PEPTIDES AND METHODS FOR THE CARBON-CARBON BOND FORMATION

Abstract

The present invention relates to methods for the preparation of α-hydroxyacyl compounds, and peptides for the catalyzed formation of α-hydroxyacyl compounds as well as their use in the preparation of α-hydroxyacyl compounds.

Claims

1. A method for preparation of a compound of the formula (I′) ##STR00466## wherein R represents ##STR00467## ##STR00468## ##STR00469## ##STR00470## ##STR00471## ##STR00472## ##STR00473## ##STR00474## ##STR00475## ##STR00476## ##STR00477## ##STR00478## ##STR00479## ##STR00480## ##STR00481## wherein R.sup.1, R.sup.2, R.sup.3, R.sup.4, R.sup.5, R.sup.6, R.sup.7, R.sup.8, and R.sup.9 represent independently of each other —H, —F, —Cl, —Br, —I, -Ph, —OH, —OCH.sub.3, —OC.sub.2H.sub.5, —OC.sub.3H.sub.7, —OCH(CH.sub.3).sub.2, —OC.sub.4H.sub.9, —OCH(CH.sub.3)CH.sub.2CH.sub.3, —OCH.sub.2CH(CH.sub.3).sub.2, —OC(CH.sub.3).sub.3, —O-cyclo-C.sub.3H.sub.5, —O-cyclo-C.sub.4H.sub.7, —O-cyclo-C.sub.5H.sub.9, —O-cyclo-C.sub.6H.sub.11, —OPh, —OCH.sub.2-Ph, —OCH.sub.2CH.sub.2-Ph —OCH═CH.sub.2, —OCH.sub.2—CH═CH.sub.2, —OCH.sub.2CH.sub.2—CH═CH.sub.2, —OCF.sub.3, —OC.sub.2F.sub.5, —SCH.sub.3, —SC.sub.2H.sub.5, —SC.sub.3H.sub.7, —SCH(CH.sub.3).sub.2, —SC.sub.4H.sub.9, —SCH(CH.sub.3)CH.sub.2CH.sub.3, —SCH.sub.2CH(CH.sub.3).sub.2, —SC(CH.sub.3).sub.3, —NO.sub.2, —NH.sub.2, —NHCH.sub.3, —NHC.sub.2H.sub.5, —NHC.sub.3H.sub.7, —NH-cyclo-C.sub.3H.sub.5, —NHCH(CH.sub.3).sub.2, —NHC(CH.sub.3).sub.3, —N(CH.sub.3).sub.2, —N(C.sub.2H.sub.5).sub.2, —N(C.sub.3H.sub.7).sub.2, —N(C.sub.4H.sub.9).sub.2, —N(cyclo-C.sub.3H.sub.5).sub.2, —N[CH(CH.sub.3).sub.2].sub.2, —N[C(CH.sub.3).sub.3].sub.2, —CH.sub.3, —C.sub.2H.sub.5, —C.sub.3H.sub.7, —CH(CH.sub.3).sub.2, —C.sub.4H.sub.9, —CH.sub.2—CH(CH.sub.3).sub.2, —CH(CH.sub.3)—C.sub.2H.sub.5, —C(CH.sub.3).sub.3, —C.sub.5H.sub.11, —CH(CH.sub.3)—C.sub.3H.sub.7, —CH.sub.2—CH(CH.sub.3)—C.sub.2H.sub.5, —CH.sub.2—CH.sub.2—CH(CH.sub.3).sub.2, —CH(CH.sub.3)—CH(CH.sub.3)—CH.sub.3, —C(CH.sub.3).sub.2—C.sub.2H.sub.5, —C.sub.6H.sub.13, —CH(CH.sub.3)—C.sub.4H.sub.9, —CH.sub.2—CH(CH.sub.3)—C.sub.3H.sub.7, —CH.sub.2—CH.sub.2—CH.sub.2—CH(CH.sub.3).sub.2, —CH(CH.sub.3)—CH(CH.sub.3)—C.sub.2H.sub.5, —C(CH.sub.3).sub.2—C.sub.3H.sub.7, —C.sub.7H.sub.15, —CH(CH.sub.3)—C.sub.5H.sub.11, —CH.sub.2—CH(CH.sub.3)—C.sub.4H.sub.9, —CH.sub.2—CH.sub.2—CH.sub.2—CH.sub.2—CH(CH.sub.3).sub.2, —C.sub.8H.sub.17, —C.sub.9H.sub.19, —C.sub.10H.sub.21, —CH(CH.sub.3)—C.sub.3H.sub.7, —CH.sub.2—CH(CH.sub.3)—C.sub.2H.sub.5, —CH(CH.sub.3)—CH(CH.sub.3).sub.2, —C(CH.sub.3).sub.2—C.sub.2H.sub.5, —CH.sub.2—C(CH.sub.3).sub.3, —CH(C.sub.2H.sub.5).sub.2, —C.sub.2H.sub.4—CH(CH.sub.3).sub.2, —CH.sub.2F, —CF.sub.2I, —CHF.sub.2, —CF.sub.3, —CH.sub.2I, —CH.sub.2Br, —CH.sub.2I, —CH.sub.2—CH.sub.2F, —CH.sub.2—CHF.sub.2, —CH.sub.2—CF.sub.3, —CH.sub.2—CH.sub.2I, —CH.sub.2—CH.sub.2Br, —CH.sub.2—CH.sub.2I, —CH═CH.sub.2, —CH.sub.2—CH═CH.sub.2, —C(CH.sub.3)═CH.sub.2, —CH═CH—CH.sub.3, —C.sub.2H.sub.4—CH═CH.sub.2, —C.sub.7H.sub.15, —C.sub.8H.sub.17, —CH.sub.2—CH═CH—CH.sub.3, —CH═CH—C.sub.2H.sub.5, —CH.sub.2—C(CH.sub.3)═CH.sub.2, —CH(CH.sub.3)—CH═CH, —CH═C(CH.sub.3).sub.2, —C(CH.sub.3)═CH—CH.sub.3, —CH═CH—CH═CH.sub.2, —C.sub.3H.sub.6—CH═CH.sub.2, —C.sub.2H.sub.4—CH═CH—CH.sub.3, —CH.sub.2—CH═CH—C.sub.2H.sub.5, —CH═CH—C.sub.3H.sub.7, —CH.sub.2—CH═CH—CH═CH.sub.2, —CH═CH—CH═CH—CH.sub.3, —C≡CH, —C≡C—CH.sub.3, —CH.sub.2—C≡CH, —C.sub.2H.sub.4—C≡CH, —CH.sub.2—C≡C—CH.sub.3, —C≡C—C.sub.2H.sub.5, —C.sub.3H.sub.6—C≡CH, —C.sub.2H.sub.4—C≡C—CH.sub.3, —CH.sub.2—C≡C—C.sub.2H.sub.5, —C≡C—C.sub.3H.sub.7, —CHO, —COCH.sub.3, —COC.sub.2H.sub.5, —COC.sub.3H.sub.7, —CO-cyclo-C.sub.3H.sub.5, —COCH(CH.sub.3).sub.2, —COC(CH.sub.3).sub.3, —COOH, —COOCH.sub.3, —COOC.sub.2H.sub.5, —COOC.sub.3H.sub.7, —COO-cyclo-C.sub.3H.sub.5, —COO-cyclo-C.sub.4H.sub.7, —COO-cyclo-C.sub.5H.sub.9, —COO-cyclo-C.sub.6H.sub.11, —COOCH(CH.sub.3).sub.2, —COOC(CH.sub.3).sub.3, —O—COOCH.sub.3, —O—COOC.sub.2H.sub.5, —O—COOC.sub.3H.sub.7, —O—COO-cyclo-C.sub.3H.sub.5, —O—COOCH(CH.sub.3).sub.2, —O—COOC(CH.sub.3).sub.3, —CONH.sub.2, —CONHCH.sub.3, —CONHC.sub.2H.sub.5, —CONHC.sub.3H.sub.7, —CONH-cyclo-C.sub.3H.sub.5, —CONH[CH(CH.sub.3).sub.2], —CONH[C(CH.sub.3).sub.3], —CON(CH.sub.3).sub.2, —CON(C.sub.2H.sub.5).sub.2, —CON(C.sub.3H.sub.7).sub.2, —CON(cyclo-C.sub.3H.sub.5).sub.2, —CON[CH(CH.sub.3).sub.2].sub.2, —CON[C(CH.sub.3).sub.3].sub.2, —NHCOCH.sub.3, —NHCOC.sub.2H.sub.5, —NHCOC.sub.3H.sub.7, —NHCO-cyclo-C.sub.3H.sub.5, —NHCO—CH(CH.sub.3).sub.2, —NHCO—C(CH.sub.3).sub.3, —SOCH.sub.3, —SOC.sub.2H.sub.5, —SOC.sub.3H.sub.7, —SO-cyclo-C.sub.3H.sub.5, —SOCH(CH.sub.3).sub.2, —SOC(CH.sub.3).sub.3, —SO.sub.2CH.sub.3, —SO.sub.2C.sub.2H.sub.5, —SO.sub.2C.sub.3H.sub.7, —SO.sub.2-cyclo-C.sub.3H.sub.5, —SO.sub.2CH(CH.sub.3).sub.2, —SO.sub.2C(CH.sub.3).sub.3, —SO.sub.3H, —SO.sub.3Na, —SO.sub.3K, —SO.sub.3CH.sub.3, —SO.sub.3C.sub.2H.sub.5, —SO.sub.3C.sub.3H.sub.7, —SO.sub.3-cyclo-C.sub.3H.sub.5, —SO.sub.3CH(CH.sub.3).sub.2, —SO.sub.3C(CH.sub.3).sub.3, —SO.sub.2NH.sub.2, —NHCO—OCH.sub.3, —NHCO—OC.sub.2H.sub.5, —NHCO—OC.sub.3H.sub.7, —NHCO—O-cyclo-C.sub.3H.sub.5, —NHCO—OCH(CH.sub.3).sub.2, —NHCO—OC(CH.sub.3).sub.3, —NH—CO—NH.sub.2, —NH—CO—NHCH.sub.3, —NH—CO—NHC.sub.2H.sub.5, —NH—CO—NHC.sub.3H.sub.7, —NH—CO—NH-cyclo-C.sub.3H.sub.5, —NH—CO—NH[CH(CH.sub.3).sub.2], —NH—CO—NH[C(CH.sub.3).sub.3], —NH—CO—N(CH.sub.3).sub.2, —NH—CO—N(C.sub.2H.sub.5).sub.2, —NH—CO—N(C.sub.3H.sub.7).sub.2, —NH—CO—N(cyclo-C.sub.3H.sub.5).sub.2, —NH—CO—N[CH(CH.sub.3).sub.2].sub.2, —NH—CO—N[C(CH.sub.3).sub.3].sub.2, —NH—CS—NH.sub.2, —NH—CS—NHCH.sub.3, —NH—CS—NHC.sub.2H.sub.5, —NH—CS—NHC.sub.3H.sub.7, —NH—CS—NH-cyclo-C.sub.3H.sub.5, —NH—CS—NH[CH(CH.sub.3).sub.2], —NH—CS—NH[C(CH.sub.3).sub.3], —NH—CS—N(CH.sub.3).sub.2, —NH—CS—N(C.sub.2H.sub.5).sub.2, —NH—CS—N(C.sub.3H.sub.7).sub.2, —NH—CS—N(cyclo-C.sub.3H.sub.5).sub.2, —NH—CS—N[CH(CH.sub.3).sub.2].sub.2, —NH—CS—N[C(CH.sub.3).sub.3].sub.2, —B(OH).sub.2, —B(OR.sup.22), ##STR00482## ##STR00483## ##STR00484## ##STR00485## ##STR00486## ##STR00487## R.sup.11 to R.sup.16 represents independently of each other: —H, —NH.sub.2, —OH, —OCH.sub.3, —OC.sub.2H.sub.5, —OC.sub.3H.sub.7, —OCF.sub.3, —CF.sub.3, —F, —Cl, —Br, —I, —CH.sub.3, —C.sub.2H.sub.5, —C.sub.3H.sub.7, —CH(CH.sub.3).sub.2, -Ph, or —CN; R.sup.10, R.sup.17, R.sup.18, R.sup.19, R.sup.20, and R.sup.21 represents independently of each other —H, —CH.sub.3, —C.sub.2H.sub.5, —C.sub.3H.sub.7, —CH(CH.sub.3).sub.2, —C.sub.4H.sub.9, —CH.sub.2—CH(CH.sub.3).sub.2, —CH(CH.sub.3)—C.sub.2H.sub.5, —C(CH.sub.3).sub.3, —C.sub.5H.sub.11, —CH(CH.sub.3)—C.sub.3H.sub.7, —CH.sub.2—CH(CH.sub.3)—C.sub.2H.sub.5, —CH(CH.sub.3)—CH(CH.sub.3).sub.2, —C(CH.sub.3).sub.2—C.sub.2H.sub.5, —CH.sub.2—C(CH.sub.3).sub.3, —CH(C.sub.2H.sub.5).sub.2, —C.sub.2H.sub.4—CH(CH.sub.3).sub.2, CH.sub.2—CH.sub.2F, —CH.sub.2—CHF.sub.2, —CH.sub.2—CF.sub.3, —CH═CH.sub.2, —CH.sub.2—CH═CH.sub.2, —C(CH.sub.3)═CH.sub.2, —CH═CH—CH.sub.3, —C.sub.2H.sub.4—CH═CH.sub.2, —C.sub.7H.sub.15, —C.sub.8H.sub.17, —CH.sub.2—CH═CH—CH.sub.3, —CH═CH—C.sub.2H.sub.5, —CH.sub.2—C(CH.sub.3)═CH.sub.2, —CH(CH.sub.3)—CH═CH, —CH═C(CH.sub.3).sub.2, —C(CH.sub.3)═CH—CH.sub.3, —CH═CH—CH═CH.sub.2, —C.sub.3H.sub.6—CH═CH.sub.2, —C.sub.2H.sub.4—CH═CH—CH.sub.3, —CH.sub.2—CH═CH—C.sub.2H.sub.5, —CH═CH—C.sub.3H.sub.7, —CH.sub.2—CH═CH—CH═CH.sub.2, —CH═CH—CH═CH—CH.sub.3, —C≡CH, —C≡C—CH.sub.3, —CH.sub.2—C≡CH, —C.sub.2H.sub.4—C≡CH, —CH.sub.2—C≡C—CH.sub.3, —C≡C—C.sub.2H.sub.5, —C.sub.3H.sub.6—C≡CH, —C.sub.2H.sub.4—C≡C—CH.sub.3, —CH.sub.2—C≡C—C.sub.2H.sub.5, —C≡C—C.sub.3H.sub.7, -cyclo-C.sub.3H.sub.5, -cyclo-C.sub.4H.sub.7, -cyclo-C.sub.5H.sub.9, -cyclo-C.sub.6H.sub.11, —CHO, —COCH.sub.3, —COC.sub.2H.sub.5, —COC.sub.3H.sub.7, —CO-cyclo-C.sub.3H.sub.5, —COCH(CH.sub.3).sub.2, —COC(CH.sub.3).sub.3, —COOH, —COOCH.sub.3, —COOC.sub.2H.sub.5, —COOC.sub.3H.sub.7, —COO-cyclo-C.sub.3H.sub.5, —COOCH(CH.sub.3).sub.2, —COOC(CH.sub.3).sub.3, —CONH.sub.2, —CONHCH.sub.3, —CONHC.sub.2H.sub.5, —CONHC.sub.3H.sub.7, —CONH-cyclo-C.sub.3H.sub.5, —CONH[CH(CH.sub.3).sub.2], —CONH[C(CH.sub.3).sub.3], —CON(CH.sub.3).sub.2, —CON(C.sub.2H.sub.5).sub.2, —CON(C.sub.3H.sub.7).sub.2, —CON(cyclo-C.sub.3H.sub.5).sub.2, —CON[CH(CH.sub.3).sub.2].sub.2, —CON[C(CH.sub.3).sub.3].sub.2, preferably R.sup.21 cannot be —H; R.sup.22 represents —CH.sub.3, —C.sub.2H.sub.5, —C.sub.3H.sub.7, —CH(CH.sub.3).sub.2, —C.sub.4H.sub.9, —CH.sub.2—CH(CH.sub.3).sub.2, —CH(CH.sub.3)—C.sub.2H.sub.5, —C(CH.sub.3).sub.3, —C.sub.5H.sub.11, —CH(CH.sub.3)—C.sub.3H.sub.7, —CH.sub.2—CH(CH.sub.3)—C.sub.2H.sub.5, —CH(CH.sub.3)—CH(CH.sub.3).sub.2, —C(CH.sub.3).sub.2—C.sub.2H.sub.5, —CH.sub.2—C(CH.sub.3).sub.3, —CH(C.sub.2H.sub.5).sub.2, —C.sub.2H.sub.4—CH(CH.sub.3).sub.2, CH.sub.2—CH.sub.2F, —CH.sub.2—CHF.sub.2, —CH.sub.2—CF.sub.3, —CH═CH.sub.2, —CH.sub.2—CH═CH.sub.2, —C(CH.sub.3)═CH.sub.2, —CH═CH—CH.sub.3, —C.sub.2H.sub.4—CH═CH.sub.2, —C.sub.7H.sub.15, —C.sub.8H.sub.17, —CH.sub.2—CH═CH—CH.sub.3, —CH═CH—C.sub.2H.sub.5, —CH.sub.2—C(CH.sub.3)═CH.sub.2, —CH(CH.sub.3)—CH═CH, —CH═C(CH.sub.3).sub.2, —C(CH.sub.3)═CH—CH.sub.3, —CH═CH—CH═CH.sub.2, —C.sub.3H.sub.6—CH═CH.sub.2, —C.sub.2H.sub.4—CH═CH—CH.sub.3, —CH.sub.2—CH═CH—C.sub.2H.sub.5, —CH═CH—C.sub.3H.sub.7, —CH.sub.2—CH═CH—CH═CH.sub.2, —CH═CH—CH═CH—CH.sub.3, —C≡CH, —C≡C—CH.sub.3, —CH.sub.2—C≡CH, —C.sub.2H.sub.4—C≡CH, —CH.sub.2—C≡C—CH.sub.3, —C≡C—C.sub.2H.sub.5, —C.sub.3H.sub.6—C≡CH, —C.sub.2H.sub.4—C≡C—CH.sub.3, —CH.sub.2—C≡C—C.sub.2H.sub.5, —C≡C—C.sub.3H.sub.7, -cyclo-C.sub.3H.sub.5, X represents —OH, —O.sup.− or —SCoA, wherein n is an integer 1, 2, or 3; or a salt thereof, comprising: A) providing a compound R—CHO (1) and an acyl-CoA (2) ##STR00488## wherein Y represents —H or —COO.sup.−, in a buffer solution with a pH value in the range of 5.0 to 7.5 optionally together with at least one co-solvent; B) performing a coupling reaction of R—CHO (1) and acyl-CoA (2) by an oxalyl-CoA decarboxylase (3a) in the presence of thiamine diphosphate (3b) and a magnesium (II) cation to obtain the α-hydroxyacyl-CoA thioester (I′) ##STR00489## wherein X represents —S—CoA, and the oxalyl-CoA decarboxylase (3a) is a peptide encoded by a gene selected from Methylobacterium extorquens or a mutant thereof; and optionally comprising the additional step C) C) cleaving the coenzyme A moiety of said α-hydroxyacyl-CoA thioester (I′) by a hydrolase, α-hydroxyacyl-CoA:oxalate CoA-transferase and/or α-hydroxyacyl-CoA:formate CoA-transferase to obtain the compound of the formula (I′) or a salt thereof ##STR00490## wherein X represents —OH or O.sup.−, and the hydrolase is acyl-CoA thioester hydrolase YciA (4) encoded by a gene selected from E. coli.

