Methods for producing 3-hydroxy-3-methylbutyric acid

Abstract

Described is a method for the conversion of 3-methylcrotonyl-CoA into 3-hydroxy-3-methylbutyric acid comprising the steps of: (a) enzymatically converting 3-methylcrotonyl-CoA into 3-hydroxy-3-methylbutyryl-CoA; and (b) further enzymatically converting the thus produced 3-hydroxy-3-methylbutyryl-CoA into 3-hydroxy-3-methylbutyric acid wherein the enzymatic conversion of 3-hydroxy-3-methylbutyryl-CoA into 3-hydroxy-3-methylbutyric acid according to step (b) is achieved by first converting 3-hydroxy-3-methylbutyryl-CoA into 3-hydroxy-3-methylbutyryl phosphate and then subsequently converting the thus produced 3-hydroxy-3-methylbutyryl phosphate into 3-hydroxy-3-methylbutyric acid.

Claims

1. A method for the conversion of 3-methylcrotonyl-CoA into 3-hydroxy-3-methylbutyric acid comprising the steps of: (a) enzymatically converting 3-methylcrotonyl-CoA into 3-hydroxy-3-methylbutyryl-CoA by a first enzyme having at least 95% sequence identity to one of SEQ ID NOs:1-3 or 6 wherein said first enzyme is capable of converting 3-methylcrotonyl-CoA into 3-hydroxy-3-methylbutyryl-CoA; and (b) enzymatically converting the thus produced 3-hydroxy-3-methylbutyryl-CoA into 3-hydroxy-3-methylbutyryl phosphate by a second enzyme having at least 95% sequence identity to one of SEQ ID NOs:28 or 29 wherein said second enzyme is capable of converting 3-hydroxy-3-methylbutyryl-CoA into 3-hydroxy-3-methylbutyryl phosphate; and (c) enzymatically converting the thus produced 3-hydroxy-3-methylbutyryl phosphate into 3-hydroxy-3-methylbutyric acid by a third enzyme having at least 95% sequence identity to one of SEQ ID NOS:30 or 31 wherein said third enzyme is capable of converting 3-hydroxy-3-methylbutyryl phosphate into 3-hydroxy-3-methylbutyric acid.

2. The method of claim 1, wherein the first enzyme, the second enzyme and the third enzyme are in a composition.

3. The method of claim 2, wherein 3-hydroxy-3-methylbutyric acid is recovered.

4. The method of claim 2, wherein the 3-hydroxy-3-methylbutyric acid is further converted into isobutene.

5. The method of claim 4, wherein the isobutene is recovered.

6. The method of claim 1, wherein the first enzyme, the second enzyme and the third enzyme are expressed in a microorganism.

7. The method of claim 6, wherein the microorganism produces 3-hydroxy-3-methylbutyric acid.

8. The method of claim 7, wherein 3-hydroxy-3-methylbutyric acid is recovered.

9. The method of claim 7, wherein the 3-hydroxy-3-methylbutyric acid is further converted into isobutene.

10. The method of claim 9, wherein the isobutene is recovered.

11. The method of claim 6, wherein 3-hydroxy-3-methylbutyric acid is recovered.

12. The method of claim 6, wherein the 3-hydroxy-3-methylbutyric acid is further converted into isobutene.

13. The method of claim 12, wherein the isobutene is recovered.

14. The method of claim 1, wherein 3-hydroxy-3-methylbutyric acid is recovered.

15. The method of claim 1, wherein the 3-hydroxy-3-methylbutyric acid is further converted into isobutene.

16. The method of claim 15, wherein the isobutene is recovered.

Description

(1) FIG. 1 shows schematically the reaction of the conversion of 3-methylcrotonyl-CoA into 3-methylcrotonic acid catalyzed by a thioesterase.

(2) FIG. 2 shows schematically the reaction of the conversion of 3-methylcrotonyl-CoA into 3-methylcrotonic acid catalyzed by a CoA-transferase.

(3) FIG. 3 shows schematically the reaction of the conversion of 3-methylcrotonic acid into 3-hydroxy-3-methylbutyric acid.

(4) FIG. 4 shows the conversion of 3-methylglutaconyl-CoA into 3-methylcrotonyl-CoA via a decarboxylation reaction.

(5) FIG. 5 shows conversion of 3-hydroxy-3-methylglutaryl-CoA into 3-methylglutaconyl-CoA.

(6) FIG. 6 shows the condensation of acetyl-CoA and acetoacetyl-CoA into 3-hydroxy-3-methylglutaryl-CoA.

(7) FIG. 7 shows possible pathways for producing acetoacetyl-CoA from acetyl-CoA.

(8) FIG. 8 shows the conversion of 3-methylcrotonyl-CoA into 3-hydroxy-3-methylbutyryl-CoA.

(9) FIG. 9 shows the conversion of 3-hydroxy-3-methylbutyryl-CoA into 3-hydroxy-3-methylbutyric acid catalyzed by a thioesterase.

(10) FIG. 10 shows the conversion of 3-hydroxy-3-methylbutyryl-CoA into 3-hydroxy-3-methylbutyric acid catalyzed by a CoA-transferase.

(11) FIG. 11 shows the conversion of 3-methylcrotonyl-CoA into 3-methylcrotonyl-phosphate and the subsequent conversion of 3-methylcrotonyl-phosphate into 3-methylcrotonic acid.

(12) FIG. 12 shows the activity of several studied enzymes for 3-methylcrotonyl-CoA hydration monitored by recording the decrease of absorbance of 3-methylcrotonyl-CoA at 263 nm (Example 2).

(13) FIG. 13 shows the time course of the formation of 3-hydroxy-3-methylbutyryl-CoA acid (3-hydroxyisovaleryl-CoA) by hydration of 3-methylcrotonyl-CoA catalyzed by the 3-hydroxypropionyl-CoA dehydratase from Metallosphaera sedula and enoyl-CoA hydratase from Bacillus anthracis (Example 3).

(14) FIG. 14 shows a chromatogram obtained for the enzymatic hydration of 3-methylcrotonyl-CoA with 3-hydroxypropionyl-CoA dehydratase from Metallosphaera sedula as outlined in Example 4. The assay was performed with 4 mM 3-methylcrotonyl-CoA and incubated for 16 min.

(15) FIG. 15 shows the formation of 3-hydroxyisovaleryl-CoA from 3-methylcrotonyl-CoA as function of time at different substrate concentrations. The conversion was catalyzed by the 3-hydroxypropionyl-CoA dehydratase from Metallosphaera sedula (Example 4).

(16) FIG. 16 shows in part A the percentage of the conversion of 3-hydroxy-3-methylglutaryl-CoA into 3-methylglutaconyl-CoA by different enzymes shown in part B (Example 5).

(17) FIG. 17 shows a schematic of exemplary biochemical pathways leading to isobutene from acetyl-CoA as outlined in Example 6.

