a higher SCS for Succinate, isovalerate, and 3-methylthiopropionate pH-optimum than soluble acyl-CoA synthetase from the Suggest the involvement of this enzyme in the catabolism of same organism. Glu, Leu, and Met. However, considering that ACS II also Dicarboxylate-CoA Ligase from Rattus norvegicus displays high levels of activity for the latter two substrates, the (0407 Dicarboxylic acids (C5-C16) were converted into major physiological role of SCS is most likely in the gen their CoA esters by a dicarboxylyl-CoA synthetase from the eration of ATP from succinyl-CoA. SCS is a reversible microsomal protein fraction from a rat liver Vamecq J. de enzyme and it exhibited with adipate 59% of the activity of Hoffmann E. Van Hoof F. (1985) The microsomaldicarboxy the native substrate (Shikata et al., 2007. J Biol Chem, 282: lyl-CoA synthetase. Biochem J. 230: 683-693.]. The enzyme 26963-26970), therefore it seems to be a suitable candidate. uses ATP, but not GTP. The reaction is inhibited by its prod ucts, and AMP is more inhibitory than pyrophosphate. The enzyme showed the highest activity for the C12 substrate. In O Some assays of dicarboxylyl-CoA synthetase, the plateau of OH + ATP + CoA -- CoA consumption was followed by CoA release. This is an HO important observation because it suggests that the reaction might be reversible. However, this suggestion should be O treated with some caution, because the protein sample may COOO42 have been contaminated with CoA thioesterase. Whilst this Succinate enzyme may be promising, BLAST searching of the R. nor O vegicus genome sequence did not reveal any bioW or pauA HO homologues. An alternative approach would be needed to S-CoA + ADP + phosphate identify the gene, for example, enzyme purification, N-termi O nal sequencing, primer design and PCR amplification. COOO92 Pimeloyl-CoA Synthetase from Unidentified Bacterium, LP-1 Strain Succinyl-CoA 0408. The activity was found in environmental cultures of denitrifying bacteria from activated sludge enriched to grow Pimeloyl-CoA Synthetase from Bacillus megaterium on sodium nitrate and pimelate Gallus C. Schink B. (1994) 0403. The enzyme was purified about 34-fold with 79% Anaerobic degradation of pimelate by newly isolated denitri recovery Izumi Y. Morita H, Tani Y, Ogata K. (1974) The fying bacteria. Microbiology. 140: 409–416... The enzyme pimeloyl-CoA synthetase responsible for the first step in was active in the presence of ATP and Mg". The product of biotin biosynthesis by microorganisms. Agr: Biol. Chem. 38. the reaction was AMP rather than ADP. The enzyme did not 2257-2262. The reaction required pimelate, ATP, CoASH accept benzoic acid. and MgCl2. The enzyme was specific towards pimelate only, 04.09 CoA transferases are less attractive candidates than and other dicarboxylic acids were not accepted. Interestingly, thioesterases, due to either a narrow Substrate specificity or a ATP could be replaced with ADP, suggesting that pyrophos lack of gene/protein characterization Such as for the Succi phate cleavage is not essential. The amino acid and nucleotide nate-hydroxymethylglutarate CoA transferase (EC 2.8.3.13. sequences are not known. However three complete genomes despite showing activity towards the C-6 derivative of of Bacillus megaterium have been published since 1974. pimeloyl-CoA, adipyl-CoA-(Substrate specificity of a dicar Therefore, a BLAST search for bioW homologs was under boxyl-CoA:dicarboxylic acid coenzyme A transferase from taken. Unfortunately, this did not show any matches. rat liver mitochondria: Deana et al. Biochem Int. 1992, 26, Plant Pimeloyl-CoA Synthase from Lavandula vera L. 767-773). Only one succinyl-CoA: pimeloyl-CoA trans 0404 Pimeloyl-CoA synthase activity inlavender cell cul ferase has been reported but the gene responsible for activity tures has been described (6). Although H-pimelic acid was is not available and the organism reported was not identified US 2015/0337275 A1 Nov. 26, 2015

(Gallus and Schink (1994) Microbiol. 10:409-416 Anaerobic 0411. Other suitable CoA transferases include the CoA degradation of pimelate by newly isolated denitrifying bac transferase from Rattus norvegious (Succinate-hydroxym teria). Some enzymes from this family may still be useful, and ethylglutarate CoA-transferase EC 2.8.3.13), since it accepts the following enzymes were identified as potential candi adipyl-CoA as substrate, with higher affinity than that dates: observed for succinyl-CoA, malonyl-CoA, glutaryl-CoA Glutaconate Coenzyme A-Transferase (Gct) from Acidami (Substrate specificity of a dicarboxyl-CoA:dicarboxylic acid nococcus fermentans (UniProt Q59111) coenzyme A transferase from rat liver mitochondria (Deana 0410 Glutaconate coenzyme A-transferase (Gct) from Biochem Int. 1992, 26, 767-773). Similarly, the acetyl-CoA: Acidaminococcus fermentans is an enzyme catalysing the 5-hydroxypentanoate CoA-transferase from Clostridium formation of (R)-2-hydroxyglutaryl-CoA from acetyl-CoA aminovalericum (EC 2.8.3.14) has not been tested for activity and the free acid. Gct was characterized as a heterooctamer towards C7 substrates, but does show activity towards C3-05 (CB) composed of two different subunits (GctA, 35725 Da Substrates and is specific for Saturated Substrates (Eikmanns and GctB, 29168 Da). Both subunits: GctA, and GctB have and Buckel, 1990. Properties of 5-hydroxyvalerate CoA been co-expressed in E. coli and shown to be functional as transferase from Clostridium aminovalericum. Biol. Chem. glutaconate CoA-transferase (EC 2.8.3.12). Gct is also active 371: 1077-1082). The gene(s) responsible for this activity on unsaturated and Saturated as well as hydroxylated Sub were not listed.

O O O O --- -- ~s. a-e- -- -- ~s- COOO24 CO2804 COOO33 CO3237 strates (2-Hydroxyglutarate), but is not tested against C, co Exemplification of Reversible Hydrolysis of Pimeloyl-CoA enzyme derivatives (Mack et al., 1994. Location of the two to Pimelic Acid by Acid-Thiol Ligases and CoA Transferases genes encoding glutaconate coenzyme A-transferase at the beginning of the hydroxyglutarate operon in Acidaminococ 0412. The enzymes below was selected from the list above cus fermentans.). Eur. J. Biochem, 226:41-51). This seems to were selected as examples: be a promising target, since it had been possible to change the activity of the enzyme glutaconate CoA transferase from Acidaminococcus fermentans (GctAB; EC 2.8.3.12) (Gluta Gene Synthesis UniProt conate CoA-transferase from Acidaminococcus fermentans: Source Reference Access Buckel et al. Eur. J. Biochem. 1981, 118, 315-321) from a Organism Number Protein Name Code CoA transferase to CoA esterase by site-directed mutagenesis Thermococcus BDIGENE0041 Succinyl-CoA Synthetase Q5JEN7 (Mack M. Buckel W. (1997) Conversion of glutaconate CoA kodaikaraensis large subunit transferase from Acidaminococcus fermentans into an acyl EC 6.2.1.5 BDIGENE0042 Acetyl-CoA synthetase II Q5JLA9 CoA hydrolase by site-directed mutagenesis. FEBS Lett. 405: subunit beta Acidaminococcus BDIGENE0043 Glutaconate CoA-transferase Q59111 209-212). This enzyme was successfully expressed in E. coli fermenians subunit A (GctA) and used for enzymatic biosynthesis of glutaconyl-CoA, EC 2.83.12 BDIGENE0044 Glutaconate CoA-transferase Q59112 3-methylglutaconyl-CoA, butynedioyl-CoA, 2-hydroxyad subunit B (GctB) pioyl-CoA, oxalocrotonyl-CoA and muconyl-CoA Pseudomonas BDIGENE0045 Pimeloyl-CoA synthetase Q97ER2 (Parthasarthy A, Pierik A, Kahnt J, Zelder O, Buckel W. mendocina (pauA) (2011) Substrate specificity of 2-hydroxyglutaryl-CoA dehy EC 6.2.1.14 dratase from Clostiridium symbiosum: Toward a bio-based production of adipic acid. Biochemistry. 50: 3540-35500). The crystal structure of GctAB was also published (Jacob U. 0413 Genes were codon optimized for E. coli and Mack M, Clausen T. Huber R, Buckel W. Messerschmidt A. expressed in BL21 (DE3) as previously reported (Binieda et (1997) Glutaconate CoA-transferase from Acidaminococcus al. (1999) Biochem.J. 340: 793-801; Shikata et al. (2007). A fermentans: the crystal structure reveals homology with other novel ADP-forming succinyl-CoA synthetase in Thermococ CoA-transferases. Structure. 5:415-426). This transferase cus kodakaraensis structurally related to the archaeal nucleo binds the terminal carboxylate anion via several hydrogen side diphosphate-forming acetyl-CoA synthetases. J Biol bonds to serine residues, which may adjust to the different Chem. 282:26963-26970). The His-tagged proteins were chain lengths. purified on 1 ml Histrap columns. The enzyme from

O O O O O O

------e-e- -> 3 -- ---. COOO24 CO2214 COOO33 CO2411 US 2015/0337275 A1 Nov. 26, 2015

Pseudomonas mendocina was entirely insoluble and only -continued trace amounts could be refolded, thus only the enzymes from Acidaminococcus and Thermococcus were used in activity Protein 43f44 41f42 45 43f44 41f42 45 assays with their native substrates. Activity for both enzymes 5 mM sodium acetate 4 4 was confirmed using colorimetric assays monitoring the for HO 624 544 544 624 544 544 mation of CoA esters of their natural substrates, (E)-gluta 2.5 mM MgCl2 (ADD LAST) 2 2 2 2 2 2 conate and Succinate, by measuring absorbance at 520 nm and 405 nm, respectively. Total, ul 800 800 800 800 800 800 0414 Biotransaformation reactions to assay the disap 0415 Negative controls contained 30 ul elution buffer in pearance of Adipoyl-CoA and Pimeloyl-CoA and formation the place of the enzyme solution. Reaction mixtures were of adipic acid orpimelic acid was performed in 1 ml reactions incubated at 70° C. (41/42) or room temperature (43/44; 45) by adding 30 ul protein solution to 300 ul reaction mixture for 5 hours, diluted with 570 ul water and assayed by LC-MS. that contained (per ml) in the reaction mixtures as described 0416 All 3 enzymes could hydrolyse pimeloyl-CoA to below: pimelic acid (See, FIG. 26). Despite the fact that only trace amounts of soluble enzyme of PauA could be produced, this enzyme formed pimelic acid from pimeloyl-CoA. This con firmed that acid-thiolligases of EC 6.2.1.14 are indeed revers Protein 43.44 41f42 45 43f44 41f42 45 ible, and can be used to prepare pimelic acid from pimeloyl Add volume of the stock to get: A0 AD AM PO PD PM CoA. Furthermore, the CoA transferases can be used to 20 mM sodium phosphate, pH 7.0 16 16 16 16 16 16 produce pimelic acid from pimeloyl-CoA. 1 mM Pimeloyl-CoA 16O 160 160 0417 3.1.5.3 Thioester Hydrolases (EC 3.1.2.-).-) 1 mM Adipoyl-CoA 16O 16O 16O - SMAMP 80 — — 80 3.1.5.3.1 Selection of Suitable Enzymes SMADP 80 — 80 — 0418. Thioestrases are hydrolytic enzymes that catalyze the hydrolysis of CoA or acp thioesters:

O O

0419. Thioesterases are ubiquitous in all organisms acting on CoA or acp activated carboxylic acids and belong to EC 3.1.2, including EC 3.1.2.2: EC 3.1.2.28; EC 3.1.2.19; EC 3.1.2.20 for CoA ester hysdorlases, while acp thioester hydrolases are found in EC 3.1.2.14 and EC 3.1.2.21. Only two thioesterases (ACOT8 and ACOT4) have been tested for activity towards dicarboxylyl-mono-CoA esters such as Suc cinyl-CoA.

Size Enzyme Source Substrates (kDa) Purification

ACOT8 3.1.2.27 Mus Broad range 50 Yes, (PTE-2) miscuits straight chain CoA, his-tag fusion, bile acids CoA and homogeneity branched-chain acyl-CoA MOTS 3.1.2.2 Mus Medium chain acyl- 46.6 Yes, (PTE-Ic) palmitoyl musculius CoA his-tag fusion, CoA homogeneity hydrolase ACOT3 3.1.2.2 Mus Long chain acyl- 47.4 Yes, (PTE-Ia) palmitoyl- musculius CoAS his-tag fusion, CoA homogeneity hydrolase ACOT4 3.12.3 Mus Succinyl-CoA and 36 Yes, (PTE-1b) Succinyl-CoA musculus glutaryl-CoA (active as his-tag fusion, thioester-ase ONLY. dimer) homogeneity US 2015/0337275 A1 Nov. 26, 2015 37