2. The method for the preparation of the compounds according to claim 1 represented by formula (I) ##STR00491## wherein in step B) a (S)-α-hydroxyacyl-CoA thioester (I) ##STR00492## wherein X represents —S—CoA is obtained, or in step C) the compound of the formula (I) or a salt thereof ##STR00493## wherein X represents —OH or O.sup.− is obtained.

3. The method for preparation of the compounds according to claim 2, wherein the acyl-CoA (2) in step A) and step B) has the following structure ##STR00494## wherein Y represents —COO.sup.−; and wherein in step C) the coenzyme A moiety is cleaved by a hydrolase, α-hydroxyacyl-CoA:oxalate CoA-transferase and/or α-hydroxyacyl-CoA:formate CoA-transferase.

4. The method for the preparation of the compounds according to claim 2 represented by formula (II) ##STR00495##

5. The method according to claim 4, wherein steps A), B) and C) are carried out as cascade reaction in one-pot system.

6. The method according to claim 3, further comprising preparing the oxalyl-CoA (2′) by the following step A′) A′) reacting an oxalate (C.sub.2O.sub.4.sup.2−) (2a′) with a free coenzyme A (2b) by oxalyl-CoA synthetase (2c′) in the presence of ATP (2e) to produce an oxalyl-CoA (2′).

7. The method according to claim 6, wherein steps A′), A), B), and C) are carried out as cascade reaction in one-pot system.

8. The method according to claim 7, wherein in step A′) a ratio of the oxalate (2a′) and the free coenzyme A (2b) is 10 to 10,000.

9. The method according to claim 1, wherein in step B), a concentration of the oxalyl-CoA decarboxylase is in the range of 1 to 25 μM and/or a concentration of the magnesium (II) cation is in the range of 2.5 mM to 20 mM.

10. The method according to claim 1, wherein in step B), the mutant comprises OXC.sub.Me-Y497A, OXC.sub.Me-S568A, or OXC.sub.Me-Y497A-S568A.

11. The method according to claim 6, wherein in step A′) the oxalate is selected from the group consisting of (NH.sub.4.sup.+).sub.2C.sub.2O.sub.4, Li.sub.2C.sub.2O.sub.4, Na.sub.2C.sub.2O.sub.4, K.sub.2C.sub.2O.sub.4, Rb.sub.2C.sub.2O.sub.4, Cs.sub.2C.sub.2O.sub.4, MgC.sub.2O.sub.4, CaC.sub.2O.sub.4, ZnC.sub.2O.sub.4, FeC.sub.2O.sub.4, MnC.sub.2O.sub.4, NiC.sub.2O.sub.4, CuC.sub.2O.sub.4, and CoC.sub.2O.sub.4.

12. The method according to claim 1, wherein R represents ##STR00496## and R.sup.1, R.sup.2, R.sup.3, R.sup.4, R.sup.5, R.sup.6 and R.sup.7 represents independently of each other —H, —Cl, —OH, —OCH.sub.3, —NO.sub.2, —N(CH.sub.3).sub.2, —C≡CH, —CHO, —COCH.sub.3 or —COOH, wherein n represents an integer 1.

13. A peptide OXC.sub.Me-Y497A having the sequence SEQ ID NO. 2, OXC.sub.Me-S568A having the sequence SEQ ID NO. 3 or OXC.sub.Me-Y497A-S568A having the sequence SEQ ID NO. 4.

14. A composition comprising a peptide OXC.sub.Me having the sequence SEQ ID NO. 1, OXC.sub.Me-Y497A having the sequence SEQ ID NO. 2, OXC.sub.Me-S568A having the sequence SEQ ID NO. 3 or OXC.sub.Me-Y497A-S568A having the sequence SEQ ID NO. 4 and a thioesterase YciA encoded by a gene selected from E. coli.