(18) FIG. 18 shows a mass spectrum of 3-hydroxyisovaleryl phosphate (3-hydroxy-3-methylbutyryl phosphate) produced from 3-hydroxyisovaleryl-CoA (3-hydroxy-3-methylbutyryl-CoA) and phosphate catalyzed by phosphate butyryltransferase from Bacilllus subtilis (Example 9).

(19) FIG. 19 shows an overlay of typical HPLC-chromatograms obtained for

(20) a) the enzymatic Assay A in Example 10; and

(21) b) the enzyme-free Assay H in Example 10.

(22) In the HPLC-chromatograms 3-hydroxyisovaleryl-CoA (3-hydroxy-3-methylbutyryl-CoA) and CoSH were analyzed.

(23) The consumption of 3-hydroxyisovaleryl-CoA (3-hydroxy-3-methylbutyryl-CoA) with simultaneous production of CoA-SH was observed in the enzymatic assay combining phosphate butyryltransferase with butyrate kinase.

(24) FIG. 20 shows an overlay of typical HPLC-chromatograms obtained for

(25) a) enzymatic assay (Assay A, Example 10)

(26) b) enzyme-free assay (Assay H, Example 10).

(27) In the HPLC-chromatograms 3-hydroxyisovaleric acid, ADP and ATP were analyzed.

(28) The consumption of ADP with simultaneous production of ATP was observed in the enzymatic assay combining phosphate butyryltransferase with butyrate kinase.

(29) FIG. 21 shows the results of the production of 3-hydroxyisovaleric acid (3-hydroxy-3-methylbutyric acid) and ATP in the enzymatic assays wherein phosphate butyryltransferase from Bacilllus subtilis is combined with different butyrate kinases. Further, the results of different control assays as indicated in Example 10 are shown.

(30) FIG. 22 shows the results of the production of 3-hydroxyisovaleric acid (3-hydroxy-3-methylbutyric acid) and ATP in the enzymatic assays wherein phosphate butyryltransferase from Enterococcus faecalis is combined with different butyrate kinases. Further, the results of different control assays as indicated in Example 11 are shown.

(31) FIG. 23 shows an example of a typical HPLC-chromatogram obtained for the enzymatic assay with acyl-CoA thioesterase II from Pseudomonas putida.

(32) FIG. 24a shows an overlay of typical HPLC-chromatograms (analysis of 3-methylcrotonyl-CoA, 3-methylcrotonic acid and CoA-SH) obtained for

(33) a) enzymatic assay (assay A, Example 13)

(34) b) enzyme-free assay (assay H, Example 13).

(35) The consumption of 3-methylcrotonyl-CoA with simultaneous production of CoA-SH and 3-methylcrotonic acid was observed in the enzymatic assay combining phosphate butyryltransferase with butyrate kinase.

(36) FIG. 24b shows an overlay of typical HPLC-chromatograms obtained for (analysis of ADP and ATP) obtained for

(37) a) enzymatic assay (assay A, Example 13)

(38) b) enzyme-free assay (assay H, Example 13).

(39) The consumption of ADP with simultaneous production of ATP was observed in the enzymatic assay combining phosphate butyryltransferase with butyrate kinase.

(40) FIG. 25 shows the results of the production of 3-methylcrotonic acid and ATP in enzymatic assays comprising phosphate butyryltransferase from Bacilllus subtilis combined with different butyrate kinases, as well as in different control assays.

(41) FIG. 26 shows the results of the production of 3-methylcrotonic acid and ATP in enzymatic assays comprising phosphate butyryltransferase from Enterococcus faecalis combined with different butyrate kinases, as well as in different control assays.

(42) In this specification, a number of documents including patent applications are cited. The disclosure of these documents, while not considered relevant for the patentability of this invention, is herewith incorporated by reference in its entirety. More specifically, all referenced documents are incorporated by reference to the same extent as if each individual document was specifically and individually indicated to be incorporated by reference.

(43) The invention will now be described by reference to the following examples which are merely illustrative and are not to be construed as a limitation of the scope of the present invention.

EXAMPLES

Example 1: Cloning, Expression and Purification of Enzymes

(44) Gene Synthesis, Cloning and Expression of Recombinant Proteins

(45) The sequences of the studied enzymes inferred from the genomes of prokaryotic and eukaryotic organisms were generated by oligonucleotide concatenation to fit the codon usage of E. coli (genes were commercially synthesized by GeneArt). A stretch of 6 histidine codons was inserted after the methionine initiation codon to provide an affinity tag for purification. The genes thus synthesized were cloned in a pET-25b(+) expression vector (vectors were constructed by GeneArt).

(46) Competent E. coli BL21(DE3) cells (Novagen) were transformed with these vectors according to standard heat shock procedure. The transformed cells were grown with shaking (160 rpm) using ZYM-5052 auto-induction medium (Studier F W, Prot. Exp. Pur. 41, (2005), 207-234) for 6 h at 30 C. and protein expression was continued at 28 C. or 18 C. overnight (approximately 16 h). The cells were collected by centrifugation at 4 C., 10,000 rpm for 20 min and the pellets were stored at 80 C.

(47) Protein Purification and Concentration

(48) The pellets from 200 ml of culture cells were thawed on ice and resuspended in 5 ml of 50 mM HEPES buffer pH 7.0 containing 500 mM NaCl, 10 mM MgCl.sub.2, 10 mM imidazole and 1 mM DTT. Twenty microliters of lysonase (Novagen) were added. Cells were incubated 10 minutes at room temperature and then returned to ice for 20 minutes. Cell lysis was completed by sonication for 230 seconds. The bacterial extracts were then clarified by centrifugation at 4 C., 4000 rpm for 40 min. The clarified bacterial lysates were loaded onto a PROTINO-2000 Ni-TED column (Macherey-Nagel) allowing adsorption of 6-His tagged proteins. Columns were washed and the enzymes of interest were eluted with 6 ml of 50 mM HEPES buffer pH 7.0 containing 300 mM NaCl and 250 mM imidazole. Eluates were then concentrated, desalted on Amicon Ultra-4 10 kDa filter unit (Millipore) and enzymes were resuspended in 50 mM HEPES pH 7.0, containing 100 mM NaCl. The purity of proteins thus purified varied from 70% to 90% as estimated by SDS-PAGE analysis. Protein concentrations were determined by direct UV 280 nm measurement on the NanoDrop 1000 spectrophotometer (Thermo Scientific) or by Bradford assay (BioRad).

Example 2: Screening of a Collection of Hydro-Lyases Using 3-Methylcrotonyl-CoA as Substrate for the Production of 3-Hydroxy-3-Methylbutyryl-CoA (3-Hydroxyisovaleryl-CoA)

(49) The genes coding for acyl-CoA dehydratases and enoyl-CoA hydratases were synthesized and the corresponding enzymes were further produced according to the procedure described in Example 1. The stock solution of 3-methylcrotonyl-CoA (Sigma-Aldrich) was prepared in water.