ACOT8 Peroxisomal Acyl-CoA Thioesterase 8 (Formerly dicarboxylyl-CoA substrates (The identification of a succi PTE-2) from Mus musculus nyl-CoA thioesterase suggests a novel pathway for Succinate 0420 ACOT8 is a homologue of human PET1 production in peroxisomes: Westin J. et al. Biol. Chem 2005, thioesterase. The enzyme was overexpressed as a His-Tag 280,38125-38132: Divergence of function in the hotdog fold fusion protein in E. coli and purified to homogeneity using enzyme superfamily—the bacterial thioesterase YCiA: HisTrap column chromatography Westin MA, Hunt MC, Zhuanget al. Biochemistry 2008, 47,2789-2796). Both genes Alexson S. E. (2005) The identification of a succinyl-CoA were identified from their respective host genome and Suc thioesterase Suggests a novel pathway for Succinate produc cessfully cloned and expressed in Escherichia coli (Charac tion in peroxisomes. J Biol Chem. 280: 38125-38132). The terization of an acyl-CoA thioesterase that functions as a activity was inhibited at substrate concentrations >5-10 uM major regulator of peroxisomal : Hunt et al. for long-chain acyl-CoAs, but addition of BSA prevented J. Biol. Chem. 277, 1128-1138; Divergence of function in the inhibition. The substrate range of the thioesterase was tested hot dog fold enzyme Superfamily—the bacterial thioesterase using saturated and unsaturated Straight-chain acyl-CoAS YciA: Zhuang et al. Biochemistry 2008, 47,2789-2796). (C2-C20). The highest activity was for C10-C18, bile acids (trihydroxycoprostanoyl-CoA, choloyl-CoA and chenode 0426 Both genes were selected as wild-type sequences for oxycholoyl-CoA and branched-chain acyl-CoAS. K. Values the construction of the corresponding E. coli expression vec for most of the substrates tested were rather low, strongly tors and used to transform E. coli BL21 (DE3). Expression of suggesting that ACOT8 has very broad substrate specificity Haemophilus influenzae YciA thioesterase was achieved in and can accept most of them. ACOT8 was strongly inhibited E. coli, (FIG. 14.h) but not of the Mus musculus Acot8 by CoASH with an ICs of 10-15uM.ACOT8 was also tested thioesterase. A proteinband at the expected size (17 kDa) was for activity towards dicarboxylyl-mono-CoA derivatives and clearly detected after IPTG induction in the soluble fraction preferentially hydrolyzed the longer-chain substrates (activ (Lanes 2 to 3) and in the insoluble fraction (Lanes 6 and 7). No ity increase with C chain length from C3 to C12) Hunt MC, corresponding protein was observed in the negative control Solaas K, Kase B F Alexson SE. (2002) Characterization of experiment (Lanes 4 and 8) suggesting that the expected 17 an acyl-coA thioesterase that functions as a major regulator of kDa Haemophilus influenzae YciA thioesterase protein was peroxisomal lipid metabolism. J Biol Chem. 277: 1128 Successfully expressed in E. coli from the assembly construct 1138). 18. Soluble expression of the YciA protein appeared to ACOT3 (Formerly PTE-1a) from Mus musculus increase over time between 4h and 24 h post-induction incu 0421 ACOT3 was expressed in E. coli and purified to bation (Lanes 2 and 3) whereas insoluble expression homogeneity 19. The enzyme showed activity for monocar appeared to decrease in the same period (Lanes 6 and 7). boxylic acid esters from C8 to C26, with the highest activity Therefore, the 24 h soluble sample seemed to be the best towards longer chain substrates (C16). The enzyme did not candidate for monitoring activity of the Haemophilus influ accept bile acid esters. There was no inhibition with free enzae YciA thioesterase. CoASH. Dicarboxylic acid esters were not tested. ACOT5 (Formerly PTE-1c) from Mus musculus 3.1.5.3.3 Synthesis of Pimeloyl-CoA 0422 ACOT5 was expressed in E. coli and purified to homogeneity Westin MA, Alexson SE, Hunt MC. (2004) 0427 Coenzyme A sodium salt (150 mg, 195umol) was Molecular cloning and characterization of two mouse peroxi suspended in a mixture of acetone (22.5 mL) and water (225 some proliferator-activated receptor alpha (PPARalpha)- LL). The Suspension was cooled to 0 degC. on ice and regulated peroxisomal acyl-CoA thioesterases. J Biol Chem. adjusted to pH 8 with 200 mM sodium bicarbonate solution. 279: 21841-21848. The enzyme showed activity for mono Acid chloride (586 umol. 3eq) was added slowly to the sus carboxylic acid esters from C6 to C20, with the highest activ pension with stirring whilst maintaining the reaction pH at 8 ity towards medium chain substrates (C10). The enzyme did by further addition of sodium bicarbonate solution. The reac not accept bile acid esters. There was no inhibition with free tion mixture was allowed to stir at 0°C. for a further 1 h. Once CoASH. Dicarboxylic acid esters were not tested. coenzyme A had been consumed, as determined by HPLC ACOT4 Peroxisomal Acyl-CoA Thioesterase 4 (Formerly reaction monitoring, the reaction mixture was concentrated in PTE-1b) from Mus musculus vacuo and the white residue was resuspended in water (20 0423 ACOT4 was expressed in E. coli and purified to mL), adjusted to pH 3 with 5M phosphoric acid and extracted homogeneity 18. The enzyme showed a narrow substrate with diethyl ether (3x20 mL). The remaining aqueous layer range, accepting Succinyl-CoA and glutaryl-CoA. The reac was then purified by solid phase extraction (SPE) using a tion was not inhibited by CoASH at concentrations up to 500 Strata X33 um polymeric reverse phase 1 g/20 mL cartridge uM. supplied by Phenomenex. 0424 ACOT8. ACOT5 and ACOT3 are thus excellent can 0428 The SPE cartridge was primed with methanol and didates for the hydrolysis of dicarboxylyl-CoAs and other equilibrated with 10 mM sodium phosphate (pH4.2) buffer acyl-CoAs that are involved in the pimelic acid biosynthesis prior to use. The aqueous layer containing the crudeacyl-CoA pathway. product was loaded onto the cartridge. 10 mM sodium phos 3.1.5.3.2 Cloning, Expression and Biotransformations with phate (pH4.2) buffer washes (3x5 mL) were carried out, Selected Thioesterases followed by a stepwise gradient using 5, 10 and 20% MeOH/ 0425 Thioesterases from Mus musculus Acot3 (WT buffer elutions (3x5 mL). Each fraction was collected and mRNA sequence GeneBank NM 133240; mRNA fragment analysed by HPLC. The fractions containing product only 65-1027) and the Haemophilus influenzae YciA (WT gene were combined and concentrated in vacuo. The isolated prod sequence (GeneBank L42023 fragment 878596-879060, EC ucts were used directly in the biotransformations without 3.1.2.20) were selected as examples for the conversion of further purification. 1H NMR or LCMS were used to charac pimeloyl-CoA to pimelic acid, due to their specificity towards terise the products. US 2015/0337275 A1 Nov. 26, 2015

3.1.5.3.4 Biotransformation of Pimeloyl-CoA to Pimelic EC 2.6.1.19 Acid 0433 4-Aminobutyrate Transaminase from Streptomyces 0429 Biotransformations were carried out in a final vol griseuS ume of 2 mL. The biotransformations comprised of 1 mM 0434. This enzyme was first studied in 1985 and was found substrate, 200 mM KCl and 50 mM potassium Hepes (pH to catalyse the reaction 7.5). Negative controls comprised of cells harbouring the 4-aminobutanoate-2-oxoglutarate->4-oxobutanoate empty vector The biocatalyst was charged as a cell free glutamate. extract to give a final concentration of ~15 g/L whole cell The enzyme has a relatively wide Substrate range, including equivalents. The reactions were carried out at ambient tem 6-aminohexanoate and 7-aminoheptanoate, using 2-oxoglu perature (20°C.) in 15 mL falcon tubes with stirring. Each tatarate as the amino acceptor, and forming the corresponding biotransformation was sampled after 1 h, 4 hand 24 h and semialdehydes (Yonaha, K. K. Suzuki, and S. Toyama, analysed by HPLC (Agilent 1100) on a Phenomenex Luna 4-Aminobutyrate:2-oxoglutarate aminotransferase of Strep C18250x4.6 mm, 5 micron column. Prior to injection, each tomyces griseus. purification and properties. EurJ Biochem, sample was heated to 100° C. for 5 mins to stop the reaction, 1985. 146(1): p. 101-6). Although this is very promising, the spun at 14000 rpm for 2 minutes and filtered through a 0.2 reversibility of these reactions has not been tested. Therefore, micron syringe filter. The filtrates were analysed by HPLC at the reverse reactions are tested in vitro, using glutamate as the 260 nm for the disappearance of pimeloyl-CoA (Rt 12.39 amino donor, since this would provide 7-aminoheptanoic mins) and the formation of CoA (Rt 7.98 minutes), both acid from 7-oxoheptanoate. The enzyme was purified in S. which absorbs strongly at this wavelength, while pimlic acid griseus IFO3102 (S. griseus subsp. griseus) and was found to was not detectable by UV at 260 nm. Pimelic acid formation be 100 kDa, consisting of two identical subunits approx. 50 was confirmed by LC-MS. kDa each. The cofactor is pyridoxal 5'-phosphate. A protein is 0430. A low level of background hydrolysis was observed deposited on the Pubmed database (Table 1). The gene encod in the negative control, which can be ascribed to low levels of ing this (SGR 1829), annotated as a putative 4-aminotrans endogenous thioesterase activity in the E. coli harbouring the ferase, is present in the genome sequence of S. griseus IFO empty vector. Complete hydrolysis of pimeloyl-CoA to 13350, which is closely related to IFO 3102. Although not pimelic acid and acetyl-CoA was observed after 1 hour of clear, it is highly likely that this protein is a close homologue reaction. of the enzyme described and studied (Yonaha, K. K. Suzuki, and S. Toyama, 4-Aminobutyrate. 2-oxoglutarate ami 3.1.6 Conversion of Pimelic Acid Semialdehyde to notransferase of Streptomyces griseus. purification and 7-Aminoheptanoic Acid by (O-Transaminases (EC properties. EurJBiochem, 1985. 146(1): p. 101-6). In light of 2.6.1.-) and Conversion of C-Ketosuberate to the Substrate range data available, this protein is a very good C.-AminoSuberate by Amino Acid Aminotransferases candidate for testing as a suitable aminotransferase enzyme. (EC 2.6.1.-) EC 2.6.1.39 0431 Transaminases or aminotransferases are enzymes 0435 Aminoadipate Transaminase from Thermus ther that catalyse the reaction between an amino acid and a keto mophilus (Q72LL6) acid. This reaction involves removing the amino group from 0436 This is an enzyme with homology to the alpha the amino donor, leaving behind an O-keto acid, and transfer aminoadipate aminotransferases (AAAAT) from mammals ring it to the acceptor C-keto acid and converting it into an and is thought to be involved in lysine synthesis in the bacte amino acid. They are important enzymes in the synthesis of rium Thermus thermophilus HB272, 3. Kinetic data are amino acids and are ubiquitous throughout nature. Many of available for Substrates Such as 2-oxoadipate, 2-oxoisoca these enzymes contain pyridoxal-phosphate (PLP) as cofac proate, alpha-aminoadipate and even aromatic amino acids tor. These enzymes belong to a large diverse family covering 3. The protein has been expressed and purified in E. coli and a wide range of cellular functions, using a wide range of was found to exist in vitro mainly as a homodimer, with each Substrates. Subunit ~44 kDa in size (Miyazaki, T., et al., alpha-Aminoa dipate aminotransferase from an extremely thermophilic bac 3.1.6.1 Selection of Transaminases terium, Thermus thermophilus. Microbiology, 2004. 150(Pt 7): p. 2327–34). The broad substrate specificity, its prokary 0432 A number of candidates, identified by literature otic origin and the availability of the encoding gene Suggest searching, BLAST searching and protein database searching that it may be a good candidate for testing with 1. It is a very are Suitable enzymes for to conversion of pimelic acid semi well studied enzyme and its crystal structure is available aldehyde to 7-aminoheptanoic acid. (PDB ID: 27P7), which offers mechanistic insights into sub strate specificity (Ouchi, T., et al., Dual roles of a conserved pair; Arg23 and Ser20, in recognition of multiple substrates in Enzyme alpha-aminoadipate aminotransferase from Thermus ther Name? Accession Gene Protein mophilus. Biochem Biophy's Res Commun, 2009. 388(1): p. EC No. number Source available purified 21-7: Tomita, T., et al., Mechanism for multiple-substrates EC YP OO1823341.1 Streptomyces Yes Yes recognition of alpha-aminoadipate aminotransferase from 26.1.19 grisetts Thermus thermophilus. Proteins, 2009. 75(2): p. 348-59). LysN Q72LL6 Thermits Yes Yes thermophilius Aminoadipate Aminotransferase from Rattus norvegicus AADaT Q64602 Rattus norvegicus Yes Yes (Q64602) 0437. This enzyme is similar to the enzyme from Thermus thermophilus described previously. It is annotated as a US 2015/0337275 A1 Nov. 26, 2015 39 kynurenine/aminoadipate aminotransferase, although there is Activity Assay of IlvE-Omega Vf Fusion (plNG2022) and Some evidence to suggest that two distinct proteins carry out IlvE-Ad Optimized Omega Fusion (plNG2030) Aminotrans these reactions (Mawal, M. R. and D. R. Deshmukh, Alpha ferase Proteins aminoadipate and kynurenine aminotransferase activities from rat kidney. Evidence for separate identity. J Biol Chem, 0444 Recombinant Vf aminotransferase was used to 1991. 266(4): p. 2573-5). The protein has been purified from assay for activity on 7-aminoheptanoic acid. Three biotrans rat kidney and expressed in human embryonic kidney fibro formations were prepared examining the effect of enzyme blast cells and was found to be around 100 kDa in size. The loading on conversion of 7-aminoheptanoic acid and Sodium enzyme can accept 2-oxoadipate as a Substrate. Earlier stud pyruvate to L-alanine. Detection of L-alanine in the biotrans ies (presumably on the same enzyme) indicate that it is able to formation would provide positive evidence for the desired use DL-2-aminopimelic acid as an amino donor, with 2-oxo aminotransferase specificity on 7-aminoheptanoic acid. The glutarate as acceptor in a reaction that is apparently reversible reaction scheme depicting aminotransferase conversion of (Deshmukh, D. R. and S. M. Mungre, Purification and prop 7-aminoheptanoic acid to pimelic acid semialdehydeis erties of 2-aminoadipate: 2-oxoglutarate aminotransferase depicted below: from bovine kidney. Biochem J, 1989. 26.1(3): p. 761-8). The same paper indicated that pimelic acid may be an inhibitor of activity with a 30% reduction in activity in the presence of pimelic acid. CO2H 0438. The above three enzymes should be suitable for the pyruvic acid L-alanine final transamination to yield 7-aminoheptanoic acid. The transaminase enzyme from S. griseus is a good candidate and the process of He cloning, expression and purification would not be particularly -- -e- -- difficult. The same is true of LysN from T. thermophilus, which seems to have a broad Substrate range. The enzyme H from Rattus norvegicus would be more challenging but activ ity data Suggest that this enzyme may be able to use the 11a1n-CO2H --~-coil specific Substrates required. HN O 7-aminoheptanoic acid pimelic acid semialdehyde 5-Aminovalerate Transaminase (EC 2.6.1.48) 0439 Diamino aminotransferases from EC 2.6.1.48 such 0445. The three biotransformations (3 ml volume) con as the 5-aminovalerate transaminases are also suitable cata tained 50 mM sodium pyruvate and 10 mM 7-aminohep lysts for the preparation of 7-aminoheptanoic acid form tanoic acid. The reactions were held in a 30° C. orbital incu pimelic acid semialdehyde. bator and aliquots removed periodically from the 0440 3.1.6.2 Expression of Selected Aminotransferases biotransformations for analysis. 0441. Two genes were selected for the conversion of both pimelic acid semialdehyde to 7-aminoheptanoic acid, and the Sodium 7-aminoheptanoic Enzyme conversion of C.-ketosuberate to C.-aminosuberate: (1) IlvE Sample ID pyruvate?mM acid mM loading viv% Omega Vffusion (SEQID 1 and 2) and (2) IlvE-Adoptimized omega fusion (SEQID 3 and 4). These genes were prepared rc0212-40-1 50 10 O.33 in plasmids that incorporate fusion of the omega aminotrans rc0212-40-2 50 10 O.66 ferase with the 5'-end of the branched chainamino transferase rcO212-40-3 50 10 1.67 IlvE, which previously showed high expression level in E. coli (See, FIG. 23 and FIG. 24). It was demonstrated that introduction of the IlvE fragment increases significantly the 0446. The aliquots (1 ml) were removed from the biotrans expression of the aminotransferase genes. Another heat formations and heat to 95°C. for 5 minutes to precipitate induced plasmid containing a branched chain aminotrans soluble protein. The sample was Subsequently centrifuged ferase gene with the IlvEfragment was also tested (ING66) in and the supernatant filtered through a 0.2 micron filterprior to the activity assays. HPLC analysis. The HPLC analysis used a precolumn deri 0442. A further 5 genes (Bacillus weihenstephanensis vatisation method based on chemical adduct formation with KBAB4, Pseudomonas aeruginosa (gi9951072), Bacillus beta-mercaptoethanol and orthophthaldehyde. External stan subtilis (gi16078032), Pseudomonas syringae class III, dards of L-alanine and 7-aminoheptanoic acid were used to Rhodobacter sphaeroides class III) were selected (DSM determine the concentration of 7-aminoheptanoic acid and patent (Preparation of 6-aminocaproic acid from 5-formyl L-alanine in the biotransformations. valeric acid: DSM-US2011-0171699 A1 patent). These 0447. The results (FIG. 16) clearly show the formation of enzymes showed some activity in the conversion of C.-keto L-alanine in the biotransformation providing compelling evi pimelic acid into O-aminopimelic acid and may also be active dence for aminotransferase activity and specificity with Vf in the conversion of pimelic acid semialdehyde into 7-amino aminotransferase. The 7-aminoheptanoic acid concentration heptanoic acid. (See, SEQID 5-14). decreases over time which is consistent with the expected 0443 SDS-PAGE analysis of IlvE-Omega Vf fusion aminotransferase activity. The rate of 7-aminoheptanoic acid (pNG2022) (FIG. 23) and IlvE-Adoptimized omega fusion increases with increased concentration of enzyme which is (pNG2030) (FIG. 24) proteins expressed in E. coli BL21 as again consistent with expected transminase activity. The before. A clear band of the expected size is observed in both the IlvE-Omega Vffusion (plNG2022) (FIG. 15, Lanes 5 and amount of L-alanine formed is not matched by the 7-amino 6) and IlvE-Adoptimized omega fusion (pNG2030) (FIG. heptanoic consumed, this could be caused by alternative bio 15, Lanes 8 and 9) proteins. There is no corresponding band chemical degradation of the L-alanine formed by other non in the negative control, confirming the expression of the IlvE aminotransferase enzymes present in the E. coli extract. Omega Vffusion and the IlvE-Ad optimized omega fusion 0448. A further examination of the aminotransferases was proteins. performed to assess for activity towards 2-aminoSuberate. US 2015/0337275 A1 Nov. 26, 2015 40