15. The composition according to claim 14 further comprising an oxalyl-CoA synthetase.

Description

DESCRIPTION OF THE FIGURES

[0812] FIG. 1 shows the reaction of an aldehyde R—CHO with formyl-CoA or oxalyl-CoA to an α-hydroxyacyl-CoA thioester, and a possible subsequent formation of an α-hydroxy carboxylic acid or an α-hydroxy carboxylate.

[0813] FIG. 2 shows the hydrolysis of an α-hydroxyacyl-CoA thioester to the α-hydroxy carboxylate (first reaction), and the transfer of the coenzyme A moiety from the α-hydroxyacyl-CoA thioester to a formate (second reaction) resulting in the formation of an α-hydroxy carboxylate and a formyl-CoA thioester as well as the transfer of the coenzyme A moiety from α-hydroxyacyl-CoA thioester to an oxalate resulting in the formation of an α-hydroxy carboxylate and an oxalyl-CoA thioester.

[0814] FIG. 3 shows the catalytic cycle of a one-pot system performed as a cascade reaction using a thioesterase YciA in order to hydrolyse an α-hydroxyacyl-CoA thioester to an α-hydroxy carboxylate.

[0815] FIG. 4 shows the generation of oxalyl-CoA thioester by a transfer of the coenzyme A moiety from formyl-CoA thioester by means of a formyl-CoA:oxalate CoA transferase (FRC).

[0816] FIG. 5 shows the catalytic cycle of a one-pot system performed as a cascade reaction using a thioesterase YciA in order to hydrolyse an α-hydroxyacyl-CoA thioester to an α-hydroxy carboxylate, wherein a formyl-CoA thioester is transferred to a oxalyl-CoA thioester by means of a formyl-CoA:oxalate CoA-transferase (FRC).

[0817] FIG. 6 shows the catalytic cycle of a one-pot system performed as a cascade reaction using a transferase in order to transfer the coenzyme A moiety from an α-hydroxyacyl-CoA thioester to an oxalate resulting in the formation of an α-hydroxy carboxylate and oxalyl-CoA thioester.

[0818] FIG. 7 shows the catalytic cycle of a one-pot system performed as a cascade reaction using a transferase in order to transfer the coenzyme A moiety from an α-hydroxyacyl-CoA thioester to a formate resulting in the formation of an α-hydroxy carboxylate and formyl-CoA thioester.

[0819] FIG. 8 shows the peptide sequence SEQ ID No. 1 of oxalyl-CoA decarboxylase encoded by a gene selected from Methylobacterium extorquens (OXC.sub.Me).

[0820] FIG. 9 shows the peptide sequence SEQ ID No. 2 (OXC.sub.Me-Y497A) of a mutation of oxalyl-CoA decarboxylase encoded by a gene selected from Methylobacterium extorquens.

[0821] FIG. 10 shows the peptide sequence SEQ ID No. 3 (OXC.sub.Me-S568A) of a mutation of oxalyl-CoA decarboxylase encoded by a gene selected from Methylobacterium extorquens.

[0822] FIG. 11 shows the peptide sequence SEQ ID No. 4 (OXC.sub.Me-Y497A-S568A) of a mutation of oxalyl-CoA decarboxylase encoded by a gene selected from Methylobacterium extorquens.

[0823] FIG. 12 shows the MS/MS spectra of the α-hydroxyacyl-CoA thioester products. Calculated m/z values of the fragmentation products are indicated on top. Spectra on the left show the parent ion. Spectra on the right show the fragmentation products of the parent ion (asterisk). The numbers in the spectrum indicate measured m/z values. Mandelyl-CoA was produced from deuterated benzaldehyde; the deuterium is retained on the α-carbon in the product.

[0824] FIG. 13 shows Michaelis-Menten graphs of OXC.sub.Me and mutants thereof according to example 6. Error bars show standard deviation of three replicates.

[0825] FIG. 14 shows competing reaction pathways of OXC. Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP and 25 mM benzaldehyde, 1 μM OXC.sub.Me or 1 μM OXC.sub.Me-Y497A. The reaction was started by adding either 1 μM formyl-CoA (A) or 1 μM oxalyl-CoA (B). Samples were taken after 0, 2, 5, 10 and 20 minutes and analyzed with the LC-MS detection of CoA esters. CoA esters were quantified by comparison to synthetic standards of formyl-CoA, oxalyl-CoA and mandelyl-CoA, respectively. Concentration of CoA-esters was determined by comparison to a standard curve obtained from chemically synthesized formyl-CoA, oxalyl-CoA and mandelyl-CoA, respectively. Error bars show standard deviation of three replicates. C, apparent turnover numbers were determined by linear regression of the mandelyl-CoA formation rate over 20 min. wt=wild type=OXC.sub.Me; Y497A=OXC.sub.Me-Y497A.

[0826] FIG. 15 shows the screen for mandelyl-CoA Thioesterase activity. CoA formation was detected with the Ellman's reagent (5,5′-dithiobis-(2-nitrobenzoic acid)), which reacts with free thiols under release of 2-nitro-5-thiobenzoate (ε.sub.412nm=14, 150 M.sup.−1 cm.sup.−1). Assays were carried out at 30° C. in 50 mM MES-KOH pH 6.8 and contained 1 mM Ellman's reagent and 0.5 μM YciA or 0.5 μM TesB or 0.5 μM Paal. The reaction was started by adding 0.1 mM mandelyl-CoA (time point indicated with the arrow).

[0827] FIG. 16 shows the thioesterase activity of YciA towards oxalyl-CoA, formyl-CoA and racemic mandelyl-CoA. Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP and 25 mM benzaldehyde, 2 μM YciA (A) or no YciA (B). The reaction was started by adding either 0.5 mM formyl-CoA or 0.5 mM oxalyl-CoA or 0.5 mM mandelyl-CoA. Samples were taken after 0, 1, 5 and 30 minutes and analyzed with the LC-MS detection of CoA esters. CoA esters were quantified by comparison to synthetic standards of formyl-CoA, oxalyl-CoA and mandelyl-CoA, respectively. Error bars show standard deviation of two.

[0828] FIG. 17 shows calibration curves of commercially obtained rac. mandelic acid, rac. 3-phenyllactic acid, rac. 4-chloromandelic acid and (S)-2-chloromandelic acid. The lines show linear fit to the data.

[0829] FIG. 18 shows Michaelis-Menten graph of OXS with oxalate as substrate according to example 5. Oxalyl-CoA production was followed by coupling OXS to purified PanE2, an NADPH-dependent oxalyl-CoA reductase from M. extorquens. An assay containing 50 mM potassium phosphate pH 6.5, 0.3 mM NADPH, 10 mM MgCl.sub.2, 1 mM ATP, 2.5 mM CoA, 1.2 μM PanE2, 176 nM OXS was preincubated for 2 min and the reaction started by adding disodium oxalate to a final concentration of 5, 10, 25, 100, 250, 1,000 μM, respectively. Reaction procedure was monitored by following the oxidation of NADPH at 340 nm. Error bars show standard deviation of three replicates.

[0830] FIG. 19 shows mandelic acid formation of the cascade over time. ee of (S)-mandelic acid is indicated for the last time point (22 h) according to example 13. Mandelic acid formation of the OXS-OXC-YciA cascade over time. Enantiomeric excess of (S)-mandelic acid is indicated for the last time point (22 h). Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 5 μM OXS, 5 μM OXC and 2 μM YciA. The reaction was started by adding 10 mM oxalate. As negative controls, YciA, OXS or OXC were omitted in separate reactions. Samples were taken after 0, 0.05, 0.25, 0.66, 1, 2, 3, 22 hours and analyzed with the LC-MS detection of mandelic acid derivatives. Mandelic acid was quantified by comparison to commercially obtained standards of mandelic acid.

[0831] FIG. 20 shows the oxalyl-CoA decarboxylation activity of OXC.sub.Me and HACL.sub.Hs. Assays were carried out at 30° C. in 100 mM MES-KOH pH 6.8 and contained 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.05 μM OXC.sub.Me or 1 μM HACL.sub.Hs or no enzyme. The reaction was started by adding 0.5 mM oxalyl-CoA. Samples were taken after 1, 15, 30, 60 min and analyzed with the LC-MS detection of CoA esters.

[0832] FIG. 21 shows the peptide sequence of oxalyl-CoA synthetase OXS encoded by a gene selected by Methylobacterium extorquens (SEQ ID NO. 5).

[0833] FIG. 22 shows the peptide sequence of YciA by a gene selected from E. coli (SEQ ID NO.6).

[0834] FIG. 23 shows the amino acid sequence SEQ ID NO. 18 of acyl-CoA synthetase (eryACS-1) from Erythrobacter sp. NAP1 used as a formyl-CoA synthetase.

[0835] FIG. 24 shows the amino acid sequence SEQ ID NO. 19 of a mutant of acyl-CoA synthetase used as a formyl-CoA synthetase.

[0836] FIG. 25 shows the nucleotide sequence of eryACS-1 (ENA|EAQ27819|EAQ27819.1acetyl-CoA synthetase from Erythrobactersp. NAP1 a; herein SEQ ID NO. 20).

[0837] FIG. 26 shows the nucleotide sequence of eryACS-1 (ENA|EAQ27819|EAQ27819.1 acetyl-CoA synthetase from Erythrobacter sp. NAP1; herein SEQ ID NO. 21) being codon-optimized for E. coli.

[0838] FIG. 27 shows the primers used for site directed mutagenesis of eryACS-1 (from Erythrobacter sp. NAP1)

[0839] FIG. 28 shows the DNA sequence SEQ ID NO. 25 of the thioesterase YciA encoded by a gene selected from E. coli.

[0840] FIG. 29 shows the one-pot synthesis of mandelyl-CoA from oxalate, CoA and benzaldehyde, by combining OXS and OXC.sub.Me. Assays were carried out at 30° C. in reaction buffer containing 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.85 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 5 μM OXS and 5 μM OXC.sub.Me. The reaction was started by adding 10 mM disodium oxalate. Samples were taken after 0, 10, 60, 240 min and analyzed with the CoA ester method described above. Error bars show standard deviation of two replicates.

[0841] FIG. 30 shows Michaelis-Menten graph of eryACS-1 and variants thereof with formate as substrate according to example 6. AMP production was followed at 30° C. by coupling eryACS-1 to purified adenylate kinase (ADK), pyruvate kinase (PK) and lactate dehydrogenase (LDH). An assay containing 200 mM MOPS-KOH pH 7.8, 0.3 mM NADH, 10 mM MgCl.sub.2, 2 mM ATP, 0.5 mM CoA, 2.5 mM PEP, 0.2 mg/mL ADK, 15 U/mL PK from rabbit muscle (Sigma), 23 U/mL LDH from rabbit muscle (Sigma), and an appropriate concentration of eryACS-1 was preincubated for 2 min and the reaction started by adding sodium formate to a final concentration of 10, 25, 50, 100, 150, 200 mM, respectively. Reaction procedure was monitored by following the oxidation of NADH at 340 nm. Error bars show standard deviation of three replicates. eryACS-1 V379I=eryACS-1, wherein at position 379 valine is replaced by isoleucine. eryACS-1 V379I Δpatz=eryACS-1, wherein at position 379 valine is replaced by isoleucine, and the enzyme were expressed by an E. coli strain lacking lysine acetyltransferase patZ.

[0842] FIG. 31 shows the one-pot synthesis of mandelyl-CoA from formate, CoA and benzaldehyde, by combining FCS and OXC.sub.Me according to example 21. Assays were carried out at 30° C. in reaction buffer containing 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.85 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 5 μM FCS and 5 μM OXC.sub.Me. The reaction was started by adding 50 mM sodium formate. Samples were taken after 0, 10, 60, 240 min and analyzed with the CoA ester method described above. Error bars show standard deviation of two replicates.