(50) Standard reaction mixture contained:

(51) 50 mM HEPES pH 7.0

(52) 0.25 mM 3-methylcrotonyl-CoA

(53) 5 mM MgCl.sub.2

(54) 5 mM NaCl

(55) 0.002 mg/ml of purified enzyme

(56) Assays were performed in 96-well plates at 30 C. in a total volume of 0.12 ml.

(57) Each reaction was started by the addition of 3-methylcrotonyl-CoA. The samples were then continuously monitored for the decrease of the absorbance of 3-methylcrotonyl-CoA at 263 nm (Fukui T et al. J. Bacteriol. 180 (1998), 667-673) on a SpectraMax Plus 384 UV/Vis Microplate Reader (Molecular Devices).

(58) Several enzymes showed activity with 3-methylcrotonyl-CoA as a substrate (FIG. 12). No modification of the absorbance at 263 nm was observed for the control assay without enzyme.

Example 3: HPLC-Based Analysis of Products of the Enzymatic Hydration of 3-Methylcrotonyl-CoA

(59) The enzymatic assays were conducted in total reaction volume of 0.2 ml

(60) The standard reaction mixture contained:

(61) 50 mM HEPES pH 7.0

(62) 4 mM 3-methylcrotonyl-CoA

(63) 20 mM MgCl.sub.2

(64) 20 mM NaCl

(65) 0.02 mg/ml of purified 3-hydroxypropionyl-CoA dehydratase from Metallosphaera sedula (Uniprot Accession number: A4YI89) or purified enoyl-CoA hydratase Bacillus anthracis (Uniprot Accession number: Q81YG6)

(66) Assays were incubated for 0, 5 and 30 min with shaking at 30 C.

(67) After an incubation period, the reactions were stopped by the addition of 0.1 ml of acetonitrile. The amount of 3-hydroxy-3-methylbutyryl-CoA (3-hydroxyisovaleryl-CoA) was quantified using a HPLC-based procedure.

(68) The samples were centrifuged, filtered through a 0.22 m filter and the clarified supernatants were transferred into a clean vial for further analysis.

(69) HPLC analysis was performed using 1260 Infinity LC System (Agilent), equipped with column heating module and UV detector (260 nm). 5 l of samples were separated on Zorbax SB-Aq column (2504.6 mm, 5 m particle size, column temp. 30 C.), with a mobile phase flow rate of 1.5 ml/min. The separation was performed using mixed A (H.sub.2O containing 8.4 mM sulfuric acid) and B (acetonitrile) solutions in a linear gradient (0% B at initial time 0 min.fwdarw.70% B at 8 min). 3-hydroxyisovaleryl-CoA was chemically synthesized from 3-hydroxyisovaleric acid, upon request, by a company specialized in custom synthesis (Syntheval, France).

(70) The retention time of 3-hydroxyisovaleryl-CoA and 3-methylcrotonyl-CoA under these conditions was 4.40 and 5.25 min, respectively.

(71) Significant production of 3-hydroxyisovaleryl-CoA was observed in enzymatic assays (FIG. 13). No 3-hydroxyisovaleryl-CoA production was observed in enzyme-free control assay.

(72) Thus, acyl-CoA dehydratase and enoyl-CoA hydratase were able to efficiently catalyze the hydration of 3-methylcrotonyl-CoA into 3-hydroxyisovaleryl-CoA.

Example 4: Kinetic Studies of the Hydration Reaction of 3-Methylcrotonyl-CoA

(73) The enzymatic assays were conducted in total reaction volume of 0.2 ml

(74) The standard reaction mixture contained:

(75) 50 mM HEPES pH 7.0

(76) 0-4 mM 3-methylcrotonyl-CoA

(77) 20 mM MgCl.sub.2

(78) 20 mM NaCl

(79) 0.01 mg/ml 3-hydroxypropionyl-CoA dehydratase from Metallosphaera sedula.

(80) Assays were incubated for 0, 2, 4, 8, 16 min with shaking at 30 C. and the reactions were stopped by the addition of 100 l of acetonitrile. The amount of 3-hydroxyisovaleryl-CoA was quantified according to HPLC-based procedure described in Example 3.

(81) FIG. 14 shows an example of typical chromatogram obtained for enzymatic assay with substrate concentration of 4 mM after 16 min of incubation.

(82) The graph depicting 3-hydroxyisovaleryl-CoA formation as a function of time at different substrate concentrations is shown in the FIG. 15. 3-Hydroxypropionyl-CoA dehydratase from Metallosphaera sedula was found to have a K.sub.M of 0.3 mM and a k.sub.cat of 10 s.sup.1 for 3-methylcrotonyl-CoA.

Example 5: Enzyme-Catalyzed Dehydration of 3-Hydroxy-3-Methylglutaryl-CoA into 3-Methylglutaconyl-CoA

(83) The genes coding for acyl-CoA dehydratases and enoyl-CoA hydratases were synthesized and the corresponding enzymes were further produced according to the procedure described in Example 1. Stock solution of 3-hydroxy-3-methylglutaryl-CoA (Sigma-Aldrich) was prepared in water. The enzymatic assays were conducted in total reaction volume of 0.2 ml.

(84) The standard reaction mixture contained:

(85) 50 mM HEPES pH 7.0

(86) 4 mM 3-hydroxy-3-methylglutaryl-CoA

(87) 20 mM MgCl.sub.2

(88) 20 mM NaCl

(89) 0.01 mg/ml of purified enzyme

(90) After an incubation period of 16 min the assays were stopped by the addition of 0.1 ml of acetonitrile.

(91) The samples were centrifuged, filtered through a 0.22 m filter and the clarified supernatants were transferred into a clean vial for HPLC analysis. HPLC analysis was performed according to the procedure described in Example 3. The retention time of 3-hydroxy-3-methylglutaryl-CoA in these conditions was 4.20 min

(92) The progress of the reaction was followed by measuring consumption of substrate.

(93) Several acyl-CoA dehydratases and enoyl-CoA hydratases were shown to be able to catalyze the dehydration of 3-hydroxy-3-methylglutaryl-CoA into 3-methylglutaconyl-CoA (FIG. 16).

Example 6: Microorganism for the Production of 3-Hydroxyisovaleric Acid from Acetyl-CoA

(94) This working example shows the production of 3-hydroxyisovaleric acid by recombinant E. coli, expressing several exogenous genes.

(95) Like most organisms, E. coli converts glucose to acetyl-CoA. The enzymes used in this study to convert acetyl-CoA into 3-hydroxy-3-methylbutyric acid (aka 3-hydroxyisovaleric acid) following pathway 1 (FIG. 17) are summarized in Table 1.