chemical adduct formation with Boc-cysteine and orthoph thaldehyde. External standards of DL-2-aminosuberic acid and L-alanine were used to determine the retention times of O1. CO2H HN s CO2H the 2-aminoSuberic acid enantiomers. pyruvic acid L-alanine 0452. The results indicate a reduction in the L-2-amino concentration and an enantioenrichment in the D-2-aminosuberic acid across all six biotransformations. transaminase This effect is consistent with the action of the aminotrans Hs -- a- -- ferase on the L-enantiomer, since the aminotransferases tested here are known to exhibit L-stereoselectivity. No mea Surable L-alanine was observed by HPLC which should result COH COH from the transamination of pyruvate, however it is likely that * COH --~-coil the L-alanine formed has been degraded further by the other HNY N--- 2 O host proteins present in the biotransformation as has been previously observed in the 7-aminoheptanoic acid reactions. L-2-aminosuberic acid 2-ketosuberic acid 0453 The ING66 biotransformations (rc0212-46-1 and -2) showed significantly higher levels of L-2-aminosuberic 0449 An aminotransferase conversion of L-2-amino acid degradation compared to the plNG2022 and plNG2030 Suberic acid to 2-ketoSuberic acid provides an opportunity to aminotransferases. ING 66 branched chain aminoacid ami modulate biotransformations. notransferase is already known to exhibit excellent selectivity 0450 Six biotransformations were carried out examining for L-2-aminoglutarate (L-glutamic acid) which is a shorter pING 2022, pING2030 and ING66 (Ilve branched chainami (C5) homologue of L-2-aminosuberic acid so this preference notransferase, E.C.2.6.1.42, construct already in strain col in activity for NG 66 is not unexpected. lection). Each biotransformation contained 10 mM DL-2- Production of 7-Aminoheptanoic Acid from Pimelic Acid aminosuberic acid, 50 mM sodium pyruvate and high or low Semialdehyde loading of cell free extract (CFE) as indicated below, N.B. 0454 Based on the observed activity of the Vibrio fluvialis ING 66 used as whole cells: ()-aminotransferase (Accession number HQ418483.1 (WT gene sequence: fragment 325-1686)) activity on 7-aminohep tanoic acid, the reaction was performed in the opposite direc whole cell tion, i.e. pimelic acid semialdehyde conversion to 7-amino equivalents heptanoic acid. Both the Vibrio fluvialis ()-amino acid lab book ID aminotransferase g/L transaminase enzyme, as well as the N-ter-His-Tag purified Bacillus weihenstephanensis were evaluated. The Bacillus rcO212-44-5 pING2022 5 weihenstephanensis co-amino acid transaminase (Accession rc0212-44-6 pING2030 5 rcO212-44-7 pING2022 2O number JA114119.1) was expressed and purified using stan rc0212-44-8 pING2030 2O dard techniques. Reactions were performed with both adipic rc0212-46-1 ING66 25 acid semialdehyde and pimelic acid semialdehyde and prod rc0212-46-2 ING66 5 uct formation (6-aminocaproic acid and 7-aminoheptanoic acid) was confirmed by HPLC (Agilent 1200, GraceSmartRP C18 column). 0451. The six biotransformations (3 ml volume) were held in a 30° C. orbital incubator and aliquots removed periodi General Procedure for the Synthesis of Adipic Acid cally from the biotransformations for analysis. The aliquots Semialdehdye and Pimelic Acid Semialdehyde: (1 ml) were removed from the biotransformations and heat to 95°C. for 5 minutes to precipitate soluble protein. The sample 0455 Synthesis of the C6 & C7 semialdehydes was per was Subsequently centrifuged and the Supernatant filtered formed according to the general reaction scheme below, through a 0.2 micron filterprior to HPLCanalysis. The HPLC based on published methods (Synthetic communications, 21 analysis used a precolumn derivatisation method based on (8&9), 1075-1081 (1991):

O

H2SO4 S. O MeOH C PCC la ls---pi OMe DCM 2a n = 1 methyl 6-hydroxyhexanoate 2b n = 2 methyl 7-hydroxyhepanoate

H2SO4 C) MeOH 1b US 2015/0337275 A1 Nov. 26, 2015

-continued --~~. 3a n = 1 methyl 6-oxohexanoate 3b n = 2 methyl 7-oxoheptanoate

Lipase pH8 --~~ss.H O 4a n = 1 6-oxohexanoate (Nasalt) 4b n = 27-oxoheptanoate (Nasalt)

Methyl 6-Hydroxyhexanoate Preparation (2a) =98 degC., 8 mbar, 3b b.p=106 degC., 7 mbar) to afford a clear, colourless distillate (3a 16.9 g, 3b 4.9 g). The products 0456 e-caprolactone la (50 g, 0.425 mol. 1 eq) was dis were characterised by Hand 'C NMR analysis. solved in methanol (500 mL). Concentrated sulphuric acid (1 mL) was charged and the Solution was heated to reflux over General Procedure for Hydrolysis of 3a/b to Make C6/C7 night. Complete conversion to 2a was determined by GC-MS. Acid Semi Aldehydes 4a/b Using Immobilised Lipase The reaction was allowed to cool to RT and water (700 mL) Acrylic Resin from Candida antartica was charged. The product was extracted into diethyl ether 0460 Each ester aldehyde 3a/b (1 g, 6.9 mmol. 1 eq) was (3x200 mL), dried over magnesium sulfate and concentrated suspended in 50 mM sodium phosphate buffer (pH 8) (20 to afford a clear, colourless liquid (40.7g). The crude product mL). Immobilised lipase resin (100 mg) was charged and the was distilled under high vacuum (b.p.-120° C., 7 mbar) to suspension was heated to 50 degC. and stirred for 45 mins. afford a colourless oil 2a (32.9 g) which was used directly in After this time, GC-MS indicated the ester aldehyde starting the next stage. material 3a/b had been consumed. The Suspension was cooled to RT, the lipase beads were filtered and the resultant filtrate Methyl 7-Hydroxyheptnoate Preparation (2b) was concentrated in vacuo to afford the crude C6/C7 semi 0457 Concentrated sulphuric acid (1 mL), water (120 mL) aldehydes 4a/4b in phosphate buffer. The products were char and methanol were cooled to 15 degC. on ice. Potassium acterised by "H NMR. persulfate (365.6 g. 1.34 mol, 3 eq) was slowly added with General Procedure for Hydrolysis of 3a/b to Make C6/C7 with stirring. 1b (50 g., 0.447 mmol. 1 eq) in Acid Semi Aldehydes 4a/b Using Immobilised Lipase methanol (150 mL) was added dropwise over 45 mins. The Acrylic Resin from Candida antartica reaction was allowed to warm to RT overnight with stirring. 0461) Each ester aldehyde 3a/b (1 g, 6.9 mmol. 1 eq) was Complete conversion to 2b was determined by GC-MS. At suspended in 50 mM sodium phosphate buffer (pH 8) (20 this point the work-up product described for 2a was followed. mL). Immobilised lipase resin (100 mg) was charged and the General Procedure for PCC Oxidation of Hydroxy Esters suspension was heated to 50 degC. and stirred for 45 mins. 2a/b to Prepare Ester Aldehydes 3a/3b After this time, GC-MS indicated the ester aldehyde starting 0458 Pyridinium chlorochromate (PCC)(60 g) and silica material 3a/b had been consumed. The Suspension was cooled gel (60 g) were charged to a 1 LRB flask along with dichlo to RT, the lipase beads were filtered and the resultant filtrate romethane (200 mL). The Suspension was concentrated to was concentrated in vacuo to afford the crude C6/C7 semi dryness to obtain a fine powder. Dichloromethane (500 mL) aldehydes 4a/4b in phosphate buffer. The products were char was charged along with 3 A molecular sieves (70 g). The acterised by "H NMR. Suspension was cooled to 0 degC. on ice. 0459 Each hydroxy ester (2a 32.6 g. 2b 37 g) was dis V.f. (D-Amino Acid Transaminase Biotransformation solved in dichloromethane (30 mL) and added dropwise to the Conditions PCC/silica suspension over 30 mins. After further stirring for 30 mins, complete conversion to ester aldehyde 3a/b was observed by GC-MS. The crude reaction mixture was filtered 0462 through a celite pad. Diethyl ether washes (5x200 mL) were done to ensure all the product was washed into the clear, dark Lab book reference kg1112-021-10 brown filtrate. The filtrate was then passed through a silica Substrate concentration 5 mM (6-oxohexanoic acid) pad to remove the coloured impurities. Further diethyl ether 5 mM (7-oxoheptanoic acid) L-monosodium glutamate (amino donor) 100 mM washes (3x100 mL) of the silica pad were done. The com Biocatalyst loading (gL) based on whole 15 g/L whole cells bined filtrates were concentrated in vacuo to afford a pale cell equivalents (WCE) (cell free extract) equivalents' yellow green liquid (3a 24.5 g. 3b 7.9 g). The crude product pH 7 was purified by short path high vacuum distillation (3a b.p. US 2015/0337275 A1 Nov. 26, 2015 42