[0843] FIG. 32 shows (S)-mandelic acid formation of the cascade over time. (S)-Mandelic acid formation of the FCS-OXC-YciA cascade over time. Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 5 mM ATP, 25 mM benzaldehyde, 1 μM FCS, 2 μM OXC and 2 μM YciA. The reaction was started by adding 50 mM sodium formate. Samples were taken after 0, 2, 15, 30, 60, 120, 300 min and analyzed with the LC-MS detection of mandelic acid derivatives. Mandelic acid was quantified by comparison to commercially obtained standards of mandelic acid.

[0844] FIG. 33 shows formyl-CoA ‘proofreading’ by FRC. FRC converts formyl-CoA and oxalate into formate and oxalyl-CoA. Because mandelyl-CoA formation rate proceeds faster with oxalyl-CoA than with formyl-CoA as C.sub.1 donor, addition of FRC results in increased mandelyl-CoA formation rate. Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP and 25 mM benzaldehyde, 1 μM OXC.sub.Me (A,C) or 1 μM OXC.sub.Me-Y497A (B,D). In B and D the reaction additionally contained 20 μM FRC. The reaction was started by adding 1 mM oxalyl-CoA. Samples were taken after 0, 2, 5, 10 and 20 minutes and analyzed with the LC-MS detection of CoA esters. CoA esters were quantified by comparison to synthetic standards of formyl-CoA, oxalyl-CoA and mandelyl-CoA, respectively. Error bars show standard deviation of three replicates. E, mandelyl-CoA formation rates were determined by linear regression (slope of mandelyl-CoA in the graphs). Final concentration refers to the last time point (20 minutes). wt=wild type=OXC.sub.Me; Y497A=OXC.sub.Me-Y497A.

[0845] FIG. 34 shows formyl-CoA ‘proofreading’ by FRC in the OXS-OXC.sub.Me-YciA cascade according to example 27. Addition of FRC results in increased mandelic acid and 3-phenyllactic acid formation rate. Assays were carried out at 30° C. in reaction buffer containing 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 7.6 mM benzaldehyde or phenylacetaldehyde, 2 μM FRC, 2 μM YciA, 10 μM OXC.sub.Me-Y497A and 5 μM OXS. The reaction was started by adding 10 mM disodium oxalate. Samples were taken after 0, 15, 30, 60 min and analyzed with the mandelic acid derivatives method described above.

[0846] FIG. 35 shows the one-pot synthesis of mandelyl-CoA from oxalate, CoA and benzaldehyde, by combining OXS, OXC.sub.Me and FRC according to example 29, which acts as formyl-CoA proofreading enzyme. Assays were carried out at 30° C. in reaction buffer containing 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.85 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 1 μM OXC.sub.Me, 5 μM OXS and 20 μM FRC and were mixed and the reaction was initiated by the addition of 10 mM disodium oxalate. Samples were taken after 0, 2, 5, 10, 30 min and quenched with 4% formic acid and products analyzed with the CoA ester method described above.

EXAMPLES

[0847] Abbreviations [0848] LC-MS liquid chromatography mass spectrometry [0849] rcf relative centrifugal force [0850] HPLC high performance liquid chromatography [0851] UPLC ultra performance liquid chromatography [0852] TOF time of flight [0853] ESI electron spray ionization [0854] HEPES 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid [0855] MES 2-(N-morpholino)ethanesulfonic acid [0856] MOPS 3-Morpholinopropane-1-sulfonic acid [0857] TES 2-[[1,3-dihydroxy-2-(hydroxymethyl)propan-2-yl]amino]ethanesulfonic acid [0858] ThDP thiamine diphosphate [0859] UV ultraviolet [0860] CoA coenzyme A [0861] ATP adenosine triphosphate [0862] dNTP deoxy nucleoside triphosphate [0863] NADH nicotinamide adenine dinucleotide [0864] NADPH nicotinamide adenine dinucleotide phosphate [0865] CDI carbonyldiimidazole [0866] THF tetrahydrofuran [0867] ACN acetonitrile [0868] DMF dimethylformamide [0869] EG ethylene glycol [0870] DMSO dimethyl sulfoxide [0871] FCS formyl-CoA synthetase [0872] OXS oxalyl-CoA synthetase [0873] OXC oxalyl-CoA decarboxylase [0874] OXC.sub.Me oxalyl-CoA decarboxylase from Methylobacterium extorquens [0875] HACL hydroxyacyl-CoA lyases [0876] HACL.sub.Hs human hydroxyacyl-CoA lyases [0877] HACOT α-hydroxyacyl-CoA:oxalate transferase [0878] HAFT α-hydroxyacyl-CoA:formate transferase [0879] FRC formyl-CoA:oxalate transferase [0880] ADK adenylate kinase [0881] LDH lactate dehydrogenase [0882] PK pyruvate kinase [0883] NdeI endonuclease isolated from Neisseria denitrificans [0884] BamHI endonuclease isolated from Bacillus amyloliquefaciens [0885] DpnI endonuclease isolated from Diplococcus pneumoniae [0886] PCR polymerase chain reaction [0887] LB lysogeny broth [0888] TB terrific broth [0889] IPTG Isopropyl β-D-1-thiogalactopyranoside [0890] OD.sub.600 optical density at 600 nm [0891] FPLC fast protein liquid chromatography [0892] SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresis

Material and Methods

[0893] Chemicals

[0894] Unless stated otherwise, standard laboratory reagents were obtained from Sigma-Aldrich® (Steinheim, Germany) or Carl Roth GmbH & Co. KG (Karlsruhe, Germany) with the highest purity available. Propionaldehyde, Vanillin, (R)-mandelic acid were (S)-mandelic acid were obtained from Tokyo Chemical Industry (Zwijndrecht, Belgium). 4-chloromandelic acid was obtained from Alfa Aesar (ThermoFisher (Kandel) GmbH; Kandel, Germany).

[0895] Method for Detection of α-Hydroxyacyl-CoA Thioester (that is the Method Described in LC-MS Detection of CoA Esters) or α-Hydroxy Carboxylic Acids (that is the Method Described in LC-MS Detection of Mandelic Acid Derivatives)

[0896] LC-MS analyses was used to investigate the formation of the product and side products.

[0897] LC-MS Analyses

[0898] Samples were prepared for LC-MS analysis by quenching an aliquot of a reaction with formic acid (final concentration 4%) and centrifuging for 10 min at 17,000 rcf, to remove precipitated proteins. LC-MS data were analyzed and quantified using MassHunter Qualitative Navigator and Quantitative Analysis software (Agilent, Waldbronn, Germany).

[0899] LC-MS Detection of CoA Esters

[0900] Samples were diluted 1:10 in H.sub.2O. UPLC-high resolution MS of CoA esters was performed as described previously..sup.[5] CoA esters were analyzed using an Agilent 6550 iFunnel Q-TOF LC-MS system equipped with an electrospray ionization source set to positive ionization mode. Compounds were separated on a RP-18 column (50 mm×2.1 mm, particle size 1.7 μm, Kinetex EVO C18, Phenomenex) using a mobile phase system comprised of 50 mM ammonium formate pH 8.1 (A) and methanol (B). Chromatographic separation was carried out using the following gradient condition at a flow rate of 250 μl/min: 0 min 2.5% B; 2.5 min 2.5% B; 8 min 23% B; 10 min 80% B; 11 min 80%; 12 min 2.5% B; 12.5 min 0% B. The column oven was set to 40° C. and autosampler was maintained at 10° C. Standard injection volume was 1 μl. Capillary voltage was set at 3.5 kV and nitrogen gas was used as nebulizing (20 psig), drying (13 l/min, 225° C.) and sheath gas (12 l/min, 400° C.). The time of flight (kindly confirm) (TOF) was calibrated using an ESI-L Low Concentration Tuning Mix (Agilent) before measurement (residuals less than 2 ppm for five reference ions) and was recalibrated during a run using 922.0908 m/z as reference mass. The scan range for MS and MS/MS data is 100-1000 m/z and 50-1000 m/z respectively. Collision energy used for MS/MS fragmentation was 35 eV.

[0901] The MS/MS spectra are depicted in FIG. 12.

[0902] LC-MS Detection of Mandelic Acid Derivatives

[0903] Samples were diluted 1:10 in H.sub.2O. UPLC-high resolution MS analyses were performed on an Agilent 6550 iFunnel QTOF LC/MS system equipped with an electrospray ionization source to negative ionization mode. The analytes were isocratically chromatographed on a chiral column (100 mm×2.1 mm, particle size 2.7 μm, Poroshell 120 Chiral-T, Agilent) kept at ambient temperature using a mobile phase system comprised of 30:70 20 mM ammonium formate pH 4/methanol at a flow rate of 250 μl/min for 10 min. Samples were held at 10° C. and injection volume was 1 μl. Capillary voltage was set at 3.5 kV and nitrogen gas was used as nebulizing (20 psig), drying (13 l/min, 225° C.) and sheath gas (12 l/min, 400° C.). MS data were acquired with a scan range of 100-1100 m/z. For quantification the calculated m/z value of [M-H].sup.− was used to obtain the extracted ion count from the total ion count.

[0904] Enzyme Assays with LC-MS Detection

[0905] Unless noted otherwise, all LC-MS assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP.

[0906] Kinetic Investigations

[0907] Spectrophotometric enzyme assays Assays were performed on a Cary-60 UV/Vis spectrophotometer (Agilent) at 30° C. using quartz cuvettes (10 mm path length; Hellma, Müllheim, Germany). For the determination of steady-state kinetic parameters, each concentration was measured in triplicates and the obtained curves were fit using GraphPad Prism 7. Hyperbolic curves were fit to the Michaelis-Menten equation to obtain apparent k.sub.cat and K.sub.M values.

Synthesis of the Starting Materials and Reference Samples

Example 1—Formyl-CoA Synthesis

[0908] Formyl-CoA was synthesized as described previously (W. S. Sly, E. R. Stadtman, J. Biol. Chem. 1963, 238, 2632-2638; b) S. Jonsson, S. Ricagno, Y. Lindqvist, N. G. Richards, J. Biol. Chem. 2004, 279, 36003-36012.). After Extraction with diethylether formyl-CoA was purified by preparative HPLC-MS with an acetonitrile gradient in 25 mM ammonium formate pH 4.2. The fractions containing the product were lyophilized and stored at −20° C. Formyl-CoA was dissolved in aq. HCl (pH 4). The concentration was determined by enzymatic depletion with PduP (J. Zarzycki, M. Sutter, N. S. Cortina, T. J. Erb, C. A. Kerfeld, Sci. Rep. 2017, 7, 42757.), following NADH consumption at 340 nm.

Example 2—Oxalyl-CoA Synthesis

[0909] Oxalyl-CoA was synthesized enzymatically with OXS. A 5 mL reaction containing 50 mg mM CoA (0.064 mmol, 1 eq.), 52 mg ATP (0.086 mmol, 1.3 eq.), 10 mg disodium oxalate (0.075 mmol, 1.2 eq.) in buffer (100 mM MES-KOH pH 6.8, 15 mM MgCl.sub.2) was started by adding OXS to a final concentration of 0.4 mg/mL and incubated at 30° C. for 1 hour. The reaction was quenched with 250 μL formic acid and the enzyme removed by centrifugation (4,000×g, 4° C., 10 min). Oxalyl-CoA was purified by preparative HPLC-MS with an acetonitrile gradient in 25 mM ammonium formate pH 4.2. The fractions containing the product were lyophilized and stored at −20° C. Oxalyl-CoA was dissolved in 10 mM acetate buffer pH 4.5. The concentration was determined spectrophotometrically, using the extinction coefficient for saturated CoA esters (ε.sub.260nm=16.4 cm.sup.−1 mM.sup.−1) or by enzymatic depletion with PanE2, following NADPH consumption at 340 nm.