(96) TABLE-US-00001 TABLE 1 Uniprot Gene Accession Step Enzyme abbreviation number A Acetyl-CoA acetyltransferase CA_C2873 P45359 from Clostridium acetobutylicum B Hydroxymethylglutaryl-CoA synthase Hcs1 P54874 from Schizosaccharomyces pombe C Hydroxybutyryl-CoA dehydratase MXAN_3757 Q1D5Y4 from Myxococcus xanthus (LiuC) D Glutaconate CoA-transferase MXAN_4264 Q1D4I4 (subunits AibA & AibB) from MXAN_4265 Q1D4I3 Myxococcus xanthus E Hydroxybutyryl-CoA dehydratase MXAN_3757 Q1D5Y4 from Myxococcus xanthus (LiuC)
Expression of 3-Hydroxyisovaleric Acid Biosynthetic Pathway in E. coli

(97) The modified version of pUC18 (New England Biolabs), containing a modified Multiple Cloning Site (pUC18 MCS) (WO 2013/007786), was used as an expression vector. A terminator sequence was inserted into pUC18 MCS between the HindIII and NarI restriction sites and the resulting vector was termed pGBE1992.

(98) The corresponding genes were codon optimized for expression in E. coli and synthesized by GeneArt (Life Technologies)

(99) The acetyl-coA acetyltransferase (CA_C2873) gene was PCR amplified from the pMK-RQ-thl_adc vector (master plasmid provided by GeneArt) using primers 3211 and 3212. A PacI restriction site at the 5 end of the PCR product was introduced. At the 3 end of the PCR product a NotI restriction site and a ClaI restriction site were introduced. The resulting 1.6 kbp PCR product and pGBE1992 were digested with the PacI and NotI restriction enzymes and then ligated together resulting in the pGBE2101 plasmid. The recombinant pGBE2101 plasmid was verified by sequencing.

(100) The 3-hydroxybutyryl-CoA dehydratase (MXAN_3757) gene was PCR amplified from the pMK-T-ACoADH_MX vector (master plasmid provided by GeneArt) using primers 3327 and 3328. An EcoRI restriction site at the 5 end and a KpnI restriction site at the 3 end were inserted by PCR. The amplified gene comprised a full-length MXAN_3757 coding sequence with a stretch of 6 histidine codons after methionine initiation codon to provide an affinity tag for purification. The resulting 0.8 kbp PCR product was digested with the restriction enzymes and ligated into the pGBE2101 plasmid previously digested with the EcoRI and KpnI restriction endonucleases. The resulting plasmid was termed pGBE2326 and verified by sequencing.

(101) The plasmid pMK-RQ-AibA_AibB (master plasmid provided by GeneArt) was digested with the restriction enzymes ClaI and NotI to create a 1.6 Kbp product. The 1.6 kbp restriction fragment, contained MXAN_4264 and MXAN_4265 genes, was ligated into cut pGBE2326 plasmid. The resulting recombinant plasmid pGBE2360 was verified by sequencing.

(102) The Hcs1 gene coding for hydroxymethylglutaryl-CoA synthase from S. pombe was PCR amplified from the pET-25b(+)-A_129 (master plasmid provided by GeneArt) with primers 3329 and 3330. A NotI restriction site at the 5 end and a HindIII restriction site at the 3 end were thereby introduced by PCR. The amplified gene comprised a full-length Hcs1 coding sequence with a stretch of 6 histidine codons after methionine initiation codon to provide an affinity tag for purification. The resulting 1.4 kbp PCR product was digested with the NotI and HindIII restriction enzymes and ligated with the digested pGBE2360 plasmid. The resulting plasmid pGBE2396 was verified by sequencing.

(103) TABLE-US-00002 TABLE2 Primer names Primerssequences 3211 CCCGCGGCCGCCCTATCGATTTATTAGC (SEQIDNO:22) 3212 TTAATTAATGAAAGAAGTGGTGATTGC (SEQIDNO:23) 3327 GGGGAATTCAGGAGGTGTACTAGATGCATCATCATCATCAC CACATGCC(SEQIDNO:24) 3328 CCCGGTACCTTATTAGCGACCTTTATAAACCGG (SEQIDNO:25) 3329 GGGGCGGCCGCAGGAGGTGTACTAGATGCACCATCATCATC ATCACAGC(SEQIDNO:26) 3330 CCCAAGCTTTTATTACGGTTTAACGCTATAGC (SEQIDNO:27)
Culture Medium and Flask Fermentation Conditions

(104) Strain MG1655 E. coli was made electrocompetent. MG1655 electrocompetent cells were then transformed with the expression vector pGBE2396. An empty plasmid pUC18 was transformed as well to create a strain used as a negative control in the assay.

(105) The transformed cells were then plated on LB plates, supplied with ampicillin (100 g/ml). Plates were incubated overnight at 30 C. Isolated colonies were used to inoculate LB medium, supplemented with ampicillin, followed by incubation at 30 C. overnight. 1 ml of this overnight culture was used to inoculate 300 ml of ZYM-5052 auto-inducing media (Studier F W, Prot. Exp. Pur. 41, (2005), 207-234). This culture was grown for 20 h at 30 C. and 160 rpm shaking.

(106) A volume of cultures corresponding to OD.sub.600 of 30 was removed and centrifuged. The pellet was resuspended in 30 ml of MS medium (Richaud C., Mengin-Leucreulx D., Pochet S., Johnson E J., Cohen G N. and Marlire P, The Journal of Biological Chemistry, 268, (1993), 26827-26835) containing glucose (45 g/L), and MgSO4 (1 mM).

(107) The cultures were then incubated in 160 ml bottles, sealed with a screw cap, at 30 C. with shaking for 3 days. The pH value of the cultures was adjusted to 8.5 twice per day using 30% NH.sub.4OH.

(108) At the end of incubation 1 ml of culture mediums was removed and centrifuged at 4 C., 10,000 rpm for 5 min. The supernatants were filtered through a 0.22 m filter and diluted with an equal volume of H.sub.2O. The production of 3-hydroxyisovaleric acid was then analyzed.

(109) Analysis of 3-Hydroxyisovaleric Acid Production

(110) The amount of 3-hydroxyisovaleric acid produced was measured using a HPLC-based procedure. HPLC analysis was performed using a 1260 Infinity LC System Agilent, equipped with column heating module, and refractometer. 10 l of samples were separated using 3 columns connected in series as follows: 1. Hi-Plex guard column (1007.7 mm, 8 m particle size) (Agilent) 2. Hi-Plex column (1007.7 mm, 8 m particle size) (Agilent) 3. Zorbax SB-Aq column (2504.6 mm, 5 m particle size, column temp. 65 C.) (Agilent).

(111) The mobile phase consisted of aqueous sulfuric acid (1 mM), mobile phase flow rate was 1.5 ml/min. Commercial 3-hydroxyisovaleric acid (TCl) was used as reference. Retention time of 3-hydroxyisovaleric acid under these conditions was 7.7 min. About 2.2 mM 3-hydroxyisovaleric acid was produced in these shake-flask experiments by engineered E. coli, contained the genes of 3-hydroxyisovaleric acid biosynthetic pathway. No 3-hydroxyisovaleric acid production was observed with the control strain, contained empty vector.

(112) Bioreactor Fermentation Conditions

(113) Strain E. coli MG1655 agp aphA was made electrocompetent. The electrocompetent cells were then transformed with the expression vector pGBE2396.