-continued were determined by correlating the retention times of their respective external reference standards. Temperature (degC) 30 degC 0466 It was observed that significantly more 6-amino Final reaction volume 5 mL. hexanoic acid was generated under these conditions, provid *The term whole cell equivalents (WCE) is a term that is used to estimate the amount of ing strong evidence that this transaminase has higher activity biocatalyst present in a cell free extract after cell lysis. This is calculated by resuspending a known weight of wet whole cells in a defined lysis buffer volume. For example, if 1 g of spun towards 6-oxohexanoic acid compared to 7-OXO heptanoic wet whole cell pellet is resuspended in 10 mL of lysis buffer, the WCE concentration is 100 acid. gL. Typically, we would prepare a stock suspension of biocatalyst in this way and dilute this stock as necessary to obtain the desired biocatalyst concentration in each biotransformation, 0467. In both cases, significant amounts of each product were also observed in the negative control experiments using the empty vector cells i.e. the E. coli strain not expressing the Protocol for In-Situ Preparation of C6/C7 Semi Aldehyde target transaminase. We predict these results are due to back Substrates Used in Biotransformations ground transaminase activity in the host cell free extract. However, the HPLC results clearly show that for both sub 0463. 10 mM stock solutions of C6 and C7 semialdehydes strates, higher levels of product were formed in the experi were prepared prior to testing. The following protocol was ments containing cells with the over-expressed transaminase used to prepare these stock solutions: 10 mM Methyl ester enzyme compared to the negative controls. No product for aldehyde was suspended in 50 mM sodium phosphate (pH mation was observed in either of the negative control experi 7)(50 mL). Immobilised lipase acrylic resin from Candida ments using water as a replacement for cell free extract. antarctica (10 mg) was charged. The Suspension was heated to 50 degC. and stirred for 1 h. TLC analysis indicated com B.W. ()-Amino Acid Transaminase (N-ter-His-Tag) plete consumption of the starting material after this time. The Biotransformation Conditions reaction was cooled to RT, the beads were filtered and the filtrate, containing ~10 mM semialdehyde, was used directly 0468 in each biotransformation. V.f. () Amino Acid Transaminase Biotransformation Protocol Substrate concentration 10 mM L-monosodium glutamate (amino donor) 100 mM 0464 6 biotransformations were then carried out each in a Biocatalyst loading (gL) 100 uL purified N-ter-His-Tag final volume of 5 mL. Each biotransformation comprised of 5 10 uL purified N-ter-His-Tag mM semi aldehyde, 100 mM L-monosodium glutamate and 1 uL purified N-ter-His-Tag pH 7 the biocatalyst, which was charged as cell free extract to give Temperature (degC) 30 degC. a final concentration of 15 g/L whole cell equivalents. The Final reaction volume 5 mL. reactions were carried out at 30 degC. in 15 mL falcon tubes with stirring at pH 7. Each biotransformation was sampled periodically and analysed by HPLC. Prior to injection, each sample was heated to 100 degC. for 5 mins, spun at 14000 rpm B.W. ()-Amino Acid Transaminase (N-ter-His-Tag) for 2 mins and filtered through a 0.2 micron syringe filter. Biotransformation Protocol List of V.f. () Amino Acid Transaminase Biotransformations 0469 14 biotransformations were then carried out in a Carried Out final volume of 1 mL. Each biotransformation comprised of 10 mM semi aldehyde, 100 mM L-monosodium glutamate and a range of purified enzyme aliquot Volumes (100, 10 and Experiment ID Substrate biocatalyst comment 1 LL). This was done to evaluate transaminase activity at KG 1112-021-10-86 6-oxo- V.f. () amino acid different protein concentrations. Each biotransformation was hexanoic transaminase cell carried out at pH 7 and placed in a shaking incubator at 30 80 free extract degC. (250 rpm). Sampling was done periodically and analy KG 1112-021-10-96 6-oxo- E. coli BW 25113 Negative control hexanoic AdadAX using host cells sed by HPLC. Prior to injection, each sample was heated to aci with empty vector 100 degC. for 5 mins, spun at 14000 rpm for 2 mins and KG1012-021-8O-6 6-oxo- Water Negative control filtered through a 0.2 micron syringe filter. hexanoic using water aci List of B.w.. (p-Amino Acid Transaminase (N-ter-His-Tag) KG 1112-021-10-87 7-oxo- V.f. () amino acid heptanoic transaminase cell Biotransformation Experiments aci free extract KG 1112-021-10-97 7-oxo- E. coli BW 25113 Negative control 0470 heptanoic AdadAX using host cells aci with empty vector KG1012-021-80-3 7-oxo- Water Negative control Volume of heptanoic using water No Substrate biocatalyst biocatalyst comment 1 6-oxo B.W. (a) amino acid 100 L N-ter-His-Tag hexanoic acid transaminase purified enzyme Results from V.f. () Amino Acid Transaminase Biotransfor aliquot used 2 6-oxo E. coli (BL21 DE3) 100 IL Negative control mations hexanoic acid using host cells with empty vector. 0465. It was successfully demonstrated that V.f. ()-amino Subjected to the acid transaminase has activity towards 6-oxohexanoic acid same purification and 7-OXO heptanoic acid. Formation of the corresponding process as the over 6-amino hexanoic acid and 7-amino heptanoic acid products expressed strain. were measured by HPLC. Identification of the product peaks US 2015/0337275 A1 Nov. 26, 2015 43

-continued HPLC Analytical Conditions 0474 Instrument: Agilent 1200; Column: GraceSmart RP Volume of C18 150x4.6 mm, 5 micron; Column temp: 40 degC. Flow No substrate biocatalyst biocatalyst comment rate: 1 mL/min: Detection: UV 338 nmi; Injection volume: 14 uIL 3 6-oxo- Water 100 IL Negative control Mobile phase: Line A: 5 mM sodium phosphate pH 7: Line B: hexanoic acid using water instead acetonitrile of cells. 4 6-oxo- B.W. (a) amino acid 10 L.L Gradient: hexanoic acid transaminase 5 6-oxo- E. coli (BL21 DE3) 10 L.L 0475 hexanoic acid 6 6-oxo- B.W. (a) amino acid 1 LIL hexanoic acid transaminase Time (mins) % B 7 6-oxo- E. coli (BL21 DE3) 1 LIL O 5 hexanoic acid 4 25 6 8O 8 7-oxo- B.W. (a) amino acid 100 IL N-ter-His-Tag 8 8O heptanoic acid transaminase purified enzyme 8.5 5 aliquot used 12 5 9 7-oxo- E. coli (BL21 DE3) 100 IL Negative control heptanoic acid using host cells Non Chiral HPLC Pre-Column Derivatisation was Carried with empty vector. Out as Follows for Each Sample that was Taken: Subjected to the 0476 A 5 mL solution containing B-mercaptoethanol (50 same purification uL), o-phthaldialdehyde (50 mg), 400 mM potassium borate process as the over buffer pH10 (4.5 mL) and methanol (0.5 mL) was prepared. expressed strain. Derivatisation was done in the autosampler of the HPLC 10 7-oxo- water 100 IL Negative control using the following injector program: 2 LL Sample, 6 LL heptanoic acid using water instead derivatisation mixture, 6 uL potassium borate buffer pH10 of cells. was mixed in the injector loop and held for 4 mins prior to 11 7-oxo- B.W. (a) amino acid 10 IL injection. heptanoic acid transaminase Reference std Concentrations: 12 7-oxo- E. coli (BL21 DE3) 10 IL heptanoic acid 0477 10 mM 6-amino caproic acid 13 7-oxo- B.W. (a) amino acid 1 LL 10 mM 7-amino hexanoic acid heptanoic acid transaminase 50 mM L-monosodium glutamate 14 7-oxo- E. coli (BL21 DE3) 1 LL 0478 Conclusion: the conversion of adipic acid semialde heptanoic acid hyde and pimelic acid semialdehyde to the corresponding 6-aminohexanoic and 7-aminoheptanoic acids was measured by HPLC. The semi-aldehydes methyl esters were prepared Results from B.W. ()-Amino Acid Transaminase Biotransfor chemically, the final stage of the synthesis was the hydrolysis mations of the appropriate methyl ester using an immobilised lipase 0471. It was demonstrated that purified N-ter-His-Tag B.w giving rise to a semialdehyde Solution which was used with ()-amino acid transaminase has low activity towards 6-oxo out further purification. Product (co-amino acid) formation hexanoic acid and 7-oxoheptanoic acid. The HPLC results was observed for both the C6 and C7 semialdehydes using show that the desired amino acid products were formed in the omega aminotransferases from both Vibrio fluvialis and experiments where 100 uL of the purified enzyme was Bacillus weihenstephanensis. The Vibrio fluvalis enzyme had charged. slightly higher activity. 0472. There did not appear to be a significant difference D- and L-Specific Amino Acid Aminotransferases for the between C6 and C7 substrate specificity, as the measured Production of D- and L-2-Aminosuberate (Substrates for D product levels in each experiment were comparable. How or L-Specific Decarboxylases to Produce 7-Aminoheptanoic ever, in both cases, the level of product formed was very low. Acid) 0473. To ensure correct identification of the low level 0479 IlvE L-AAAT from E. coli (WT gene: X02413.1 product peaks in the related chromatograms, a series of (fragment 301 . . . 1230); WT protein: 1104250A) was sample spikes using the appropriate external reference stan selected as the L-selective aminotransferase and Bacillus dards were carried out. In each case, a clear increase in the sphaericus D-AAAT (WT gene: U26732.1 (fragment 427 . . product peak size was observed. The external reference stan . 1278) WT protein: P54693.1) as an example of a D-selective dard spike Volume was proportional to the increase in peak alpha amino acid aminotransferase to demonstrate the con magnitude observed. version of C-ketoSuberate to C.-aminoSuberate. US 2015/0337275 A1 Nov. 26, 2015 44

General Procedure for Preparation 2-Oxo Acid Substrates 0480

O O

HO EtO pi OH, -e- OEt

O O 1 n = 1 adipic acid n = 1 diethyl adipate n = 2 pimelic acid n = 2 diethyl pimelate

O O

HO O-Na+ -a- EtO pi pi OEt

O O O O O 4 OEt n = 1 2-oxo-pimelate monosodium salt n = 2 2-oxo-Suberate monosodium salt 3 n = 1 triethyl oxalyl adipate n = 2 triethyl oxalyl pimelate

Literature Refs: during addition which was cooled on ice to 0 degC. After 2 h, the solids were filtered and filter cake washes were done using 0481. J. Org. Chem. 1986, 51, 2389-2391 and Organic diethyl ether (5x50 mL). The filter cake (1' crop) was col syntheses, Coll. Vol. 3, p510, 1955 lected and the filtrate was stored at 4 degC. overnight. Further solid precipitated overnight (2" crop). This was filtered and Diethyl Ester Preparation (2) combined with the 1 crop to afford a yellow solid (33.7g). This was suspended in 5M hydrochloric acid and diethyl ether 0482. A RB flask containing absolute ethanol (880 mL) extractions (3x100 mL) were carried out. The combined was cooled to 0 degC. Acetyl chloride (124 mL, 1.71 mol, 7 organics were washed with brine (1x100 mL), dried over eq) was added dropwise over 1 h. The dioc acid (1) (0.245 magnesium Sulfate, filtered and concentrated to afford the mol. 1 eq) was charged and the resultant Suspension was triethyl oxalyl intermediate (3) as a crude yellow oil (24.5 g). allowed to warm to RT. Once the reaction had gone to comple tion, as determined by thin layer chromatography, water (300 mL) was charged and the ethanol was removed by evapora Hydrolysis and Product Isolation (4) tion. The remaining aqueous solution was neutralised to pH 7 with sat sodium bicarbonate solution. The product was 0484 The crude triethyl oxalyl intermediate (3) (24.5 g) extracted with methyl tert butyl ether (3x250 mL), dried over was dissolved in 10M hydrochloric acid (100 mL) and left sodium sulfate, filtered and concentrated in vacuo to afford stirring overnight at RT. The reaction mixture was concen the crude diethyl ester (2) as a colourless oil in 70% isolated trated in vacuo to afford a gummy orange oil. This was dis yield. The diethyl ester was purified by high vacuum distilla Solved in acetone (100 mL) and activated charcoal (4 g) was tion prior to the next stage. charged. The Suspension was left stirring overnight at RT and filtered through a celite pad. The filter cake was washed with Triethyl Oxalyl Intermediate Preparation (3) acetone (5x10 mL) and the resultant pale yellow filtrate was concentrated to afford a pale orange oil (20.5 g). 0483 To a flame dried flask containing anhydrous diethyl ether (16.1 mL) was charged potassium metal lumps (8g. 206 0485 The product was redissolved in water (10 mL) (pH mmol. 1 eq) under a nitrogen atmosphere. Anhydrous ethanol 0.2) and neutralised to pH 4 with 10M sodium hydroxide. A (32 mL, 536 mmol. 2.6 eq) was added dropwise over 30 mins. pale yellow suspension formed which was cooled to 0 degC. Once the potassium metal had completely dissolved (gentle Isopropyl alcohol (10 mL) was charged with stirring. The heating to 30 degC. was necessary) a pale yellow solution was resultant suspension was filtered to afford an off-white filter observed. Diethyl (27.5 mL. 206 mmol. 1 eq) was cake (7.4 g) which was transferred to a vacuum dessicator and added quickly with good stirring. At this point, the diethyl dried overnight. After this time the product was isolated as an ester intermediate (2) (206 mmol. 1 eq) in diethyl ether (50 off-white solid (3.77 g). NMR was used to characterise the mL) was added over 10 mins. An orange Suspension formed final product (see Appendices Section for spectra). US 2015/0337275 A1 Nov. 26, 2015