Example 3—Mandelyl-CoA Synthesis (Reference Sample)

[0910] Mandelyl-CoA and (S)-mandelyl-CoA were synthesized chemically with the carbonyldiimidazole (CDI) CoA-acylation method described previously (D. M. Peter, B. Vögeli, N. S. Cortina, T. J. Erb, Molecules 2016, 21, 517.). 21 mg CDI (0.127 mmol, 4 eq.) was dissolved in 1 mL tetrahydrofuran, mandelic acid or (S)-mandelic acid was added (0.127 mmol, 4 eq.) and the mixture stirred at 22° C. for 15 min. 25 mg CoA (0.032 mmol, 1 eq.) was dissolved in 1 mL 0.5 M NaHCO.sub.3 and added to the reaction mixture, followed by stirring at 22° C. for 30 min. THF was removed by applying vacuum (100 mbar) for 5 min. The mixture was then purified by preparative HPLC with a methanol gradient in 25 mM ammonium formate pH 8.0. The fractions containing the product were lyophilized and stored at −20° C. Mandelyl-CoA was dissolved in 10 mM acetate buffer pH 4.5. The concentration was determined spectrophotometrically, using the extinction coefficient for saturated CoA esters (ε.sub.260nm=16.4 cm.sup.−1 mM.sup.−1). The concentration was confirmed by enzymatic depletion with YciA, detecting the liberated CoA with Ellman's reagent (ε.sub.412nm=14.15 cm.sup.−1 mM.sup.−1).

Example 4—Synthesis of the Peptides

Example 4a—Cloning and Mutagenesis

[0911] Oligonucleotides were obtained from Eurofins Genomics (Ebersbach, Germany). oxc (MexAM1_META1p0990), oxs (MexAM1_META1p2130), and panE2 (MexAM1_META1p3141) were PCR-amplified from M. extorquens chromosomal DNA using the corresponding primers (Table 1). The purified PCR products were digested with NdeI and BamHI and ligated into pET-16b. Correct cloning was confirmed by sequencing (Eurofins Genomics).

[0912] The gene encoding eryACS-1 (GenBank: EAQ27819.1) was obtained by gene synthesis performed by the Joint Genome Institute (Walnut Creek, Calif., USA). The gene was codon optimized for E. coli and sub-cloned into pSEVA141.

[0913] Human Hacl1 (GenBank: CAB60200.1) was obtained by gene synthesis (Table 2), performed by BaseClear (Leiden, The Netherlands). The gene was codon optimized for E. coli and sub-cloned into pET-16b.

[0914] adk, paal, tesB and yciA were obtained from the ASKA collection ([M. Kitagawa, T. Ara, M. Arifuzzaman, T. Ioka-Nakamichi, E. Inamoto, H. Toyonaga, H. Mori, DNA Res. 2005, 12, 291-299).

[0915] Point mutations were introduced into oxc by PCR using mismatch primers (Table 1). A 50 μL reaction contained 60 ng of pET-16b_OXC.sub.Me, 0.25 μM forward and reverse primer, 200 μM dNTP, 5 μL 10× reaction buffer, 1 μL Phusion polymerase (2 U/μL). Template plasmid was removed by DpnI digest (10 U) at 37° C. immediately after PCR amplification. Mutations were confirmed by sequencing.

TABLE-US-00002 TABLE 1 Primers used for cloning of OXC and OXS and site-directed mutagenesis of OXC. Nucleotide sequence (5′ to 3′) restriction Primer name sites or mismatches are underlined) oxc_fw_Ndel GTTCACATATGACCGTCCAGGCCCAG (SEQ ID NO. 7) oxc_rv_BamHI CGCTGGATCCTCACTTCTTCTTCAAGGTGCTC (SEQ ID NO. 8) oxs_fw_Ndel GTTCACATATGACGATGCTTCTGCC (SEQ ID NO. 9) oxs_rv_BamHI CAATGGATCCTCAGACCAGCCCGAG (SEQ ID NO. 10) panE2_fw_Ndel GCGCACATATGAGCATCGCGATCGTCG (SEQ ID NO. 11) panE2_rv_BamHI CAGAGGATCCTCATGCTCCCTGGATCGC (SEQ ID NO. 12) oxc_Y497A_fw CAACAACGGCATCGCTCGCGGCACCGAC (SEQ ID NO. 13) oxc_Y497A_rv GTCGGTGCCGCGAGCGATGCCGTTGTTG (SEQ ID NO. 14) oxc_S568A_fw CCGGCAGCGAGGCCGGCAATATCGG (SEQ ID NO. 15) oxc_S568A_rv CCGATATTGCCGGCCTCGCTGCCGG (SEQ ID NO. 16)

TABLE-US-00003 TABLE 2 Synthetic gene sequences Gene name DNA sequence hacl1 ATGCCGGACAGTAACTTCGCAGAGCGCAGCGAGGAGCAGG (Gen Bank: TGTCTGGTGCTAAAGTCATCGCTCAGGCCCTGAAAACGCAA CAB60200) GATGTGGAGTACATATTTGGCATCGTAGGCATCCCAGTGAC (SEQ ID NO. CGAAATCGCCATTGCTGCCCAGCAGCTAGGCATCAAGTACA 24) TCGGGATGAGGAATGAGCAAGCGGCTTGTTATGCTGCCTCC GCGATTGGATATCTGACAAGCAGGCCAGGAGTCTGCCTTGT TGTTTCTGGCCCAGGTCTCATCCATGCCTTGGGCGGTATGG CAAATGCAAACATGAACTGCTGGCCCTTGCTTGTGATTGGTG GTTCCTCTGAAAGAAACCAAGAAACAATGGGAGCTTTCCAGG AGTTTCCTCAGGTTGAAGCTTGTAGATTATATACCAAGTTCTC TGCCCGCYCAAGCAGCATAGAAGCTATTCCTTTTGTTATTGA AAAGGCAGTGAGAAGCAGTATCTATGGTCGTCCAGGTGCTT GCTATGTTGACATACCAGCAGATTTTGTGAACCTTCAGGTGA ATGTGAATTCTATAAAGTACATGGAACGCTGCATGTCACCTC CTATTAGCATGGCAGAAACCTCTGCTGTGTGCACGGCGGCT TCTGTTATTAGGAATGCCAAACAACCCCTTCTTATCATCGGG AAAGGTGCTGCTTACGCTCATGCAGAAGAGAGTATCAAGAAA TTGGTGGAGCAATATAAACTGCCATTTTTGCCCACCCCTATG GGAAAGGGTGTTGTCCCTGACAACCATCCATACTGTGTAGG TGCAGCCAGATCCAGGGCTTTGCAATTTGCTGATGTAATTGT GTTATTTGGTGCCAGACTAAATTGGATTTTACATTTTGGACTG CCTCCAAGATATCAGCCAGATGTGAAGTTTATCCAGGTTGAT ATCTGTGCAGAAGAATTGGGGAATAATGTAAAGCCCGCTGTT ACTTTGCTAGGAAACATACATGCTGTCACTAAGCAGCTTTTA GAGGAACTTGATAAAACACCATGGCAGTATCCTCCAGAGAG CAAGTGGTGGAAAACTCTGAGAGAAAAAATGAAGAGCAATG AAGCTGCATCCAAGGAACTAGCTTCTAAAAAATCCCTGCCTA TGAATTATTACACAGTATTCTACCATGTTCAAGAACAACTACC TAGAGACTGTTTCGTGGTAAGTGAAGGAGCAAATACTATGGA CATTGGACGGACTGTGCTTCAGAACTACCTTCCTCGTCACAG GCTTGATGCTGGTACTTTCGGAACAATGGGAGTTGGTTTGG GATTTGCTATTGCAGCTGCCGTGGTGGCTAAAGATAGAAGC CCTGGGCATTGGATCATCTGTGTGGAAGGAGACAGTGCATT TGGGTTTTCTGGCATGGAGGTAGAAACCATCTGCAGGTACA ACTTGCCAATCATACTGTTGGTAGTGAATAACAATGGAATTTA CCAAGGTTTTGATACAGATACTTGGAAAGAAATGTTAAAATTT CAAGATGCTACTGCAGTGGTCCCTCCAATGTGTTTGCTGCCA AATTCACATTATGAGCAAGTCATGACTGCATTTGGAGGCAAA GGGTATTTTGTACAAACACCAGAAGAACTCCAAAAATCCCTG GAGCAGAGCCTAGCAGACACAACTAAACCTTCTCTTATCAAC ATCATGATTGAGCCACAAGCCACACGGAAGGCCCAGGATTT TCATTGGCTGACCCGCTCTAATATGTAA hacl1 codon ATGCCGGATTCAAACTTTGCTGAACGTTCAGAAGAACAGGTT optimized for TCTGGCGCAAAAGTGATTGCGCAAGCGCTAAAAACCCAGGA E. coli (SEQ ID TGTCGAATATATCTTCGGTATAGTGGGTATTCCGGTGACGGA NO. 17) AATCGCCATCGCGGCTCAGCAACTCGGTATTAAATACATCGG TATGCGTAATGAGCAGGCGGCTTGTTACGCAGCAAGCGCGA TTGGCTATCTGACCTCAAGACCGGGGGTATGCCTTGTTGTCA GCGGCCCGGGCCTGATTCATGCCCTGGGTGGTATGGCGAA TGCTAACATGAACTGCTGGCCGCTGTTAGTTATAGGCGGCA GCAGTGAACGCAATCAGGAGACCATGGGCGCATTCCAAGAG TTTCCTCAGGTGGAAGCCTGCCGCCTGTATACCAAATTTTCT GCTAGACCTTCGTCAATAGAAGCCATCCCCTTCGTGATTGAA AAAGCGGTCCGTAGTTCTATCTATGGGCGCCCTGGGGCTTG CTATGTTGACATTCCCGCCGACTTCGTAAACCTCCAAGTGAA TGTCAATTCCATTAAATACATGGAACGCTGTATGTCGCCGCC GATCAGTATGGCCGAGACCAGCGCCGTGTGCACAGCGGCG AGTGTTATCCGCAACGCCAAACAGCCATTACTAATAATTGGG AAAGGAGCCGCCTATGCTCACGCTGAAGAGTCCATCAAAAA ACTGGTGGAACAATACAAGTTGCCCTTTTTACCTACTCCCAT GGGTAAAGGTGTTGTACCTGATAATCATCCATACTGCGTCGG TGCAGCCCGTTCCCGCGCACTACAGTTCGCTGACGTGATTG TGCTGTTTGGTGCGCGTTTAAACTGGATACTCCACTTCGGTT TACCGCCACGCTACCAGCCGGATGTGAAATTTATACAGGTTG ACATCTGTGCCGAAGAACTGGGCAATAATGTGAAACCAGCA GTCACTTTGCTGGGGAACATCCATGCAGTCACCAAACAACTG CTTGAAGAGCTGGACAAGACGCCGTGGCAATATCCGCCAGA GTCAAAATGGTGGAAGACGCTACGTGAGAAAATGAAGAGCA ACGAAGCGGCTTCGAAAGAGTTGGCCTCCAAAAAGTCATTG CCTATGAATTACTATACCGTCTTTTATCATGTCCAAGAGCAGC TGCCGCGGGATTGCTTTGTCGTGTCGGAGGGCGCCAACACC ATGGATATAGGTCGCACTGTCCTTCAGAACTATCTGCCCCGG CATCGCCTGGACGCTGGTACCTTCGGAACTATGGGCGTAGG GCTGGGCTTTGCTATTGCGGCGGCAGTGGTAGCAAAGGATC GCAGCCCGGGGCAGTGGATTATTTGTGTAGAAGGGGACTCC GCATTTGGATTCTCTGGTATGGAGGTAGAGACAATATGCCGT TATAATTTGCCAATTATCCTTCTAGTGGTGAATAACAATGGTA TTTATCAGGGCTTTGATACAGATACGTGGAAGGAAATGCTGA AATTTCAAGATGCCACAGCCGTGGTACCCCCCATGTGTTTGC TGCCGAACTCTCATTACGAGCAAGTGATGACAGCATTCGGC GGCAAGGGTTACTTTGTGCAGACACCGGAAGAACTTCAAAA GAGCCTTCGCCAGAGCTTGGCGGACACCACGAAACCGAGTC TGATTAATATAATGATTGAACCACAGGCCACAAGGAAGGCAC AAGACTTTCACTGGCTCACGCGTTCAAATATGTAA