(114) The transformed cells were then plated on LB plates and the preculture was prepared according to the procedure described in the section Culture medium and flask fermentation conditions.

(115) The fermentation was performed in a 1 L bioreactor with pH and temperature control (Multifors 2, Infors HT). Cells of a preculture in LB medium were used to inoculate a 900 ml of MS liquid medium containing MgSO4 (1 mM), yeast extract (2 g/L) and ampicilline (100 g/ml) at an initial optical density (OD.sub.600) of 0.5 Concentration of glucose over the fermentation run was maintained between 0 g/L and 10 g/L using feed pumps. Temperature and pH were maintained constant (30 C. and 6.5, respectively). Dissolved oxygen was maintained to 20% (100% is obtained in air). Aliquots of culture medium were taken over the fermentation period and centrifuged at 4 C., 10 000 rpm for 5 min. The supernatant were then filtered through a 0.22 m filter and diluted with an equal volume of H.sub.2O. The amount of produced 3-hydroxyisovaleric acid was measured according to the HPLC-based procedure described above.

(116) After 143 hours of fermentation the 3-hydroxyisovaleric acid concentration reached 12 mM.

Example 7: Enzyme-Catalyzed Hydrolysis of 3-Hydroxyisovaleryl-CoA into 3-Hydroxyisovaleric Acid

(117) The gene coding for acyl-CoA thioesterase II from Pseudomonas putida was synthesized and the corresponding enzymes were further produced according to the procedure described in Example 1.

(118) Vectors pCAN contained genes coding for acyl-CoA thioester hydrolase YciA and acyl-CoA thioesterase 2 (TesB) from Escherichia coli were purchased from NAIST (Nara Institute of Science and Technology, Japan, ASKA collection). Provided vectors contained a stretch of 6 histidine codons after the methionine initiation codon. The corresponding enzymes were further produced according to the procedure described in Example 1.

(119) The enzymatic assays were conducted in total reaction volume of 0.2 ml

(120) The standard reaction mixture contained:

(121) 50 mM HEPES pH 7.0

(122) 10 mM 3-hydroxyisovaleryl-CoA

(123) 20 mM MgCl.sub.2

(124) 20 mM NaCl

(125) 1 mg/ml of purified thioesterase

(126) The assays were incubated for 30 min with shaking at 30 C. and the reactions were stopped by the addition of 0.1 ml of acetonitrile. The amount of 3-hydroxyisovaleryl-CoA was quantified according to HPLC-based procedure described in Example 3.

(127) In these conditions, the 3-hydroxyisovaleryl-CoA retention time was 4.40 min and coenzyme A (CoA) retention time was 3.96 min. A significant decrease of 3-hydroxyisovaleryl-CoA peak was observed in conjunction with increased coenzyme A peak.

(128) Additionally, 3-hydroxyisovaleric acid production was analyzed according to the procedure described en Example 6.

(129) All the studied thioesterases catalyzed the hydrolysis of 3-hydroxyisovaleryl-CoA with the formation of 3-hydroxyisovaleric acid (Table 3).

(130) No 3-hydroxyisovalerate signal was observed in control assay without enzyme.

(131) TABLE-US-00003 TABLE 3 Uniprot Accession Conversion, Gene names Organism number % yciA Escherichia coli P0A8Z0 25 tesB Escherichia coli P0AGG2 52 tesB Pseudomonas putida Q88DR1 58

Example 8: Cloning and Overexpression of Recombinant Phosphate Butyryltransferases and Butyrate Kinases

(132) Gene Synthesis, Cloning and Expression of Recombinant Proteins

(133) The sequences of phosphate butyryltransferase genes from Bacillus subtilis (strain 168) and Enterococcus faecalis MTUP9 (Uniprot Accession number: P54530 and A0A038BNC2, respectively) and butyrate kinase from Lactobacillus casei W56 and Geobacillus sp. GHH01 (Uniprot Accession number: K0N529 and L8A0E1, respectively) were generated by oligonucleotide concatenation to fit the codon usage of E. coli (genes were commercially synthesized by GeneArt). The expression of corresponding proteins was conducted following the method described in Example 1.

(134) The cells were collected by centrifugation at 4 C., 10.000 rpm for 20 min and the pellets were stored at 80 C.

(135) Protein Purification and Concentration

(136) The pellets from 200 ml of cultured cells were thawed on ice and resuspended in 5 ml of 50 mM potassium phosphate buffer pH 7.5 containing 100 mM NaCl, 10 mM MgCl.sub.2, 10 mM imidazole and 1 mM DTT. Twenty microliters of lysonase (Novagen) were added. Cells were incubated 10 minutes at room temperature and then returned to ice for 20 minutes. Cell lysis was completed by sonication for 230 seconds. The bacterial extracts were then clarified by centrifugation at 4 C., 4000 rpm for 40 min. The clarified bacterial lysates were loaded onto a PROTINO-2000 Ni-TED column (Macherey-Nagel) allowing adsorption of 6-His tagged proteins. Columns were washed and the enzymes of interest were eluted with 6 ml of 50 mM potassium phosphate buffer pH 7.5 containing 100 mM NaCl and 250 mM imidazole. Eluates were then concentrated, desalted on an Amicon Ultra-4 10 kDa filter unit (Millipore) and enzymes were resuspended in 50 mM potassium phosphate buffer pH 7.5, containing 100 mM NaCl. The purity of proteins thus purified varied from 70% to 90% as estimated by SDS-PAGE analysis. Protein concentrations were determined by direct UV 280 nm measurement on the NanoDrop 1000 spectrophotometer (Thermo Scientific) or by Bradford assay (BioRad).

Example 9: Formation of 3-Hydroxyisovaleryl Phosphate from 3-Hydroxyisovaleryl-CoA and Phosphate Catalyzed by Phosphate Butyryltransferase from Bacillus subtilis

(137) The studied enzymatic reaction was carried out under the following conditions:

(138) 50 mM Tris-HCl pH 7.5

(139) 10 mM MgCl.sub.2

(140) 10 mM NaCl

(141) 10 mM 3-hydroxyisovaleryl-CoA (3-hydroxy-3-methylbutyryl-CoA)

(142) The purified phosphate butyryltransferase from B. subtilis was resuspended in 5 mM potassium phosphate pH 7.5.

(143) A control assay without enzyme was performed in parallel.

(144) The enzymatic assay was initiated by the addition of 20 l of enzyme preparation into 20 l of reaction mixture. The formation of 3-hydroxyisovaleryl phosphate (3-hydroxy-3-methylbutyryl phosphate) was analyzed by Mass Spectrometry (MS). 5 l of the assay was introduced into the mass spectrometer via a loop injection.