Biotransformation Conditions -continued

0486 IlvE (L-Selective Aminotransferase) Experiment ID Substrate biocatalyst comment KG 1012-59-3 2-oxo- water Negative control Lab book reference RHO912-021-56 KG 1012-59-4 2-oxo-suberate B.S. D-AAAT cell Substrate concentration 10 mM (sodium salt) free extract L-monosodium glutamate (amino donor) 50 mM KG 1012-59-5 2-oxo-suberate I17 empty vector Negative control Biocatalyst loading (gL) 100 g/L whole cells (IlvE) (sodium salt) pH 7-7.4 KG 1012-59-6 2-oxo-suberate Water Negative control Temperature (degC) 30 degC. (sodium salt) Final reaction volume 10 mL. KG 1012-93-1 2-oxo-pimelate B.S. D-AAAT cell (sodium salt) free extract KG 1012-93-2 2-oxo-pimelate I17 empty vector Negative control 0487 B.S. D-AAAT (D-Selective Aminotransferase) (sodium salt) KG 1012-93-3 2-oxo-pimelate Water Negative control (sodium salt) Lab book reference KG 1012-021-59 (2-oxo-Suberate) KG 1012-021-93 (2-oxo-pimelate) Substrate concentration 10 mM D-alanine (amino donor) 50 mM Conclusions Pyridoxal phosphate co factor O.4 mM Biocatalyst loading (gL) based on 21 g/L whole cell equivalents 0491. Using E. Coli IlvE it was successfully demonstrated whole cell equivalents (WCE) pH 7 that L-2-amino Suberic acid and L-2-amino pimelic acid Temperature (degC) 30 degC. products were generated from 2-oxo Suberate and 2-oxo Final reaction volume 5 mL. pimelate respectively. 0492. The biotransformation results were determined by 0488 Biotransformation Protocols correlating HPLC retention times of the external reference standards to the observed product peaks formed in each IlvE (L-AAAT) biotransformation. 0489. 4 biotransformations were carried out eachina final 0493 A non-chiral derivatisation HPLC method was Volume of 5 mL. Each biotransformation comprised of 10 employed to enable detection of the corresponding deriva mM substrate and 50 mM L-monosodium glutamate. The tised product adducts at 338 nm on a reverse phase C18 biocatalyst was charged as whole cells to give obtain a final column. IlvE is well-known to be enantiospecific forming concentration of 100 g/L. The reactions were carried out at 30 only L-amino acids from keto-acids so an achiral analytical degC. in 15 mL falcon tubes with stirring. Each biotransfor method was used for simplicity. The details for the non-chiral mation was sampled periodically and analysed by HPLC. derivatisation protocol are outlined in the Appendices sec Prior to injection, each sample was heated to 100 degC. for 5 tion. mins, spun at 14000 rpm for 2 mins and filtered through a 0.2 0494. It appears that the enzyme activity towards L-2- micron Syringe filter. amino pimelic acid/2-oxo pimelate measured by HPLC is considerably lower than L-2-amino suberic acid/2-oxo sub erate under these conditions. To investigate this Surprising Experiment ID Substrate biocatalyst comment result further, the biotransformation was repeated using a RHO912-56-1 2-oxo-pimelate IlvE whole cells different batch of 2-oxo pimelate, however the same result RHO912-S6-2 2-oxo-suberate IlvE whole cells was obtained, suggesting the activity towards the C8 com RHO912-56-3 2-oxo-pimelate water Negative control pounds really is higher than towards the C7 compounds. This RHO912-56-4 2-oxo-suberate water Negative control is particularly Surprising given that the C5 amino-diacid, glutamic acid, is the common amino donor for all these reac B.S. D-AAAT tions. 0490 9 biotransformations were carried out in a final vol 0495 Negative control experiments were conducted by ume of 5 mL. Each biotransformation comprised of 10 mM replacing the whole cells with water. No background activity substrate, 50 mM D-Alanine and 0.4 mM pyridoxal phos was observed for either substrate in these experiments. phate. The biocatalyst was charged as a cell free extract to 0496 Using Bacillus sphaericus D-AAAT it was success obtain a final concentration of 21 g/L whole cell equivalents. fully demonstrated that D-2-amino glutamic acid (positive The reactions were carried out at 30 degC. in 15 mL falcon control) and D-2-amino-Suberic acid and D-2-amino-pimelic tubes in a shaking incubator. Each biotransformation was acid products were generated from 2-oxoglutarate, 2-oxo sampled periodically and analysed by HPLC. Prior to injec Suberate and 2-oxopimelate respectively. tion, each sample was heated to 100 degC. for 5 mins, spun at 14000 rpm for 2 mins and filtered through a 0.2 micron 0497. The biotransformation results were determined by Syringe filter. correlating HPLC retention times of external reference stan dards to the observed product peaks formed in each biotrans formation. Experiment ID Substrate biocatalyst comment 0498. In addition, HPLC was also employed to demon KG 1012-59-1 2-oxo-glutaric acid B.S. DAAAT cell Positive control strate the enantioselectivity of the enzyme. Using a pre-col free extract umn derivatisation protocol for each sample in the presence KG 1012-59-2 2-oxo-glutaric acid I17 empty vector Negative control of a chiral derivatisation reagent, it was possible to separate the corresponding diastereomeric adducts on a reverse phase US 2015/0337275 A1 Nov. 26, 2015 46

C18 column at 338 nm wavelength. The details for the chiral Reference Standard Concentrations: derivatisation protocol are outlined in the Appendices sec tion. 0506 10 mM D-glutamic acid; 10 mM DL-2-amino 0499. Two negative control experiments were run for each suberic acid; 10 mMDL-2-amino pimelic acid; 10 mM L-2- substrate; I 17 empty vector cells i.e. the E. coli strain not amino Suberic acid; 50 mM D-Alanine; 50 mM L-MSG expressing the target aminotransferase and water. 0500. A small amount of background conversion was 3.1.7 Conversion of C.-Amino Suberic Acid to observed in the I17 empty vector negative control experi 7-Amino Heptanoic Acid by C.-Amino ments for each substrate. We expect this result is due to low Decarboxylases (EC 4.1.1.-) (See, FIG. 10) level background aminotransferase activity in the host cell free extract. This is most likely a combination of endogenous 0507 Suitable catalysts were selected from the C-amino alanine racemase activity generating some L-alanine which acid decarboxylases as candidates for the decarboxylation of Subsequently acts as amino donor for endogenous L-AAAT C.-amino Suberic acid, in particular the aspartate 4-decar enzymes in E. coli. boxylase (EC 4.1.1.11), the glutamate decarboxylase (EC 4.1.1.15), the lysine decarboxylase (EC 4.1.1.18), the diami 0501. No detectable levels of the expected amino acids were observed in any of the water negative control experi nopimelate decarboxylase (EC 4.1.1.20), and the diaminobu ments under these conditions. tyrate decarboxylase (EC 4.1.1.86). 0508. The Escherichia coli GadA glutamate decarboxy HPLC Analytical Conditions: lase and Escherichia coli LySA diaminopimelate decarboxy lase were initially selected from those groups of proteins. 0502. Instrument: Agilent 1100; Column: Phenomenex Other suitable enzymes include LySA from Methanocaldo Kinetex 150x4.6 mm, 5 micro GraceSmart 150x4.6 mm, 5 coccus jannaschii (AAB991.00) (EC 4.1.1.20) (Ray, S. S., et micron; Column temp: 40 degC. Flow rate: 1 mL/min: Detec al., Cocrystal structures of diaminopimelate decarboxylase: tion: UV 338 nmi; Injection volume: 14 uL mechanism, evolution, and inhibition of an antibiotic resis Mobile phase: Line A: 5 mM sodium phosphate pH 7: Line B: tance accessory factor. Structure, 2002. 10(11): p. 1499-508), acetonitrile and the lysine decarboxylases present in E. coli: CadA (AAA23536), which is inducible under anaerobic conditions and by lysine, and has a very high activity and Ldc (accession Gradient: D87518) that is constitutively expressed but has lower activ 0503 ity for lysine (Kikuchi, Y., et al., Characterization of a second lysine decarboxylase isolated from Escherichia coli. J. Bac teriol, 1997. 179(14): p. 4486-9). Time (mins) % B 0509. The DNA sequences corresponding to the selected O 5 genes as well as their related protein sequence are shown in 4 25 SEQID 15-18. Sequence of the Escherichia coli LySA diami 6 8O 8 8O nopimelate decarboxylase was selected as an E. coli codon 8.5 5 optimized gene as previously described (Preparation of 12 5 6-aminocaproic acid from 5-formyl Valeric acid: DSM US2011-0171699 A1 patent). The Escherichia coli GadA glutamate decarboxylase was selected as wild type sequence. The genes were cloned and expressed using methods Non Chiral HPLC Pre-Column Derivatisation was Carried described in previous examples. Out as Follows for IlvE Experiments: 0510. In addition to the previously prepared constructs 0504. A 5 mL solution containing B-mercaptoethanol (50 (I29:pET21-GadA and I30:pET21-LySA, 2 further constructs uL), o-phthaldialdehyde (50 mg), 400 mM potassium borate were prepared by introduction of the ilvE fragment (assem buffer pH10 (4.5 mL) and methanol (0.5 mL) was prepared. blies I31 and I32 below) using the in ABLE technology. Intro Derivatisation was done in the autosampler of the HPLC duction of the 15 amino acid Nter peptide of the E. coli using the following injector program: 2 LL sample, 6 ul branched amino acid aminotransferase (IlvE) upstream of the derivatisation mixture, 6 uL potassium borate buffer pH10 protein of interest has previously been shown to result in an was mixed in the injector loop and held for 4 mins prior to increase in expression of Some protein targets in E. coli. injection.

Chiral HPLC Pre-Column Derivatisation was Carried Out as Genes Accession numbers Follows for D-AAAT Experiments: I29 E. coli GadAWT sequence M84024.1 I30 E. coli LySA (E. coli JA114145.1 0505. A 5 mL solution containing Boc-L-Cysteine (128 codon optimised) mg), o-phthaldialdehyde (50 mg), 400 mM potassium borate I31 E. Coi GadA Lw. First 45 bases of the iLvE gene were fused to the GadA gene buffer pH10 (4.5 mL) and methanol (0.5 mL) was prepared. I32 E. coli LySA iLVE First 45 bases of the iLvE gene Derivatisation was done in the autosampler of the HPLC were fused to the LySA gene using the following injector program: 2 LL sample, 6 ul iLVE (wild type sequence) X02413.1 (wild type sequence) derivatisation mixture, 6 uL potassium borate buffer pH10 fragment 301-1230. was mixed in the injector loop and held for 4 mins prior to injection. US 2015/0337275 A1 Nov. 26, 2015 47

0511. The expression results are summarized below: -continued OH

Soluble fraction Insoluble fraction HN ~~~ O Assembly Protein 4h 24h 4h 24h n=1 6-amino caproic acid I29 E. coli GadA ------n =27-amino heptanoic acid I30 E. coli LySA ------I31 E. coli GadA ilvE ------I32 E. coli LySA ilvE ------Reaction mixtures were analysed by HPLC as follows: Instrument: Agilent 1100; Column: GraceSmart RP C18 150x4.6 mm, 5 micron; Column temp: 40 degC. Flow rate: 1 Preparation of Protein Extracts for Activity Assay mL/min. Detection: UV 338 nm. Injection volume: 14 uL. Mobile phase: Line A: 5 mM sodium phosphate pH 7: Line B: 0512. The previously frozen cell pellets from the 24 hour acetonitrile post-induction samples were re-suspended in bead lysis buffer consisting of 0.1 mM Tris, 1 mg/ml Pepstatin A and Gradient: 200 mM PMSF protease inhibitors (161 uL of the mixture should be used for the lysis of the cell pellet from 1 mL of 0516 culture at an ODoo of 5), Supplemented with 1.5 mg lysozyme. Acid washed 212-300 um glass beads (0.4 g per 1 mL of lysis buffer) were added to each sample. Lysis was Time (mins) % B performed by 30 second bursts of Vortexing on full speed, O 5 followed by 30 second incubation on ice, repeated to a total 4 25 time of 10 minutes. 6 8O 8 8O 0513. 1 mL samples were taken and the remainder of the 8.5 5 culture used for the activity assays. The 1 mL samples were 12 5 centrifuged at 13000 rpm, for 2 minutes and the supernatant separated. SDS samples were prepared and analyzed on SDS PAGE gels as described above (FIG. 18). Following lysis, Non Chiral HPLC Pre-Column Derivatisation was Carried there is a large band in the soluble fraction for the E. coli Out as Follows for Each Sample that was Taken: GadA ilvE (I31) protein (FIG. 18, Lane 4) that was not 0517. A 5 mL solution containing B-mercaptoethanol (50 present in the 1 mL sample, lysed by bugbuster. The E. coli uL), o-phthaldialdehyde (50 mg), 400 mM potassium borate LySA (I32) protein, which showed high soluble expression buffer pH10 (4.5 mL) and methanol (0.5 mL) was prepared. previously, did not show much soluble expression in the bead Derivatisation was done in the autosampler of the HPLC lysis (FIG. 18, Lane 5). This may indicate a large variability using the following injector program: 2 LL sample, 6 ul in the lysis methods. Crude cell extracts were then used to derivatisation mixture, 6 uL potassium borate buffer (pH10) monitor the activity of proteins expressed from the 4 con was mixed in the injector loop and held for 4 mins prior to Structs I29 to I32. injection. 0514 For biotransformation reactions, larger scale culti vations were performed and cell free extracts containing 20 Reference Standard Concentrations: g/L whole cell equivalents were prepared. Host cells harbour 0518) 10 mM 6-amino hexanoic acid ing the empty vectors were used as negative controls. 0519) 10 mM 7-amino heptanoic acid 0515 Biotransformations were performed to assay activ 0520 10 mM DL-2-aminosuberic acid ity for 2-aminoSuberate as well as 2-aminoheptanoic acid. 0521. 10 mM DL-2-aminopimelic acid Each biotransformation comprised of 10 mM substrate, 10 0522. It was successfully demonstrated that all the ug/mL pyridoxal phosphate and 1 mM f-mercaptoethanol, C.-amino acid decarboxylases have activity towards 2-amino adjusted to pH 7 with 1M sodium hydroxide. The biocatalyst pimelic acid and 2-amino Suberic acid. Formation of the was charged as cell free extract to obtain a final concentration corresponding 6-amino hexanoic acid and 7-aminoheptanoic of 20 g/L whole cell equivalents. The reactions were carried acid products were measured by HPLC. Identification of the out at 37 degC. in 15 mL falcon tubes with stirring. Each product peaks were determined by correlating the retention biotransformation was sampled periodically and analysed by times of their respective external reference standards. HPLC. Prior to injection, each sample was heated to 100° C. 0523. In the experiments which evaluatedI29, I30 and I31 for 5 mins, spun at 14000 rpm for 2 mins and filtered through strains, lower level activity was observed for both substrates. a 0.2 micron Syringe filter. A number of samples were spiked with the corresponding external reference standard to ensure the correct peak was identified. 0524 For both substrates, I32 E. Coli LySA iLVE gave the best result, demonstrating higher activity compared to the HO OH alpha AAD pi -e- other strains. 0525. In addition, very low levels of each product were also observed in the negative control experiments using the n = 1 DL-2-aminopimelic acid empty vector cells i.e. the E. coli strain not expressing the n =2DL-2-aminosuberic acid target decarboxylase. We predict these results are due to back ground decarboxylase activity in the host cell free extracts. US 2015/0337275 A1 Nov. 26, 2015 48