Example 4b—Protein Production and Purification

[0916] All proteins except HACL.sub.Hs and FCS were heterologously produced in E. coli BL21 (DE3). 500 mL TB containing 100 μg/mL ampicillin (in the case of OXC.sub.Me, OXS, FRC, PduP, PanE2, eryACS-1) or 34 μg/mL chloramphenicol (AdK, Paal, TesB and YciA) was inoculated with freshly-transformed cells and incubated at 37° C. After reaching an OD.sub.600 of 0.8, expression was induced by adding IPTG to a final concentration of 0.25 mM and the incubation temperature was lowered to 25° C. FCS was produced in E. coli BL21-AI (ThermoFisher Scientific) in which the peptidyl-lysine N-acetyltransferase gene patZ had been replaced by a kanamycin resistance cassette kanR. 1 L TB containing 100 μg/mL ampicillin and 50 μg/mL kanamycin was inoculated with freshly-transformed cells and incubated at 37° C. After reaching an OD.sub.600 of 0.6, L-arabinose was added to a final concentration of 0.2% (w/v) and the incubation temperature was lowered to 25° C. After 30 minutes, expression was induced by IPTG to a final concentration of 0.5 mM. HACL.sub.Hs was produced in E. coli ArcticExpress (DE3). 1 L LB containing 100 μg/mL ampicillin and 15 μg/mL gentamycin was inoculated with freshly-transformed cells and incubated at 37° C. After reaching an OD.sub.600 of 0.8 the culture was cooled on ice for 15 min. Then expression was induced by adding IPTG to a final concentration of 0.1 mM and the incubation temperature was lowered to 15° C. Cells were harvested after 16 h (24 h for HACL.sub.Hs) by centrifugation (4500×g, 10 min) and resuspended in buffer A (500 mM KCl, 50 mM HEPES-KOH pH 7.6). If not used immediately, cell pellets were flash-frozen in liquid nitrogen and stored at −20° C. The cell lysate obtained by sonication was clarified by centrifugation 75,000×g at 4° C. for 45 min. The supernatant was filtered through a 0.4 μm syringe tip filter (Sarstedt, Nümbrecht, Germany). Ni-affinity purification was performed with an Äkta FPLC system from GE Healthcare (GE Healthcare, Freiburg, Germany). The filtered soluble lysate was loaded onto a 1 mL Ni-Sepharose Fast Flow column (HisTrap FF, GE Healthcare, Little Chalfont, UK) that had been equilibrated with 10 mL buffer A. After washing with 20 mL 85% buffer A, 15% buffer B (500 mM KCl, 50 mM HEPES-KOH pH 7.6, 500 mM imidazole), the protein was eluted with 100% buffer B. Fractions containing purified protein were pooled and the buffer was exchanged to storage buffer (150 mM KCl, 50 mM HEPES-KOH pH 7.6) with a desalting column (HiTrap, GE Healthcare). Proteins were concentrated by ultrafiltration (Amicon Ultra). Concentration was determined on a NanoDrop 2000 Spectrophotometer (Thermo Scientific, Waltham, Mass., USA) using the extinction coefficient at 280 nm, as calculated by protparam (https://web.expasy.org/protparam/). Enzyme purity was confirmed by SDS-PAGE. The purified proteins were stored in 50 vol % glycerol at −20° C. OXC.sub.Me wild-type and mutants were flash-frozen in liquid nitrogen and stored at −80° C.

[0917] Kinetic Investigations

Example 5—Michaelis-Menten Kinetics of OXS

[0918] Oxalyl-CoA production was followed by coupling OXS to purified PanE2, an NADPH-dependent oxalyl-CoA reductase from M. extorquens. An assay containing 50 mM potassium phosphate pH 6.5, 0.3 mM NADPH, 10 mM MgCl.sub.2, 1 mM ATP, 2.5 mM CoA, 1.2 μM PanE2, 176 nM OXS was preincubated for 2 min and the reaction started by adding disodium oxalate to a final concentration of 5, 10, 25, 100, 250, 1′000 μM, respectively. Reaction procedure was monitored by following the oxidation of NADPH at 340 nm.

[0919] Result: The steady-state kinetic parameters are shown in FIG. 18 (k.sub.cat=1.30±0.02 s.sup.−1; K.sub.M(oxalate)=9±1 μM)

Example 6—Michaelis-Menten Kinetics OXC.SUB.Me.-Decarboxylation

[0920] Formyl-CoA production was followed by coupling OXC to purified PduP, a promiscuous CoA-dependent aldehyde dehydrogenase that reduces formyl-CoA to formaldehyde under NADH consumption. An assay containing 50 mM MES-KOH pH 6.5, 0.3 mM NADH, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 5 μM PduP, and OXC.sub.Me or mutant thereof like OXC.sub.Me-Y497A, OXC.sub.Me-S568A or OXC.sub.Me-Y497A-S568A (concentration depending on the mutant) was preincubated for 2 min and the reaction started by adding oxalyl-CoA (concentrations depending on the mutant). Reaction procedure was monitored by following the oxidation of NADH at 340

[0921] The formyl-CoA formed after the decarboxylation of the oxalyl-CoA by ThDP and OXC.sub.Me was removed from the reaction mixture in order to investigate the decarboxylation without the influence of the equilibrium.

[0922] Results: The results are depicted in the following table 3 and in curve of the kinetic investigations in FIG. 13.

TABLE-US-00004 TABLE 3 Steady-state kinetic parameters of oxalyl- CoA decarboxylation catalyzed by OXCMe..sup.[a] Mutation k.sub.cat (s.sup.−1) K.sub.M (μM) k.sub.cat/K.sub.M (s.sup.−1 M.sup.−1) wild-type 98 ± 3  105 ± 11 9.3 × 10.sup.5 Y497A 1.32 ± 0.04 180 ± 17 7.3 × 10.sup.3 (OXC.sub.Me-Y497A) S568A 5.53 ± 0.11 23 ± 2 1.1 × 10.sup.5 (OXC.sub.Me-S568A) Y497A S568A 0.32 ± 0.01 103 ± 15 3.1 × 10.sup.3 (OXC.sub.Me-Y497A-S568A) .sup.[a]Michaelis-Menten graphs are shown in FIG. 13.

[0923] In OXC.sub.Me variants OXC.sub.Me-Y497A having the sequence SEQ ID NO. 2, and OXC.sub.Me-S568A having the sequence SEQ ID NO. 3, formyl-CoA formation was decreased 20 to 50-fold, while the K.sub.M of both variants was largely unaffected (Table 3 and FIG. 13). When starting with oxalyl-CoA and benzaldehyde OXC.sub.Me-Y497A showed 5-fold increased mandelyl-CoA production rate (20 min.sup.−1; FIG. 14C), suggesting that activity of the enzyme was successfully redirected towards carboligation by suppressing protonation of thiamine-α-hydroxyl-CoA-adduct (see FIG. 3).

[0924] The mutation Y497A increased the ratio of carboligation to decarboxylation by a factor of approximately 400 compared to the wild-type

Example 7—OXC.SUB.Me .Aldehyde and Ketone Screen

[0925] In reaction buffer, 1 mM formyl-CoA and 10 mM aldehyde (formaldehyde, acetaldehyde, glycolaldehyde, propionaldehyde, glyceraldehyde, glyoxylate, succinic semialdehyde, benzaldehyde, phenylacetaldehyde, and acetone) was mixed with 5 μM OXC.sub.Me or HACL.sub.Hs. The reaction was stopped after 1 h and products analyzed with the CoA ester method described above.

[0926] Result: Glyceraldehyde, glyoxylate and acetone were not accepted by OXC.sub.Me, i.e. a carbon-carbon-bond was not formed. The results are listed in table 4:

TABLE-US-00005 TABLE 4 Comparison of the aldehyde substrate scope of OXC.sub.Me [00397] [00398] Entry R Product name OXCMe.sup.[c] HACL.sub.Hs.sup.[c] 1 H glycolyl-CoA 100 11 2 CH.sub.3 lactyl-CoA 74 100 3 CH.sub.2CH.sub.3 2-hydroxybutyryl- 5 100 CoA 4 CH.sub.2OH glyceryl-CoA 1 100 5 CHOHCH.sub.2OH erythronyl-CoA n.d..sup.[b] n.d..sup.[b] 6 COOH tartronyl-CoA n.d. n.d..sup.[b] 7 (CH2)2COOH 2- 1 100 hydroxyglutaryl- CoA 8 Ph mandelyl-CoA 100 3 9 CH2Ph phenyllactyl-CoA 22 100 [a]The reaction contained 2a-2i (10 mM), formyl-CoA (1 mM), OXC.sub.Me. Products were analyzed by LC-MS after 1 h reaction time. .sup.[b]Product not detected. .sup.[c]Relative activity in %. Relative activity refers to the comparison of OXC and HACL.

Example 8—Comparison of the Decarboxylation Activity of OXC.SUB.Me .and HACL.SUB.Hs

[0927] Oxalyl-CoA decarboxylation activity of OXC.sub.Me and HACL.sub.Hs. Assays were carried out at 30° C. in 100 mM MES-KOH pH 6.8 and contained 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.05 μM OXC.sub.Me or 1 μM HACL.sub.Hs or no enzyme. The reaction was started by adding 0.5 mM oxalyl-CoA. Samples were taken after 1, 15, 30, 60 min and analyzed with the LC-MS detection of CoA esters.

[0928] Results: The results are depicted in FIG. 20. In the control experiment without an enzyme the concentration of the oxalyl-CoA does not change, and thus formyl-CoA is not formed.

Example 9—Comparison OXC Wild Type and OXC.SUB.Me-Y497A .Starting Form Formyl-CoA or Oxalyl-CoA

[0929] Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP and 25 mM benzaldehyde, 1 μM OXC.sub.Me or 1 μM OXC.sub.Me-Y497A. The reaction was started by adding either 1 mM formyl-CoA or 1 mM oxalyl-CoA. Samples were taken after 0, 2, 5, 10 and 20 minutes and analyzed with the LC-MS detection of CoA esters. CoA esters were quantified by comparison to synthetic standards of formyl-CoA, oxalyl-CoA and mandelyl-CoA, respectively. Concentration of CoA-esters was determined by comparison to a standard curve obtained from chemically synthesized formyl-CoA, oxalyl-CoA and mandelyl-CoA, respectively.

[0930] Results: The results are depicted in FIGS. 14A and 14B. Starting from the formyl-CoA the formation is faster by means of OXC.sub.Me-Y497A than OXC.sub.Me (wild type). When oxalyl-CoA is used as a C.sub.1-source, the mandelyl-CoA is formed much faster in case of the OXC.sub.Me-Y497A.

Example 10—YciA Substrate Screen (Selectivity of YciA Towards Formyl-CoA, Oxalyl-CoA and Mandelyl-CoA)

[0931] Thioesterase activity of YciA towards oxalyl-CoA, formyl-CoA and racemic mandelyl-CoA. Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP and 25 mM benzaldehyde, 2 μM YciA (A) or no YciA (B). The reaction was started by adding either 0.5 mM formyl-CoA or 0.5 mM oxalyl-CoA or 0.5 mM mandelyl-CoA. Samples were taken after 0, 1, 5 and 30 minutes and analyzed with the LC-MS detection of CoA esters. CoA esters were quantified by comparison to synthetic standards of formyl-CoA, oxalyl-CoA and mandelyl-CoA, respectively.

[0932] Results: The results are depicted in FIG. 16.

[0933] YciA from E. coli has shown a relatively high α-hydroxyacyl-CoA in particular mandelyl-CoA thioesterase activity, low activity with formyl-CoA and no activity with oxalyl-CoA.