(145) Flow injection analyses were performed using a Dionex Ultimate chromatographic system (Thermo Fisher Scientific) at a flow rate of 100 L/min with a mobile phase composed of H.sub.2O containing 10 mM ammonium formate pH 9.45 and acetonitrile 75:25 v/v. Detection was performed with a Q-exactive spectrometer (Thermo Fisher Scientific) fitted with an electrospray ionization source (negative ionization mode at a resolution of 70000 m/m, FWHM at m/z 200). Non-resonant induced dissociation experimentsHigher-energy C-trap dissociation (HCD)were acquired at normalized collision energy of 10%. Raw data were manually inspected using the Qualbrowser module of Xcalibur version 3.0 (Thermo Fisher Scientific).

(146) The formation of new ion with m/z at 197.0213, corresponding to C.sub.5H.sub.10O.sub.6P.sup., was observed in the enzymatic assay (FIG. 18).

(147) Structural elucidation and complete assignment of this newly formed ion (m/z at 197.0213) were further investigated using MS/MS analysis. The fragment ion with m/z value of 96.968, corresponding to the H.sub.2PO.sub.4.sup. species, was generated under MS/MS experiment. Thus, the generation of 3-hydroxyisovaleryl phosphate in the enzyme catalyzed assay was proved by MS technique.

Example 10: Conversion of 3-Hydroxyisovaleryl-CoA and ADP into 3-Hydroxyisovaleric Acid and ATP Catalysed by the Combined Action of Phosphate Butyryltransferase from Bacillus subtilis and Butyrate Kinase from Lactobacillus casei or Geobacillus sp.

(148) The enzymatic assays were conducted in a total reaction volume of 0.2 ml.

(149) The standard reaction mixture contained:

(150) 50 mM potassium phosphate buffer pH 7.5

(151) 4 mM 3-hydroxyisovaleryl-CoA

(152) 4 mM ADP

(153) 10 mM MgCl.sub.2

(154) 10 mM NaCl

(155) 0.2 mg/ml of purified phosphate butyryltransferase from Bacillus subtilis

(156) 0.2 mg/ml of purified butyrate kinase from Lactobacillus casei or Geobacillus sp.

(157) A series of controls were performed in parallel (Assays C-H; Table 4).

(158) TABLE-US-00004 TABLE 4 Assay composition A B C D E F G H 3-hydroxyisovaleryl-CoA + + + + + + + + ADP + + + + + + phosphate butyryltransferase + + + + + from Bacillus subtilis butyrate kinase from + + + Lactobacillus casei butyrate kinase from Geobacillus + + + sp

(159) The assays were then incubated for 20 min with shaking at 30 C.

(160) After an incubation period, the reactions were stopped by heating the reaction medium 4 min at 90 C. The samples were centrifuged, filtered through a 0.22 m filter and the clarified supernatants were transferred into a clean vial for further analysis. The consumption of ADP and 3-hydroxyisovaleryl-CoA, and the formation of ATP, 3-hydroxyisovaleric acid and free coenzyme A (CoA-SH) were followed by using HPLC-based methods.

(161) HPLC-Based Analysis of ADP, ATP and 3-Hydroxyisovaleric Acid

(162) HPLC analysis was performed using 1260 Infinity LC System (Agilent), equipped with column heating module and RI detector. 2 l of samples were separated on Polaris C18-A column (1504.6 mm, 5 m particle size, column temp. 30 C.) with a mobile phase flow rate of 1.5 ml/min. The separation was performed using 8.4 mM sulfuric acid in H.sub.2O/MeOH mixed solution (99/1) (V/V). In these conditions, the retention time of ADP, ATP and 3-hydroxyisovaleric acid were 2.09 min, 2.26 min and 5.01 min, respectively.

(163) HPLC Based Analysis of 3-Hydroxyisovaleryl-CoA and Free Coenzyme A (CoA-SH)

(164) HPLC analysis was performed using 1260 Infinity LC System (Agilent), equipped with column heating module and UV detector (260 nm). 1 l of samples were separated on Zorbax SB-Aq column (2504.6 mm, 5 m particle size, column temp. 30 C.), with a mobile phase flow rate of 1.5 ml/min. The separation was performed using mixed A (H.sub.2O containing 8.4 mM sulfuric acid) and B (acetonitrile) solutions in a linear gradient (0% B at initial time 0 min.fwdarw.70% B at 8 min). In these conditions, the retention time of 3-hydroxyisovaleryl-CoA and free coenzyme A (CoA-SH) were 4.53 and 4.10 min, respectively.

(165) Typical chromatograms obtained for the enzymatic assay A and the enzyme-free assay H are shown in FIGS. 19 and 20.

(166) The results of the HPLC analysis are summarized in FIG. 21.

(167) The obtained data indicate that 3-hydroxyisovaleryl-CoA was converted into 3-hydroxyisovaleric acid with the concomitant generation of ATP from ADP in a two-step reaction, catalyzed respectively by two enzymes (assays A and B). Thus, the conversion occurred through the formation of the intermediate 3-hydroxyisovaleryl phosphate followed by transfer of the phosphate group from this intermediate on ADP, thereby releasing ATP.

(168) A significant production of 3-hydroxyisovaleric acid, without simultaneous generation of ATP, was observed when phosphate butyryltransferase was used alone (assay E). This production is due to a spontaneous hydrolysis of the 3-hydroxyisovaleryl phosphate generated by the action of phosphate butyryltransferase.

(169) The production of 3-hydroxyisovaleric acid was observed in the same manner for the control assays without ADP (assays C and D). This production was also due to a hydrolysis of the 3-hydroxyisovaleryl phosphate generated by the action of phosphate butyryltransferase.

Example 11: Conversion of 3-Hydroxyisovaleryl-CoA and ADP into 3-Hydroxyisovaleric Acid and ATP Catalysed by the Combined Action of the Phosphate Butyryltransferase from Enterococcus faecalis and Butyrate Kinase from Lactobacillus casei or Geobacillus sp.

(170) The enzymatic assays were conducted in a total reaction volume of 0.2 ml.

(171) The standard reaction mixture contained:

(172) 50 mM potassium phosphate buffer pH 7.5

(173) 4 mM 3-hydroxyisovaleryl-CoA

(174) 4 mM ADP

(175) 10 mM MgCl.sub.2

(176) 10 mM NaCl

(177) 0.2 mg/ml of purified phosphate butyryltransferase from Enterococcus faecalis

(178) 0.2 mg/ml of purified butyrate kinase from Lactobacillus casei or Geobacillus sp.

(179) A series of controls were performed in parallel (Assays C-H Table 5).

(180) TABLE-US-00005 TABLE 5 Assay composition A B C D E F G H 3-hydroxyisovaleryl-CoA + + + + + + + + ADP + + + + + + phosphate butyryltransferase + + + + + from Enterococcus faecalis butyrate kinase from + + + Lactobacillus casei butyrate kinase from Geobacillus + + + sp

(181) The assays were then incubated for 20 min with shaking at 30 C.