0526. However, the HPLC results clearly show that higher White White R H (1989) Biosynthesis of the 7-mercapto levels of product were formed in the experiments containing heptanoic acid subunit of component B (7-mercaptohep cells with the over-expressed decarboxylase enzyme com tanoyl)threonine phosphate of methanogenic bacteria, Bio pared to the negative controls. No product formation was chemistry, 28: 860-865 based on the analysis of other observed in either of the negative control experiments using metabolic pathways in Archea and the mass spectroscopic water as a replacement for cell free extract. analysis of incorporation of 2H and 13C labeled precursors 0527 . The concentration of the observed peak corre into the final product. The pathway contains a nonoxidative sponding to 7-aminoheptanoic acid is shown in FIG. 19. decarboxylation step of 2-ketosuberic acid to 7-oxoheptanoic acid (pimelate semialdehyde) mediated by a postulated 2-ke 3.1.8 Conversion of C-Keto Suberic Acid to Pimelic toacid decarboxylase. In the next step, 7-mercaptoheptanoic Acid Semialdehyde by C.-Keto Decarboxylases (EC acid is formed. Since this product only is formed (and not the 4.1.1.-) C5 and C6 equivalents), this indicates that at least one of the 0528 3.1.8.1 Selection of Suitable Catalysts upstream enzymes in the pathway will only accept C8, but not 0529 Amongst the family of decarboxylase, the O.-keto C6 and C7 substrates. acid decarboxylases emerged as the best candidates for the 0535 The pathway was verified by confirmation of the decarboxylation of O-keto Suberic acid, in particular the pyru 2-ketodicarboxylic acid elongation reaction from 2-ketoglu vate decarboxylase EC 4.1.1.1), the benzoylformate decar tarate to 2-ketosuberate using GC-MS White R H (1989) A boxylase (EC 4.1.1.7) and the branched-chain-2-oxoacid novel biosynthesis of medium chain length alpha-ketodicar decarboxylase (EC 4.1.1.72). boxylic acids in methanogenic archaebacteria. Archivers of 0530 Genes encoding proteins from each of those four Biochemistry and Biophysics, 270: 691-697 and synthesis of groups have been identified from a broad range of hosts and 7-mercaptoheptanoic acid and coenzyme B from 7-oxohep their products characterized biochemically and in Some cases tanoic acid White R H (1989) Steps in the conversion of structurally (The crystal structure of benzoylformate decar a-ketosuberate to 7-mercaptoheptanoic acid in methanogenic boxylase at 1.6A resolution Diversity of catalytic residues bacteria, Biochemistry, 28:9417-9423; White R H (1994) in ThDP-dependent enzymes: Hasson et al. Biochemistry Biosynthesis of 7-mercaptoheptanoylthreonine phosphate, 1998, 37, 9918-9930; High resolution crystal structure of Biochemistry, 33: 7077-7081. The genes encoding enzymes pyruvate decarboxylase from Zymomonas mobilis: responsible for the some of the steps of the pathway (akSA, Dobritzsch et al. J. Biol. Chem. 1998, 273, 20196-20204; aksD, aksh, and aksh) have been identified, cloned and Structure of the branched-chain keto acid decarboxylase expressed in E. coli, but not the gene encoding 2-ketosuberic (KdcA) from Lactococcus lactis provides insights into the acid decarboxylase Howell DM, Harich K, Xu H, White R structural basis for the chemoselective and enantioselective H. (1998) Alpha-keto acid chain elongation reactions carboligation reaction: Berthold et al. Acta Crystallographica involved in the biosynthesis of coenzyme B (7-mercaptohep Sec. D 2007. D63, 1217-1224). tanoyl threonine phosphate) in methanogenic Archaea. Bio 0531. The following enzymes were then selected amongst chemistry, 37: 101.08-101 17: Howell DM, Graupner M, Xu those groups according to their substrate specificity: Lacto H. White RH. (2000) Identification of enzymes homologous coccus lactis kiv) 2-ketoisovalerate decarboxylase to isocitrate dehydrogenase that are involved in coenzyme B (GeneBank JA114157); Lactococcus lactis kdcA branched and leucine biosynthesis in methanoarchaea. JBacteriol. 182: chain alpha-keto acid decarboxylase (GeneBank JA114154), 5013-5016; Drevland RM, Jia Y. Palmer DR, Graham D. E. Saccharomyces cerevisiae pdc1 pyruvate decarboxylase (2008) Methanogen homoaconitase catalyzes both hydrol (GeneBank JA114148), Zymomonas mobilis pdc (I472A) yase reactions in coenzyme Bbiosynthesis. J Biol Chem. 283: pyruvate decarboxylase mutant (GeneBank JA114151), 28888-28896. Nothing is known about nonoxidative decar Pseudomonas putida mdlC (A460I) benzoylformate mutant boxylation of 2-ketosuberic acid to 7-oxoheptanoic acid and (GeneBank AY143338; fragment 2860-4446). this activity has not been reported in vitro. However 2-keto 0532. The first four enzymes previously showed some suberic acid was converted to coenzyme B by whole cells activity in the decarboxylation of the O.-keto pimelic acid, the White R H (1989). C-7 derivative of the target compound, O-keto suberic acid (Preparation of 6-aminocaproic acid from 5-formyl Valeric Methodology acid: DSM-US2011-0171699 A1 patent). 0533 Characterization of the A460I mutant of the ben 0536 Work done by White and Graham showed that Zoylformate decarboxylase from Pseudomonas putida dis Archaebacteria (like Methanocaldococcus jannaschii) are played increased selectivity towards 2-keto hexanoic acid likely to possess an enzyme responsible for nonoxidative (Exchanging the Substrate specificities of pyruvate decar decarboxylation of 2-ketosuberic acid to 7-oxoheptanoic boxylase from Zymomonas mobilis and benzoylformate acid. Other enzymes involved in the coenzyme B pathway decarboxylase from Pseudomonas putida: Siegert et al. Port. were identified by analysis of the genomes of M. jannaschii Eng. Des. Sel. 2005, 18, 345-357) compared to the wild-type Bult C J, White O, Olsen GJ, Zhou L, Fleischmann R D, enzyme which showed low activity towards 2-keto octanoic Sutton G. G. Blake J A, FitzGerald L. M. Clayton R A, acid (Comparative characterisation of ThPP-dependent Gocayne J. D. Kerlavage A. R. Dougherty BA, Tomb J. F. decarboxylases: Gocke et al. J. Mol. Cat. B: Enzymatic 2009, Adams MD, Reich CI, Overbeek R, Kirkness E. F. Weinstock 61, 30-35). KG, Merrick J. M. Glodek A, Scott J. L. Geoghagen NS, Venter J.C. (1996) Complete genome sequence of the metha 3.1.9 Conversion of C-Ketosuberic Acid to Pimelic nogenicarchaeon, Methanococcus jannaschii. Science. 273: Acid Semialdehyde by Putative Acetolactate 1058-1073 and Methanobacterium thermoautotrophicum Synthases with Decarboxylase Activity AH Smith DR, Doucette-Stamm LA, Deloughery C. Lee H. 0534. The pathway for synthesis of coenzyme B (7-her Dubois J. Aldredge T. Bashirzadeh R. Blakely D, Cook R. captoheptanoyl)threonine phosphate has been postulated by Gilbert K. Harrison D. Hoang L. Keagle P. Lumm W. Pothier US 2015/0337275 A1 Nov. 26, 2015 49

B. Qiu D, Spadafora R, Vicaire R, Wang Y. Wierzbowski J. coenzyme M pathway Graupner M. Xu H and White R H. Gibson R, Jiwani N. Caruso A, Bush D, Reeve J. N. (1997) (2000) Identification of the gene encoding sulfopyruvate Complete genome sequence of Methanobacterium ther decarboxylase, an enzyme involved in biosynthesis of coen moautotrophicum deltaH: functional analysis and compara Zyme M. J. Bacteriol. 182: 4862-4867. tive genomics. J Bacteriol. 179: 7135-7155. The genomes were analyzed for the presence of the target gene clusters and were BLAST searched for proteins showing homology to O O existing enzymes. We took a similar approach to identify the \/ 1. P O –se gene encoding 2-ketoSuberic acid decarboxylase. O 0537. Results O Analysis of the Gene Clusters phosphoenolpyruvate 0538 Bacterial genomes can show a high level of organi O O Q CO Zation in which genes encoding enzymes playing a role in \/ specific pathways are localized in operons and/or gene clus O1 S O ters. Because other enzymes of the coenzyme B pathway have com) comE been identified and the sequences of Some archaebacterial genomes are published, an attempt was made to identify O 2-ketosuberic acid decarboxylase. The table below shows the 3-sulfopyruvate genes and enzymes involved in the pathway.

Genes

Meihanocaidococcus Meihanobacterium Enzyme Reaction iannaschii thermoautotrophicum AkSA condensation of C-ketoglutarate and M.JOSO3 MTH1630 (R)-homocitrate acetyl-CoA: condensation of C.- synthase ketoadipate and C.-ketopimelate to acetyl-CoA: AksDAksE dehydration of (R)-homocitrate and M10O3 MTH1113 homocitrate analogs to cis-homoaconitate or the M1271 MTH1631 synthase respective analogs, and the dehydratase and hydration of the products to cis- homoisocitrate and its respective homoaconitate- analog hydratase Aksh NAD-dependent decarboxylation of M1596 MTHO184 NAD-dependent (-)-threo-isohomocitrate, (-)-threo threo-isocitrate iso(homo) citrate and (-)-threo dehydrogenase iso(homo) citrate 2-ketosuberic 2-ketOSuberic acid decarboxylation 9. 99 acid decarboxylase mercaptoheptanoate 9. 99 synthase mercaptoheptanoyl 9. 99 threonine synthase N-(7- 9. 99 mercaptoheptanoyl)- 3-O- phosphono-L- threonine synthase

0539 The coenzyme B biosynthetic genes from M. jann -continued aschii and M. thermoautotrophicum did not appear to reside O O O O in gene clusters, except for the genes encoding (R)-homoci VW VW trate synthase (MTH 1630) and homocitrate synthase/dehy -O1 S N-1s -R -o-1N1S Nshi dratase large subunit (MTH1631) in M. thermoautotrophi cum. The analysis of hypothetical operons containing any of sulfoacetaldehyde coenzyme M the genes listed above did not show the presence of any protein with homology to known decarboxylases. Therefore, a search was made for analogous decarboxylations in related 0541. Nonoxidative decarboxylation of 3-sulfopyruvate metabolic pathways in M. jannaschii. to sulfoacetaldehyde is catalyzed by sulfopyruvate decar Analysis of Methanocaldococcus jannaschii Genome boxylase (EC.4.1.1.79) containing two subunits, com) (SEQ 0540 Analysis of other metabolic pathways in M. jann ID 19) and comE (SEQ ID 20). As both coenzyme B and aschii showed that a similar decarboxylation is present in the coenzyme M are involved in methanogenesis and the reac US 2015/0337275 A1 Nov. 26, 2015 50 tions downstream of the decarboxylation are analogous, it is -continued possible that the decarboxylases evolved together and show high homology to each other. QUERY 0542 ComD, comE and their orthologues found in the (Name & BRENDA database were used as queries to BLAST search UNIPROT the Methanocaldococcus jannaschii. The best match was entry) Substrate Source Hits (E-value < 10') obtained using com/comE decarboxylase from Methano fom2 phosphonopyruvate Erwinia P58416.1 Sulfopyruvate cella paludicola (SEQID 21). Q6D9X5 decarboxylase Caiolov'Oia decarboxylase subunit Subsp. beta 3e-20 (0543. Only two open reading frames (MJO277 and atroSeptica P58415.1 Sulfopyruyate MJO663) encoded by the M. jannaschii genome showed sig (Pectobacterium decarboxylase subunit nificant homology to comD/comE with the E-value <0.0001. atrosepictim) alpha 2e-11 Both of them were annotated as acetolactate synthases (SEQ NP 247516. ID 22 and SEQID 23). 2-oxoglutarate 0544 Another search was performed using sequences of ferredoxinoxido decarboxylases accepting similar Substrates that are depos reductase ited in the BRENDA database (2-ketoglutarate decarboxy subunit beta 2e-05 lase, oxaloacetate decarboxylase, C.-ketoisovalerate decar NP 247250. boxylase, transaminated amino acid decarboxylase, acetolactate benzoylformate decarboxylase, 2-ketoarginine decarboxy lase, phosphonopyruvate decarboxylase, pyruvate decar PDC6 Pyruvate Saccharomyces NP 247250. boxylase and indole-3-pyruvate decarboxylase). P26263 decarboxylase cerevisiae acetolactate isozyme 3 4.1.1.1 subunit 6e-23 QUERY (Name & UNIPROT entry) Substrate Source Hits (E-value < 10) P58416.1 Sulfopyruvate Kgd 2-ketoglutarate Mycobacterium No hits decarboxylase subunit Q9CC97 decarboxylase leprae beta 6e-04 EC 4.1.1.71 citM Oxaloacetate Lactococcits No hits NP 247310. Q7X4Z5 decarboxylase iaciis hypothetical protein kivd C-ketoisovalerate Lactococcits NP 247250. MJ 03379e-04 Q684J7 decarboxylase iaciis acetolactate synthase IpdC Indole-3-pyruvate Pantoea NP 247250. EC 4.1.1.1 catalytic Subunit F2EPZ5 decarboxylase ananatis acetolactate synthase E-value: 8e-29 catalytic subunit 6e-22 NP 247647. acetolactate synthase NP 247647. large subunit IlvB5e-19 acetolactate synthase ARO10 transaminated Saccharomyces NP 247250. large subunit 7e-12 Q064.08 amino acid cerevisiae acetolactate synthase NP 247516. decarboxylase catalytic subunit 2e-17 2-oxoglutarate EC 4.1.1.- NP 247647. ferredoxinoxido acetolactate synthase reductase large subunit IlvB9e-12 subunit beta 9e-05 P58416.1 Sulfopyruvate decarboxylase subunit beta 4e–04 McC benzoylformate Sulfolobus NP 247250. (0545 MJ0277 (accession number NP 247250) and Q97XR3 decarboxylase Solfatarict is acetolactate synthase EC 4.1.1.7 catalytic subunit 3e-23 MJ0663 (NP 247647) were good hits for almost all the NP 247O65. queries used to search the genome. The protein alignment signal recognition using the ClustalW algorithm showed significant homology particle protein Srp54 6e-04 between MJ0277 and comD/comE proteins (FIG. 20). 2V3C C Crystal Structure Of The (0546) However, MJ0277 showed even higher homology to Srp54-Srp 19-7s.S LACLA acetolactate synthase from Lactococcus lactis (FIG. SrpRnaée-04 21). 3NDB B Crystal Structure Of A 0547 Analysis of the functional domains in LACLA Signal Sequence Bound acetolactate synthese, com/comE from Methanocella palu To The Signal 6e-04 aruI 2-ketoarginine Pseudomonas NP 247250.1 dicola and the hypothetical acetolactate synthase MJO277 B7V368 decarboxylase aeruginosa acetolactate synthase showed that all of them have a similar structure containing a catalytic subunit 4e-71 thiamine phosphate binding site (FIG. 22). Almost identical NP 247647.1 acetolactate synthase results were obtained for the second open reading frame large subunit 2e-38 MJO663 from Methanocaldococcus jannaschii. NP 248425.1 cobyrinic acid a,c-diamide 0548. The acetolactate synthase (ALS) enzyme (also synthase 4e–05 known as acetohydroxy acid synthase) catalyzes the conver sion of pyruvate to O-acetolactate. US 2015/0337275 A1 Nov. 26, 2015