Example 11—Comparison of the Activity of the Thioesterase Activity of YciA, TesB and Paal

[0934] Screen for mandelyl-CoA Thioesterase activity. CoA formation was detected with the Ellman's reagent (5,5′-dithiobis-(2-nitrobenzoic acid)), which reacts with free thiols under release of 2-nitro-5-thiobenzoate (ε.sub.412nm=14, 150 M.sup.−1 cm.sup.−1). Assays were carried out at 30° C. in 50 mM MES-KOH pH 6.8 and contained 1 mM Ellman's reagent and 0.5 μM YciA or 0.5 μM TesB or 0.5 μM Paal. The reaction was started by adding 0.1 mM mandelyl-CoA.

[0935] Results: The result is depicted in FIG. 15. YciA exhibits a higher thioesterase activity than TesB and Paal.

Example 12—OXS-OXC.SUB.Me.-YciA Cascade Prototyping

[0936] In a 1.5 mL microfuge tube, reaction buffer containing 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 2 μM YciA, 5 μM OXC.sub.Me and 5 μM OXS were mixed and the reaction was initiated by the addition of 10 mM disodium oxalate. For the negative controls each enzyme was omitted in a separate reaction. Samples were taken after 0, 3, 15, 40 min, 1, 2, 3 and 22 h and analyzed with the mandelic acid derivatives method described above. For quantification commercial, racemic mandelic acid was diluted in reaction buffer to appropriate concentrations to obtain a calibration curve

Example 13—Comparison of OXC.SUB.Me .with its Mutant

[0937] Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 5 μM OXS, 5 μM OXC and 2 μM YciA. The reaction was started by adding 10 mM oxalate. As negative controls, YciA, OXS or OXC were omitted in separate reactions. Samples were taken after 0, 0.05, 0.25, 0.66, 1, 2, 3, 22 hours and analyzed with the LC-MS detection of mandelic acid derivatives.

[0938] Mandelic acid was quantified by comparison to commercially obtained standards of mandelic acid.

[0939] Results: The results are depicted in FIG. 19.

[0940] When OXC.sub.Me is replaced by the OXC.sub.Me-Y497A variant having the sequence SEQ ID NO. 2, mandelic acid production rate increased 5-fold and yield increased 4-fold.

Example 14—OXS-OXC.SUB.Me.-YciA Cascade Aldehyde Scope

[0941] The aldehyde substrate screen of the OXS-OXC.sub.Me-Y497A-YciA cascade was carried out as described above (example 13), except that the aromatic aldehydes were prepared as 33 mM stocks in 20 vol % DMSO and diluted to final concentration of 25 mM into the assay. Samples were analyzed with the mandelic acid derivatives method described above. Mandelic acid (3a), 4-chloromandelic acid (3c), 2-chloromandelic acid (3d) and 3-phenyllactic acid (3b) were quantified by comparison to commercial standards.

[0942] The results are depicted in the following table 5.

TABLE-US-00006 TABLE 5 Scope of the OXS-OXC.sub.Me-Y497A-YciA cascade for the synthesis of chiral α-hydroxy acids..sup.[a] [00399] [00400] Conversion ee Entry R Product name [%].sup.[b] [%].sup.[b] 1 Ph Mandelic acid 75 99 2 CH.sub.2Ph 3-phenyllactic acid 58 n.d..sup.[c] 3 4-OH—Ph 4-hydroxymandelic acid n.d. 95 4 4-NO.sub.2—Ph 4-nitromandelic acid n.d. 80 5 4-Cl—Ph 4-chloromandelic acid 57 95 6 3-Cl—Ph 3-chloromandelic acid n.d. 98 7 2-Cl—Ph 2-chloromandelic acid 93 44 8 4-CHO—Ph 4-aldehydemandelic acid n.d. 46 9 4-COOH—Ph 4-formylmandelic acid n.d. n.d. 10 4-COCH.sub.3Ph 4-acetylmandelic acid n.d. 51 11 4-N(CH.sub.3).sub.2Ph 4-(dimethylamino)mandelic n.d. n.d. acid 12 4-C≡CH—Ph 4-ethynylmandelic acid n.d. 75 13 3-OCH.sub.3-4- 4-hydroxy-3- n.d., product n.d. OH—Ph methoxymandelic acid detected 14 2-furyl Hydroxy(2-furyl)acetic acid n.d. 75 15 2-thiopheneyl Hydroxy(2- n.d. 65 thiopheneyl)acetic acid 16 3-pyridinyl Hydroxy(3-pyridinyl)acetic n.d., product n.d. acid detected 17 2-naphthyl Hydroxy(2-naphthyl)acetic n.d. 80 acid .sup.[a]The reaction contained aldehyde (25 mM), disodium oxalate (10 mM), ATP (10 mM), OXS(5 μM), OXC.sub.Me-y497A (5 μM), YciA (2 μM). .sup.[b]Determined by chiral HPLC (see Figure S5-7 for detailed quantification). .sup.[c]Not determined.

Example 14b—OXS-OXC.SUB.Me.-YciA Cascade

[0943] Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 10 mM oxalate, 5 μM OXS, 5 μM OXC and 2 μM YciA. The reaction was started by adding 25 mM aromatic aldehyde (prepared as 33 mM stock solution containing 20 vol % DMSO). The tested aromatic aldehydes are listed in Table 6. The reactions were quenched by the addition of 4% formic acid after 24 h and analyzed with the LC-MS detection of mandelic acid derivatives.

[0944] Results: The results are depicted in Table 6.

TABLE-US-00007 TABLE 6 Extended aldehyde scope of the OXS-OXC.sub.Me-Y497A-YciA cascade for the synthesis of chiral α-hydroxy acids. [00401] [00402] [00403] Entry R Product 18 [00404] [00405] 19 [00406] [00407] 20 [00408] [00409] 21 [00410] [00411] 22 [00412] [00413] 23 [00414] [00415] 24 [00416] [00417] 25 [00418] [00419] 26 [00420] [00421] 27 [00422] [00423] 28 [00424] [00425] 29 [00426] [00427] 30 [00428] [00429] 31 [00430] [00431] 32 [00432] [00433] 33 [00434] [00435] 34 [00436] [00437] 35 [00438] [00439] 36 [00440] [00441] 37 [00442] [00443] 38 [00444] [00445] 39 [00446] [00447] 40 [00448] [00449] 41 [00450] [00451] 42 [00452] [00453] 43 [00454] [00455] 44 [00456] [00457] 45 [00458] [00459] 46 [00460] [00461] 47 [00462] [00463] 48 [00464] [00465]

Example 14c: OXS-OXC.SUB.Me.-YciA Cascade on Semi-Preparative Scale

[0945] In a glass vial (25 mL reaction volume) 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP (0.0125 mmol, 5.3 mg), 0.15 mM ThDP (0.00375 mmol, 1.7 mg), 0.5 mM CoA (0.0125 mmol, 9.8 mg), 25 mM disodium oxalate (0.625 mmol, 84 mg), 25 mM aldehyde (0.625 mmol), 2 μM YciA, 10 μM OXC.sub.Me-Y497A, 5 μM OXS, 1.3 μM adenylate kinase and 25 units creatine kinase (Roth) were mixed and the reaction was initiated by the addition of 10 mM creatine phosphate. The same amount of creatine phosphate was added after 1, 2, 4 and 6 h to reach a final concentration of 50 mM (1.25 mmol, 409 mg). The vials were incubated without shaking at 30° C. for 24 h. Then the pH was lowered to 3 by adding HCl. The quenched reaction was saturated with NaCl and extracted with 25 mL diethylether. The organic phase was dried over MgSO.sub.4 and filtered. After evaporation of the ether under vacuum, 50 mg (0.331 mmol, 53%) of a white solid remained, which was confirmed to be mandelic acid by UV, HPLC and NMR. .sup.1H NMR (300 MHz, DMSO-d.sub.6) δ 7.4 (m, 5H); 5.05 (s, 1H). .sup.13C NMR (300 MHz, DMSO-d.sub.6) δ 174.5, 140.7, 128.6, 128.1, 127.1, 72.9.

[0946] The aldehydes 1-46 were used to prepare α-hydroxy carboxylic acid on the basis of the aforementioned procedure.

Synthesis of α-Hydroxyacyl-CoA Thioester and α-Hydroxy Carboxylic Acids

Example 15a—Preparation of α-Hydroxyacyl-CoA Thioester in a One Step Procedure According to the Invention—Starting from Formyl-CoA and Oxalyl-CoA (One Pot)

[0947] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 10 mM aldehyde (formaldehyde, acetaldehyde, glycolaldehyde, propionaldehyde, glyceraldehyde, glyoxylate, succinic semialdehyde, benzaldehyde, and phenylacetaldehyde) or 10 mM acetone and 5 μM OXC.sub.Me or HACL.sub.Hs, respectively. The reaction was started by adding 1 mM formyl-CoA or oxalyl-CoA. The reaction was stopped after 1 h by quenching with 4% formic acid and products analyzed with the CoA ester method described above.

Example 15b—Preparation of α-Hydroxyacyl-CoA Thioesters a Two Step Procedure Starting from Oxalate by Enzymatic Catalysis (One Pot)

[0948] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.85 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 5 μM OXC.sub.Me-Y497A and 5 μM OXS were mixed and the reaction was initiated by the addition of 10 mM disodium oxalate. Samples were taken after 0, 10, 60, 240 min and quenched with 4% formic acid and products analyzed with the CoA ester method described above.

[0949] The aldehydes 1-48 according to example 14 were used to prepare α-hydroxyacyl-CoA thioesters on the basis of the aforementioned procedure.

[0950] Results: The results for the formation of mandelyl-CoA are depicted in FIG. 29. CoA was converted to mandelyl-CoA. With OXC.sub.Me-Y497A the reaction was complete after 60 min, with OXC.sub.Me after 120 min.

Example 16—Preparation of α-Hydroxyacyl-CoA Thioesters in a Two Step Procedure Starting from Formate by Enzymatic Catalysis (One Pot)

[0951] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 5 μM OXC.sub.Me-Y497A and 5 μM FCS were mixed and the reaction was initiated by the addition of 50 mM sodium formate. Samples were taken after 0, 10, 60, 240 min and quenched with 4% formic acid and products analyzed with the CoA ester method described above.

[0952] The aldehydes 1-46 were used to prepare α-hydroxyacyl-CoA thioesters on the basis of the aforementioned procedure.

Example 17—Preparation of α-Hydroxy Carboxylic Acid Starting from Oxalyl-CoA by Enzymatic Catalysis in a Two Step Procedure According to the Invention (One Pot)

[0953] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 10 mM aldehyde, 5 μM OXC.sub.Me or OXC.sub.Me-Y497A, 2 μM YciA. The reaction was started by adding 1 mM formyl-CoA or oxalyl-CoA. The reaction was stopped after 1 h by quenching with 4% formic acid and products analyzed with the mandelic acid derivatives method described above.

Example 18—Preparation of α-Hydroxy Carboxylic Acid Starting from Oxalate by Enzymatic Catalysis in a Three Step One Pot Procedure According to the Invention (One Pot)

[0954] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 15 vol % DMSO, 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 25 mM aldehyde, 5 μM OXS, 5 μM OXC.sub.Me or OXC.sub.Me-Y497A and 2 μM YciA. The reaction was started by adding 10 mM oxalate. Samples were taken after 0, 0.05, 0.25, 0.66, 1, 2, 3, 22 hours, quenched with 4% formic acid and analyzed with the LC-MS detection of mandelic acid derivatives.

[0955] The aldehydes 1-46 were used to prepare α-hydroxy carboxylic acid on the basis of the aforementioned procedure.

Example 19—Michaelis-Menten Kinetics of the Formation of Formyl-CoA Catalyzed by Formyl-CoA Synthetase (Acetyl-CoA Synthetase)

[0956] The Michaelis-Menten kinetic of eryACS-1 (an acyl-CoA synthetase from Erythrobacter sp.) and variants thereof as formyl-CoA synthetase (FCS) were investigated with formate as substrate according to example 6.