(182) After an incubation period, the reactions were stopped by heating the reaction medium 4 min at 90 C. The samples were centrifuged, filtered through a 0.22 m filter and the clarified supernatants were transferred into a clean vial for further analysis. The consumption of ADP and 3-hydroxyisovaleryl-CoA, and the formation of ATP and 3-hydroxyisovaleric acid and free coenzyme A (CoA-SH) were followed by HPLC analysis according to the methods described in Example 10.

(183) The results of HPLC analysis are summarized in FIG. 22.

(184) The obtained data indicate that the 3-hydroxyisovaleryl-CoA was converted into 3-hydroxyisovaleric acid with the concomitant generation of ATP from ADP in a two-step reaction, catalyzed respectively by two enzymes (assays A and B). Thus, the conversion occurred through the formation of the intermediate 3-hydroxyisovaleryl phosphate followed by transfer of the phosphate group from this intermediate on ADP, thereby releasing ATP.

(185) A significant production of 3-hydroxyisovaleric acid, without simultaneous generation of ATP, was observed when phosphate butyryltransferase was used alone (assay E). This production was due to a hydrolysis of the 3-hydroxyisovaleryl phosphate generated by the action of phosphate butyryltransferase.

(186) The production of 3-hydroxyisovaleric acid was observed in the same manner for the control assays without ADP (assays C and D). This production was also due to a hydrolysis of the 3-hydroxyisovaleryl phosphate generated by the action of phosphate butyryltransferase.

Example 12: Enzyme-Catalyzed Hydrolysis of 3-Methylcrotonyl-CoA into 3-Methylcrotonic Acid and Free Coenzyme-A

(187) The gene coding for acyl-CoA thioesterase II from Pseudomonas putida was synthesized according to the procedure as described in Example 1.

(188) The vector pCA24N which contained the gene encoding acyl-CoA thioesterase 2 (TesB) from Escherichia coli was purchased from NAIST (Nara Institute of Science and Technology, Japan, ASKA collection). This vector provided contained a stretch of 6 histidine codons after the methionine initiation codon.

(189) The corresponding enzymes were produced according to the procedure described in Example 1.

(190) The enzymatic assays were conducted in a total reaction volume of 0.2 ml.

(191) The standard reaction mixture contained:

(192) 50 mM HEPES pH 7.0

(193) 10 mM 3-methylcrotonyl-CoA (Sigma-Aldrich)

(194) 20 mM MgCl.sub.2

(195) 20 mM NaCl

(196) 1 mg/ml of purified recombinant thioesterase.

(197) Control assays were performed in which either no enzyme was added or no substrate was added.

(198) The assays were incubated for 30 min with shaking at 30 C., the reactions were stopped by the addition of 0.1 ml of acetonitrile and the samples were then analyzed by HPLC-based procedure.

(199) HPLC Based Analysis of the Consumption of 3-Methylcrotonyl-CoA and the Formation of 3-Methylcrotonic Acid and Free Coenzyme a (CoA-SH)

(200) HPLC analysis was performed using an 1260 Infinity LC System (Agilent), equipped with a column heating module and a UV detector (210 nm). 5 l of samples were separated on Zorbax SB-Aq column (2504.6 mm, 5 m particle size, column temp. 30 C.) with a mobile phase flow rate of 1.5 ml/min. The separation was performed using mixed A (H.sub.2O containing 8.4 mM sulfuric acid) and B (acetonitrile) solutions in a linear gradient (0% B at initial time 0 min.fwdarw.70% B at 8 min). Commercial 3-methylcrotonyl-CoA, 3-methylcrotonic acid (Sigma-Aldrich) and CoA-SH (Sigma-Aldrich) were used as references. In these conditions, the retention time of free coenzyme A (CoA-SH), 3-methylcrotonyl-CoA and 3-methylcrotonic acid were 4.05, 5.38 and 5.83 min, respectively.

(201) No 3-methylcrotonic acid signal was observed in control assays.

(202) Both studied thioesterases catalyzed the hydrolysis of 3-methylcrotonyl-CoA with the formation of 3-methylcrotonic acid. An example of a chromatogram obtained with acyl-CoA thioesterase II from Pseudomonas putida is shown in FIG. 23.

(203) The degree of production of 3-methylcrotonic acid as observed in the enzymatic assays is shown in Table 6.

(204) TABLE-US-00006 TABLE 6 Uniprot 3-methylcrotonic Accession acid produced, Gene names Organism number mM tesB Escherichia coli P0AGG2 0.6 tesB Pseudomonas putida Q88DR1 3.1

Example 13: Conversion of 3-Methylcrotonyl-CoA and ADP into 3-Methylcrotonic Acid and ATP Catalysed by the Combined Action of Phosphate Butyryltransferase from Bacillus subtilis and Butyrate Kinase from Lactobacillus casei or Geobacillus sp.

(205) In the assays described in the following, the following enzymes were used:

(206) TABLE-US-00007 TABLE 7 Uniprot Accession Gene names Organism number Ptb (yqiS) Bacillus subtilis P54530 Buk Lactobacillus casei K0N529 BuK Geobacillus sp. L8A0E1

(207) The enzymatic assays were conducted in a total reaction volume of 0.2 ml.

(208) The standard reaction mixture contained:

(209) 50 mM potassium phosphate buffer pH 7.5

(210) 4 mM 3-methylcrotonyl-CoA

(211) 4 mM ADP

(212) 10 mM MgCl.sub.2

(213) 10 mM NaCl

(214) 0.2 mg/ml of purified phosphate butyryltransferase from Bacillus subtilis

(215) 0.2 mg/ml of purified butyrate kinase from Lactobacillus casei or Geobacillus sp.

(216) A series of controls was performed in parallel (Assays C-H as shown in Table 8).

(217) TABLE-US-00008 TABLE 8 Assay composition A B C D E F G H 3-methylcrotonyl-CoA + + + + + + + + ADP + + + + + + phosphate butyryltransferase + + + + + from Bacillus subtilis butyrate kinase from + + + Lactobacillus casei butyrate kinase from Geobacillus + + + sp

(218) Assays were then incubated for 20 min with shaking at 30 C.

(219) After an incubation period, the reactions were stopped by heating the reaction medium 4 min at 90 C. The samples were centrifuged, filtered through a 0.22 m filter and the clarified supernatants were transferred into a clean vial for further analysis. The consumption of ADP and 3-methylcrotonyl-CoA, and the formation of ATP, 3-methylcrotonic acid and free coenzyme A (CoA-SH) were followed by using HPLC-based methods.

(220) HPLC-Based Analysis of ADP and ATP

(221) HPLC analysis was performed using the 1260 Infinity LC System (Agilent) equipped with a column heating module and an RI detector. 2 l of samples were separated on a Polaris C18-A column (1504.6 mm, 5 m particle size, column temp. 30 C.) with a mobile phase flow rate of 1.5 ml/min. The consumption of ADP and the formation of ATP were followed by HPLC analysis according to the methods described in Example 10.