0553. The proteins annotated as acetolactate synthases O O MJ0277 (NP 247250) and MJ0663 (NP 247647) were acetolactate synthase identified as prime candidates for nonoxidative decarboxyla OH + OH --e- tion of 2-ketoadipic acid, 2-ketopimelic acid and 2-keto Suberic acid. Acetolactate synthases had been reported to O O catalyse decarboxylation of 2-ketoacids (Atsumi, S. et al., O O 2009. Acetolactate synthase from Bacillus subtilis serves as a 2-ketoisovalerate decarboxylase from isobutanol synthesis in Escherichia coli. Applied and Environmental Microbiology, S 75(19): 6306-6311). In addition: 3-sulfopyruvate decarboxy w lase comD/comE (P58415/P58416), Kgd (Q9CC97), citM (Q7X4Z5), kivd (Q684J7), AR010 (Q06408), MdlC 0549. The first step of the reaction is decarboxylation of (Q97XR3), arul (B7V368), fom2 (Q6D9X5), PDC6 pyruvate (which is a 2-ketoacid) to an enzyme-bound inter (P26263) and IpdC(F2EPZ5) are excellent candidates for the mediate that is further condensed with the second pyruvate conversion of 2-ketosuberic acid to pimelic acid semialde molecule resulting in acetolactate. Thus, 2-ketoacid decar hyde (SEQ ID 24-40). boxylases and acetolactate synthases are likely to show high homology to each other. 3.1.10 Conversion of C-Keto Pimelic Acid to C-Keto Suberic Acid by C.-Keto Acid Chain Elongation (See, Conclusions FIG. 10) 0550 Analysis of the M. jannaschii genome showed the 0554 3.1.10.1 Selection of Enzyme Gene Targets presence of two proteins annotated as acetolactate synthases 0555 Conversion of C.-ketopimelic acid into O-keto (MJO277 and MJ0663) that may be the previously unidenti suberic acid was previously reported to be part of the biosyn fied 2-ketosuberic acid nonoxidative decarboxylase involved thesis of the Coenzyme B in methanogenic archaebacteria, in coenzyme B biosynthesis. Only Methanogenic spp. appar Such as Methanocaldococcus jamaschii, Methanocaldococ ently have this second version of acetolactate synthase cus vannieli, Methanocaldococcus maripaludis, Methano (MJO663) with, for example, Methanocaldococcus fervens bacterium thermoautotrophicum, Methanosarcina termo (NC 013156.1) (Locus YP 003128272); Methanotorris phila and others (White, Arch. Biochem. Biophys. 270: 691 igneus Kol 5 (NC 015562.1) (Locus YP 004.483786) and 697 (1989)). Genes encoding the enzymes involved in the Methanocaldococcus infernus ME (NC 014122.1) (Locus multi cycles of chain elongation from C.-ketoglutaric acid to YP 003615749) all having proteins with greater than 75% Suberic acid, homologous to enzymes involved in the leucin positives to MJ0663. No non-methanogenic spp. has greater and lysine biosynthesis, were identified from Methanocaldo than 50% positives to MJ0663. The other spp. have a second coccus jamaschii and their corresponding proteins character gene with homology to MJ0277, Methanocaldococcus fer ized (O-keto acid chain elongation reactions involved in the vens (Locus YP 003127480); Methanotorris igneus Kol 5 biosynthesis of coenzyme B (7-mercaptoheptanoyl threonine (Locus YP 004484320) and Methanocaldococcus infernus phosphate) in Howell et al., Biochemistry 37: 101.08-101 17 ME (Locus YP 003615922) and this gene is annotated as an (1998); Drevland et al., J. Bacteriol. 189: 4391-4400 (2007): acetolactate synthase. This gene has homology to acetolactate Howell et al., J. Bacteriol. 182:5013-5016 (2000); Drevland synthase in other non-methanogenic organisms. et al., J. Biol. Chem. 283: 28888-28896 (2008); and Jeyakan 0551. It is very likely that one of them is the real acetolac tate synthase. This enzymatic activity has been shown in than et al., Biochemistry 49: 2687-2696 (2010). many archaebacteria, and M. jannaschii encodes the aceto 0556 Orthologous genes were also identified from lactate synthase regulatory subunit, MJO161. However, one of Methanobacterium thermoautotrophicum, but their products the enzymes may be the unidentified 2-ketoacid decarboxy have not been characterized yet. At least two isogenes were lase and has simply been misannotated. It is likely that the reported in this archaebacteria for each of the required activ regulatory subunit MJO161 or ILVN (Methanocaldococcus ity (AskA, AksD/E and Aksh) as previously reported in jannaschii Locus NP 247129), the small subunit of aceto Methanocaldococcus janaschii (Smith et al. 1997, 179,7135 lactate synthase is required for decarboxylase activity of 7155). The table below presents the orthology of the corre MJO277 or M.JO663. This is a Small subunit - 190aa that binds sponding Methanobacterium thermoautotrophicum proteins to the dimer or possibly tetramer holoenzyme of AHAS/ILVB with the well-characterized Methanocaldococcus janaschii in equal number. ILVN responds to Valine concentration act proteins based on the literature previously mentioned and ing as a feedback to prevent over production (Barak and their potential deduced function Chipman 2012). The ILVB holoenzyme is only 5-15% effi cient in the absence of ILVN, the regulatory unit has no M. Orthologous M. activity on its own (M.Vyazmensky, C. Sella, Z. Barak, D. M. ianaschii Functions of M. thermoautotrophicum Chipman. Isolation and Characterization of Subunits of Enzyme activity proteins ianaschi proteins proteins Acetohydroxy Acid Synthase Isozyme III and Reconstitution AkSA: M.JOSO3 Condensation of MTH1630 of the Holoenzyme. Biochemistry, 35 (1996), pp. 10339 2-isopropylmalate C.-ketoglutarate 10346). synthase and acetyl-CoA 0552. Alternatively, 2-ketosuberic acid might be decar homocitrate to form trans synthase homoaconitate; boxylated by the enzyme responsible for decarboxylation of functions in the 3-sulfopyruvate in the coenzyme M pathway. Therefore, Ci-ketosuberate comD/comE decarboxylase should also be considered as a synthesis target enzyme. US 2015/0337275 A1 Nov. 26, 2015 52

-continued genes AkSA(1)-MTH1630, AksD(2)-MTH1631 and Aks (2)-MTH1388 in which the starting codons GTG or TTG M. Orthologous M. were changed to ATG. ianaschii Functions of M. thermoautotrophicum 0559) 3.1.10.2 Cloning Expression Vectors Containing Enzyme activity proteins ianaschi proteins proteins Gene Targets M1195 Formation of 2- MTH1481 0560. The selected gene targets from Methanobacterium isopropylmalate from acetyl-CoA and thermoautotrophicum were then cloned into the IPTG induc 2-oxoisovalerate in ible pET21-a backbone using in ABLE technology, as leucine biosynthesis described in detail in Section 3.1.1.2) to generate I9 (M. AksD: M10O3 Formation of MTH1631 thermoautotrophicum AkSA(1)-MTH1630 gene in pRT21 3-isopropylmalate cis-homo, dehydratase? aconitate (n = 1-4) a), I1 0 (M. thermoautotrophicum AkSA(2)-MTH 1481 gene homoaconitase when coupled in pRT21-a), I11 (M. thermoautotrophicum AksD(1)- (Large subunit) with MJ1271; MTH 1386 gene pET21-a gene in pET21-a), I12 (M. ther involved in moautotrophicum AksD(2)-MTH 1631 gene inpET21-a), I13 the production of C.-keto Suberic acid (M. thermoautotrophicum Aksh(1)-MTHO829 gene in MJO)499 Involved in the MTH1386 pET21-a). I14 (M. thermoautotrophicum Aksh (2)- leucine biosynthesis MTH 1387 gene in pRT21-a). I15 (M. thermoautotrophicum Aksh: M1271 Formation of MTHO829: Aks (1)-MTHO184 gene in pET21-a), and 116 (M. ther 3-isopropylmalate cis-homo, Contains the dehydratase or aconitate (n = 1-4) YLTR consensus moautotrophicum AksF(2)-MTH 1388 gene in pRT21-a). isomerasef when coupled with sequence 0561 Briefly, potential Earl sites in the selected genes and homoaconitase MJ1003; involved characteristic of the vector backbone were disrupted to prevent interference (Small subunit) in the production of homoaconitase with the part/linker fusion preparation involving Earl diges C.-keto Suberic acid MJ1277 Involved in the MTH1387: tion/ligation cycles. No Earl sites were observed in AksD(1)- leucine biosynthesis Contains the MTH1386, AksD(2)-MTH1631 and Aksh (2)-MTH1387. YLVM consensus One Earl site was observed in each of the AkSA(1)- SeleCe. characteristic of MTH1630, AksA(2)-MTH1481, Aksh (1)-MTHO829 and isopropylmalate Aksh (2)-MTH1388 genes, and two Earl sites were observed isomerase in the Aks (1)-MTHO184 gene three Earl sites were detected AksF: MJ1596 Involved in the MTH1388 in the vector backbone pET21 -a. The Earl sites were dis 3-isopropylmalate production of dehydrogenase? C.-keto Suberic acid rupted by introducing a silent mutation(s) into the restriction homoisocitrate MJO720 Involved in the MTHO184 site. The three Earl sites in the vector were disrupted by dehydrogenase leucine biosynthesis mutating the last guanidine base of the restriction site into a thymine. 0562. The Earl free sequences were then used as input 0557. Identification of orthologs to the chain elongation sequences in the part designer Software to design the corre proteins in Methanobacterium thermoautotrophicum was sponding truncated parts flanked by Earl sites, as discussed in reported in the literature from comparative genomics data Section 3.1.1.2. The synthesized Methanobacterium ther between Methanocaldococcus jamaschii and Methanobacte moautotrophicum AkSA(1)-MTH1630 (TP48), AkSA(2)- rium thermoautotrophicum genomes, sequence analysis of MTH1481 (TP49), AksD(1)-MTH1386 (TP50), AksD(2)- the corresponding proteins and activity assay. Potential appli MTH1631 (TP51), Aksh (1)-MTHO829 (TP52), Aksh (2)- cation of the alternative Methanobacterium thermoau MTH1387 (TP53), Aksh (1)-MTHO184 (TP54) and Aksh (2)- totrophicum gene products and their corresponding Metha MTH1388 (TP55) truncated parts were inserted into the nocaldococcus jamaschii orthologs for the synthesis of alpha DNA2.0 pJ201 vector. The pET21-a truncated part vector was keto pimelic acid was reported (WO2010-104391 A2). fully synthesized with introduction of a fragment including Therefore, eight Methanobacterium thermoautotrophicum the chloramphenicol resistance gene to produce the vector proteins were selected for the exemplification of the chain TP31. elongation reaction between pimelic acid and Suberic acid, 0563 Phosphorylation and subsequent annealing of the i.e., AkSA 2-isopropylmalate synthase/homocitrate synthases inABLE oligonucleotides were performed as described in MTH1630 (GeneBank AE000666.1; fragment 1494539 section 3.1.1.2. Briefly, the oligos were incubated with T4 1495714) and MTH1481, (GeneBank AE000666.1; frag kinase at 37° C. for 30 minutes, followed by inactivation of ment 1337346-1338863); AksD 3-isopropylmalate dehy the enzyme at 65° C. for 20 minutes. Next, part linker fusions dratase/homoaconitases (Large subunit) MTH 1386 were prepared by ligating the 5'-end of the truncated part with (GeneBank AE000666.1; fragment 1253910-1255169); and its corresponding part oligo annealed fragment, and the MTH1631 (GeneBank AE000666.1; fragment 1495715 3'-end of the truncated part with the linker oligo annealed 1497001); Aksh. 3-isopropylmalate dehydratase or fragment, as previously described in section 3.1.1.2. Part isomerase/homoaconitases (Small subunit) MTH 0829 linker fusions corresponding to the selected genes (Methano ((GeneBank AE000666.1; fragment 752586-753098); and bacterium thermoautotrophicum AkSA(1)-MTH1630, AkSA MTH1387 (GeneBank AE000666.1; fragment 1255181 (2)-MTH1481, AksD(1)-MTH1386, AksD(2)-MTH1631, 1255669); and Aksh 3-isopropylmalate dehydrogenase/ho Aksh(1)-MTHO829, Aksh (2)-MTH1387, Aksh (1)- moisocitrate dehydrogenases MTHO184 (GeneBank MTHO184 and Akse (2)-MTH1388) were prepared by liga AE000666.1; fragment 130396-131391); and MTH 1388 tion of their respective annealed part oligos (POA48, POA49, (GeneBank AE000666.1; fragment 1255666-1256655) POA50, POA51, POA52, POA53, POA54 or POA55), their 0558 Wild-type genes were selected for the construction truncated part (TP48, TP49, TP50, TP51, TP52, TP53, TP54 of the corresponding E.coli expression vectors, except for the or TP55) and the annealed linker oligos from the vector back US 2015/0337275 A1 Nov. 26, 2015