[0957] AMP production was followed at 30° C. by coupling eryACS-1 to purified adenylate kinase (ADK), pyruvate kinase (PK) and lactate dehydrogenase (LDH). An assay containing 200 mM MOPS-KOH pH 7.8, 0.3 mM NADH, 10 mM MgCl.sub.2, 2 mM ATP, 0.5 mM CoA, 2.5 mM PEP, 0.2 mg/mL ADK, 15 U/mL PK from rabbit muscle (Sigma), 23 U/mL LDH from rabbit muscle (Sigma), and an appropriate concentration of eryACS-1 was preincubated for 2 min and the reaction started by adding sodium formate to a final concentration of 10, 25, 50, 100, 150, 200 mM, respectively. Reaction procedure was monitored by following the oxidation of NADH at 340 nm.

[0958] Results: The results are depicted in FIG. 30.

[0959] Acyl-CoA synthetase from Erythrobacter sp. NAP1 (eryACS-1) exhibits a favorable activity towards formate (k.sub.cat=0.67±0.03 s.sup.−1; K.sub.M(formate)=73±8 mM). The enzyme was engineered by replacing valine 379 by isoleucine, which increased catalytic efficiency, mainly through an improved turnover number. Activity of the enzyme was further increased by expressing the gene in an E. coli strain lacking lysine acetyltransferase patZ, which inhibits acetyl-CoA synthetase, to finally obtain a highly active variant (k.sub.cat=16±1 s.sup.−1; K.sub.M(formate)=100±10 mM).

Example 20—Preparation of Formyl-CoA by FCS

[0960] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM CoA, 10 mM ATP and 0.1 μM FCS were mixed and the reaction was initiated by the addition of 50 mM sodium formate. Samples were taken after 0, 1, 5, 20 min and quenched with 4% formic acid and products analyzed with the CoA ester method described above.

Example 21—Preparation of α-Hydroxyacyl-CoA Thioesters a Two Step Procedure Starting from Formate by Enzymatic Catalysis (One Pot)

[0961] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.85 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 5 μM OXC.sub.Me and 5 μM FCS (eryACS-1—an acyl-CoA synthetase from Erythrobacter sp., FIG. 24) were mixed and the reaction was initiated by the addition of 50 mM sodium formate. Samples were taken after 0, 10, 60, 240 min and quenched with 4% formic acid and products analyzed with the CoA ester method described above.

[0962] Results: The results are depicted in FIG. 31. CoA was converted to mandelyl-CoA. The reaction proceeded at the same rate for OXC.sub.Me and OXC.sub.Me-Y497A. 40% conversion was achieved after 240 min.

Example 22—FCS-OXC.SUB.Me.-YciA Cascade Prototyping

[0963] The investigation of (S)-mandelic acid formation of the cascade FCS-OXC.sub.Me-YciA using formate as starting material were performed according to example 12.

[0964] Assays were carried out at 30° C. in reaction buffer consisting of 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 5 mM ATP, 25 mM benzaldehyde, 1 μM FCS, 2 μM OXC and 2 μM YciA. The reaction was started by adding 50 mM sodium formate. Samples were taken after 0, 2, 15, 30, 60, 120, 300 min and analyzed with the LC-MS detection of mandelic acid derivatives. Mandelic acid was quantified by comparison to commercially obtained standards of mandelic acid.

[0965] Results: The results are depicted in FIG. 32.

[0966] The formyl-CoA synthetase can be used in a cascade reaction in a one-pot system along with a thioesterase YciA and oxalyl-CoA synthetase for obtaining mandelic acid.

[0967] The aldehydes 1-46 were used to prepare α-hydroxy carboxylic acid on the basis of the aforementioned procedure.

Example 23—OXC.SUB.Me-Y497A., FRC and Formyl-CoA

[0968] Formyl-CoA:oxalate CoA-transferase (FRC) from Oxalobacter formigenes (EC 2.8.3.16) catalyzes the reaction formyl-CoA+oxalate=oxalyl-CoA+formate. This reaction is fully reversible. Converting formyl-CoA into oxalyl-CoA in the context of the cascade should have 2 beneficial effects: First, YciA has (low) thioesterase activity towards formyl-CoA, but none towards oxalyl-CoA (FIG. 16). This means that no (or less) ATP equivalents are lost to formyl-CoA hydrolysis, resulting in higher yield. Secondly, OXC.sub.Me-Y497A is faster in the carboligation with oxalyl-CoA than with formyl-CoA (FIG. 14). Therefore, turning formyl-CoA into oxalyl-CoA should increase the rate of OXC, and thus the rate of the whole cascade. In order to test these hypothesis FRC was added to a reaction containing OXC, benzaldehyde, oxalate and oxalyl-CoA.

[0969] Results: The results are depicted in FIG. 33. OXC.sub.Me-Y497A with FRC gave mandelyl-CoA formation rate of 33 μM/min, a 1.65-fold improvement over no FRC. Also note how the formyl-CoA is not accumulating in the reaction with OXC.sub.Me-Y497A, whereas it is still formed rapidly in the case of OXC.sub.Me. This can be explained by the k.sub.cat for decarboxylation (OXC.sub.Me, 98 s.sup.−1 vs. OXC.sub.Me-Y497A, 1.3 s.sup.−1).

Example 24—Preparation of Oxalyl-CoA by FRC

[0970] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 10 mM disodium oxalate and 0.1 μM FRC were mixed and the reaction was initiated by the addition of 1 mM formyl-CoA. Samples were taken after 0, 1, 5, 20 min and quenched with 4% formic acid and products analyzed with the CoA ester method described above.

Example 25—Preparation of Formyl-CoA by FRC

[0971] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 10 mM sodium formate and 0.1 μM FRC were mixed and the reaction was initiated by the addition of 1 mM oxalyl-CoA. Samples were taken after 0, 1, 5, 20 min and quenched with 4% formic acid and products analyzed with the CoA ester method described above.

Example 26—Preparation of α-Hydroxy Carboxylic Acid Starting from an Oxalate Using OXS-OXC.SUB.Me.-YciA with FRC (One Pot)

[0972] In a 1.5 mL microfuge tube, reaction buffer containing 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 0.5 μM FRC, 2 μM YciA, 5 μM OXC.sub.Me and 5 μM OXS were mixed and the reaction was initiated by the addition of 10 mM disodium oxalate. For the negative control FRC was omitted in a separate reaction. Samples were taken after 0, 3, 15, 40 min, 1, 2, 3 and 22 h and analyzed with the mandelic acid derivatives method described above. For quantification commercial, racemic mandelic acid was diluted in reaction buffer to appropriate concentrations to obtain a calibration curve.

[0973] The aldehydes 1-46 were used to prepare α-hydroxy carboxylic acid on the basis of the aforementioned procedure.

Example 27—Preparation of α-Hydroxy Carboxylic Acid Starting from an Oxalate Using OXS-OXC.SUB.Me.-YciA with FRC (One Pot)

[0974] In a 1.5 mL microfuge tube, reaction buffer containing 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 10 mM ATP, 7.6 mM benzaldehyde or phenylacetaldehyde, 2 μM FRC (Formyl-CoA:oxalate CoA-transferase from Oxalobacter formigenes EC 2.8.3.16), 2 μM YciA, 10 μM OXC.sub.Me-Y497A and 5 μM OXS were mixed and the reaction was initiated by the addition of 10 mM disodium oxalate. For the negative control FRC was omitted in a separate reaction. Samples were taken after 0, 15, 30, 60 min and analyzed with the mandelic acid derivatives method described above.

[0975] Results: The results are depicted in FIG. 34. Addition of FRC increased product formation rate 1.4-fold and 2.3-fold for mandelic acid and 3-phenyllactic acid, respectively.

Example 28—Preparation of α-Hydroxy Carboxylic Acid Starting from an Oxalate Using OXS-OXC.SUB.Me.-HACOT without a Hydrolase (One Pot)

[0976] In a 1.5 mL microfuge tube, reaction buffer containing 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2. 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 0.25 mM ATP, 25 mM benzaldehyde, 10 μM HACOT, 5 μM OXC.sub.Me and 0.1 μM OXS were mixed and the reaction was initiated by the addition of 10 mM disodium oxalate. For the negative controls FRC was omitted in a separate reaction. Samples were taken after 0, 3, 15, 40 min, 1, 2, 3 and 22 h and analyzed with the mandelic acid derivatives method described above. For quantification commercial, racemic mandelic acid was diluted in reaction buffer to appropriate concentrations to obtain a calibration curve.

[0977] The aldehydes 1-46 were used to prepare α-hydroxy carboxylic acid on the basis of the aforementioned procedure.

Example 29—OXS-OXC.SUB.Me .with Formyl-CoA Proofreading by FRC (One Pot)

[0978] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.85 mM CoA, 10 mM ATP, 25 mM benzaldehyde, 1 μM OXC.sub.Me, 5 μM OXS and 20 μM FRC and were mixed and the reaction was initiated by the addition of 10 mM disodium oxalate. Samples were taken after 0, 2, 5, 10, 30 min and quenched with 4% formic acid and products analyzed with the CoA ester method described above.

[0979] Results: The results are depicted in FIG. 35. CoA was converted to mandelyl-CoA. Addition of FRC increased the reaction rate, indicating that formyl-CoA proofreading has a beneficial effect. The effect is greater with the wild-type OXC.sub.Me, which is expected from the higher formyl-CoA release rate compared to OXC.sub.MeY497A.

Example 30—Preparation of α-Hydroxy Carboxylic Acid Starting from Oxalyl-CoA Using OXC.SUB.Me.-YciA-FRC

[0980] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 25 mM benzaldehyde, 10 mM sodium oxalate, 10 μM OXC.sub.Me-Y497A, 2 μM YciA and 2 μM FRC and were mixed and the reaction was initiated by the addition of 5 mM oxalyl-CoA. Samples were taken after 0, 15, 30, 60 min and quenched with 4% formic acid and products analyzed with the mandelic acid derivatives method described above.

Example 31—Preparation of α-Hydroxy Carboxylic Acid Starting from Oxalyl-CoA Using OXC.SUB.Me.-HACOT without a Hydrolase

[0981] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 25 mM benzaldehyde, 10 mM sodium oxalate, 10 μM OXC.sub.Me-Y497A, 10 μM HACOT and were mixed and the reaction was initiated by the addition of 5 mM oxalyl-CoA. Samples were taken after 0, 15, 30, 60 min and quenched with 4% formic acid and products analyzed with the mandelic acid derivatives method described above.

Example 32—Preparation of α-Hydroxy Carboxylic Acid Starting from Formyl-CoA Using OXC.SUB.Me.-YciA

[0982] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 25 mM benzaldehyde, 10 μM OXC.sub.Me-Y497A, 2 μM YciA and were mixed and the reaction was initiated by the addition of 5 mM formyl-CoA. Samples were taken after 0, 15, 30, 60 min and quenched with 4% formic acid and products analyzed with the mandelic acid derivatives method described above.

Example 33—Preparation of α-Hydroxy Carboxylic Acid Starting from Formate Using FCS-OXC.SUB.Me.-HAFT (α-Hydroxyacyl-CoA:Formate Transferase) without a Hydrolase

[0983] Assays were carried out at 30° C. in a 1.5 mL microfuge tube. Reactions contained 50 mM TES-KOH pH 6.8, 10 mM MgCl.sub.2, 0.5 mM ADP, 0.15 mM ThDP, 0.5 mM CoA, 0.25 mM ATP, 25 mM benzaldehyde, 0.1 μM FCS, 10 μM OXC.sub.Me-Y497A, 10 μM HAFT and were mixed and the reaction was initiated by the addition of 50 mM sodium formate. Samples were taken after 0, 30, 60, 120 min and quenched with 4% formic acid and products analyzed with the mandelic acid derivatives method described above.