(222) HPLC Based Analysis of 3-Methylcrotonyl-CoA, 3-Methylcrotonic Acid and Free Coenzyme A (CoA-SH)

(223) HPLC analysis was performed using the 1260 Infinity LC System (Agilent) equipped with a column heating module and a UV detector (260 nm). 1 l of samples were separated on a Zorbax SB-Aq column (2504.6 mm, 5 m particle size, column temp. 30 C.), with a mobile phase flow rate of 1.5 ml/min.

(224) The consumption of 3-methylcrotonyl-CoA and the formation of 3-methylcrotonic acid and free coenzyme A (CoA-SH) were followed according to the procedure described in Example 12. Under these conditions, the retention time of 3-methylcrotonyl-CoA, 3-methylcrotonic acid and free coenzyme A (CoA-SH) was 5.38 min, 5.73 min and 4.07 min, respectively.

(225) Typical chromatograms obtained for the enzymatic assay A and enzyme-free assay H are shown in FIGS. 24a and 24b.

(226) The results of the HPLC analysis are summarized in FIG. 25.

(227) The obtained data indicate that 3-methylcrotonyl-CoA was converted into 3-methylcrotonic acid with the concomitant generation of ATP from ADP in a two-step reaction, catalyzed by two enzymes, respectively (assays A and B). Thus, the conversion occurred through the formation of the intermediate 3-methylcrotonyl phosphate followed by transfer of a phosphate group from this intermediate to ADP, thereby releasing ATP.

(228) A significant production of 3-methylcrotonic acid without a simultaneous generation of ATP was observed when phosphate butyryltransferase was used alone (assay E).

(229) This production is due to a spontaneous hydrolysis of 3-methylcrotonyl phosphate generated by the action of phosphate butyryltransferase.

(230) The production of 3-methylcrotonic acid was observed in the same manner for the control assays without ADP (assays C and D). This production was also due to a hydrolysis of 3-methylcrotonyl phosphate generated by the action of phosphate butyryltransferase.

Example 14: Conversion of 3-Methylcrotonyl-CoA and ADP into 3-Methylcrotonic Acid and ATP Catalysed by the Combined Action of the Phosphate Butyryltransferase from Enterococcus faecalis and Butyrate Kinase from Lactobacillus casei or Geobacillus sp.

(231) In the assays described in the following, the following enzymes were used:

(232) TABLE-US-00009 TABLE 9 Uniprot Accession Gene names Organism number Ptb Enterococcus faecalis A0A038BNC2 Buk Lactobacillus casei K0N529 BuK Geobacillus sp. L8A0E1

(233) The enzymatic assays were conducted in a total reaction volume of 0.2 ml.

(234) The standard reaction mixture contained:

(235) 50 mM potassium phosphate buffer pH 7.5

(236) 4 mM 3-methylcrotonyl-CoA

(237) 4 mM ADP

(238) 10 mM MgCl.sub.2

(239) 10 mM NaCl

(240) 0.2 mg/ml of purified phosphate butyryltransferase from Enterococcus faecalis

(241) 0.2 mg/ml of purified butyrate kinase from Lactobacillus casei or Geobacillus sp.

(242) A series of controls was performed in parallel (Assays C-H Table 10).

(243) TABLE-US-00010 TABLE 10 Assay composition A B C D E F G H 3-methylcrotonyl-CoA + + + + + + + + ADP + + + + + + phosphate butyryltransferase + + + + + from Enterococcus faecalis butyrate kinase from + + + Lactobacillus casei butyrate kinase from Geobacillus + + + sp

(244) Assays were then incubated for 20 min with shaking at 30 C.

(245) After an incubation period, the reactions were stopped by heating the reaction medium 4 min at 90 C. The samples were centrifuged, filtered through a 0.22 m filter and the clarified supernatants were transferred into a clean vial for further analysis. The consumption of ADP and 3-methylcrotonyl-CoA and the formation of ATP and 3-methylcrotonic acid and free coenzyme A (CoA-SH) were followed by HPLC analysis according to the methods described in Example 10 and Example 12.

(246) The results of the HPLC analysis are summarized in FIG. 26.

(247) The obtained data indicate that the 3-methylcrotonyl-CoA was converted into 3-methylcrotonic acid with the concomitant generation of ATP from ADP in a two-step reaction, catalyzed by two enzymes, respectively (assays A and B). Thus, the conversion occurred through the formation of the intermediate 3-methylcrotonyl phosphate followed by the transfer of a phosphate group from this intermediate on ADP, thereby releasing ATP.

(248) A significant production of 3-methylcrotonic acid without simultaneous generation of ATP was observed when phosphate butyryltransferase was used alone (assay E).

(249) This production was due to a hydrolysis of 3-methylcrotonyl phosphate generated by the action of phosphate butyryltransferase.

(250) The production of 3-methylcrotonic acid was observed in a similar manner for the control assays without ADP (assays C and D). This production was also due to a hydrolysis of 3-methylcrotonyl phosphate generated by the action of phosphate butyryltransferase.

Example 15: Screening for Hydro-Lyases for the Production of 3-Hydroxy-3-Methylbutyric Acid from 3-Methylcrotonic Acid

(251) Hydro-lyases classified as enzymes belonging to the family of 2-methylcitrate dehydratases (EC 4.2.1.79) have already been described as being capable of converting 3-methylcrotonic acid into 3-hydroxy-3-methylbutyric acid. This has been described for a maleate hydratase from Pseudomonas pseudoalcaligenes which can use 3-methylcrotonic acid as a substrate (van der Werf et al., Appl. Environ. Microbiol. 59 (1993), 2823-2829).

(252) Further hydro-lyases which show a corresponding reactivity for the enzymatic conversion of 3-methylcrotonic acid into 3-hydroxy-3-methylbutyric acid can be identified by screening known hydro-lyases for this reactivity as outlined in the following:

(253) Corresponding genes encoding for enzymes belonging to the family of hydro-lyases classified as EC 4.2.1.-, preferably encoding for enzymes belonging to the family of aconitate hydratases (EC 4.2.1.3), maleate hydratases (EC 4.2.1.31) or 2-methylcitrate dehydratases (EC 4.2.1.79), can be derived from commonly available resources. The corresponding gene of a candidate enzyme can be synthesized and the enzyme can be produced according to the procedure as described in Example 1.

(254) Once the enzyme is produced and purified in accordance with the above description, the respective hydro-lyase can be tested with respect to its reactivity for the enzymatic conversion of 3-methylcrotonic acid into 3-hydroxy-3-methylbutyric acid. For this hydratase assay, a reaction mixture containing MgCl.sub.2, NaCl and 0-100 mM 3-methylcrotonic acid is used. Control assays are performed in which either no enzyme is added or no substrate is added. Each sample is monitored for the consumption of 3-methylcrotonic acid and/or for the formation of 3-hydroxy-3-methylbutyric acid by HPLC-based procedure.

(255) A hydro-lyase will be identified as a suitable enzyme capable of enzymatically converting 3-methylcrotonic acid into 3-hydroxy-3-methylbutyric if it shows in the above assay an increased formation of 3-hydroxy-3-methylbutyric acid over the control assays.