bone. Part linker fusions corresponding to the vector back junctions between the 3'-end of the vector backbone and the bone were prepared by ligation of its annealed part oligos, its 5'-end of the genes as well as between the 3'-end of the gene truncated part, and the annealed linker oligos from either and the 5'-end of the vector backbone were sequenced to genes. A negative control gene-free assembly was also pre confirm the construction. pared by ligation of the vector backbone annealed part oligos, 0567 3.1.10.3 Expression of Exogenous Thioesterase its truncated part, and its annealed linker oligos, resulting in Genes in E. coli self-assembly of the vector backbone. 0568 Expression of exogenous Methanobacterium ther 0564) A ten and twenty-fold molar excesses of linker and moautotrophicum AkSA(1)-MTH1630, AkSA(2)-MTH 1481, part oligos to the truncated part was used in the gene and AksD(1)-MTH1386, AksD(2)-MTH1631, Aksh (1)- vector reactions to favor ligation between the truncated part MTHO829, Aksh (2)-MTH1387, Aksh (1)-MTHO184 and and the oligonucleotides during each Earl digestion/ligation Aksh (1)-MTHO184 chain elongation genes gene products in cycle. The 50LL Earl digestion/ligation reactions were incu E. coli was performed as described in section 3.1.1.3. Briefly, bated in a thermocycler (Eppendorf Mastercycler Gradient), electrocompetent BL21 (DE3) cells were transformed with alternating the temperature between 37° C. and 16°C. Upon 10 ng DNA, and the transformation samples were plated on completion of the cycles, samples were loaded on a 0.7% LB-Amp-Agar. Transformants were obtained after overnight agarose gels, and the expected size fragments were observed incubation at 37° C. A single clone was picked from each for preparation of the part/linker fusions. The correct size assembly, and 5 ml of LB-Amp medium were inoculated as a bands were then excised from the gel, and the DNA was gel starting culture. The ODoo of the culture was measured after extracted using the QIAGEN QIAquick Gel Extraction Kit. overnight incubation at 37° C. and 250 rpm, 2 mL culture The DNA concentration ranged from 13.6 ng/ul to 24.8 ug/ul (approximately 8x10) were used to inoculate 100 mL of in a total volume of 30 ul (DNA conc. range 4.9 nM to 53.9 LB-Amp, which were incubated in a 500 mL baffled shake nM). Self-assembly of the part/linker fusion (pET21-a part/ flask at 37° C. and 250 rpm. pET21-a linker) for the generation of the negative control 0569 Protein expression was induced by addition of 1 gene-free construct was also performed. mMIPTG (final conc.), and cultures were incubated at 37°C. 0565 Combination of the gene part/linker fusions with and 250 rpm. A 1 mL sample was taken prior to induction and their respective vector part/linker fusion was then carried out after 4 hand 24 h post-induction incubation, and the ODoo through a 2-part assembly, as described in section 3.1.1.2. measured to determine cell growth. After 24h post-induction Briefly, the gene part linkers and the vector part linkers were incubation, the remaining culture was transferred to a 50 mL incubated at room temperature for 30 min prior to transfor falcon tube, harvested by centrifugation at 5000 rpm for 10 mation of high efficiency chemically competent NEB10B E. minutes, and the cell pellets used for further analysis or stored coli cells using 2 ul of the assembly reaction and 10 ul of the at -20° C. competent cells. Transformed cells were plated on LB-Amp (0570 For SDS-PAGE analysis, the samples were centri agar, and incubated at 37° C. overnight. About one thousand fuged at 13000 rpm for 2 min, the supernatant was removed, clones were obtained for each assembly. Two random clones and the cells were lysed using the Bugbuster protein extrac were selected from each assembly, and the corresponding tion reagent Supplemented with lysozyme (15 mg/mL) and vectors were isolated using the QIAGEN QIAprep Miniprep benzonase (3.4 U/LL). The lysis reactions were then centri Kit. fuged at 13000 rpm for 2 min, the soluble fraction was trans 0566 Restriction was performed to confirm vector con ferred to a new tube, and the insoluble fraction re-suspended struction, as described in section 3.1.1.2. Briefly, clones in water. 20 ul of each fraction was mixed with 80 ul SDS obtained from I9 were analyzed using BglII and XmnI to Sample buffer (SDS-Loading Buffer, 9% DTT and water), identify positive clones for the insertion of the Methanobac heated in a heatblock for 5 minutes at 95°C., and 10ul of each terium thermoautotrophicum AkSA(1)-MTH1630 gene into SDS-sample preparation was then loaded on SDS-PAGE pET21-a, clones obtained from I10 assembly were analyzed 4-20% Tricene gels to analyze the protein content of the using PstI and Xmal/Xbal to identify insertion of AkSA(2)- soluble and insoluble fractions. MTH 1481 gene into pET21-a, clones obtained from Ill (0571. Only the Akse(2)-MTH1387 protein was success assembly were analyzed using HincII and AlwNI to identify fully expressed in soluble form, i.e., an 18 kD bank in the insertion of AksD(1)-MTH 1386 gene, clones obtained from soluble fraction after IPTG induction. The AkSA(1)- I12 assembly were analyzed using NcoI and BglII to identify MTH1630, AksD(2)-MTH1631, Aksh (1)-MTHO184 and insertion of AksD(2)-MTH 1631 gene, clones obtained from Aksh (2)-MTH1388 proteins were expressed in insoluble I13 assembly I 13 were analyzed using Mlul and XmnI to forms, i.e., a 44, 36, and 46 kD bands, respectively, were identify insertion of Aksh (1)-MTHO829 into pET21-a, detected in the insoluble portion. No expression was detected clones obtained from I14 assembly I 14 were analyzed using for the AkSA(2)-MTH1481, AksD(1)-MTH1386 and Aks Psil and Xmal/AlwNI to identify insertion of Aksh (2)- (2)-MTH 1387 proteins. Therefore, expression optimization MTH 1387 gene into the E. coli expression vector, clones was performed for AkSA(1)-MTH1630, AksD(2)-MTH1631, obtained from I15 assembly were analyzed using AlwNI and Aks (1)-MTHO184, Aksh (2)-MTH1388, AkSA(2)- HincII to identify insertion of Aks (1)-MTHO184 gene into MTH1481, AksD(1)-MTH1386, and Aksh (2)-MTH1387, the E. coli expression vector, and clones obtained from 116 based on reports of Successful expression of Methanocaldo assembly were analyzed using Bsal and HindIII/AlwNI to coccus jannaschii orthologs in the literature. In particular, identify insertion of the Aksh (2)-MTH1388. The restriction literature data on the expression of the M. janaschii orthologs products were run on an agarose gel, and the expected band revealed a set of different conditions in particular in terms of pattern was observed for the clones tested from each assem inducer concentration and induction temperature. The IPTG bly, confirming the construction of E. coli pET21-a expres conc. was varied between 0.1 and 1 mM (initial conditions 1 sion vectors harbouring the Methanobacterium thermoau mM) and the temperature of induction was varied between 16 totrophicum chain elongation genes. In addition, the part and 37 C (initial conditions 37 C). The same expression US 2015/0337275 A1 Nov. 26, 2015 54 protocol and the same SDS-PAGE sample preparation, as 0574. The final constructs were then prepared incubating described in the section 3.1.1.3 was used for the solubility the 4-part assembly reaction, i.e., the ADH1 promoter, each of optimization experiments. the Methanobacterium thermoautotrophicum genes AkSA (1)-MTH1630, AksA(2)-MTH1481, AksD(2)-MTH1631 0572. The level of AkSA(1)-MTH1630 protein expression and Akse(1)-MTHO829, the CYC1 terminator and the yeast was improved after decreasing 10-fold the IPTG concentra vector backbone, at 37 C. Then, 10ul high efficiency chemi tion, but the protein remained insoluble; optimization of the cally competent NEB10B E. coli cells were transformed with temperature did not affect expression or solubility. No sig 2 Jul of the assembly reaction, plated on LB-Kan-agar, and nificant improvement in AksD(2)-MTH1631 and Aks(2)- incubated at 37°C. overnight. Between two hundred and two MTH1388 solubility by lowering 10-fold the IPTG concen hundred and fifty clones were obtained for the 4-part assem tration since both proteins were only visible in the insoluble blies. Five random clones were then picked from the plates, fractions after induction. No significant improvement in Solu and the corresponding vectors were isolated using the bility was observed for AksD(2)-MTH1631 and Aks(2)- QIAGEN QIAprep Miniprep Kit. MTH 1388 by lowering the induction temperature to 16° C. 0575. The isolated constructs were analyzed using restric since both proteins were only visible in the insoluble fractions tion analysis. A preliminary analysis was performed using after induction. However, Aksh (1)-MTHO184 solubility was PmeI/SapI to identify any truncated vector contaminants. The improved by lowering the induction temperature to 30° C. clones did not contain truncated vector contaminants, so they Therefore, AkSA(1)-MTH1630 (19), AksD(2)-MTH1631 were then analyzed using the restriction enzymes XhoI/Ndel, (I12) and Akse (2)-MTH1388 (I16) may be expressed in dif Bsal and Mlul in order to confirm the insertion of the 4 ferent hosts, e.g., eukaryotic system like Saccharomyces cer part/linker fusions. The expected band pattern was observed evisiae or in a thermophile bacteria like Bacillus Stearother for both clones confirming the construction of S. cerevisiae mophilus. expression vector harbouring the ADH1 promoter, the gene of 0573. Since the Methanobacterium thermoautotrophicum interest, the CYC1 terminator and the yeast vector backbone. AkSA(1)-MTH1630, AkSA(2)-MTH1481, AksD(2)- Lastly, the junctions between the 4 parts were sequenced to MTH1631 and Akse (1)-MTH0829 gene products were not confirm proper assembly. expressed in a soluble form in E. coli using the pET21-a/ 0576 Transformation with the S. cerevisiae 124-127 IPTG expression system, these genes was cloned for expres expression vectors can be performed as described herein, and sion using the Saccharomyces cerevisiae expression system. expression from the constitutive promoter ADH1p can moni The expression vector was assembled using the inABLE tech tored after cell lysis using SDS-PAGE. The 124 construct nology described above, except the Sap restriction enzyme includes the ADH1 promoter, the Methanobacterium ther instead of Earl in the preparation of the part/linker fusion of moautotrophicum AkSA(1)-MTH 1630 gene, and the CYC1 the yeast vector backbone, comprising the constitutive pro terminator inserted in the yeast vector backbone. The 125 moter ADH1p, the transcription terminator CYC1t, and the construct includes the ADH1 promoter, the Methanobacte high copy number vector backbone derived from the rium thermoautotrophicum AkSA(2)-MTH 1481 gene, and YEplac195 plasmid. Electrocompetent TOP10 E. coli cells the CYC1 terminator. The 126 construct includes the ADH1 were transformed with each part/linker assembly and a large promoter, the Methanobacterium thermoautotrophicum DNA stock for was prepared using the QIAGEN Hi-Speed AksD(2)-MTH1631 gene, and the CYC1 terminator. The 127 Plasmid Purification Midi Kit (67.4-173.0 nM in 120 ul). The construct includes the ADH1 promoter, the Methanobacte expected size fragment was observed for the preparation of rium thermoautotrophicum Aks(1)-MTHO829 gene, and the yeast vector part/ADH1 promoter linker, despite the use the CYC1 terminator. of SapI, after agarose gel electrophoresis. The correct bands 0577 Patents, patent applications, publications, product were then excised from the gel, and the DNA was gel descriptions, and protocols cited throughout this application extracted using the QIAGEN QIAquick Gel Extraction Kit are incorporated herein by reference in their entireties for all (14.1-138.1 nM in 30 ul). purposes.

SEQUENCE LISTING

<16 Os NUMBER OF SEO ID NOS: 43

<21 Os SEQ ID NO 1 &211s LENGTH: 1362 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: sequence encoding IlvE-Omega Vf fusion

<4 OOs SEQUENCE: 1 atgaacaaac cqcagagctg ggaggcacgc gCagaalacct acagcctgta tigttt cact 60 gatatgc.cga gCttgcacca gcgcggcacg gttgttgtta ct cacggcga gggit cogtac 12O

attgtggatgtcaatggtcg tcgittatctg gacgcgaat a gcggcctgtg gaatatggtc 18O

gcgggttittg accatalaagg cctgattgac gcggcaaagg cacaatacga gCdCtttc.ca 24 O

ggittat catg Ctttctttgg togcatgagc gaccagaccg. tcatgttgtc. cdagaaactg 3 OO

US 2015/0337275 A1 Nov. 26, 2015 56

- Continued

Thr Gly Llys Pro Tyr Asn Ser Val Phe Gly Lieu Pro Leu Pro Gly Phe 1.65 17O 17s Val His Lieu. Thr Cys Pro His Tyr Trp Arg Tyr Gly Glu Glu Gly Glu 18O 185 19 O Thr Glu Glu Glin Phe Val Ala Arg Lieu Ala Arg Glu Lieu. Glu Glu Thir 195 2OO 2O5 Ile Glin Arg Glu Gly Ala Asp Thir Ile Ala Gly Phe Phe Ala Glu Pro 21 O 215 22O Val Met Gly Ala Gly Gly Val Ile Pro Pro Ala Lys Gly Tyr Phe Glin 225 23 O 235 24 O Ala Ile Lieu Pro Ile Lieu. Arg Llys Tyr Asp Ile Pro Val Ile Ser Asp 245 250 255 Glu Val Ile Cys Gly Phe Gly Arg Thr Gly Asn. Thir Trp Gly Cys Val 26 O 265 27 O Thr Tyr Asp Phe Thr Pro Asp Ala Ile Ile Ser Ser Lys Asn Lieu. Thr 27s 28O 285 Ala Gly Phe Phe Pro Met Gly Ala Val Ile Leu Gly Pro Glu Lieu. Ser 29 O 295 3 OO Lys Arg Lieu. Glu Thir Ala Ile Glu Ala Ile Glu Glu Phe Pro His Gly 3. OS 310 315 32O Phe Thr Ala Ser Gly His Pro Val Gly Cys Ala Ile Ala Lieu Lys Ala 3.25 330 335 Ile Asp Val Val Met Asn. Glu Gly Lieu Ala Glu Asn Val Arg Arg Lieu. 34 O 345 35. O Ala Pro Arg Phe Glu Glu Arg Lieu Lys His Ile Ala Glu Arg Pro Asn 355 360 365 Ile Gly Glu Tyr Arg Gly Ile Gly Phe Met Trp Ala Lieu. Glu Ala Val 37 O 375 38O Lys Asp Lys Ala Ser Llys Thr Pro Phe Asp Gly Asn Lieu. Ser Val Ser 385 390 395 4 OO Glu Arg Ile Ala Asn. Thir Cys Thr Asp Lieu. Gly Lieu. Ile Cys Arg Pro 4 OS 41O 415 Lieu. Gly Glin Ser Val Val Lieu. Cys Pro Pro Phe Ile Lieu. Thr Glu Ala 42O 425 43 O Glin Met Asp Glu Met Phe Asp Llys Lieu. Glu Lys Ala Lieu. Asp Llys Val 435 44 O 445

Phe Ala Glu Wall Ala 450

<210s, SEQ ID NO 3 &211s LENGTH: 1323 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: sequence encoding IlvE-Ad optimized omega fusion

<4 OOs, SEQUENCE: 3 atgagcgc.cg Ctaaactgcc agacctgagc cacctgtgga tigc.cgtttac C9C Caaccgc 6 O

Cagtttaaag caatcc.gcg cctgttggcg agcgcaaagg gCatgtact a tacgagctt C 12 O gatggcc.gcc aaattctgga C9gcacggca ggtctgtggit gcgtgaatgc aggcc attgc 18O cgtgaagaaa ttgttagcgc gatcgc.ct co Caagcgggcg tatggatta CdCaccgggc 24 O